USA --> Metric

[echen1024]

Well, I got fed up of using imperial and decided it was time for a change. I started a petition on whitehouse.gov, but it probably won't go anywhere, but thought it would be worth a try. I figured I would post it on this forum, and would appreciate it if you could help and sign it. I know, it is annoying as hell and needs an account 

http://wh.gov/lsq6F

[miguelvp]

So we can't use mils anymore?

It's been tried again and again, it just affects too many industries starting with cars, also temperature people are just too used to the values and ranges. Conversions are really easy in any event 2.54cm per inch. Also what about drag racers, they just couldn't race to the quarter mile (1320 feet) * 12 inches/feet * 2.54cm/inch * 1m/100cm = 402.336 meters, it just sounds odd

Also what, change comas and periods? it's just too much change and it will never happen.

Edit: I was raised with metric (Spain) but I had no trouble adapting to imperial at all.

[echen1024]

Probably. *sigh* It's just that conversions in imperial are a pain in the ass. /5280, x1760 WTF?

[rexxar]

There's a reason we're the only country on the planet still using the imperial system. It's because 'merica and freedom or some such.

Really it's because Americans in general are xenophobic. The slightest mention of change seems to send people into fits.

[miguelvp]

Probably. *sigh* It's just that conversions in imperial are a pain in the ass. /5280, x1760 WTF?

5280 I will guess is feet in a mile, don't know what 1760 is, but 1.609 is kilometers per mile.

[GeoffS]

1760 yards per kilometre.

I remember yards ...

[IanB]

1760 yards per kilometre.

I remember yards ...

1760 yards in a mile 

[GeoffS]

1760 yards per kilometre.

I remember yards ...

1760 yards in a mile 

See? There you go, after a while you forget all about conversions. 

[miguelvp]

duh, i should have just divided the numbers and 3 would be a good indication

Edit, So:
1/5280 mile/feet * 1760 yards/mile = 1/3 yards/feet

Also I always liked the fraction use instead of the decimal point, it's just more precise.

[IanB]

Well, I got fed up of using imperial...

In America, engineers tend to use a metric version of customary units when it helps. For example, an inch divides into 10ths, 100ths and thousandths (thou)--look at an engineer's ruler sometime. Pounds multiply by 1000 when there are a lot of them (giving, for example, thousands of pounds per hour).

Britain only went metric when the government made it illegal to use the old units. Britain was pressured into change when all of her trading partners were using metric units. America does not seem to have the same pressure, so change will be slow in coming.

[mariush]

Have some bad news for you, it's been done already:

https://petitions.whitehouse.gov/petition/make-metric-system-standard-united-states-instead-imperial-system/FndsKXLh

It even got a reply from some employee.

And you should know president Ford made an act promoting metrification of America, but it was then ignored : http://en.wikipedia.org/wiki/Metric_Conversion_Act

Lots of good articles about this issue here:
http://www.vox.com/2014/5/29/5758542/time-for-the-US-to-use-the-metric-system
http://mentalfloss.com/article/50160/years-ken-butcher-was-sole-employee-us-metric-program
http://boards.straightdope.com/sdmb/showthread.php?t=695662

and this issue shows up on reddit every few weeks, and there's even a subreddit dedicated to it: http://www.reddit.com/r/metric

[TheBorg]

I agree with this, totally.

My one concern is that America will simply not do this due to cost. Metric is already taught in our schools and all, but the cost to change *every* form of measurement to metric is pretty crazy.

The bigger concern with cost is the price of parts. I encourage myself to use metric parts in my projects, but my last one ended up using English units (I won't mix metric and Imperial...) simply due to the cost of parts. I could get some gears for it with metric bores for $30 each, or near the same part in Imperial units for $5. That's messed up.

[HighVoltage]

Having lived in the US, I studied the reasoning for them keeping the imperial system.
What was a most interesting fact: They have a already a law in place since the 1970th, that the metric system must be used.
But this law has a wide exception clause that applies to everyone and everything.
Therefore, I think this will never change.

Big things happened because of the US using the imperial system:
Remember that one of the NASA probes on Mars crashed?
Allegedly the crash was based on a wrong conversion from metric to imperial units.
Even this huge mistake did not make the US change units.

It will probably never happen?

[miguelvp]

Having lived in the US, I studied the reasoning for them keeping the imperial system.
What was a most interesting fact: They have a already a law in place since the 1970th, that the metric system must be used.
But this law has a wide exception clause that applies to everyone and everything.
Therefore, I think this will never change.

Big things happened because of the US using the imperial system:
Remember that one of the NASA probes on Mars crashed?
Allegedly the crash was based on a wrong conversion from metric to imperial units.
Even this huge mistake did not make the US change units.

It will probably never happen?

It wouldn't have happened if they only used imperial

Kidding aside, it was a mistake not because of the unit system:

The problem here was not the error, it was the failure of NASA's systems engineering, and the checks and balances in our processes to detect the error. That's why we lost the spacecraft.

Then again, that's what happens when you keep on shrinking space exploration budgets, so they lost a $125 million craft, the military has more expensive airplanes than that. $150 million for an F22 after $67 Billion on development.

I wonder if those are metric

[T3sl4co1l]

We ARE metric.  1 in = 2.54 cm exactly, which is in turn defined by the speed of light and a resonance of atomic cesium.

In my work, I have no problem switching between units.  Especially since most CAD has a switch for what the user sees.  Imperial are more familiar to me, but I have a good idea what some mm length is, or how heavy some kilos are, without having to think about it.  PCB footprints are normally in mm; SI units are easy for transformer design (I conventionally use MHz, uH, mm^2 and etc., but that's just multipliers on SI base units).

What would really burn your brain, I suspect, is if everyone still used fractional measurements.  Older datasheets (e.g. vacuum tubes) are specified this way.  But even then, it's easy to remember what the decimals of a given fraction look like; there are only a few to remember (down to, say, 1/32nds) that anyone bothers to use (okay, drills are commonly listed in 1/64ths, and also come in numbered wire gauges, but to know that, you'd look at a table, which shows you which sizes are available -- still true in metric, since they don't make just any decimal size mm drill after all).

By and large, almost everything is in decimals these days, which means the numbers read the exact same whether metric or not.  The only difference is how many zeroes past the decimal point, and which one is in parenthesis, "in (mm)" or "mm (in)".

Supposedly, aerospace remains cursed with imperial units, to a surprising degree; not just length and mass, but also force, pressure and various other physical units.  I don't know how much so, as I haven't worked in aerospace and don't know any engineers there.  If true, it's certainly a disturbing trend for a field so otherwise highly developed.

Tim

[TMM]

Also I always liked the fraction use instead of the decimal point, it's just more precise.

The problem there is that fractions are only precise at representing perfect fractions.

1.2 centimetres = 0.472440945 inches
15/16 inches = 2.38125 centimetres

neither is a more 'precise' system unless whatever you are measuring fits perfectly into the units that you are measuring.

[G7PSK]

Imperial/ metric is still a problem here in the UK and most likely in the rest of Europe and the world for that matter. When I purchase aluminium section it's size depends on whether it was made in the US,Canada and Europe such as Denmark or Norway. US, Canadian is usually imperial dimensions and the rest is metric so a length of 100x100x10mm could be exactly that or it could arrive as 4x4x3/8 inch. the stockholders also tend to get the stuff mixed up so if several lengths are required you could get 3 metric and 2 imperial, leading to much cursing etc.

[HighVoltage]

It wouldn't have happened if they only used imperial

Very true.

Also, it is much easier to remember, that 13 inches is almost exactly 33 cm
There are plenty of hidden systems between the metric and imperial system.

[HighVoltage]

We ARE metric.  1 in = 2.54 cm exactly, which is in turn defined by the speed of light and a resonance of atomic cesium.

Very interesting, I did not know that in regards to speed of light and the inch.

I read a German report where some guy did some research in to the imperial system and he found that 1 inch is NOT exactly 2.54 cm
In his calculations, the inch is 2,54000508001016002032...
So, since there are three zeros behind the 4, we always assumed it is exactly 2.54

[miguelvp]

It wouldn't have happened if they only used imperial

Very true.

Also, it is much easier to remember, that 13 inches is almost exactly 33 cm
There are plenty of hidden systems between the metric and imperial system.

0.2 mm is a big error.

will be better to remember that 254 inches is exactly 100 cm.

Edit: No matter, still mils are used in electronics world wide, and that is 1000 th of an inch. (or 25.4 um)

[lewis]

I am of the generation where we transitioned from imperial to metric. As such, I can't reliably think in either!

I like to buy my beer in pints, but my wine in centilitres. I buy diesel for the car in litres, but measure its consumption in miles per gallon (that's a British gallon, my Yank friends!). I have no idea how far 100km is, but if I've got to drive 100 miles I know exactly how long that will take me and how much fuel it will use. We're the only country in Europe still using miles on the roads, DON'T change it thanks very much!

I'm 6' 4" tall and 19 1/2 stone, no idea what that is in metres, or even pounds, but I know it's about 120kg. When Jay Leno says a car is 3500lb, I have no idea if that is heavy or light, but I can easily visualise 2 tonnes (not tons).

I weigh ingredients in grams if the quantity required is less than 200g, then I change to ounces and pounds for the larger quantities. Liquids (except beer) are litres and millilitres, I have no idea how big a fluid ounce is. I have absolutely no idea about Fahrenheit, temperature is always Celsius. When a yank says 'that's 150 degrees' I have no idea if that's hot or cold or what.

When I'm cutting timber to build something, I always measure it in inches and sixteenths, but when I'm fitting studwork the timber is always metric distances apart. Timber is always '3x2' or '4x4' and sheet wood is always '8x4' although I know that works out to be 2440mm x 1220mm. Steel RSJ and copper pipe are always metric, but pipe fittings are in inches BSP. Cable size is always metric: millimetres squared please, not SWG or AWG, or 'gauge'. Don't get me started on Newton-metres and pound-feet, or BTUs.

Nuts and bolts are ALWAYS metric, M6 thread with 10mm head I can understand, not 3/8 whitworth with a head that measures 18mm but you need a 32.6mm spanner or whatever it is. No idea about BA, or UNC or UNF or any of that. (I'm sure the proper engineers like Robrenz on here will be:  ).

I know how big an 0805 resistor is, but no idea what it is in metric; all my PCB designs have a metric board profile and metric fixing holes but an imperial grid. But critical component centres are placed on a metric grid, but tracks are routed on an imperial grid, with imperial widths and spacings, and imperial pad sizes, but metric drill holes, on 1 or 2 ounce copper.....   

It's a fecking nightmare!

[max666]

Oh how many times have I cursed 

Standardisation is a good thing!

[mariush]

It wouldn't have happened if they only used imperial

Very true.

Also, it is much easier to remember, that 13 inches is almost exactly 33 cm
There are plenty of hidden systems between the metric and imperial system.

0.2 mm is a big error.

will be better to remember that 254 inches is exactly 100 cm.

Edit: No matter, still mils are used in electronics world wide, and that is 1000 th of an inch. (or 25.4 um)

You're wrong.

1 m (meter) = 100 cm (centi-meter) = 1000 mm  (milli - meter)
1 km (kilo - meter) = 1000 m

1 inch = 2.54 cm ( centimeter, centime = one hundred of a meter) = 25.4 mm (millimeter or thousands of a meter)

100 inch = 254 cm . Since there's 100 cm in a meter, you can say 2.54 meters without any problems.

Compare that with   

1 foot = 12 inches
1 yard = 3 feet = 36 inches
1 rod  = 5.5 yards = 16.5 feet = 198 inches =  1/320 statute mile  = 1/4 surveyor's chain
1 chain = 66 feet
1 furlong = 10 chains = 40 rods = 660 feet = 220 yards = 1/8 mile

[miguelvp]

Ok, got it backwards 100 inches is exactly 254 cm I just wanted to say that the conversion is exact.

My dad was born and raised in the States, me in Spain. The thing is that we know when it's hot or cold in either measure, distances etc, you don't even have to convert units. It's like speaking many languages, you don't have to translate from one to the other, you just think on it. Same thing with measurements.

Edit: I like the I'll race you 320 rods on my hot rod!

sounds better than a quarter mile

[GeoffS]

All my schooling was done using the imperial system as metrification happened after I left school. (except for decimal currency which started on my first day of high school)
While there was a period where you converted everything back and forth between metric and imperial, it soon became perfectly natural to think in metric units.
You'd be amazed at how quickly you adapt.

[GeoffS]

Edit: I like the I'll race you 320 rods on my hot rod!

sounds better than a quarter mile

320 rods to the mile so a drag race is only 80 rods (or 2 furlongs.) 

[T3sl4co1l]

We ARE metric.  1 in = 2.54 cm exactly, which is in turn defined by the speed of light and a resonance of atomic cesium.

Very interesting, I did not know that in regards to speed of light and the inch.

I read a German report where some guy did some research in to the imperial system and he found that 1 inch is NOT exactly 2.54 cm
In his calculations, the inch is 2,54000508001016002032...
So, since there are three zeros behind the 4, we always assumed it is exactly 2.54

IIRC, the accepted measure of an inch (before metric was adopted) was very slightly off (I certainly don't remember the difference, but I think it was on the order of your number above).  And after adoption, the entire breadth of the United States only changed a few inches or something, not enough even cumulatively to bother the surveyors.

The US is about 3000 mi across, depending on how you want to measure it (e.g., a quick glance at the map provides a route from NY to LA of 2795 mi, so I'll go with that).  That's *5280*12 or around 177 million inches.  The change produces a difference of 354 inches, or 29 feet, if your figure is correct.

If property lots were measured from a single federal datum, that would be a problem -- by the end, they'd all be off by about a house-width.  Fortunately, that's not the case anyway.  (I think lots are usually defined by the county?  That being the level (hierarchically and geographically) below state, state being equivalent to an EU member country for example.  So, whatever datum is chosen depends on state (~500mi) and local (~50mi) considerations.  So you wouldn't expect the error to accumulate more than the spacing of those possible standards.)

Tim

[HighVoltage]

The sequence of the allegedly exact inch to cm conversion is kind of interesting and we find the binary exponent in there.
I have no proof that this is even true, just found it years ago in a German book.

See attachment...
 

[nanofrog]

Supposedly, aerospace remains cursed with imperial units, to a surprising degree; not just length and mass, but also force, pressure and various other physical units.

IME, it can be mixed on the design side, but strictly Imperial units for manufacturing in the US (due to the tooling).

[aroby]

I switch between mm and inches.  I use mm for small measurements and feet/inches for large measurements.  Much easier to remember I need something 6' long or 3mm wide than the equivalent in the other units!  In the UK (where I am from) it always used to be the case with temperature to use Fahrenheit for warmer temps and Celsius for lower temps - people love to complain about the weather so want to make it sound as extreme as possible ...

Anthony

[John Coloccia]

The only advantage the metric system has over imperial is converting between units.  This is very rarely important.  For example, it's not all that important to know how many inches from here to the sun...or how many miles wide your garage is.  While it's cute, it's uninteresting.

Also, a good deal of imperial measurement centers around very human centric, usable quantities.  Ever try following a recipe in metric?  25ml of this, 33ml of that, 90ml of this, 52ml of that.  It's a nightmare.  In imperial measure, you get  1tsp of this, 1/2 tbsp of that.  Much much simpler for a simple human to remember.

A lot of people confuse DECIMAL for metric.  We use decimal measure all the time.  You don't need the metric system for that.

And anyhow, all you guys don't even USE the stupid system the way it's designed to be used.  When's the last time you used the decameter?  All you use is km, m, cm and mm.  Of those, the only two that are ever really mixed are cm and mm, and I wouldn't allow such a document to see the light of day if it ever came across my desk.  Go fix it and make everything the same so we don't confuse everyone.

This whole converting units thing is just not all that compelling.  If I start in inches, it's because inches is the appropriate measurement.  Why would I then start talking about yards, or miles?  It makes no sense, nor does it make sense that you'd start with cm, and then start talking about meters.  If meters are appropriate, then use meters.

And anyhow, no one is stopping you from saying things like "kilo-inch", and indeed the milli-inch (mil, .001") is a standard, accepted unit of measure.

Lots of blah blah blah about nothing outside of an academic atmosphere, IMHO.  It would be nice to get on the metric system here, but for the only reason that it would be nice to simply get everything standardized, and we're kind of the odd-balls at this point.  The "superiority" of the metric system really doesn't manifest itself in any practical way.

[GK]

That White House government site is a hoot. So any numbnuts can sign up and petition Mr Obama about anything that gets on their goat? Like.........

https://petitions.whitehouse.gov/petition/reverse-our-nation%E2%80%99s-%E2%80%9Cno-child-left-white%E2%80%9D-policy/TTtQCq62
 
https://petitions.whitehouse.gov/petition/fund-solar-roadways-create-national-clean-energy-surplus-and-end-wintery-road-conditions/3hjb7XbS
 
https://petitions.whitehouse.gov/petition/build-military-outpost-moon-advanced-tech-will-help-keep-our-country-safe-growing-superpowers/n6rTHfff
 

So, what if the 100,000 signatures are attained? Will the White House be obliged to act? huh? What a mind-bogglingly stupid public relations initiative.

[Monkeh]

I just wish Eagle could default to metric.

It can.

[zapta]

Well, I got fed up of using imperial and decided it was time for a change. I started a petition on whitehouse.gov, but it probably won't go anywhere, but thought it would be worth a try. I figured I would post it on this forum, and would appreciate it if you could help and sign it. I know, it is annoying as hell and needs an account 

http://wh.gov/lsq6F

echen, what is the cost of the change you propose?

This petition site is intended for Americans only, right? If so, worth mentioning it when posting to an international forum.

[miguelvp]

And anyhow, all you guys don't even USE the stupid system the way it's designed to be used.  When's the last time you used the decameter?  All you use is km, m, cm and mm.  Of those, the only two that are ever really mixed are cm and mm, and I wouldn't allow such a document to see the light of day if it ever came across my desk.  Go fix it and make everything the same so we don't confuse everyone.

Well, they do get used sometimes, the decimeter is used to define the liter which is equal to 1 cubic decimeter and equal to one kilogram.

Decameters is used for radio waves.

Hectameters are used to measure land, one hectare is 1 square hectameter.

Most of the things we manipulate in regular life are in the cm range. But yeah, it all doesn't matter much, the imperial system is as good as the metric system and I wouldn't change it.

[nctnico]

The metric system is based on scientific quantities. With the imperial system you always have to use clunky conversions. Calculate power from torque and rpm for example. Same goes for all kinds of electric field formulas.

[zapta]

If you switched today then in a few decades you wouldn't need two sets of spanners

Two sets of spanners?  This is so 20th century.

http://www.metrinch.tv/

[SeanB]

I have broken those on cheap Chinese bolts..............

You get the curse as well of fasteners meant for the US market, where the thread is a metric size, but the head is an imperial measure, so you get things like a M8 bolt with a 14mm head, which is not the correct head of 13mm. Or worse a thread that is metric in diameter but imperial in pitch.

My solution in those cases is a good set of taps and dies and make it a proper thread and use a proper metric fastener, though on the one machine I do have a near complete set of imperial taps and dies to use for working on it, along with imperial thread inserts  and taps. Funny enough those are cheap, as the stock has been sitting around in a store since the late 1960's.

[zapta]

I have broken those on cheap Chinese bolts..............

Broke Metrinch?   You must be kidding. I have a set for almost 20 years and it  easily survived all my car works.

[SeanB]

I work on some stuff with hard to reach fasteners......

This was today's victim, the motor did not like drinking all the gearbox oil when it popped the base seal. New motor, new seals, new bearings on the input shaft and a little manufacturing of new gaskets along with using lots of Loctite gasket sealer on it and it will work again when I am finished. Not happy with the preload on the worm drive, but at least now it has a breather so I filled the gearbox up enough so that the top bearing would be oiled. Putting it back is fun, as it literally has to have the mounting bolts fastened from the bottom of the motor, no access to then in the machine otherwise. Had to go buy a M10 tap to clear some bolt holes as well, they were full of crud from new, and the bolts were seizing. Fun with the new motor having different cable gland threads as well, and I bought new glands as well for it.

[miguelvp]

The metric system is based on scientific quantities. With the imperial system you always have to use clunky conversions. Calculate power from torque and rpm for example. Same goes for all kinds of electric field formulas.

Physics doesn't care about units, but do you know how many kilo watts your car is capable of?  Mechanical HP is still imperial, Metric HP is a bit under, who made it to be 75Kg was it just the best round figure to make it close to the imperial one, I mean was that just an arbitrary value?

So now the standard should be KWatts but no one uses that for cars do they?

[miguelvp]

Er, no, the main reason is to make maths easier. Say you have a triangle that is 300mm on one side and 400mm on the other. How long do you need to cut the third bit of wood to make the hypotenuse? This is a question anyone designing or building a sloped roof has had to answer.

Why is that any different, math is math and doesn't care about units. the hypotenuse is still going to be the square root of the sum of the squares regardless of the unit.

[Dave]

The sequence of the allegedly exact inch to cm conversion is kind of interesting and we find the binary exponent in there.
I have no proof that this is even true, just found it years ago in a German book.

See attachment...

Hate to break it to you, but the image you have posted is incorrect.

From Wikipedia:

Traditional standards for the exact length of an inch have varied in the past, but now the imperial or US customary inch is defined to be exactly 25.4 mm.

The fraction 1 cm = 1/0.3937 in is just a good approximation, the exact value of the denominator is 0.393700787401...

[Corporate666]

There's a reason we're the only country on the planet still using the imperial system. It's because 'merica and freedom or some such.

Really it's because Americans in general are xenophobic. The slightest mention of change seems to send people into fits.

Nonsense.

You won't find a country elsewhere on this planet that has more of a mix of everything and everyone than the USA.  It is simply that we are so entrenched in Imperial measurements that changing would be difficult.  It is not simply an issue of teaching our children and "getting used to" a different measurement system, it is the change of billions of dollars worth of tooling, machinery, processes, fasteners and programs that would have to change.  There is not a compelling reason to change - metric is somewhat easier for conversions and proportions, but it's not so much easier as to make switching a no-brainer.  It's just that the benefits are dwarved by the costs.  As long as that remains true, we won't change.

[ajb]

That White House government site is a hoot. So any numbnuts can sign up and petition Mr Obama about anything that gets on their goat? Like.........

https://petitions.whitehouse.gov/petition/reverse-our-nation%E2%80%99s-%E2%80%9Cno-child-left-white%E2%80%9D-policy/TTtQCq62
 
https://petitions.whitehouse.gov/petition/fund-solar-roadways-create-national-clean-energy-surplus-and-end-wintery-road-conditions/3hjb7XbS
 
https://petitions.whitehouse.gov/petition/build-military-outpost-moon-advanced-tech-will-help-keep-our-country-safe-growing-superpowers/n6rTHfff
 

So, what if the 100,000 signatures are attained? Will the White House be obliged to act? huh? What a mind-bogglingly stupid public relations initiative.

My understanding is that they're committed to respond, not necessarily to act.  I'm sure many initiatives that attain the requisite 100,000 signatures (98.542354 kilosignatures in Metric) will receive a politely worded 'we appreciate your concern over this ridiculous issue, but that's too dumb even for us to go along with, so thanks but no thanks'.

The only advantage the metric system has over imperial is converting between units. This is very rarely important.  For example, it's not all that important to know how many inches from here to the sun...or how many miles wide your garage is.

Er, no, the main reason is to make maths easier. Say you have a triangle that is 300mm on one side and 400mm on the other. How long do you need to cut the third bit of wood to make the hypotenuse? This is a question anyone designing or building a sloped roof has had to answer.

  The math is the same so the ratio is the same.  It's going to be 3:4:5 whether you measure in mm or in inches. 

All in all the blend doesn't bother.  Most CAD packages make it easy to do dual dimensioning and it's simple to scale objects as needed to suit either system.  CNC machines don't care if a part is 2.000" or 2.124" or whatever, they'll just do as they're told.  Really the biggest pain is fasteners.  We have to keep twice as many fasteners in stock because some of our components are metric and some are imperial, and because some metric thread sizes are very close to imperial sizes at a glance (6-32 & M3, 8-32 & M4), it can be a huge pain in the ass if two bins happen to get mixed up. 

[Corporate666]

The math is the same so the ratio is the same.  It's going to be 3:4:5 whether you measure in mm or in inches. 

All in all the blend doesn't bother.  Most CAD packages make it easy to do dual dimensioning and it's simple to scale objects as needed to suit either system.  CNC machines don't care if a part is 2.000" or 2.124" or whatever, they'll just do as they're told.  Really the biggest pain is fasteners.  We have to keep twice as many fasteners in stock because some of our components are metric and some are imperial, and because some metric thread sizes are very close to imperial sizes at a glance (6-32 & M3, 8-32 & M4), it can be a huge pain in the ass if two bins happen to get mixed up.

I think you hit on an important point...personally, with the rise of CAD packages and calculators, I wonder if imperial vs metric is even less of an issue than ever before.

I switch between them all the time... even my CNC machines can be programmed in either with a simple command to switch units.  The only time imperial is a pain in the ass is when using fractions... like you have something you measure at 5 and 3/8ths inches wide and you need to make it 3/32nds narrower.. how much do you need to cut off?  It's simpler having something 98mm wide that you need to make 3mm narrower.  But as was said earlier, that's more because of decimal vs fractional and decimal works for either.

And of course fasteners as you said... although there are quite a few odd-duck metric fasteners out there I've run into.  With "nice round metric numbers", it almost seems easier to make your own threads in metric than in imperial... maybe it's just me, but I don't think twice about making a 6mm bolt with .25mm thread pitch, but if it was imperial, I'd try to go with one of the (many) existing pitches.

[SeanB]

Yes, but the M6x0.25mm tap is an off the shelf part, metric wise. I had to go get a M8 by 1mm tap recently to install a proximity sensor, and the tap was off the shelf as well, and a common enough size. Made in South Africa as well, not a OHL cheapie. If I wanted something like a M6 by 0.25mm tap and die that probably would be a non stocked locally item, but would be here next day or so.

[lewis]

M6 x 0.25mm is pretty fine, standard M6 is 1.0mm thread pitch.

That's another thing! Damn metric bolt sizes aren't even standard.

[max666]

M6 x 0.25mm is pretty fine, standard M6 is 1.0mm thread pitch.

That's another thing! Damn metric bolt sizes aren't even standard.

Yes they are standard! M6 is the standard preferred thread and it has a 1 mm pitch. M6x0,25 is one of the standard metric fine threads!

But that's not the point, because independent of what system you are using you'll always have the need for finer pitch threads for special applications. But you're not using a fine thread, unless you have a good reason for it.

[IanB]

The big advantage of metric is that you can use a calculator or spreadsheet to work with it. Can you imagine how painful electronics would be if there were 12 yanks in a volt or something stupid like that? 1731 ohms in a bunt or some such nonsense?

But none of this "nonsense" is actually the case. See my earlier post. When doing engineering with U.S. customary units the calculations are done in decimal just the same as with any other set of units. I might have a dimension of 15.348 inches, or a weight of 344.6 lb. This is no different than 15.348 m or 344.6 kg.

What is an issue, as has already been mentioned, is mixing units of measure in the same formula, for example when calculating pump horsepower in terms of head in feet and flow in gallons per minute. You end up with a numerical conversion factor in the formula to put all the quantities on the same footing.

[CosPhi]

We ARE metric.  1 in = 2.54 cm exactly, which is in turn defined by the speed of light and a resonance of atomic cesium.

In my work, I have no problem switching between units.  Especially since most CAD has a switch for what the user sees.  Imperial are more familiar to me, but I have a good idea what some mm length is, or how heavy some kilos are, without having to think about it.  PCB footprints are normally in mm; SI units are easy for transformer design (I conventionally use MHz, uH, mm^2 and etc., but that's just multipliers on SI base units).

Of course on the CAD can you change easy "what the user see". But on some point the product goes to the real world and suddenly the distance between two holes is 9,2319 or whatever in cm ... and that's just annoying.

[miguelvp]

Of course on the CAD can you change easy "what the user see". But on some point the product goes to the real world and suddenly the distance between two holes is 9,2319 or whatever in cm ... and that's just annoying.

well if you have the hole distance to be 3.6346062992125984251968503937008 inches apart then what do you expect?

[T3sl4co1l]

The metric system is based on scientific quantities. With the imperial system you always have to use clunky conversions. Calculate power from torque and rpm for example. Same goes for all kinds of electric field formulas.

Physics doesn't care about units, but do you know how many kilo watts your car is capable of?  Mechanical HP is still imperial, Metric HP is a bit under, who made it to be 75Kg was it just the best round figure to make it close to the imperial one, I mean was that just an arbitrary value?

So now the standard should be KWatts but no one uses that for cars do they?

Easy: a HP is about 3/4 of a kW (oh crap, not fractions again?! ).  So an average 80-120 HP commuter car is in the 60-90kW range (which at full throttle, means fractional megawatts thermal dissipation in the radiator and exhaust -- a heck of a lot of fuel sure goes to waste in such a heat engine!).

Or if you're talking electric motors, you might as well round up for losses, so a 1 HP (746 W, but might as well call it 750) motor might require around 1kW electrical.  Bit of an overstatement, but accounts for power factor and loading.  Size fuses accordingly, eh?

Something yet further helpful: 480VAC three phase is about one HP per ampere.  That's actually 480V * sqrt(3) ~= 831W, so again a bit of an overestimate.  (The 400V in Europe comes out a bit less instead, but you guys don't have to worry about HP much,  eh?)

Tim

[HighVoltage]

Once I met a guy in the US, who just had returned from a trip to Europe and he told me the following:

"Can you believe that they are still using liter for gasoline quantity measurements."

Well, I felt at that time that the imperial system will probably never change in the US.

[PedroDaGr8]

[miguelvp]

wiki:

In 1972, the PS (Metric horsepower) was rendered obsolete by EEC directives, when it was replaced by the kilowatt as the official power measuring unit.[15] It is still in use for commercial and advertising purposes, in addition to the kW rating, as many customers are still not familiar with the use of kilowatts for engines.

So in Europe, cars are supposed to be measured in kW, a mechanical HP is 745.699881448 Watts.

Your approximation is close enough. I still think saying that you have a 500 HP ride sounds better than a 373 kW ride.



 

[GK]

Your approximation is close enough. I still think saying that you have a 500 HP ride sounds better than a 373 kW ride.

I've never heard anyone boast of being in the 229 mm club.

[ignator]

I graduated 30 years ago. All my classes were taught in both SI and SAE. So there is no excuse to not changing to the SI standard in the USA.
So started work, where the old guard, only learned SAE, and dictates all drawing standards, only does SAE dimensioning (with SI in parentheses). Even when my generation took over, we didn't change this.

Canada made the change, but anything they sell here from a materials standpoint are SAE. I assume the UK/AU made the change for trade with their closest partners.

Look at all the international design standards, in my industry it was ARINC (Avionics Radio Inc.)  Every airframe manufacture in the world uses these standards for interchangeable avionics packages. It is still SAE dimensions on LRU (line replaceable unit) package standards. This industry is not the only one plagued by this. Some of these standards are more then 50 years old, and there's no attempt to change them.
Aircraft are required to display altitude in feet, and airspeed in nautical mile references.
In the end, everything is based on arbitrary units of measure. The meter or yard, were good at measuring how far a horse could ride in a day, not that these were based on that usage. But that was the distance of interest when these became standards.
The USA will morph someday to the SI system, I see no driving reasons that will accelerate this. We tried once, and then quit.
We have bigger problems.

[Monkeh]

Aircraft are required to display altitude in feet, and airspeed in nautical mile references.

Nautical miles are defined in metric units. As for feet, that may never change, and with good reason. A mistake in altitude can be fatal.

We already had the Gimli Glider due to a human error after fuel units were switched to metric.

[AlfBaz]

As for feet, that may never change, and with good reason. A mistake in altitude can be fatal.

feet -> meters, yes jettisoned from the earth's atmosphere 

[HackedFridgeMagnet]

feet -> meters, yes jettisoned from the earth's atmosphere

the irony is we are stuck with the US spelling of meters (metres) but they themselves refuse to use them.

[flolic]

So in Europe, cars are supposed to be measured in kW, a mechanical HP is 745.699881448 Watts.

Exactly, my little car has 55kW engine and that's what is written on registration papers.

[miguelvp]

feet -> meters, yes jettisoned from the earth's atmosphere

the irony is we are stuck with the US spelling of meters (metres) but they themselves refuse to use them.

Maybe the US is trying to erase the French influence in the English language? (mètres) 

[Rigby]

It's just inertia.  The official standard in the US for almost everything is metric.  Metric is taught in schools.  It's just inertia amongst us old fuddy duddies.

[AlfBaz]

I've never heard anyone boast of being in the 229 mm club.

What is this, some obscure car thing? Or did you mean 2290mm, as in about 7.5 feet tall? See, it's just your cultural bias, in the rest of the world it's as meaningless as 229mm is to you.

Probably more gender bias than cultural

[zapta]

I've never heard anyone boast of being in the 229 mm club.

What is this, some obscure car thing? Or did you mean 2290mm, as in about 7.5 feet tall? See, it's just your cultural bias, in the rest of the world it's as meaningless as 229mm is to you.

It's an adult related reference. Wrong forum.

[John Coloccia]

Why is that any different, math is math and doesn't care about units. the hypotenuse is still going to be the square root of the sum of the squares regardless of the unit.

Right, but you put the numbers into your calculator and it says something like 5.627 inches. Your ruler doesn't have decimal divisions though, and you need to go through the extra step of converting it to a fraction you can use.

Many of my rulers have decimal divisions.  I'm sorry, but you really just don't know what you're talking about.  In unfathomable to you that you can use decimals and imperial units at the same time. 

[Tac Eht Xilef]

the irony is we are stuck with the US spelling of meters (metres) but they themselves refuse to use them.

Who's "we"? In Commonwealth countries the two spellings - "metre" & "meter" - mean two different things, and come to English via very different linguistic paths (though arguably from similar prehistoric root language).

• "Metre" - the unit of measurement - comes via the French "mètre" from the ancient Greek "metron" (a measure).
• "Meter" - the thing that measures - comes via the Middle English "mete" (dispense, allot) + Old English "-ere" (something which does), from the Old English/Germanic "metan"/"meten"/"messen" (to measure), which itself comes from Indo-European roots that predate Latin & ancient Greek.

Only the Yanks confuse themselves by spelling the distance "meter"...

[saturation]

All scientific endeavors are done in metric in the USA.  Where pure science crosses into practice such as technology, medicine and engineering, you can have bit or both, with metric usually dominating. 

[GK]

Or did you mean 2290mm, as in about 7.5 feet tall? See, it's just your cultural bias, in the rest of the world it's as meaningless as 229mm is to you.

Yes, that's it; a Charles Byrne reference. Dunno what the other guys are talking about.

[GK]

Why is that any different, math is math and doesn't care about units. the hypotenuse is still going to be the square root of the sum of the squares regardless of the unit.

Right, but you put the numbers into your calculator and it says something like 5.627 inches. Your ruler doesn't have decimal divisions though, and you need to go through the extra step of converting it to a fraction you can use.

Many of my rulers have decimal divisions.  I'm sorry, but you really just don't know what you're talking about.  In unfathomable to you that you can use decimals and imperial units at the same time.

All of my imperial/metric rulers have 1/16 divisions on the imperial edge.

[John Coloccia]

All of my imperial/metric rulers have 1/16 divisions on the imperial edge.

It's traditional to use fractions for woodworking/building, etc.  Fractions have a great advantage in the workshop.  If I have a piece of wood that's 27/32" wide, and I want to get to the center, it's simply 27/64.  When you're doing calculations for router bushings, offsets and things like that, again it's usually a double and 27/32 is simply 27/16 (1 11/16").  Or if you have two pieces that are 27 23/32" and 27 31/32", and you want to cut a piece to split the difference.  No problem...make it about 27 27/32".  Why do you think so many people still use fractions in shops and on the jobsite....because we're too dumb to know decimals exist?  It's because it's far more convenient than decimals for that kind of work.  Decimal rules and tape measures are commonly available.

When you're doing other sorts of measuring, you use decimal measure.  In fact, the different kinds of fractional and decimals rules are so common that they've basically been standardized by Starrett and you can refer to them by name.  The kind of rule I use most often is a 16R.  That's 4 scales with 32s and 64s on one side, and 50ths and 100ths on the other.  You can get all different styles, including imperial on one side, and metric on the other.

Imperial != fractions.  The scientific community has switched over to metric because in that application there IS value to be had where different units are interrelated in simple and intuitive ways.  No one else over here has switched because there is exactly zero value to be had being able to easily calculate how many guitar necks would fit in a mile, or how many 9V batteries I need in order to equal the weight of water sitting in my hot tub.

[miguelvp]

What is the diagonal distance of the monitor you are reading this?
Mine is 10.1" (Asus transformer prime with keyboard dock)

See, we can use decimal numbers with inches since we also use the decimal numbering system.

Edit: About the ruler division, maybe that's why fixed point hexadecimal math seems more natural to us. (of course I'm kidding, kind of)

[HackedFridgeMagnet]

If I have a piece of wood that's 27/32" wide, and I want to get to the center, it's simply 27/64.  When you're doing calculations for router bushings, offsets and things like that, again it's usually a double and 27/32 is simply 27/16 (1 11/16").

Must be a bitch if you want to divide into five though. Seriously doesn't seem easier than just using mm.

I never use cm just mm, metres, km.

Some advice for our imperial measurement using friends, just stop now, your descendants will be grateful.

[miguelvp]

If I have a piece of wood that's 27/32" wide, and I want to get to the center, it's simply 27/64.  When you're doing calculations for router bushings, offsets and things like that, again it's usually a double and 27/32 is simply 27/16 (1 11/16").

Must be a bitch if you want to divide into five though. Seriously doesn't seem easier than just using mm.

I never use cm just mm, metres, km.

Some advice for our imperial measurement using friends, just stop now, your descendants will be grateful.

Yes but  1 11/16" is 00000001(.)10110000 in binary  and 0x01(.) 0xB0 in hex and I didn't even need a calculator. So your tooling hardware will struggle with round off using metric system but no problem with the imperial one.

Take the 27/32, what is 27 in hex well it's 0x1B divided by 0x20 or shifted right  5 times, see, simple and the result will be binary compatible. win/win

27/64, 0x1B shifted right 6 times.

[GK]

Why do you think so many people still use fractions in shops and on the jobsite....because we're too dumb to know decimals exist? 

All I wrote is that "all of my imperial/metric rulers have 1/16 divisions on the imperial edge" 

But continuing with the theme.... if I have a piece of wood that is 50mm wide and I want to find the center, it's just 25mm. Like, duh. I'm struggling to see the "great advantage" of traditional imperial fractions here.

[Rigby]

27/64, 0x1B shifted right 6 times.

[nctnico]

What is the diagonal distance of the monitor you are reading this?
Mine is 10.1" (Asus transformer prime with keyboard dock)

That is a perfect way to demonstrate inches are inaccurate  There is nothing more deceitful than monitor sizes. Back in the CRT days they would measure the size almos between the mounting posts. LCD is a little bit better where they usually specify the edge of the panel.

[SirNick]

Sticking to an established system because it would be costly and difficult to switch is a ridiculous excuse.  The more global we get, the more we will be forced to deal with metric anyway.  Meanwhile, the rest of the world has been putting up with converting all their native measurements to our archaic system, so they can communicate with our lazy, ego-centric desire to avoid a bothersome transition.

The metric system has a very profound advantage:  It's what everyone else uses.  Maintaining two systems helps no one, not even us over the long term.  Delaying the inevitable only prolongs the pain.

I realize it's not going to change overnight, and it's also not going to come as a mandate from the government.  Everyone with an interest in getting on with it just needs to start making it a little more uncomfortable to keep up this foolishness in their own little circle of influence.  We have 2L bottles and often see things in mm and deg. C.  It won't be that hard to have metric measurements start creeping in to other parts of our lives.  Let the entrenched industries have their way until it does become cost-effective to switch.  Until then, there's no reason we can't enjoy a nice, cold (8C) beer. 

[Rigby]

That's not the fault of the inch or any other unit of measure.  Don't confuse lies with inaccuracy.

Like most things, there are different ways to do it.  If the rest of the world has standardized, then the US should, too, good or bad.  Otherwise a lot more time will be spent in discussions like this one than it would actually take to make things better by standardizing.

The point isn't that one is better or worse, its that we (the US) think that we know better than everyone else, or that every other country is wrong, or that we're too busy or important or whatever to actually make the switch.  My daughter pulls this noise whenever its time for her to do something she doesn't want to do, and she's just whining because she'd rather play Minecraft than take a shower.

The US is the same way.  We are the whiny child of the world that strong arms others until they capitulate or we leave and take our toys with us.  Any American that doesn't agree with me and HASNT lived multiple consecutive years outside of the US should do so before they disagree with me.  There's a lot to be learned by living away from your home country, no matter where you're from.  Learning that imperial vs. Metric is just a symptom was part of that for me.

[miguelvp]

27/64, 0x1B shifted right 6 times.

areyouawizard.png

LOL, nah.

In decimal if you want to divide by 10 you shift the number one to the right, divide by 1000 then 3 to the right because:
10^0=1
10^1=10
10^2=100
10^3=1000

so 234,667/10,000 = 23.4667 right?

Same thing with binary
2^0=1
2^1=2
2^2=4
2^3=8
2^4=16
2^5=32
2^6=64
...

so what is 0x43bc2a in binary 0100 0011 1011 1100 0010 1010 divided by 64 shift it 6 times to the right
0100 0011 1011 1100 00(.)10 1010 or 0x010ef0 integral part and 0xa8 for the fractional part.

Same math that applies to decimal applies to binary. and since imperial fractions are 1/2 1/4 1/8 1/16 1/32 etc.. it translates very nicely to fix point or floating point for that matter.

so on floating point where the mantissa has a binary (1.) implied
the mantissa of 0100 0011 1011 1100 00(.)10 1010 will be

(01.)0000111011110000(.)101010 x 2^16 (because the period is 16 digits to the right so
0000 1110 1111 0000 1010 1000
the exponent 16 (0x10 hex) add that to the bias (127 or 0x7F in hex) of IEEE floating point representation so 143 decimal 0x8F hex and 0 for the sign and you have your float.


s     exp             mantissa
0 | 1000 1111 | 0000 1110 1111 0000 1010 100

organize it right and convert to hex
0100 0111 1000 0111 0111 1000 0101 0100

0x47877854, that should represent 69360.65625 which is the answer of  0x43bc2a divided by 64.



So it's base 2 divisions are simple to do and they store nicely in your system, Imperial wins

I kid, I like both, and I can use either, but for tooling using based 2 fractions wins. Since binary can't divide by 5 without error. So our tool fractional sizes as odd as they seem actually have an advantage as far as computers go.

[zapta]

The metric system has a very profound advantage:  It's what everyone else uses.

By this logic the French should switch to English.

[SirNick]

No, 'cause they're all mimes.  That would make no sense at all.

[Corporate666]

Sticking to an established system because it would be costly and difficult to switch is a ridiculous excuse.  The more global we get, the more we will be forced to deal with metric anyway.  Meanwhile, the rest of the world has been putting up with converting all their native measurements to our archaic system, so they can communicate with our lazy, ego-centric desire to avoid a bothersome transition.

The metric system has a very profound advantage:  It's what everyone else uses.  Maintaining two systems helps no one, not even us over the long term.  Delaying the inevitable only prolongs the pain.

I realize it's not going to change overnight, and it's also not going to come as a mandate from the government.  Everyone with an interest in getting on with it just needs to start making it a little more uncomfortable to keep up this foolishness in their own little circle of influence.  We have 2L bottles and often see things in mm and deg. C.  It won't be that hard to have metric measurements start creeping in to other parts of our lives.  Let the entrenched industries have their way until it does become cost-effective to switch.  Until then, there's no reason we can't enjoy a nice, cold (8C) beer. 

I don't think the cost is a ridiculous excuse at all - it's actually the only excuse that really makes sense.  The costs of switching would far outweigh the benefits, otherwise we would have switched already.  So calling for the switch to be made is really calling for something that will cost alot more than we'll get out of it.

Like a high speed train from San Fran to LA, or something

[Richard Crowley]

By this logic the French should switch to English.

Well, the Metric System appears to be the only enduring after-effect of the French Revolution.

We (USA) use the same amps, volts, ohms and watts that everyone else uses. And those are "metric" aren't they?   

[zapta]

I don't think the cost is a ridiculous excuse at all - it's actually the only excuse that really makes sense. 

Well said!

A long engineering thread that discusses a proposed solution but completely ignores the cost side. Strange.

[HackedFridgeMagnet]

I don't think the cost is a ridiculous excuse at all - it's actually the only excuse that really makes sense. 

Well said!

A long engineering thread that discusses a proposed solution but completely ignores the cost side. Strange.

I don't know how you calculate those costs.
But you are also completely ignoring the costs of not changing and/or the potential savings of changing.

[Richard Crowley]

But you are also completely ignoring the costs of not changing and/or the potential savings of changing.

No, I think that is exactly what he was considering, and I agree completely.

Even nominally "metric" countries still use imperial units in many places.

[miguelvp]

Well, the Metric System appears to be the only enduring after-effect of the French Revolution.

Thank God the French republican calendar didn't go through and was abolished by Napoleon.

Imagine 12 months 30 days each, each month with 3 decades, rest was on the last day of the decade (1 day out of 10) at the end of the year having 5 days (6 on leap years) of revolution celebratory days that didn't belong to any month. and a 10h/day clock with 100 minutes per hour and one hundred seconds (probably not SI seconds).



http://en.wikipedia.org/wiki/French_Republican_Calendar#Decimal_time

¡Oh là là!

And I like the French, they did help us a lot to gain our independence.

About the cost, well it would be tremendous, all the industries will have to retooled, everything redesigned, all the infrastructure replaced, will be nuts! (Imperial nuts at that).

[Bored@Work]

Since binary can't divide by 5 without error. So our tool fractional sizes as odd as they seem actually have an advantage as far as computers go.

This is just once again making up an example so it fits the argument. We can do that to:

Counter example 1:

Divide 1010(bin) by 5(dec) and you have a binary divided by 5 without error.

Counter example 2:

You want something with a fraction? No problem: 10.1(bin)/5(dec) = 0.1(bin)

Hint: For those who want to make up some examples, the trick is to start with a multiple or fraction of the divider. Want to show that the ancient Egyptians knew some magic about space? Take some size measurement on some ancient pyramid. Doesn't matter what. Divide the distance earth-moon by that number. Not a round number? "adjust" the distance earth-moon to be a multiple of the obscure pyramid measurement in your calculation. It is anyhow not constant. Run the calculation again, publish a parascience paper about your discovery "Ancient Egyptians based Pyramid Size on Earth-Moon distance". Ignore the naysayers who reproduce your results using metric and who complain that the numbers look rather ugly.

I don't think the cost is a ridiculous excuse at all - it's actually the only excuse that really makes sense.

Not really, that argument also doesn't hold water.

a) Production in the US is low (check out current pictures of Detroit ...). There isn't simply that much there to be changed.

b) The argument always assumes it has to be done as a big-bang. Which is not true. You can gradually change. Actively taking up opportunities, instead of working twice as hard to ignore opportunities.

You need to update a process? Introduce metric for that process. Old machine broken? Replace it with a new one showing measurements in SI units. Your office runs out of supplies? Order the next batch of paper and envelopes in A and C sizes (and kick your IT guy to adjust the paper tray in your printer one notch to hold A4). A rusty roadsign needs replacement? Replace it with one showing distances in km, and write km behind each number.

c) Those countries who changed in the past didn't go bankrupt, the world didn't come to an end and hell didn't take over.

[HackedFridgeMagnet]

Well even if you don't want to change to metric. Do you think it is a better system?

As far as changing over, you are already doing it, but some of you haven't noticed.

I still think he was ignoring to costs of delaying the change to metric.
I for one really don't want to take two sets of spanners, two sets of Allen keys to a job. You surely will admit there is a cost in this.

[miguelvp]

Since binary can't divide by 5 without error. So our tool fractional sizes as odd as they seem actually have an advantage as far as computers go.

This is just once again making up an example so it fits the argument. We can do that to:

Counter example 1:

Divide 1010(bin) by 5(dec) and you have a binary divided by 5 without error.

Counter example 2:

You want something with a fraction? No problem: 10.1(bin)/5(dec) = 0.1(bin)

I said "our tool fractional sizes" so they are integers divided by base 2 integers representing values less than 1.
they have non error binary floating point representation all the way up to 1/16777216 increments or 2^-24. So you can store 16777215/16777216 and still be a non round off value.

Edit: And that's for 32 bit floats, with doubles well it goes even further.

No tricks, just coincidence that our tool fractions are base 2 fractions.

[miguelvp]

It's ok, you guys change everything to metric. We will be the only ones enjoying a pint  of beer at the bar.

Now, tell me that you'll be ok changing that and ask the barkeep  "I'll have a 473ml"

[IanB]

Now, tell me that you'll be ok changing that and ask the barkeep  "I'll have a 473ml"

Then you are definitely missing out on 95 ml 

[miguelvp]

yeah, the old imperial one not our pint, oops. But I rather tell the missus that I only had 3 pints than over a liter and a half.

[GK]

 

It's either a Middy, Pot, Handle or a Schooner down here.

[flolic]

Now, tell me that you'll be ok changing that and ask the barkeep  "I'll have a 473ml"

No, but then you'll be asking barkeeper for half liter like we all do on a Old Continent (sans UK)
So, that's even better 

[John Coloccia]

Why do you think so many people still use fractions in shops and on the jobsite....because we're too dumb to know decimals exist? 

All I wrote is that "all of my imperial/metric rulers have 1/16 divisions on the imperial edge" 

But continuing with the theme.... if I have a piece of wood that is 50mm wide and I want to find the center, it's just 25mm. Like, duh. I'm struggling to see the "great advantage" of traditional imperial fractions here.

I guess you think it's cute to pick 50mm, instead of something like 15.7mm.  Whatever.  I don't really care if anyone else sees the usefulness of fractions in the shop.  It seems rather short sighted and close minded to me, but what else is new.  I'm just trying to explain why it's still in widespread use.  It's because it's USEFUL.

[GK]

Why do you think so many people still use fractions in shops and on the jobsite....because we're too dumb to know decimals exist? 

All I wrote is that "all of my imperial/metric rulers have 1/16 divisions on the imperial edge" 

But continuing with the theme.... if I have a piece of wood that is 50mm wide and I want to find the center, it's just 25mm. Like, duh. I'm struggling to see the "great advantage" of traditional imperial fractions here.

I guess you think it's cute to pick 50mm, instead of something like 15.7mm.  Whatever.  I don't really care if anyone else sees the usefulness of fractions in the shop.  It seems rather short sighted and close minded to me, but what else is new.  I'm just trying to explain why it's still in widespread use.  It's because it's USEFUL.

And I find it "cute" that you picked a measurement that was a particularly convenient fraction of an inch! Of course measuring stuff out in easily divisible fractions of an inch makes the imperial system easier to use in some situations, but that hardly makes the imperial system better or generally easier to use than the metric system.

[John Coloccia]

Are you familiar with traditional Japanese woodworking measure?  Its even loonies than imperial units, but its VERY convenient for woodworking. The point is that standardizing for the sake of standardizing is not necessarily the best approach. There are advantages to using the most appropriate system for the task at hand.

And again, units have nothing to do with whether you use fractions or decimals. Many tasks would be much simpler with fractional metric measurement.

[vk6zgo]

It's either a Middy, Pot, Handle or a Schooner down here.

Unfortunately,there has been a "re-imperialisation by stealth",at least in WA.
Most Taverns & the like sell beer in Pints-----I think it was driven by all the little trendoids who like to pretend they are living in London!

Bunnings used to have a wide range of metric bolts,but now its a lot fewer,with most of their bolts Imperial-----I guess the Indians are still on the Imperial standard!

PS--there,that's better!--I somehow lost the "quote end" symbols!

[miguelvp]

More Ireland than London. Lot's of Irish in our melting pot.

[John Coloccia]

I guess you think it's cute to pick 50mm, instead of something like 15.7mm.  Whatever.  I don't really care if anyone else sees the usefulness of fractions in the shop.  It seems rather short sighted and close minded to me, but what else is new.  I'm just trying to explain why it's still in widespread use.  It's because it's USEFUL.

But why is it more useful than decimal, or just using multiples of 12 metric units or whatever happens to suit your project? In other words why it is better than metric?

Imperial doesn't equal fractions.  Has nothing to do with the unit of measure.

But the reason using fractions based on powers of 2 is more convenient than fractions based on powers of 10 (i.e., decimal) is that in wood working and construction you're frequently doubling and halving measurements.  You're rarely multiplying or dividing them by 10.

It's how people work in the shop.  You're constantly eyeballing and doubling things or halving things...that comes up all the time.  Even when I work in metric, I still think about things as "halfway between this and that", and rarely really care about the actual measurement.  "In the middle" is all the precision I need.  Using fractional measures based on powers of 2 means that I can very quickly note what it is and transfer it up and down a piece...or transfer it to a marking gauge.

It's not arbitrary just to make everyone's life hard.  It's the way it is because it makes things EASIER, and the imperial UNITS are also intelligently designed to be easy to work with.  12" in a foot is easily divided into 1/2s, 1/3s and 1/4s.  Quite convenient, no?  It's traditional to build Classical guitars using metric measurement, though it's traditional to build just about every other guitar with imperial measurement.  All of the classical builders I know talk in metric units, but say things like, "Oh...that's about 22 1/4mm....maybe 22 1/3mm at the most"....and then they have to convert the fraction to a number to look it up on the stupid metric ruler that is pretty much unavailable with fractions on it.  That's reality.

Metric and decimals are great when you're constantly converting back and forth between different units and you need to know things like how much cement do you need to support a swimming pool of a certain size.  Nice not to have all these crazy factors kicking around.  That's not a common everyday task.  Imperial measure, and also the fractional system that is traditionally used, is designed around people eyeballing, chopping things in half or thirds, and things like that.  It's a very practical system.

It's also why we still use nautical miles.  It's supremely convenient for doing calculations in your head.  It's also why people working with optics often use arc-seconds.  It's amazingly powerful to do back of the envelope calculations in your head.

[Rigby]

Don't confuse measurement system merit discussions with all of the benefits that come with standardization.

[John Coloccia]

Personally, I wouldn't mind if the US went all metric.  I work in both system interchangeably.  The reasons we haven't moved yet are a lot more deeply seated than simply cost or stubbornness.  That's my point.  There's a ton of engineering work that's already done in metric.  Any of my new mechanical designs where it's convenient is designed in metric with metric hardware.  I wouldn't mind as good first step to make gallons and miles disappear.  There's little value in converting between gallons and, say, tablespoons, for example, and little value in converting from miles to feet or inches, so it would be no big deal to start converting gallons and miles to liters and km/h.

I have little doubt that the inch will still be going strong long after I'm gone.

[jlmoon]

Just about punched a hole in my LCD (plastic) screen trying to mash "GeoffS's" bug icon.. 

[zapta]

Just about punched a hole in my LCD (plastic) screen trying to mash "GeoffS's" bug icon.. 

I assume it's not a self portrait.  ;-)

[jlmoon]

lol.. nah.. go check out his icon/avatar.. animated gnat .. fooled me eyes for a sec

[SeanB]

I have learned to ignore it.........

[jlmoon]

yes.. I as well!

[ajb]

Well even if you don't want to change to metric. Do you think it is a better system?

For some fields, such as science and 'abstract' engineering, sure.  For everyone else, not particularly.  For most purposes the better system is always going to be the one you're more comfortable with, both in ability to calculate and in ability to grasp the actual magnitude of the units.

I for one really don't want to take two sets of spanners, two sets of Allen keys to a job. You surely will admit there is a cost in this.

  Sure.  So let's immediately stop producing new designs in imperial and make everything in metric instead.  Now that's done, all you need to take to a job is. . . .both sets of tools because there are still millions of pieces of equipment in the field that were built with imperial fasteners.  How long do you expect it to take to naturally cycle out even a small fraction of the imperial fasteners in the field before you can confidently take only one set of tools on a service call?  Decades, surely.  In the mean time, you'll keep carrying two sets of tools and factories will continue to churn out imperial fasteners as spares if nothing else.  Now that we've spent billions of dollars to change tooling, when does switching to metric start saving us money?  For better or worse, people have a hard time planning on that time scale, especially when the upfront cost is so massive and any reward so distant.

As far as changing over, you are already doing it, but some of you haven't noticed.

Indeed.  I see metric dimensions on more and more building materials, enclosures, and so on.  Globalization helps with this, as more companies outside the US supply goods here and as US companies sell more goods overseas it only makes sense to adopt the more common units in your marketplace.  Ballpark, I'd say in another 50-100 years at the current pace the US will be where most other countries are now--metric used almost everywhere except in a few persistent traditional areas.  Maybe even longer with things like lumber where there's so much inertia in the trades and building codes built up over a hundred or more years.

[SeanB]

I have the metric and imperial tools, and use them a lot. Some machinery I work on was designed a century ago, and has remained basically unchanged since then, with spare parts from the originals still being made.  Of course the country of origin changed from USA to Mexico, then China and now India, always chasing lowest cost. User price though always has a nasty tendency to rise though. It was cheaper for many parts to simply have them made locally in an engineering shop, and they came out as better quality and a lot faster than the OEM parts to get back into service.

[SirNick]

All of the classical builders I know talk in metric units, but say things like, "Oh...that's about 22 1/4mm....maybe 22 1/3mm at the most"....and then they have to convert the fraction to a number to look it up on the stupid metric ruler that is pretty much unavailable with fractions on it.  That's reality.

You know woodworkers that speak in "about" fractions of a millimeter.  As an eyeball figure.  Y'know, give or take a few micron...   

The reason fractions are so useful in Imperial measurements is because we lack a system that makes it easy and convenient to jump to the unit that most accurately portrays scales of a base unit.  You can have a plank of wood that is a meter long, and fine-tune that with as many millimeters as it takes.  Or centimeters and then millimeters.  They're all directly and trivially comparable.

OTOH, what's the equivalent of 1mm in fractions of an inch?  Which is really easier to measure and comprehend?  3.001 meters?  Or 1 yard and 13/32"?

This isn't about gimmicky comparisons like the number of soda cans it takes to fill an amusement park aquarium.  There are real-life advantages too.  I've used Imperial all my life and I still can't immediately visualize how many yards are in a mile.  Why would that matter?  I dunno.  Imagine a construction job laying fiber optics, that come in meters, and knowing how many km you need to go.  Imagine the same with spools of 1000 feet and distance in miles.  Ugh.  Not rocket science, but not something you're likely to be able to pull off the tip of the tongue either.

b) The argument always assumes it has to be done as a big-bang. Which is not true. You can gradually change. Actively taking up opportunities, instead of working twice as hard to ignore opportunities.

Exactly my point.  No need to throw out all your old Imperial tools wholesale, that's a nonsensical way of thinking.  Just buy a metric one (if you don't already have it anyway...) and start using that for the easy stuff.  Progressive conversion.  Piece of cake.  Crises averted, and the rest of the world can stop catering to our whims.  (There's your opportunity cost BTW.)

[SirNick]

How long do you expect it to take to naturally cycle out even a small fraction of the imperial fasteners in the field before you can confidently take only one set of tools on a service call?  Decades, surely.  In the mean time, you'll keep carrying two sets of tools and factories will continue to churn out imperial fasteners as spares if nothing else.  Now that we've spent billions of dollars to change tooling, when does switching to metric start saving us money?  For better or worse, people have a hard time planning on that time scale, especially when the upfront cost is so massive and any reward so distant.

It entirely depends on what your vocation is, I suspect.  Farm equipment?  Maybe a couple decades.  Computers?  Probably by the end of the decade.  There are opportunities with enough turnover that it really doesn't matter if you switch tooling, because you're going to be spinning new designs anyway, and that's going to be the most expensive part of the bargain.

Besides, that "how does it affect ME, NOW?" attitude is one of mankind's greatest faults.  I guess you could leave this mess for later generations to sort out, when the world has become even smaller.  Or you (not "you" you, but "us" you) could man up and take steps reduce the wasteful redundancy over the course of a lifetime.  Little things can mean a lot, eventually.  It doesn't hurt too badly to do the right thing, even if you never live to see the benefit.

[John Coloccia]

OTOH, what's the equivalent of 1mm in fractions of an inch?  Which is really easier to measure and comprehend?  3.001 meters?  Or 1 yard and 13/32"?

Why would I ever convert fractions of an inch to mm?  If I was working in mm, I'd work in mm.  If I'm working in decimal anything (imperial OR metric), then I work in decimal.  If I'm working in fractions, I work in fractions.

Imperial does not mean fractions.

Anyhow, if you're comparing a bunch of pieces of wood, and all of them 1 yard and a few 32s of an inch, then it's much easier to quickly split differences using the leftover fraction, actually.  Again, for this and other reasons, many trades have stubbornly stayed with fractional measure.  The inch unit is also particularly convenient for a LOT human sized things.  Guitars....clocks....forks....bowls....shoes....etc.  They are all very conveniently described with small numbers....a few inches wide....15 inches long, etc.  These measures all came about based on the practicality of eyeballing things and quickly dividing them up in your head.  Given something that's about a foot long.....say, the average foot wearing a shoe, for example, it's downright trivial to subdivide that into inches by simply continually chopping it in half.  Any mistake is immediately apparent.  I'll bet you that I could subdivide by hand and by eye down to 1/64" with wicked accuracy because the human eye is particularly good at doing such tasks and seeing small differences in intervals.  I would never do that, but this is the basis of the system....basic measures that easily subdivide into other, useful measures.

But anyhow, over here we seem to be able to work in whatever system suits the situation best.  Given the prolific and world class science and engineering prowess that the US possesses, the notion that this metric/imperial thing is somehow an impediment seems not to be accurate.     It just doesn't seem to be a big deal.  Someone will undoubtedly bring up some satellite that crashed because of a conversion between metric and imperial, but it could just as well have crashed converting from meters to kilometers.  Why would anyone ever specify a design in mixed units?  Poor design is poor design regardless what country you happen to live in.

I see that a lot too, though.  Tons of projects are designed in inches, using metric hardware.     Seems looney to me, constantly converting back and forth to get clearance holes sizes and things like that.  THAT I'll never understand.

[SirNick]

OTOH, what's the equivalent of 1mm in fractions of an inch?  Which is really easier to measure and comprehend?  3.001 meters?  Or 1 yard and 13/32"?

Why would I ever convert fractions of an inch to mm?  If I was working in mm, I'd work in mm.

You wouldn't.  My point was that the Imperial system doesn't scale well below an inch.  There's no common, vernacular unit of measurement for small portions of an inch, except for... well... "portions of an inch", whether that be fractions or decimal.  Yes, in engineering we use "thou" or "mil"... but say that to a pedestrian and they'll think thou speaketh oddly, or in metric.  In (actual) metric, you've got mm right there.  No need to resort to fractions.  Not that fractions are hard per se, or unique to Imperial, but they are the de facto way of discussing things smaller than an inch.  Furthermore, if a mm is too big, there's another SI unit below that.  And one below that.

Imperial does not mean fractions.

*I* know that, but you have been championing Imperial for its convenient application in wood-working, specifically because of rulers and their fractional units of measure.  (Which is tied into the argument above, about there not being a common sub-inch UoM.)

The inch unit is also particularly convenient for a LOT human sized things.  Guitars....clocks....forks....bowls....shoes....etc.  They are all very conveniently described with small numbers....a few inches wide....15 inches long, etc.

They are equally convenient in metric, if you have a good frame of reference in metric.  As an American, I don't.  Oh well, learning curve.  However, I know that an inch and a cm are close enough that I can get a (very) rough idea of how big something is by equating it in inches.  It's not 1:1, but if cm is the right scale, I know if we're talking about a grapefruit size object or a house size object.  So what's wrong with measuring forks and shoes in cm?

But anyhow, over here we seem to be able to work in whatever system suits the situation best.  Given the prolific and world class science and engineering prowess that the US possesses, the notion that this metric/imperial thing is somehow an impediment seems not to be accurate.

World-class, maybe.  But not unique.  And the foreigners are gaining on us.  If you were insert-European-country-here, and you could buy a widget from the US, or an equivalent widget from, say, Singapore, except that it would be in your native unit of measure, why would you buy the Imperial one?  (That's not even considering the price disparity.)  America is on the verge of losing its place of dominance.  We should be doing everything we can to be competitive in a global marketplace, and using a proprietary unit of measure is not going to do us any favors.  It isn't 1970 anymore.

[IanB]

There's a lot of vagueness in this thread about "metric" and "imperial", and that itself indicates a source of confusion.

The USA does not in fact use the imperial system (which is used in the UK), it uses the U.S. customary system of measure. There are differences in how things are defined, most notably in fluid measure (where fluid ounces, pints and gallons are different sizes).

There are also various different metric systems of measure in use around the world, some of which relate to SI, and some which do not. (For instance in Japan it is customary to measure pressure in kg_f/cm², which is not an even multiple of the pascal or N/m².

Systems of measure are used in two ways: (1) for specifying and measuring dimensions, and (2) for doing engineering calculations.

For specifying and measuring, any agreed upon system is as good as any other.

For calculations, this is not the case. If the system of measure is not internally consistent then various odd conversion factors will appear in formulas everywhere, conversion factors that you have to look up or memorize. For instance, how many ft-lb_f/s are there in 1 hp?¹ This conversion factor will come up if you do a compressor power calculation involving pounds per square inch, cubic feet per minute, and horsepower (not to mention seconds in a minute and cubic inches in a cubic foot).

This is so much the case that engineering software typically will do its internal calculations in SI and convert units of measure on input and output.

¹ There are 550 ft-lb_f/s in 1 hp

[HackedFridgeMagnet]

As a matter of interest, what do they use in Britain and US for land areas smaller than an acre?

The old title for my land says something like 1 acre, 3 rood, 7 chains and I forgot whats after that, maybe it was square feet. WTF!
How did they ever work with that?

Luckily when the council recalculated the area they came up with 0.6 hectares but I think it is nearer 0.8. Hopefully I am paying less council rates because of it.

Those units rood and chains really had to go.

[John Coloccia]

They are equally convenient in metric, if you have a good frame of reference in metric.  As an American, I don't.  Oh well, learning curve.  However, I know that an inch and a cm are close enough that I can get a (very) rough idea of how big something is by equating it in inches.  It's not 1:1, but if cm is the right scale, I know if we're talking about a grapefruit size object or a house size object.  So what's wrong with measuring forks and shoes in cm?

There's nothing at all wrong with it.  Metric is fine system.  The point that I'm making is that the system isn't "crazy".  The fractions come about specifically because the various units conveniently divide into 2's and 3's, which is convenient for estimating by subdivision by 2s and 3s...something which is easy to do by eye, and also easy to do in your head.  It's a very useful and well thought out system for what it is.

[staxquad]

There's a lot of vagueness in this thread about "metric" and "imperial", and that itself indicates a source of confusion.

The USA does not in fact use the imperial system (which is used in the UK), it uses the U.S. customary system of measure. There are differences in how things are defined, most notably in fluid measure (where fluid ounces, pints and gallons are different sizes).

There are also various different metric systems of measure in use around the world, some of which relate to SI, and some which do not. (For instance in Japan it is customary to measure pressure in kg_f/cm², which is not an even multiple of the pascal or N/m².

Systems of measure are used in two ways: (1) for specifying and measuring dimensions, and (2) for doing engineering calculations.

For specifying and measuring, any agreed upon system is as good as any other.

For calculations, this is not the case. If the system of measure is not internally consistent then various odd conversion factors will appear in formulas everywhere, conversion factors that you have to look up or memorize. For instance, how many ft-lb_f/s are there in 1 hp?¹ This conversion factor will come up if you do a compressor power calculation involving pounds per square inch, cubic feet per minute, and horsepower (not to mention seconds in a minute and cubic inches in a cubic foot).

This is so much the case that engineering software typically will do its internal calculations in SI and convert units of measure on input and output.

¹ There are 550 ft-lb_f/s in 1 hp

In Canada, we are caught between the USA and the UK for ancient measures.  When someone discusses gallons, we have to specify American gallons or British gallons knowing that American gallons are smaller with 128 ounces and British gallons are larger with 160 ounces, but darn garnit, it's so f-ed up, even the ounces aren't equal, so basically you have to just take it for granted that you get more with British gallons and get less with American gallons, and we thank the goodness of foresight for liters.  (when we buy milk we get 3 small clear bags of milk in a 4 liter carry bag {it's slightly less than a US gallon}  Nobody really knows the exact volume of one little clear bag of milk, except 4/3 liters.)






 

[IanB]

I've never seen milk sold in plastic bags before. However do you serve it? Do you pour it into a jug and keep washing up the jug?

[staxquad]

I've never seen milk sold in plastic bags before. However do you serve it? Do you pour it into a jug and keep washing up the jug?

[John Coloccia]

Now THAT I've never seen before.

[GK]

Metric and decimals are great when you're constantly converting back and forth between different units and you need to know things like how much cement do you need to support a swimming pool of a certain size.  Nice not to have all these crazy factors kicking around.  That's not a common everyday task.  Imperial measure, and also the fractional system that is traditionally used, is designed around people eyeballing, chopping things in half or thirds, and things like that.  It's a very practical system.

Do you remember those things called multiplication tables, that they used to teach by rote?

1,2,3,4,5......
2,4,6,8,10,12.......
3,6,9,12,15,18,21,24,27,30........
4,8,12,16,20,24..........
6,12,18,24,30.........
9,18,27,36,45.......
etc..........

Make it mindlessly easy/practical to divide crap into finer fractions than just half’s or thirds........
Of course difficulty arises trying to think of imperial measures in metric terms, but nobody working with the metric system proper works in multiples or sub multiples of 25.4mm, feet or yards. I think your argument is full of crap.

[vk6zgo]

[/quote]

In Canada, we are caught between the USA and the UK for ancient measures.  When someone discusses gallons, we have to specify American gallons or British gallons knowing that American gallons are smaller with 128 ounces and British gallons are larger with 160 ounces, but darn garnit, it's so f-ed up, even the ounces aren't equal, so basically you have to just take it for granted that you get more with British gallons and get less with American gallons, and we thank the goodness of foresight for liters.  (when we buy milk we get 3 small clear bags of milk in a 4 liter carry bag {it's slightly less than a US gallon}  Nobody really knows the exact volume of one little clear bag of milk, except 4/3 liters.)


[/quote]

Converting UK to USA is often confusing because they use the same term for different quantities,& have totally different standards in such things as screw threads.

In Oz,we often had British equipment with BA ,BSF,& Whitworth screws sitting side by side with US sourced stuff using those funny screws that seem to have no known non-US equivalent-----Tektronix are full of them!

Australian made stuff usually used UNF or Whitworth & BA.
Of course,UNF are one bright point as they are used in both the UK & USA.

To anyone living outside the USA,this variation in Imperial/Customary hardware is a potent reason for going to metric.

Even between non-metric countries,it is more convenient for,say,the UK to export stuff in 200 litre drums to the USA,as each side only needs to do one well known conversion,whereas it may not be immediately obvious that a "44 gallon" drum is equivalent to a US "55 gallon " drum.

Back in the day,Australians would be bemused by the difference in mpg figures for US made cars,between local & US Road tests.
In the same way,Americans would be disappointed by the seemingly poor mpg they got from British cars,compared to UK Road Tests.

[Rigby]

You said it quite well.

It doesn't matter what people standardize on, just freaking standardize already.  Everyone speak the same measurement system, it doesn't matter what it is.

[vk6zgo]

By the way,it occurs to me that there is a nice little "niche" market supplying US threaded screws to all those of us with American equipment with missing screws.
(And the rotten things hide! )

I suppose we first have to ask:- Do Tektronix,etc use standard US sizes that can be bought over the counter by normal handymen in that country,or are they "rogue" sizes specially for Electronics?

[Richard Crowley]

Do Tektronix,etc use standard US sizes that can be bought over the counter by normal handymen in that country,or are they "rogue" sizes specially for Electronics?

All large companies that make complex products use custom hardware optimized for each application.  I live only a few minutes drive from the Tektronix campus, but that doesn't make it any easier to find oddball screws.  Unless you happen to find a bag of them at the surplus sale.

[corrado33]

I would like for the US to convert to metric. I generally use it in my daily life anyway (except for MPH and *F. I'm not nearly as intuitive with KPH and *C). However, if you ever need to get something machined here, it has to be done in inches. It's a pain, but I can see why. All of the machinists' machines have inch rulings on them. Many machinists' machines have threaded rods that are threaded so that one turn=one inch. (Or something similar.)

It'd be a royal pain to convert to the metric system for machinists.

The funny thing is, I often use both systems of measurements. If something is around an inch or larger than a few inches, I use inches. If something is smaller than an inch I generally use cm until it gets really small, then mm. If something is about 10 cm long I'll use decimeters. Anything between 1-2 feet is feet, but longer than 2 feet is divisions of a meter. I'm a bit nuts.

[IanB]

All large companies that make complex products use custom hardware optimized for each application.  I live only a few minutes drive from the Tektronix campus, but that doesn't make it any easier to find oddball screws.  Unless you happen to find a bag of them at the surplus sale.

On the surface that doesn't make a lot of sense.

If they use standard screw sizes then they have a wider choice of suppliers, can get better prices, can use standard tooling, have less chance of production holdups due alternate sources, etc.

If they use custom screws they have less flexibility, have longer lead times, have more chance of supply shortages, have higher costs due to custom orders, etc.

Spare screws may not be directly available from Tektronix, but I am not sure why they would not conform to a standard diameter, a standard thread size and pattern, and a standard head style?

[miguelvp]

Do you remember those things called multiplication tables, that they used to teach by rote?
...

Make it mindlessly easy/practical to divide crap into finer fractions than just half’s or thirds........
Of course difficulty arises trying to think of imperial measures in metric terms, but nobody working with the metric system proper works in multiples or sub multiples of 25.4mm, feet or yards. I think your argument is full of crap.

Multiplication is one thing, division is another.

What's 3*7 simple one step and easy to do on your feet without a calculator (I hope)
But what is 3/7?

The reason of the base 2 fractions is to make that simple by avoiding complex divisions and one can make them on their feet without a calculator just by dividing with a piece of string (or a math compass) not even knowing the actual length.

Then again, we are confusing math with units, totally unrelated, the math is the same in Imperial than in Metric or any other unit that you can come up with.

In the US we choose a base 2 fractional number instead of a base 10 fractional number as a standard.

Say I have a ruler in inches without divisions and I need to find 3/64 of an inch.
I can easily divide the inch in half giving me 1/2 increments, then I can take that and half it to get 1/4 increments, half it again we get 1/8 ... until 1/64 then I need three lenghts of that.

Now say I have a ruler with centimeters and I need to find 3 millimeters. I need to take that length and divide it by 10. How do you do that geometrically?

I could half the centimeter and get 5 millimeters, how do I split that into 5? I could half it and get 2.5mm half that and get 1.25mm  half it again and get .625mm

But if you task me to find 3/64 of a centimeter, then it will be easy because the base 2 fractions can be done geometrically.

So nope, John's argument is not full of crap. and statements like that don't ever make a point.

[Bored@Work]

But what is 3/7?
...
Say I have a ruler in inches without divisions and I need to find 3/64 of an inch.

Good trick to change the challenge midway, so it becomes one you can solve with your 1/64 fractions.

The only things I take from your fake challenge is for strange reasons you buy shitty rulers without divisions. And you live in a magic workshop where everything always needs to be divided by multiples of 2.

And to answer you original question, those who remember their multiplication tables can also quickly divide 3/7. In our minds we look in the table for 7 for the nearest result less-equal 30. That would be 4*7 = 28. So the result starts with 0.4. Then we look for the nearest result less-equal to 20. That would be 2 * 7 = 14. But we would also recognize that 3*7=21 comes pretty close. So we would say the result is around 0.43 and call it a day if that is accurate enough for the task at hand. Otherwise we would say the result starts with 0.42, and 0.429 would do.

That is two times using the memorized multiplication table, and be done.

[GK]

Multiplication is one thing, division is another.

What's 3*7 simple one step and easy to do on your feet without a calculator (I hope)
But what is 3/7?

Riiight. Well what is 3" / 7 ?

The reason of the base 2 fractions is to make that simple by avoiding complex divisions and one can make them on their feet without a calculator just by dividing with a piece of string (or a math compass) not even knowing the actual length.

Then again, we are confusing math with units, totally unrelated, the math is the same in Imperial than in Metric or any other unit that you can come up with.

In the US we choose a base 2 fractional number instead of a base 10 fractional number as a standard.

Say I have a ruler in inches without divisions and I need to find 3/64 of an inch.
I can easily divide the inch in half giving me 1/2 increments, then I can take that and half it to get 1/4 increments, half it again we get 1/8 ... until 1/64 then I need three lenghts of that.

Now say I have a ruler with centimeters and I need to find 3 millimeters. I need to take that length and divide it by 10. How do you do that geometrically?

I could half the centimeter and get 5 millimeters, how do I split that into 5? I could half it and get 2.5mm half that and get 1.25mm  half it again and get .625mm

But if you task me to find 3/64 of a centimeter, then it will be easy because the base 2 fractions can be done geometrically.

So nope, John's argument is not full of crap. and statements like that don't ever make a point.

I don't know what planet you live on, but I don't have to mark out one tenth of 3mm very often, and I don't need a calculator to work out that 3mm/10 is 0.3mm. I have yet to find a metric ruler with cm graduations that does not also have mm graduations, so marking out 3mm isn't particularly difficult.

3/64 of a cm is 0.46875mm. How do you accurately mark that out by manually and successively dividing down by 2 from a cm?

Now I can figure out off the top of my head that 64/3 =~20, and 10mm (1 cm) / 20 = 0.5mm. About half my metric rulers have 0.5mm graduations, so marking out the best fit to 3/64 of a cm, with about as much accuracy as it could ever be done by hand, would be a absolute doddle. 

[GK]

The only things I take from your fake challenge is for strange reasons you buy shitty rulers without divisions. And you live in a magic workshop where everything always needs to be divided by multiples of 2.

 

[vk6zgo]

All large companies that make complex products use custom hardware optimized for each application.  I live only a few minutes drive from the Tektronix campus, but that doesn't make it any easier to find oddball screws.  Unless you happen to find a bag of them at the surplus sale.

On the surface that doesn't make a lot of sense.

If they use standard screw sizes then they have a wider choice of suppliers, can get better prices, can use standard tooling, have less chance of production holdups due alternate sources, etc.

If they use custom screws they have less flexibility, have longer lead times, have more chance of supply shortages, have higher costs due to custom orders, etc.

Spare screws may not be directly available from Tektronix, but I am not sure why they would not conform to a standard diameter, a standard thread size and pattern, and a standard head style?

So if as it seems,Tek's screws are not even a standard thread size in the USA,as I had assumed,imagine how difficult it is in a country where the only common non-metric small screws are Whitworth & BA.

And it isn't just the fairly large screws used in holding things like the graticule surround on older CROs.
The Tek 651 Picture Monitors used countersunk screws,almost identical in size to a 1/8" Whitworth,but with a totally different thread.

Re Richard's comment:-What reason would they have for using a weird thread?
In what way is that optimised?

OK,other hardware may need to be specially made like brackets,bur Sony uses standard metric thread sizes,in their TV Broadcast equipment,as does NEC.

[GK]

I would like for the US to convert to metric. I generally use it in my daily life anyway (except for MPH and *F. I'm not nearly as intuitive with KPH and *C). However, if you ever need to get something machined here, it has to be done in inches. It's a pain, but I can see why. All of the machinists' machines have inch rulings on them. Many machinists' machines have threaded rods that are threaded so that one turn=one inch. (Or something similar.)

It'd be a royal pain to convert to the metric system for machinists.

Well, now just think of every American engineering company that has to cater for both international and domestic markets to remain viable. They have to absord the operational overheads to produce product to both standards.

[miguelvp]

Multiplication is one thing, division is another.

What's 3*7 simple one step and easy to do on your feet without a calculator (I hope)
But what is 3/7?

Riiight. Well what is 3" / 7 ?

Then again, we are confusing math with units, totally unrelated, the math is the same in Imperial than in Metric or any other unit that you can come up with.

You will never find a tool that is 3/7ths in the US, our tool sizes are always fractions of powers of 2 or said a different way multiple of negative powers of 2.

But I give, negative powers of 2 are just too hard to comprehend, I get it! After all they don't have a key on the calculator to figure out the exponent, like base 10 does. Well they do but you have to hit that inverse key and make things way too complicated.

Math has died in engineering, that's what I get out of all this

[TMM]

Now say I have a ruler with centimeters and I need to find 3 millimeters. I need to take that length and divide it by 10. How do you do that geometrically?

I'd go to the shops and buy a ruler that isn't designed for 5 year olds.

 

[abaxas]

Surely now is the time?

Give it the predicted 10-15 years before the States loses it's current position of dominance and mix in 19th century imperial and surely they have got to be more fooked than a $0.50 prostitute.

[miguelvp]

Now say I have a ruler with centimeters and I need to find 3 millimeters. I need to take that length and divide it by 10. How do you do that geometrically?

I'd go to the shops and buy a ruler that isn't designed for 5 year olds.

 

meh, ok say you have an A4 paper with it's fancy 1/sqrt(2) aspect ratio and I ask you to give me 3/10ths of the height.

Surely if I ask you to give me 3/8ths of it you wouldn't need a ruler or anything. or any fraction of a power of 2 of that length.

fractions of based 2 denominators are intuitive. If you don't see that you might ask a 5 year old they might be able to show it to you how it's done

[GK]

Multiplication is one thing, division is another.

What's 3*7 simple one step and easy to do on your feet without a calculator (I hope)
But what is 3/7?

Riiight. Well what is 3" / 7 ?

Then again, we are confusing math with units, totally unrelated, the math is the same in Imperial than in Metric or any other unit that you can come up with.

You will never find a tool that is 3/7ths in the US, our tool sizes are always fractions of powers of 2 or said a different way multiple of negative powers of 2.

I am aware of this fact. That. was. the. point.

[John Coloccia]

How do you find the 3/8ths point without a ruler?

If there's one takeaway point from this thread it's that there's only 2 or 3 people capable of actually understanding the concept...and we all already know how to do it.

[TMM]

Now say I have a ruler with centimeters and I need to find 3 millimeters. I need to take that length and divide it by 10. How do you do that geometrically?

I'd go to the shops and buy a ruler that isn't designed for 5 year olds.

 

meh, ok say you have an A4 paper with it's fancy 1/sqrt(2) aspect ratio and I ask you to give me 3/10ths of the height.

Surely if I ask you to give me 3/8ths of it you wouldn't need a ruler or anything. or any fraction of a power of 2 of that length.

fractions of based 2 denominators are intuitive. If you don't see that you might ask a 5 year old they might be able to show it to you how it's done

Ok and what is an actual practical application for this other than rebuilding a civilization from scratch with no rulers or measuring technology on hand?

[IanB]

So if as it seems,Tek's screws are not even a standard thread size in the USA,as I had assumed,imagine how difficult it is in a country where the only common non-metric small screws are Whitworth & BA.

And it isn't just the fairly large screws used in holding things like the graticule surround on older CROs.
The Tek 651 Picture Monitors used countersunk screws,almost identical in size to a 1/8" Whitworth,but with a totally different thread.

I haven't been convinced of this yet.

Someone who has a Tek device may like to take some of the screws and try to identify them using this guide: http://www.boltdepot.com/fastener-information/Measuring/Notation.aspx

I would be surprised if they do not have a standard diameter and thread pitch.

[miguelvp]

How do you find the 3/8ths point without a ruler?

If there's one takeaway point from this thread it's that there's only 2 or 3 people capable of actually understanding the concept...and we all already know how to do it.

That and they are also confused on how to split a dozen eggs between 2, 3 or 4 people.

[ajb]

All large companies that make complex products use custom hardware optimized for each application.  I live only a few minutes drive from the Tektronix campus, but that doesn't make it any easier to find oddball screws.  Unless you happen to find a bag of them at the surplus sale.

On the surface that doesn't make a lot of sense.

If they use standard screw sizes then they have a wider choice of suppliers, can get better prices, can use standard tooling, have less chance of production holdups due alternate sources, etc.

If they use custom screws they have less flexibility, have longer lead times, have more chance of supply shortages, have higher costs due to custom orders, etc.

Spare screws may not be directly available from Tektronix, but I am not sure why they would not conform to a standard diameter, a standard thread size and pattern, and a standard head style?

Above a certain volume, the setup cost to run custom screws instead of a standard configuration becomes negligible.  If you can eliminate metal inserts in plastic parts by using a custom thread pattern, or eliminate lock washers with a custom serrated washer head, then for a (relatively) modest upfront tooling charge you can reap large savings in other parts of the production process.  Evaluating those tradeoffs is all part of the systems design and designing for manufacture.

[miguelvp]

Just takes a string (or compass), a straight edge and a square.

[Rigby]

http://www.mathopenref.com/constdividesegment.html

[miguelvp]

That will work too, with the square and the straightedge you can skip the bottom line construction and division to draw the parallel lines that intersect the length to be divided.

Units don't matter, and both methods are more precise than using a ruler.

[TMM]

Units don't matter

QFT.

Now continue convincing us why we would want to use a unit of measurement that is other than base 10 given that is what the international numbering system is.

[SirNick]

Divisibility by 2, 3, 4, and 6 is really only useful when you happen to need to divide by 2, 3, 4, or 6.  John had a good point about /2 and /3 being good eyeball points, but frankly, does it matter if the underlying system produces nice round numbers when dividing by thirds?  That's only the case if the starting dimension is a round multiple anyway.

You can still eyeball a 100cm board into thirds.  Technically the parts will be 33.333cm if there's no kerf and no error, but so what?  Either 1) you have exact dimensions you need to cut, in which case mm should give you plenty of resolution for hand-cut dimensions; or 2) you're making arbitrary modifications and the fact that 33 1/3 doesn't reduce to an integer makes absolutely no difference at all.

Any argument for a concept steeped in tradition, habit, and legacy equipment is going to eventually devolve into a collection of corner cases.  None of this is insurmountable.  But continuing to maintain two standards is just dumb. 

[electr_peter]

Divisibility by 2, 3, 4, and 6 is really only useful when you happen to need to divide by 2, 3, 4, or 6.  John had a good point about /2 and /3 being good eyeball points, but frankly, does it matter if the underlying system produces nice round numbers when dividing by thirds?  That's only the case if the starting dimension is a round multiple anyway.

You can still eyeball a 100cm board into thirds.  Technically the parts will be 33.333cm if there's no kerf and no error, but so what?  Either 1) you have exact dimensions you need to cut, in which case mm should give you plenty of resolution for hand-cut dimensions; or 2) you're making arbitrary modifications and the fact that 33 1/3 doesn't reduce to an integer makes absolutely no difference at all.

Any argument for a concept steeped in tradition, habit, and legacy equipment is going to eventually devolve into a collection of corner cases.  None of this is insurmountable.  But continuing to maintain two standards is just dumb. 

Good points.

I would say there are two general issues:

• personal/local/company (single person or institution) standards for measuring stuff
• international standards for measuring stuff (multiple persons or institutions)

For case 1 ("personal" measurements) measuring system does not matter at all. Important thing is that (only)you understand what these measurement units relate to, because other people doesn't enter in to the picture. As I say, you can measure "from the eye" ("pamatuoti iš akies" in lithuanian) if you like. History has plenty examples of crazy measurent standards in each(!) country in Europe and, unsurprisingly, related issues. Good to know that this struggle is in the past now (well, in more advanced countries that is).

For case 2 (multiple person measurement), common measurement standard (again, does not matter what standard at this point) must be employed to achieve smooth operation of manufacturing, etc. Which standard to choose - SI (metric) or inch/foot/other sillines? Well, practically all the world already both teaches and uses base 10 counting system all throughout, so metric system it is.

People arguing for inch/foot/fractional silliness to be used continually seem to think of case 1 only. Examples given that inch/foot system is "better/easier/more natural/5 year old can do it" are just silly as well.

Usage of "pound" in US as a measurement unit is ambiguous to say the least - there are multiple significantly different in nature cases where "pound" is used. And then someone says "I got 85 pounds for this problem". What does "pound" mean in that sentence? Only person talking knew(he may forget as well) it when he said it. Why use this crazy and uncalled ambiguity? Throw away such system and use proper system with proper definitions.

You use what you (personally) are used to and do not see the irregularity here with the rest of the world. At this point it is like a religion of metrics, keep it to yourself.

[mzzj]

We ARE metric.  1 in = 2.54 cm exactly, which is in turn defined by the speed of light and a resonance of atomic cesium.

Very interesting, I did not know that in regards to speed of light and the inch.

I read a German report where some guy did some research in to the imperial system and he found that 1 inch is NOT exactly 2.54 cm
In his calculations, the inch is 2,54000508001016002032...
So, since there are three zeros behind the 4, we always assumed it is exactly 2.54

Even today US uses two slightly different inches 
"United States retains the 1
/
39.37-metre definition for survey purposes creating a slight difference between the international and US survey inches; the difference is exactly 2 parts in a million, so 1,000,000 international inches is equal to 999,998 US survey inches. This is approximately 1
/
8-inch in a mile."

2 Parts per million were not too big fuss when Imperial units were defined based on SI-system. Except for land surveyors because US coast line would have been in different location!

[mzzj]

We (USA) use the same amps, volts, ohms and watts that everyone else uses. And those are "metric" aren't they?   

You(USA) and your goddamn BTU's (british thermal unit, WTF?)

[IanB]

You(USA) and your goddamn BTU's (british thermal unit, WTF?)

It's how we (Brits) help Americans remember their heritage 

[GK]

Divisibility by 2, 3, 4, and 6 is really only useful when you happen to need to divide by 2, 3, 4, or 6.  John had a good point about /2 and /3 being good eyeball points, but frankly, does it matter if the underlying system produces nice round numbers when dividing by thirds?  That's only the case if the starting dimension is a round multiple anyway.

You can still eyeball a 100cm board into thirds.  Technically the parts will be 33.333cm if there's no kerf and no error, but so what?  Either 1) you have exact dimensions you need to cut, in which case mm should give you plenty of resolution for hand-cut dimensions; or 2) you're making arbitrary modifications and the fact that 33 1/3 doesn't reduce to an integer makes absolutely no difference at all.

John didn't have a point and I don't know why people around here seem to think that you can't get "nice round numbers" dividing by 3 or multiples of 2 with metric measurements. Timber, etc here is sold in multiples of 300mm. Standard size for a sheet of ply, gyprock, sheet steel, etc is 2400x1200.

2400 divides:
/2 = 1200
/3 = 800
/4 = 600
/6 = 400
/8 = 300
/10 = 240
/12 = 200
/16 = 150
/32 = 75
Etc.....

A standard metric 1.5mm drill bit is:
1/2 the diameter of a 3mm bit
1/3  "         "         "  "  4.5mm bit
1/4  "         "         "  "  6mm      "
1/5  "         "         "  "  7.5mm   "
1/6  "         "         "  "  9mm      "
1/7  "         "         "  "  10.5mm "
1/8  "         "         "  "  12mm    "
1/9  "         "         "  "  13.5mm "
1/10"         "         "  "  15mm    "
1/16"         "         "  "  24mm    "
Etc.....


I can't believe just how stupid this thread is getting.

[SirNick]

Easy, tiger.  It is, in fact, easy to visually divide something by half, thirds, or fourths.  Those are proportions that you can see in your mind's eye.  Now whether the unit of measure is in anyway related to that fact is moot, since as both of us are saying, it only helps if the originating dimension is a clean multiple -- and that's not a given regardless of metric or Imperial (or whatever you want to call it.)

[HackedFridgeMagnet]

Let's all at least try to understand the others point of view.
Can I just summarise some of what has been said into what I think are the two main questions.

1. When should we try to achieve standardisation of particular measurement units across different industries or nations? Or conversely when should we leave existing differing measurement systems in place?

2. What are the actually the best units for a particular type of measurement or industry, irrespective of standardisation?



One thing that has come up in this thread is that our metric system, using base10 is not really using the best numbering system.
I am not sure what is the best numbering system but I am quite sure it is not base10. Base10 is just the one that everyone uses. So as a pro-metric person I have to realise that the standards I currently want to inflict upon others aren't necessarily the best either. Tks to John C and MiguelP for pointing this out.
As in most things it is weighing up the costs vs the benefits.

Also if metric is so much better, then why do I find it easier to remember peoples heights in feet and inches?
I never think that that guy is 1650mm or 1850 mm. Nor 165cm or 185cm but I can easily say  6 foot 2 or whatever height 165cm is.
Some of the imperial measurements i find are uncannily useful.

[johansen]

Base 60 is where its at.

[HackedFridgeMagnet]

I'm from Tasmania I'm thinking Base11.

[GK]

Overall a base 10 system of measures is about as intuitive as it gets; it would be utterly ridiculous to attempt to argue for anything else.

[miguelvp]

1. When should we try to achieve standardisation of particular measurement units across different industries or nations? Or conversely when should we leave existing differing measurement systems in place?

Let it be, the US is not going to change systems anytime soon. What is next everyone drive on the right side? use the same money? Everything very Globalized centralized?. Force the French to count normal instead of counting in 20's? for example to say 99 they say "four twenties nineteen", everyone use a single programming language? all CPU'S should use the same instruction set and architecture?

Differences bring different perspectives, to me that is a good thing.

2. What are the actually the best units for a particular type of measurement or industry, irrespective of standardisation?

The best unit is to stick with the your local standards, so you can work with your local peers.

[GK]

How do you find the 3/8ths point without a ruler?

If there's one takeaway point from this thread it's that there's only 2 or 3 people capable of actually understanding the concept...and we all already know how to do it.

That and they are also confused on how to split a dozen eggs between 2, 3 or 4 people.

Says the guy who can't divide 3 by 7 using multiplication tables and thinks the question "what is 3 divided by 7" makes some kind of coherent point for the superiority of a base 2 system over a base 10 system.

[miguelvp]

Better? worse? it makes no difference, after all it's not like any of us uses exact science on a field where "close enough" is the norm and we chopped the difficult part of Maxwel's equations so that we can make things just work thanks to Oliver Heaviside.

So what if your bridge beam is 2 mm shorter or 5/64"? the heat will make it expand and contract anyways so by tolerance it can be good enough.


Edit:

And even the simplified Maxwell-Heaviside equations are simplified even more by Kirchhoff's laws where electronic components must meet three conditions:

1) Current density going into a lumped element minus the current density coming out of the lumped element is 0, meaning that the gain rate of charge over time is 0. So we make the elements follow that rule.

2) Rate of change of the magnetic flux in the lumped element must be 0 in order to make Vab = Integral over ab of E.dl true.

3) Signals in the lumped elements must be Way lower than the speed of light.

All those constrains must be met for Kirchhoff's laws to work over the lumped matter discipline. Maybe if they were not that simplified we could make faster electronics, but it is what it is.

/ Edit

[SirNick]

One thing that has come up in this thread is that our metric system, using base10 is not really using the best numbering system.
I am not sure what is the best numbering system but I am quite sure it is not base10.

I'm partial to third base.

Let it be, the US is not going to change systems anytime soon.

That's probably true.  One thing the US is not good at -- being rational.  Speaking of which....

What is next everyone drive on the right side? use the same money? Everything very Globalized centralized?. Force the French to count normal instead of counting in 20's? for example to say 99 they say "four twenties nineteen", everyone use a single programming language? all CPU'S should use the same instruction set and architecture?

Differences bring different perspectives, to me that is a good thing.

Diverse perspectives are a good thing.  Multiple standards are not.  Particularly measurement standards.

Here's diversity:  Steak or sushi.  Coke or Pepsi (or water.)  English or Portugese.  Good things.

Here's standards:  Blu-ray and HD-DVD.  Beta and VHS.  PAL and NTSC.  Do we get any benefit from having half the world's content in 29.97 fps, and the other half in 25 fps?  (And the switch to HD really missed the opportunity to define that one, for the sake of compatibility.)

[vk6zgo]

Let's all at least try to understand the others point of view.
Can I just summarise some of what has been said into what I think are the two main questions.

1. When should we try to achieve standardisation of particular measurement units across different industries or nations? Or conversely when should we leave existing differing measurement systems in place?

2. What are the actually the best units for a particular type of measurement or industry, irrespective of standardisation?



One thing that has come up in this thread is that our metric system, using base10 is not really using the best numbering system.
I am not sure what is the best numbering system but I am quite sure it is not base10. Base10 is just the one that everyone uses. So as a pro-metric person I have to realise that the standards I currently want to inflict upon others aren't necessarily the best either. Tks to John C and MiguelP for pointing this out.
As in most things it is weighing up the costs vs the benefits.

Also if metric is so much better, then why do I find it easier to remember peoples heights in feet and inches?
I never think that that guy is 1650mm or 1850 mm. Nor 165cm or 185cm but I can easily say  6 foot 2 or whatever height 165cm is.
Some of the imperial measurements i find are uncannily useful.

Actually,I find it quite easy to visualise 1.65 m or 1.85m.
The Metrication Board were so sure that we were too dumb to be able to read a decimal point that they decreed measurements should be in mm--period!
The fact that people have no problem telling the difference between $1.85 & $185.00  didn't seem to occur to them!

The Board really hated cm,but they somehow sneaked back in as a useful measure of people's heights.
I find metres & decimal parts of them,much easier for such things.

[zapta]

The world chose metric. Only the US holds out, and it costs the US money. It costs everyone who sells to the US money. Just stop being so contrary.

If it's too expensive for you to sell in the US then don't. Nobody owe you a thing.

[miguelvp]

Actually,I find it quite easy to visualise 1.65 m or 1.85m.
The Metrication Board were so sure that we were too dumb to be able to read a decimal point that they decreed measurements should be in mm--period!
The fact that people have no problem telling the difference between $1.85 & $185.00  didn't seem to occur to them!

The Board really hated cm,but they somehow sneaked back in as a useful measure of people's heights.
I find metres & decimal parts of them,much easier for such things.

Thank God all the metric system countries use comma for thousands and period for the decimal point.

Make sure you don't order from France a custom part that must measure 3.025 mm width by 2.025 mm height.

Edit: i know tiny but just to make a point

[HackedFridgeMagnet]

Yikes. If 'Moms' are 220kg including jewelry, I'm glad we don't have them in Australia.

[vk6zgo]

Base 60 is where its at.

All your base are belong to us!

[GK]

Better? worse? it makes no difference, after all it's not like any of us uses exact science on a field where "close enough" is the norm and we chopped the difficult part of Maxwel's equations so that we can make things just work thanks to Oliver Heaviside.

So what if your bridge beam is 2 mm shorter or 5/64"? the heat will make it expand and contract anyways so by tolerance it can be good enough.


Edit:

And even the simplified Maxwell-Heaviside equations are simplified even more by Kirchhoff's laws where electronic components must meet three conditions:

1) Current density going into a lumped element minus the current density coming out of the lumped element is 0, meaning that the gain rate of charge over time is 0. So we make the elements follow that rule.

2) Rate of change of the magnetic flux in the lumped element must be 0 in order to make Vab = Integral over ab of E.dl true.

3) Signals in the lumped elements must be Way lower than the speed of light.

All those constrains must be met for Kirchhoff's laws to work over the lumped matter discipline. Maybe if they were not that simplified we could make faster electronics, but it is what it is.

/ Edit

^^^^ This is what happens when english pints are confused for schooners.

[mzzj]

http://themetricsystem.info/whymetric.htm

    •  ice cream measured in pints,
    •  sand in cubic yards,
    •  backpack capacity in cubic inches,
    •  milk in gallons,
    •  strawberries in dry quarts,
    •  detergent in fluid ounces,
    •  laundry baskets in bushels,
    •  refrigerators in cubic feet, and
    •  reservoir capacity in acre-feet.

1/5 cubic feet to dry quarts anyone?   

[PE1RKI]

i often need to convert inches to metric even from euro customers and europe based companies.

us drills makes no sense for me
drill 28 is 0.1405 inch or 3.569mm
drill designated with letter "O" is  0.316 inch or 8.026mm
there is no formula to convert

http://en.wikipedia.org/wiki/Drill_bit_sizes

[Jeroen3]

Just be happy the United people of America decided to use the same time unit the British defined in Greenwich.

[magetoo]

The Metrication Board were so sure that we were too dumb to be able to read a decimal point that they decreed measurements should be in mm--period!
[...]
The Board really hated cm,but they somehow sneaked back in as a useful measure of people's heights.

Is this something that is common to English-speaking countries?  I keep seeing people using millimeters for "human scale" things where they make no sense at all.  Around here (Sweden), centimeters and decimeters are common everyday units, millimeters less so.

[HackedFridgeMagnet]

I cant remember anyone ever talking in Decimetres, Centimetres are quite common though.

In the building industry everything is either mm or metres.

[miguelvp]

Better? worse? it makes no difference, after all it's not like any of us uses exact science on a field where "close enough" is the norm and we chopped the difficult part of Maxwel's equations so that we can make things just work thanks to Oliver Heaviside.

So what if your bridge beam is 2 mm shorter or 5/64"? the heat will make it expand and contract anyways so by tolerance it can be good enough.


Edit:

And even the simplified Maxwell-Heaviside equations are simplified even more by Kirchhoff's laws where electronic components must meet three conditions:

1) Current density going into a lumped element minus the current density coming out of the lumped element is 0, meaning that the gain rate of charge over time is 0. So we make the elements follow that rule.

2) Rate of change of the magnetic flux in the lumped element must be 0 in order to make Vab = Integral over ab of E.dl true.

3) Signals in the lumped elements must be Way lower than the speed of light.

All those constrains must be met for Kirchhoff's laws to work over the lumped matter discipline. Maybe if they were not that simplified we could make faster electronics, but it is what it is.

/ Edit

^^^^ This is what happens when english pints are confused for schooners.

More like when integrals and derivatives are too complicated so Kirchhoff got rid off the gain rate of charge and the rate of  change of magnetic charge  eliminating the integrals and derivatives and turning them to regular algebraic equations that are easy and good enough.
http://en.wikipedia.org/wiki/Kirchhoff%27s_circuit_laws#Limitations

As for Heaviside work here is what he had to say:

I remember my first look at the great treatise of Maxwell's when I was a young man... I saw that it was great, greater and greatest, with prodigious possibilities in its power... I was determined to master the book and set to work. I was very ignorant. I had no knowledge of mathematical analysis (having learned only school algebra and trigonometry which I had largely forgotten) and thus my work was laid out for me. It took me several years before I could understand as much as I possibly could. Then I set Maxwell aside and followed my own course. And I progressed much more quickly... It will be understood that I preach the gospel according to my interpretation of Maxwell.

So we work with approximations of approximations, not a big deal as long as things work.

[GK]

Better? worse? it makes no difference, after all it's not like any of us uses exact science on a field where "close enough" is the norm and we chopped the difficult part of Maxwel's equations so that we can make things just work thanks to Oliver Heaviside.

So what if your bridge beam is 2 mm shorter or 5/64"? the heat will make it expand and contract anyways so by tolerance it can be good enough.


Edit:

And even the simplified Maxwell-Heaviside equations are simplified even more by Kirchhoff's laws where electronic components must meet three conditions:

1) Current density going into a lumped element minus the current density coming out of the lumped element is 0, meaning that the gain rate of charge over time is 0. So we make the elements follow that rule.

2) Rate of change of the magnetic flux in the lumped element must be 0 in order to make Vab = Integral over ab of E.dl true.

3) Signals in the lumped elements must be Way lower than the speed of light.

All those constrains must be met for Kirchhoff's laws to work over the lumped matter discipline. Maybe if they were not that simplified we could make faster electronics, but it is what it is.

/ Edit

^^^^ This is what happens when english pints are confused for schooners.

More like when integrals and derivatives are too complicated so Kirchhoff got rid off the gain rate of charge and the rate of  change of magnetic charge  eliminating the integrals and derivatives and turning them to regular algebraic equations that are easy and good enough.
http://en.wikipedia.org/wiki/Kirchhoff%27s_circuit_laws#Limitations

As for Heaviside work here is what he had to say:

I remember my first look at the great treatise of Maxwell's when I was a young man... I saw that it was great, greater and greatest, with prodigious possibilities in its power... I was determined to master the book and set to work. I was very ignorant. I had no knowledge of mathematical analysis (having learned only school algebra and trigonometry which I had largely forgotten) and thus my work was laid out for me. It took me several years before I could understand as much as I possibly could. Then I set Maxwell aside and followed my own course. And I progressed much more quickly... It will be understood that I preach the gospel according to my interpretation of Maxwell.

So we work with approximations of approximations, not a big deal as long as things work.

Ummmm........ and all of that, whatever it is, follows on from the metric/imperial, base 2/base 10 debate, how, exactly? 

However since you bring up integrals and derivatives for whatever reason; back in 1950 a significant milestone in computer development was placed onto the market. This was in the beginning of the heyday of the analog computer, when digital computers were still too large, expensive and slow for many applications. The machine was called MADDIDA. It was a digital computing machine, but not a conventional one, and was much simpler and cheaper to produce and for a short while it looked like the analog computer was imminently doomed.

MADDIDA was a digital integrating differential analyzer. In other words, a machine hard wired to do integration as its primary function. You constructed analogs of physical systems on this machine just as you did on an analog computer, but being a digital device, the results you ended up with were much more accurate. However, despite the machines potential, MADDIDA as a commercial failure. The reason? In order to get this revolutionary new machine onto the market as quickly as possible, some major short cuts were taken in it's development - the most short sighted being that the design was based on the binary number system, including all of the user I/O. There was no internal decimal conversion. Each digital integrator had to be serially programmed manually using a binary keyboard, and that made it a right royal pain in the arse to use. You would spend most of your time converting your mathematical models to binary format and the then back again the other way to interpret the results.

By the time competing manufacturers brought digital integrating differential analyzers to the market that were designed to operate on the decimal number system it was already too late. Conventional digital computers were continuing to drop in cost and increase in complexity. High level languages meant that the scientist or engineer didn't have to concern himself with 1's and 0's and tedious conversions from base 10 to base 2.
 

[miguelvp]

I brought it up because this whole thread seems to be about precision and how different units and bases causes errors.

But at the end of the day we deal with huge tolerances so it doesn't even matter. We cant deal with very high frequencies using Kirchhoff's laws because they no longer apply. They only apply within those three constrains stated above.

Engineering is the practical side of science and good enough is the norm. So what does it matter if we use inches or mm or base 2 or base 10? we don't live in a precise world.

That MADDIDA system you talk about sounds very interesting. As for base conversion, it's really not that complicated . logarithmic base conversion to any arbitrary base (even an irrational base like e) was covered in High School when I was a kid.
 

[retrolefty]

Each digital integrator had to be serially programmed manually using a binary keyboard,

 You mean they had a keyboard with just two keys? 

[Rigby]

It doesn't matter what people standardize on, just freaking standardize already.  Everyone speak the same measurement system, it doesn't matter what it is.

The world chose metric. Only the US holds out, and it costs the US money. It costs everyone who sells to the US money. Just stop being so contrary.

I wasn't being contrary, I just want the argument to end.  The world chose metric, so the US should choose metric.  My point (which was clearly missed) was that there are no inherent advantages in one measurement system over another, except in refined scopes.  Because the world has chosen metric, the entire world should choose metric, my native country as well.

To your point about it costing money, just the other day we had a $10k quadcopter with $60k worth of cameras on it crash because of a metric -> imperial conversion mixup.  The software didn't specify units, and the person sitting at the console configuring the flight assumed feet.  Well the software wanted meters.  Flew straight into the side of a building and effectively shattered.

I don't know why the software was written that way, I don't know why the operator didn't verify his assumption, I don't know why the quadcopter didn't have any kind of proximity detection, and I don't know why they didn't just use GPS or something to plan the flight.  I just know that it cost us a lot of money.

So, when I say I want everyone to standardize, I really mean that I want everyone to standardize.  And when I say that I don't give a fuck what system is used, I truly don't give a fuck what system is used.  Just fucking standardize, already.

A lot of you say that the US is a victim of it's own hubris, and I agree.  We definitely are.

[lewis]

Thank God all the metric system countries use comma for thousands and period for the decimal point.

Nope!

[SirNick]

I don't know why the software was written that way, I don't know why the operator didn't verify his assumption, I don't know why the quadcopter didn't have any kind of proximity detection, and I don't know why they didn't just use GPS or something to plan the flight.  I just know that it cost us a lot of money.

So, when I say I want everyone to standardize, I really mean that I want everyone to standardize.  And when I say that I don't give a fuck what system is used, I truly don't give a fuck what system is used.  Just fucking standardize, already.

This is a tough one for me.  I agree whole-heartedly with your entire post, but this is rightfully a casualty of human error, not measurement systems.  It could have just as easily happened if the software wanted meters and the user entered cm instead.  There's just no getting around poor user interfaces and careless assumptions.  This one particular incident would've been avoided perhaps, but it would have almost assuredly succumbed to another of equal daftness.

I still agree with everything you said though.   

[corrado33]

i often need to convert inches to metric even from euro customers and europe based companies.

us drills makes no sense for me
drill 28 is 0.1405 inch or 3.569mm
drill designated with letter "O" is  0.316 inch or 8.026mm
there is no formula to convert

http://en.wikipedia.org/wiki/Drill_bit_sizes

I really like drill bit sizes as numbers. Makes it EXTREMELY easy to find the right one, especially when you really get down into the thousands of an inch difference between them. Simple example: "Hmm, now which is bigger a drill bit which is 0.133 or 1/8th an inch?" (That answer is obvious, but it's much easier when the bits are simply numbered in integers.) Sure, it may be a pain to run through your design and convert all of the drill sizes to the numbers, but it makes it easier on the machinist and you, if you choose to work in the lab.

[Zero999]

Thank God all the metric system countries use comma for thousands and period for the decimal point.

Nope!

In most English speaking countries, the full stop character (.) is used as the decimal separator but it most of Europe it's the comma (,). I believe the official standard in the EU is a comma but it hasn't been adopted here, although at a company I used to work at there was some discussion about changing to the European standard.

I believe changing to commas would cause more confusion. At least now, if commas are used, it's normally in a non-English document, so it's obvious.

As far as the metric vs imperial debate is concerned: of course the US should slowly move over to metric because the whole world uses it.

[mzzj]

i often need to convert inches to metric even from euro customers and europe based companies.

us drills makes no sense for me
drill 28 is 0.1405 inch or 3.569mm
drill designated with letter "O" is  0.316 inch or 8.026mm
there is no formula to convert

http://en.wikipedia.org/wiki/Drill_bit_sizes

Dont get me started on US Gauges or thread "standards" 

[lewis]

I really like drill bit sizes as numbers.

Really? That's the most pain in the arse thing in the world!

"Hmm, now which is bigger a drill bit which is 0.133 or 1/8th an inch?"

Which is bigger? 1.2mm or 1.3mm?

Bloody fractions!

[John Coloccia]

re: number/letter sizes

Again, the confusion stems from a complete lack of understanding WHY they are the way they are.  It is rare that you use a number/letter size to drill a nominally sized hole.  They were sized specifically for things like clearance hole sizes, tap drill sizes, etc.  There is no set formula because as the bits get smaller, the increment gets smaller as well.  Why?  Because smaller holes needs more precise tolerances.  Simple.

Once again, it is a system developed for PRACTICAL use, not because machinists were too dumb to know what decimals were.  Do you really want to remember that a 3/32" rivet needs a .098" holes, but a 1/8" rivet needs a .128" hole?  You simply remember that the small rivets are #30 and the big rivets need #40.  If I need a tap drill, do I want to go trudging through a huge index with a magnifying glass and try to pick out the correct one?  No, I want to go to the appropriate index (letter or number) and easily grab the one I want.  Again, practical use in the machine shop, not philosophical rigidness.  It distinguishes between "normal", fractional drill sizes that you use all the time, and the "in between" sizes that you use for certain operations.  See no value in that?  2 or 3 generations of machinists disagreed.

Now will come the long string of yeah buts and other nonsense, but it's helpful to remember that these systems were put in place for very specific reasons, and they work extremely well.  For anyone curious and willing to educate themselves, you may want to research stuff like this a little bit and try to understand WHY things are as they are, why it's useful, and why it has advantages in the setting it was designed to be used.

[GK]

Once again, it is a system developed for PRACTICAL use, not because machinists were too dumb to know what decimals were.  Do you really want to remember that a 3/32" rivet needs a .098" holes, but a 1/8" rivet needs a .128" hole?  You simply remember that the small rivets are #30 and the big rivets need #40.

Why would I need to remember anything? Dimension wise, the rivets I buy are are specified in both diameter and length. A 2mm dia. rivet requires a 2mm dia. hole and a 6mm dia. rivet requires a 6mm dia. hole. Some rivets with looser tolerances (eg pop) come in specified diameters just under standard bit sizes to prevent interference fitting. A 4.8mm dia. rivet gets a 5mm hole. Is this supposed to be rocket science?

I really I don't know how all of us dumbasses in the metricated world actually manage to figure anything out, though some of us is wheely smart and can figure out stuff like dat 2mm is 1/3 of 6mm and 6mm is 1/2 of 12mm (FRACTIONS!) even so our drill bits and stuff aren't even marked as such. Go figure!

[GK]

That MADDIDA system you talk about sounds very interesting. As for base conversion, it's really not that complicated . logarithmic base conversion to any arbitrary base (even an irrational base like e) was covered in High School when I was a kid.

Who said it was complicated? It was long winded and tedious, especially when working with a lot of data.

[John Coloccia]

Once again, it is a system developed for PRACTICAL use, not because machinists were too dumb to know what decimals were.  Do you really want to remember that a 3/32" rivet needs a .098" holes, but a 1/8" rivet needs a .128" hole?  You simply remember that the small rivets are #30 and the big rivets need #40.

Why would I need to remember anything? Dimension wise, the rivets I buy are are specified in both diameter and length. A 2mm dia. rivet requires a 2mm dia. hole and a 6mm dia. rivet requires a 6mm dia. hole. Is this supposed to be rocket science?

I really I don't know how all of us dumbasses in the metricated world actually manage to figure anything out, though some of us is wheely smart and can figure out stuff like dat 2mm is 1/3 of 6mm and 6mm is 1/2 of 12mm (FRACTIONS!) even so our drill bits and stuff aren't even marked as such. Go figure!

Put a 6mm rivet in a 6mm hole, and your airplane will eventually fall apart.  You may want to just memorize #30 and #40 like everyone else. 

[SirNick]

I'm not a machinist, but something tells me there's a good chance that a 2mm rivet would not fit a 2mm hole.  It would fit a 2.1mm hole or larger.  Hence the specialized drill size thing.  (Ensuring the standard clearance requirement for a given fastener.)

That, I have no qualms with whatsoever.  Until such time that #30 is appropriate for a 3/32 rivet in the US, and a 20mm bolt everywhere else.  Then, it's freakin stupid again.

[Br0ski]

Will not happen for a loooooooooong while and here is why...

Having 2 systems generates more revenue by giving the interpreter a job.

That's the simple.

Tool manufacturers are notorious for this then you have to go buy more tools to interpret the type of screw you want to unscrew. What a great marketing strategy.

With the exception of one-way screws (to stop vandalism) none of the different screw types are needed; hex, flat head, torx, allen wrench, the star type... did I miss any?

On another note this is why racism and various language types are also great for the economy. You need to hire an interpreter and you don't have foreigners doing buisness in your country stealing ideas. It's directly related to the war days of old like a built in encryption the enemy cannot decypher. Since we are trying to stop wars this will eventually change but not without real resistance.

Br0ski
@Br0ski47

[GK]

I'm not a machinist, but something tells me there's a good chance that a 2mm rivet would not fit a 2mm hole.  It would fit a 2.1mm hole or larger.  Hence the specialized drill size thing.  (Ensuring the standard clearance requirement for a given fastener.)

That, I have no qualms with whatsoever.  Until such time that #30 is appropriate for a 3/32 rivet in the US, and a 20mm bolt everywhere else.  Then, it's freakin stupid again.

Drill a 2mm hole for a 2mm rivet with a hand drill and a standard twist bit and it will go in no worries.

I actually expanded my previous post before the replies appeared:

"Some rivets with looser tolerances (eg pop) come in specified diameters just under standard bit sizes to prevent interference fitting. A 4.8mm dia. rivet gets a 5mm hole."

The way the diameter is specified is a bit of a grey area and it depends on the type of rivet and the intended use. Sometimes the specified diameter for a metric rivet is the required hole diameter, the actual rivet being slightly smaller. 

 
EDIT:

It seems I need a metric drill set in 0.1mm increments rather than just 0.5. Those 4.8mm fencing rivets should have gotten a 4.9mm hole, not a 5!

http://www.ullrich.co.nz/fastenings/pdfs/57_blind_rivets.pdf

In any case, it's generally just as I said - simply drill a hole one standard bit size (0.1mm) up from the specified diameter of the rivet (see recommend hole sizes to the far right of the rivet dimensions charts).
 

[miguelvp]

Well, the Industrial revolution was done with imperial units. So was the 2nd industrial revolution.
Lot's of inventions later on from the States where all done using the Imperial system.

I'm not saying that it would not have happened if the Brits used the meter instead, but the point is that the Imperial system is not that bad.

here is an article for your perusal, his point is that no one converts from meters to centimeters and that in the metric system 1cm is too small and one meter is too big. Also you can eyeball easier in imperial than in metric and for cooking the imperial system makes a lot of sense, also the Fahrenheit has more granularity than the Centigrade for setting your A/C.

He doesn't mention however that the SI unit for temperature is the Kelvin or that the molar volume is used for volume (not the liter since that is not an SI unit but just one example of molar volume).

So go further and adopt the SI system for all your everyday needs and leave us alone

http://joshuaspodek.com/metric-system-isnt

Edit: and don't forget that all those definitions are at 1 atmosphere at sea level (not considering the crust density btw, or maybe they have a medium crust density in mind) while swinging a dead cat over your head at a rate of 10 sterdians per second singing kumbaya at 3 decibels.

[SeanB]

For a regular low strength pop rivet the drill size is not too important as most are used at low stress. If you are say placing aluminium sheet on an aircraft wing the hole you drill will be typically 0.05mm bigger than the rivet, and you will countersink it, and use a go-nogo gauge on each and every hole, along with a deburring on inside and outside of each hole along with cleaning the hole inside to get it clean and smooth. Then place the rivet and use a mandrel to hold the outside and use a punch to form the head on the inside. On blind holes you use a pop rivet that you set with a defined tension then cut the protruding mandrel off. I got quite good at using solid rivets, in all the assorted colours denoting different strengths. Even made my own set of mandrels to set them. You can even for light duty use just use a small ball peen hammer to set the rivet with a bigger one as a backing anvil.

[vk6zgo]

i often need to convert inches to metric even from euro customers and europe based companies.

us drills makes no sense for me
drill 28 is 0.1405 inch or 3.569mm
drill designated with letter "O" is  0.316 inch or 8.026mm
there is no formula to convert

http://en.wikipedia.org/wiki/Drill_bit_sizes

Like British "BA" screw sizes,& fish hooks!

[PE1RKI]

the drill we choose in metric before tapping a hole is multiply the screwsize by 0.83
no need to remember the drill number.

but everybody can print out a chart and problem solved
http://www.carbidedepot.com/formulas-tap-metric.htm

[Br0ski]

Quote from: rexxar on June 01, 2014, 09:49:59 PM
"There's a reason we're the only country on the planet still using the imperial system. It's because 'merica and freedom or some such.

Really it's because Americans in general are xenophobic. The slightest mention of change seems to send people into fits."

Speaking of fits there was a Silicon Valley Investor who was proposing to split California into 6 different states like the east coast. Makes complete sense both economically and politically (not to mention the number of jobs that would be created to set up the infrastructure). He made a formal proposal and put some money behind it but it went nowhere. America is hell bent on any change (if it ain't broke don't fix it). There have been proposals for over a century, 5 I am aware of, to do just this. Most were to just split it in two, Northern and Southern California. I talked it over with some locals here in Southern California and I got an ear full about how rediculous it was. With all the laws in California over such a great land mass it made perfect sense but everyone is like whatever they just are not hearing it. But they want to complain about this and that law made in Northern California that effected them here in Southern California.

"Never underestimate the power of idiots in large masses."

Br0ski
@Br0ski47

[AndersAnd]

Actually,I find it quite easy to visualise 1.65 m or 1.85m.
The Metrication Board were so sure that we were too dumb to be able to read a decimal point that they decreed measurements should be in mm--period!
The fact that people have no problem telling the difference between $1.85 & $185.00  didn't seem to occur to them!

The Board really hated cm,but they somehow sneaked back in as a useful measure of people's heights.
I find metres & decimal parts of them,much easier for such things.

Thank God all the metric system countries use comma for thousands and period for the decimal point.

Make sure you don't order from France a custom part that must measure 3.025 mm width by 2.025 mm height.

Edit: i know tiny but just to make a point

Yes that can lead to some very unfortunate confusion in some cases.
Most English speaking / imperial countries use a point "." as decimal mark and comma "," as thousands separator, while in most other / metric countries around the world it's the exact opposite with a comma "," as decimal mark and point "." as thousands separator.

http://en.wikipedia.org/wiki/Decimal_mark



Decimal marks:
Point "." – Blue
Comma "," – Green
Eastern Arabic numerals – Red
Data unavailable – Grey

[AlfBaz]

... and comma "," as thousands separator...

I was taught in school some 30+ years ago, not to use commas separating thousands because in some countries it was used as a decimal. We should use a blank space instead eg 100 000

[dannyf]

it probably won't go anywhere

People wouldn't change, until they are facing a crisis, a life-or-death situation. Now, we can live off of what our parents and their parents left for us. In 50 years, when we have exhausted that, we will change.

[HighVoltage]

I was taught in school some 30+ years ago, not to use commas separating thousands because in some countries it was used as a decimal. We should use a blank space instead eg 100 000

That is the case in Germany.
here we use the Komma as a decimal separator.
And it causes lots of confusion in international communication.

[AlfBaz]

This may explain why I've never seen a calculator or numeric keypad with a comma where the decimal point should be

[vk6zgo]

Quote from: rexxar on June 01, 2014, 09:49:59 PM
"There's a reason we're the only country on the planet still using the imperial system. It's because 'merica and freedom or some such.

Really it's because Americans in general are xenophobic. The slightest mention of change seems to send people into fits."

Speaking of fits there was a Silicon Valley Investor who was proposing to split California into 6 different states like the east coast. Makes complete sense both economically and politically (not to mention the number of jobs that would be created to set up the infrastructure). He made a formal proposal and put some money behind it but it went nowhere. America is hell bent on any change (if it ain't broke don't fix it). There have been proposals for over a century, 5 I am aware of, to do just this. Most were to just split it in two, Northern and Southern California. I talked it over with some locals here in Southern California and I got an ear full about how rediculous it was. With all the laws in California over such a great land mass it made perfect sense but everyone is like whatever they just are not hearing it. But they want to complain about this and that law made in Northern California that effected them here in Southern California.

"Never underestimate the power of idiots in large masses."

Br0ski
@Br0ski47

I've never quite understood why you have all those "postage stamp" States in the Eastern USA.

From a Western Australian perspective,California is quite a modest size.

Surely California's problems (if any) come from two things:-

(1) A very large population.

(2)A multiplicity of Local Authorities which seem to have many of the powers which,in other Federal systems, are State responsibilities.

Hence,you have City/Town Police,Fire Dept,Buiding Regulations,etc,as well as County versions of the same,& in some cases, State ones as well.

Get rid of a couple of layers,& your State Legislators may feel more bound to bring in better Laws,as they can no longer blame other layers of Government ( Except for the Feds ,of course! ), when things go wrong!

Splitting into multiple States would not cut down the existing duplication,but would add yet another layer of bureaucracy for any business which needed to operate across the whole of the former State.

[johansen]

splitting some  of these divided states into smaller ones won't really change anything because it is the lawlessness that resulted in the divisions.

fix the lawlessness and most of america's problems will go away.

[miguelvp]

Leave California alone, they homed our only self proclaimed "Emperor of these United States" and "Protector of Mexico".

http://en.wikipedia.org/wiki/Emperor_Norton

[retrolefty]

Hence,you have City/Town Police,Fire Dept,Buiding Regulations,etc,as well as County versions of the same,& in some cases, State ones as well.

 San Francisco is an example that shines upon the insanity. San Francisco's city limits are exactly the same limits as that of San Francisco County. Therefore of course there is a S.F. Police Department and a S.F. County Sheriffs Department. Makes sense, no? 

[Richard Crowley]

San Francisco is an example that shines upon the insanity. San Francisco's city limits are exactly the same limits as that of San Francisco County. Therefore of course there is a S.F. Police Department and a S.F. County Sheriffs Department. Makes sense, no? 

As in most of the US, the Sherrif forces operate the jails and guard the courts and other government properties. No difference in the County of San Francisco.  In large, rural areas, they also provide the law-enforcement forces where there are no city police.

OTOH, up here on Oregon, we love big government so much we have an EXTRA layer ("Metro") with its own taxes and bureaucracies, etc.

[miguelvp]

See the US uses "metro"

kidding aside and getting back to metric instead of going into the metrosexual tangent.

It would be worth it if we all spoke one language, have the same currency and have spring forward on the same day. But that's not going to happen either. My Esperanto is quite rusty (maybe because I never learned it)

About SirNick's mention of:

Here's standards:  Blu-ray and HD-DVD.  Beta and VHS.  PAL and NTSC.  Do we get any benefit from having half the world's content in 29.97 fps, and the other half in 25 fps?  (And the switch to HD really missed the opportunity to define that one, for the sake of compatibility.)

That is called competition and it's a good thing to have different standards and ways to do things.
As for current HD it doesn't matter, it's all digital now, and seems like 1080p and now 4K (4 times 1920x1080) will prevail.

Blue-ray is superior and won over HD-DVD. Beta was superior than VHS and VHS won because of cost, but Beta was used in the video industry.
As far as NTSC vs Pal (and dont forget Secam) well our electricity was/is 60Hz, yours is 50Hz so why did Europe go for 50Hz?

I remember as a kid that the trains going from Spain to France had to change gauge because the width of the tracks were different. Is there a standard for that or are they still different all around the EU and need special trains to cross borders?

Why doesn't Europe use the same electrical plugs and standardize?

Imagine how costly it would be to match the whole European infrastructure, changing to metric will be (at least) a hundred fold of that in the US

[AndersAnd]

Beta was superior than VHS and VHS won because of cost, but Beta was used in the video industry.

There was also Philips/Grundig's Video 2000 standard competing with Betamax and VHS: http://en.wikipedia.org/wiki/Video_2000
However only marketed in Europe, Brazil and Argentina according to Wikipedia. But I remember some from my school having Bang & Olufsen V2000 machines. It was common here since Bang & Olufsen is the only Danish brand.
V2000 could be flipped over like audio cassettes, thus doubling playing time (2x4 hours and 2x8 hours in half-speed mode).

Videotape format war http://en.wikipedia.org/wiki/Videotape_format_war

Legend has it VHS won the format war over Betamax because Sony [Betamax] refused to mass produce pornographic films.

However porn producers later chose HD-DVD over Blu-Ray, and Blu-Ray still won. But I think most watch it online now instead anyways as most don't want to be seen buying/renting porn in stores.

[HighVoltage]

There was also Philips/Grundig's Video 2000 standard competing with Betamax and VHS: http://en.wikipedia.org/wiki/Video_2000

Interesting that you mention this. I just saw a really nice documentary on GRUNDIG and they also explained that the Video 2000 standard was superior to VHS and Betamax. But Mr. Grundig refused to have pornography shown with his system and that was not only the end of Video 2000 but the end of Grundig. They never recovered, not even after Philips took them over.
 

[GeoffS]

I've just been ordering some bits for a 3d printer and on the suppliers web page for some bowden tubing was the description:

>Price is per 100mm aka 10cm, 0.1m, around 4 Inches for those from the USA, Myanmar and Liberia. 0.219 Cubits for those from Ancient Egypt.

 

[magetoo]

I remember as a kid that the trains going from Spain to France had to change gauge because the width of the tracks were different. Is there a standard for that or are they still different all around the EU and need special trains to cross borders?

I believe it's the same across Europe, and in most of the world.  Let's see... According to Wikipedia, Spain and Portugal does indeed use a different gauge.  Apparently there are high-speed lines that use standard gauge though.

No idea if there is any talk about switching completely.

Why doesn't Europe use the same electrical plugs and standardize?

There's no single standard, but we're already reasonably compatible with each other.  There are some variations on what seems to be a common theme; Schuko plugs and outlets seem to be the most widespread, but there are others like the French and Italian standards.  The Euro style plugs are compatible with most sockets.

Maybe we'll standardize completely at some point.  At least now the kind of things people are likely to bring with them to another country works. (Euro plugs for small stuff; IEC leads can be swapped for most other things)


The UK is another story.  (I see that someone has thought of a compatibility solution that could work there too.)

Imagine how costly it would be to match the whole European infrastructure, changing to metric will be (at least) a hundred fold of that in the US

We've been moving slowly in the direction of standardizing all sorts of things, from electrical plugs to the width of toilet paper, over several decades.  Doesn't seem to have done much harm.

(That's not a comment on how things would work in the US though.)

[saturation]

The lesson there too is superior technology isn't the only criteria for success.  Lower production cost is key.

Sony's licenses also raise costs to manufacturers be it memory stick vs SD card, Sony miniDisc versus CD/ mini-CD.

Regarding this thread, its one reason the US uses its own system, the cost to convert rapidly is high.  The process has been by attrition of both the older tools, and the persons used to them.  Globalization also makes metric more attractive and cheaper [e.g. working with China] and hastens the conversion without a political mandate.

Legend has it VHS won the format war over Betamax because Sony [Betamax] refused to mass produce pornographic films.

There is a tiny amount of truth to that. VHS was the cheaper format to produce, and a lot of smaller companies supported it. If you wanted only big budget stuff you could get it on Betamax, but if you wanted older re-releases, B movies, less popular TV series or, yes, porn VHS was the format of choice.

[retrolefty]

Legend has it VHS won the format war over Betamax because Sony [Betamax] refused to mass produce pornographic films.

There is a tiny amount of truth to that. VHS was the cheaper format to produce, and a lot of smaller companies supported it. If you wanted only big budget stuff you could get it on Betamax, but if you wanted older re-releases, B movies, less popular TV series or, yes, porn VHS was the format of choice.

 I lived through the Betamax Vs VHS 'standards war' and payed $700 for my first VHS Panasonic VHS recorder. It's my opinion that PORN had nothing to do with it, it was simply the market results that determined the 'winner', and it was the VHS offering up to 6 hours max recording time over the Betamax 4 hour limit that consumers voted for with their wallets. Those 6 hour recording results were terrible quality but with blank tapes costing $15 at the time the users went with the longer recording capabilities.

[SirNick]

Here's standards:  Blu-ray and HD-DVD.  Beta and VHS.  PAL and NTSC.  Do we get any benefit from having half the world's content in 29.97 fps, and the other half in 25 fps?  (And the switch to HD really missed the opportunity to define that one, for the sake of compatibility.)

That is called competition and it's a good thing to have different standards and ways to do things.

I disagree.  HD-DVD and Blu-ray aren't an example of "competition" in the capitalist sense.  The DVD Forum is a body intended to devise standards that all member companies can abide by to ensure compatibility.  The competition comes from manufacturers implementing the spec in their own players / recorders / etc.  Blu-ray is mostly a Sony product.  So there is some sense of traditional competition there, but competition between media formats isn't a good thing for anybody.  Not the consumers who buy a $1000 player at product launch and then find their format didn't survive, not the production houses that tool up for a format that becomes obsolete before the expense is paid off, not the production houses that invest in both formats to ensure at least one will prevail, not the studios that pick one and lose access to the other half of the consumers, not the studios that pick both and double their inventory and license costs, not the retail outlets that have to dedicate floor space to three copies of the same film (HD, BD, DVD)....  Even the manufacturers suffer from having to invest time, capital, and engineering resources in both, or hedge their bets on just one.

DVD-R and DVD+R was just as bad for a while, but at least now you can afford to be ambivalent to which kind you have.  Why we still have both available is beyond me.  How is that a good thing?

As for current HD it doesn't matter, it's all digital now, and seems like 1080p and now 4K (4 times 1920x1080) will prevail.

It was digital before, too.  You might call it "virtual" -- as in streamed or downloaded.  But saying it's "all" that way is not a safe assumption.  Not everywhere in the world has great broadband.  Even if they do, many places have usage caps.  There's also a lot of that great-for-everyone competition in terms of which service has what content, and what players support it.  Because God forbid we offer video in unmodified MPEG4 files that can be taken from device to device with impunity.  If that happened, people would start using torrents to pirate video!  </sarcasm>

Also:  Online streaming has a vested interest in keeping bandwidth down, which is counter to resolution, and -- more importantly -- less draconian compression.  If we actually switch to 4K (which is just about pointless unless your screen size is equivalent to an actual theater screen, keeping in mind the ratio of screen size to seating distance) you can bet the online providers will compress it to the same approximate bandwidth as current 1080 (or less) streams, undoing any benefit from the increased pixel count and potentially making quality worse except in scenes with little to no temporal activity.

Blue-ray is superior and won over HD-DVD. Beta was superior than VHS and VHS won because of cost, but Beta was used in the video industry.

BD isn't really superior.  The supported codecs are virtually identical (including several competing choices for uncompressed -- i.e., identical -- audio), although BD has a leg up on maximum storage capacity.  Few films seem to push it that far though.  The real difference comes down to authoring.  Java vs... whatever the HD-DVD VM was.

[mariush]

Also:  Online streaming has a vested interest in keeping bandwidth down, which is counter to resolution, and -- more importantly -- less draconian compression.  If we actually switch to 4K (which is just about pointless unless your screen size is equivalent to an actual theater screen, keeping in mind the ratio of screen size to seating distance) you can bet the online providers will compress it to the same approximate bandwidth as current 1080 (or less) streams, undoing any benefit from the increased pixel count and potentially making quality worse except in scenes with little to no temporal activity.

There's quite a lot of changes between the current way HD works and what's planned for 4k content...
One of the changes is for example BT2020, a new color space which works with 10bit or 12bit .. see http://en.wikipedia.org/wiki/Rec._2020   better color space along with better compression will make content more compressible with less artefacts compared to current codecs.

There's also some real time (visually lossless) compression for transferring data between video card and displays in the works from vesa : http://www.vesa.org/wp-content/uploads/2014/04/VESA_DSC-ETP200.pdf
This will make it easier for video cards to support high frame rates at 4k and higher through the classical hdmi/displayport and similar connectors.

[miguelvp]

That is called competition and it's a good thing to have different standards and ways to do things.

I disagree.  HD-DVD and Blu-ray aren't an example of "competition" in the capitalist sense.  The DVD Forum is a body intended to devise standards that all member companies can abide by to ensure compatibility.  The competition comes from manufacturers implementing the spec in their own players / recorders / etc.  Blu-ray is mostly a Sony product.  So there is some sense of traditional competition there, but competition between media formats isn't a good thing for anybody.  Not the consumers who buy a $1000 player at product launch and then find their format didn't survive, not the production houses that tool up for a format that becomes obsolete before the expense is paid off, not the production houses that invest in both formats to ensure at least one will prevail, not the studios that pick one and lose access to the other half of the consumers, not the studios that pick both and double their inventory and license costs, not the retail outlets that have to dedicate floor space to three copies of the same film (HD, BD, DVD)....  Even the manufacturers suffer from having to invest time, capital, and engineering resources in both, or hedge their bets on just one.

Blue Ray was already there, Toshiba and Microsoft among a few others decided to compete and failed.

DVD-R and DVD+R was just as bad for a while, but at least now you can afford to be ambivalent to which kind you have.  Why we still have both available is beyond me.  How is that a good thing?

That had to do with CD/DVD players. They used different lasers that work on regular CDs/DVDs but for playback of recorded media there were incompatibilities.

As for current HD it doesn't matter, it's all digital now, and seems like 1080p and now 4K (4 times 1920x1080) will prevail.

It was digital before, too.  You might call it "virtual" -- as in streamed or downloaded.  But saying it's "all" that way is not a safe assumption.  Not everywhere in the world has great broadband.  Even if they do, many places have usage caps.  There's also a lot of that great-for-everyone competition in terms of which service has what content, and what players support it.  Because God forbid we offer video in unmodified MPEG4 files that can be taken from device to device with impunity.  If that happened, people would start using torrents to pirate video!  </sarcasm>

It was transmitted in analog over the air (NTSC, Pal, Secam and all the variants of those) so TVs had to be compatible with that. Now it's digital over the air.

Also:  Online streaming has a vested interest in keeping bandwidth down, which is counter to resolution, and -- more importantly -- less draconian compression.  If we actually switch to 4K (which is just about pointless unless your screen size is equivalent to an actual theater screen, keeping in mind the ratio of screen size to seating distance) you can bet the online providers will compress it to the same approximate bandwidth as current 1080 (or less) streams, undoing any benefit from the increased pixel count and potentially making quality worse except in scenes with little to no temporal activity.

I remember an Internet where you didn't have to compress video to stream it live, but that's before everyone had access to it (AOL, MSN, Compuserve, etc..).But progress had to happen and allow everyone into the playground.

Now, you can use virtual machines with on demand virtual GPUs that run high end graphics applications that stream the content via H.264 to your home/enterprise without the user even noticing lag. You only need a G-Sync monitor with whatever input systems the application requires.

There is more latency on TVs depending on the settings, than sending a network packet from the US to the UK. I rather promote progress by increasing demand.

BD isn't really superior.  The supported codecs are virtually identical (including several competing choices for uncompressed -- i.e., identical -- audio), although BD has a leg up on maximum storage capacity.  Few films seem to push it that far though.  The real difference comes down to authoring.  Java vs... whatever the HD-DVD VM was.

HD-DVD VM was Microsoft's something.

[SirNick]

DVD-R and DVD+R was just as bad for a while, but at least now you can afford to be ambivalent to which kind you have.  Why we still have both available is beyond me.  How is that a good thing?

That had to do with CD/DVD players. They used different lasers that work on regular CDs/DVDs but for playback of recorded media there were incompatibilities.

Yeah, I remember.  Assuming I haven't mixed up which was which, DVD+R had some weird formatting restraints but was more compatible with existing devices, whereas DVD-R had a less complex format that more closely resembled a manufactured DVD, but wasn't nearly as resilient to minor read errors, and wouldn't work at all in many standalone players.  I may have mixed up one or more of those attributes.

My point though, is why do they both still exist?  Why did they both ever exist?  It would have been better for everybody if the two camps had worked together on the spec of a single format.  It did nobody any favors to have both hanging around... just a lot of consumer confusion.

As for current HD it doesn't matter, it's all digital now, and seems like 1080p and now 4K (4 times 1920x1080) will prevail.
....
It was transmitted in analog over the air (NTSC, Pal, Secam and all the variants of those) so TVs had to be compatible with that. Now it's digital over the air.

I misunderstood where you were going with that.  Apologies.

I remember an Internet where you didn't have to compress video to stream it live, but that's before everyone had access to it (AOL, MSN, Compuserve, etc..).But progress had to happen and allow everyone into the playground.

I don't.  Back in the days of AOL / MSN / CompuServe, common access was via analog modems.  There would be no uncompressed video over a 14.4Kbps dialup circuit!

I rather promote progress by increasing demand.

I like your spirit, but I'm skeptical of the reality.

HD-DVD VM was Microsoft's something.

My point exactly.  ;-)

[miguelvp]

I didn't mean stream via dial up, I used to work at the UofA and we had all the bandwidth we wanted because the Internet was just a few companies, universities and military.

The list of known internet domains fit on the glossary of a book in under 10 pages.
Sending e-mail outside of the internet was a pain, there were many networks other than the internet as well and you needed to know how to route your e-mail right.

[SirNick]

There's quite a lot of changes between the current way HD works and what's planned for 4k content...
One of the changes is for example BT2020, a new color space which works with 10bit or 12bit .. see http://en.wikipedia.org/wiki/Rec._2020   better color space along with better compression will make content more compressible with less artefacts compared to current codecs.

Interesting, I'll have to give that a look.  TBH, I've mostly given 4K a pass.  It doesn't seem to have much practical benefit in the consumer space, just a bunch of marketing BS.  Somewhat like 24bit 192kHz audio...

There's also some real time (visually lossless) compression for transferring data between video card and displays in the works from vesa : http://www.vesa.org/wp-content/uploads/2014/04/VESA_DSC-ETP200.pdf
This will make it easier for video cards to support high frame rates at 4k and higher through the classical hdmi/displayport and similar connectors.

I'm running 3*1920x1080 monitors over a DisplayPort hub at home, and it's not flawless by any stretch.  At those data rates, it's hardly surprising -- in fact, more surprising that it works as well as it does.  But no less annoying when it fails.

I'm not sure how I feel about content compression there, though.  While a lot of desktop video could stand to gain substantially from data reduction algorithms, it would make it similarly easy to mask signal integrity issues until triggered by content / codec inefficiencies.  I expect that kind of deferred failure to manifest itself in a new breed of cheap cables.

[BennyBoy]

What are you going to call a 1/4 watt resistor? a 2500 milliwatt? Thats CRAZY talk! I have no trouble using both. In my Job were always getting metric and standard parts. I grow up with it, so it's no big deal.

Well, I got fed up of using imperial and decided it was time for a change. I started a petition on whitehouse.gov, but it probably won't go anywhere, but thought it would be worth a try. I figured I would post it on this forum, and would appreciate it if you could help and sign it. I know, it is annoying as hell and needs an account 

http://wh.gov/lsq6F

[SirNick]

Metric measurements can be expressed in fractions too.  "1/4 watt" is not an Imperial anomaly.

[HackedFridgeMagnet]

What are you going to call a 1/4 watt resistor? a 2500 milliwatt?

No 250mW maybe.

[mzzj]

Metric measurements can be expressed in fractions too.  "1/4 watt" is not an Imperial anomaly.

6 3/64 BTU resistor that would be 

[echen1024]

Metric measurements can be expressed in fractions too.  "1/4 watt" is not an Imperial anomaly.

6 3/64 BTU resistor that would be 

[AndersAnd]

http://store.theonion.com/p-4960-mens-the-onion-metric-system-t-shirt.aspx


http://www.cartoonstock.com/directory/m/metric.asp

[AlfBaz]

Nothing to do with metric v imperial but my all time favourite has to be...

[Bored@Work]

Nothing to do with metric v imperial but my all time favourite has to be...

Of course this has to do with metric. All values given in cm. Try instead to calculate sqrt(1.18110236 in * 1.18110236 in + 1.57480315 in * 1.57480315 in) in your head. See? Proof that the metric system is superior.

(For the irony challenged: I am using the same stupid proof-by-example technique the fans of imperial units like to use to demonstrate superiority of the imperial system).

[mrflibble]

(For the irony challenged: I am using the same stupid proof-by-example technique the fans of imperial units like to use to demonstrate superiority of the imperial system).

You forgot to include fractions. Because while already superior, those imperial units acquire super powers when combined with fractions.

[Bored@Work]

(For the irony challenged: I am using the same stupid proof-by-example technique the fans of imperial units like to use to demonstrate superiority of the imperial system).

You forgot to include fractions. Because while already superior, those imperial units acquire super powers when combined with fractions.

Oh, I am sorry for that. Of course we don't want to deprive imperial of fraction super powers:

sqrt(75590551/64000000 in * 75590551/64000000 in + 503937008 / 320000000  in * 503937008 / 320000000 in)

[miguelvp]

(For the irony challenged: I am using the same stupid proof-by-example technique the fans of imperial units like to use to demonstrate superiority of the imperial system).

You forgot to include fractions. Because while already superior, those imperial units acquire super powers when combined with fractions.

Oh, I am sorry for that. Of course we don't want to deprive imperial of fraction super powers:

sqrt(75590551/64000000 in * 75590551/64000000 in + 503937008 / 320000000  in * 503937008 / 320000000 in)

Thank you for putting it that way.
Applying the imperial of fraction super powers the result will be 1/0.508 inches exactly

Edit: and I did it on my head, it's a fairly simple problem.

Ooops I forgot I have a decimal point, in fraction it will be 1 123/127 inches.

[miguelvp]

No snarky remarks yet?
You trying to figure it out?

Well let's just say that 508 is 254 times 2. So the only difficult part was to multiply by two and easy to do in your head
The rest is simple too.

[Rigby]

Anything is easy once you know what to do and how to do it.  "Easy" isn't justification for one measurement system over another.  You all are just freewheeling at this point. No traction.

[miguelvp]

True, and that is the point. It's easy because we know what to do and how to do it.

We have to deal with these conversions often so we have more tools than the rest and using them are second nature to us so it's easy.

Just thinking in metric and straight decimal is limiting. We are used to deal with all kinds of different bases and that's not a bad thing specially for engineering.

Regardless, I'm not trying to make fun of the metric system, I'm just saying the Imperial system is as good  and a proven system over the past centuries. we (US) can deal with both and most of us are happy with the ways thing are.

case and point:


What is the message behind statements like that?

Once someone has to resort to insults they already lost the battle

Edit: and BTW logic is boolean and binary by nature not decimal.

[vk6zgo]

The metric system is internally consistent.

To a large extent, the Imperial/Customary system is also internally consistent for lengths below a foot.
All the easily "divisible by 2" stuff gets messed up above that point.

Even the "good bit" is still a bit strange,though!
You'd think that,if we have 12" to a foot,the inch would then be divided into 1/12,1/24,1/48,etc,as this would allow a value of 1/3" to be expressed as 4/12" & so on---that would seem to follow a pattern which was "sort of" consistent with "3 feet make a yard".

Interestingly,I remember having a ruler which had  "sixteenths" on one edge & "twelfths" on the other.
On the back,it had "tenths".

Why do we use "tenths" of an inch,anyway?------Why not 3/32"?

[AlfBaz]

The metric system has it's quirks. Base unit for mass is 1 kilogram, why isn't it a gram? A kilogram seems "derived" to me

[miguelvp]

The metric system has it's quirks. Base unit for mass is 1 kilogram, why isn't it a gram? A kilogram seems "derived" to me

Originally was defined as the weight of a liter of water

A liter was defined as 1^3 decimeter of water.

All of course at sea level in the right place on earth and at one atmosphere.

[vk6zgo]

I think it's an artefact of the SI system------ there were c.g.s. & m.k.s systems in the past,SI seems to be a hybrid.

If I remember correctly,the m.k.s system was the one featured of the back of Oz "Exercise Books" when I was a kid in the 1950s,along with all the Imperial stuff like "Rod,Pole,or Perch",& "Bushel",& the little rhyme:- "A pint of clear water weighs a pound & a quarter".

They redefined the "pint" in the late '50s so the rhyme ended up a little bit "out".

[Bored@Work]

No snarky remarks yet?
You trying to figure it out?

Actually no, I am not trying to figure out EXACTLY how YOU did it. But what gives it away is that you arrived at the exact result, while the values I gave in inch were rounded. I.e. you didn't work with the rounded values or you did some fortunate rounding in your favor during your calculation, probably because you used knowledge of what the result should be. Maybe even just starting with 5/2.54

[vk6zgo]

The metric system has it's quirks. Base unit for mass is 1 kilogram, why isn't it a gram? A kilogram seems "derived" to me

Originally was defined as the weight of a liter of water

A liter was defined as 1^3 decimeter of water.

All of course at sea level in the right place on earth and at one atmosphere.

Yes,it had defined conditions under which it was valid

Just like the Gallon,(of course it was the inverse relationship in that case).

That was for the real Gallon--------I don't know about your wimpy "Gallonettes".

[IanB]

They redefined the "pint" in the late '50s so the rhyme ended up a little bit "out".

Only a tiny bit though. For practical purposes, a gallon of water at 60°F still weighs 10 lb the same as it always did.

[IanB]

The metric system has it's quirks. Base unit for mass is 1 kilogram, why isn't it a gram? A kilogram seems "derived" to me

The use of the kilogram as the base unit for mass makes the system internally consistent.

For instance a force of 1 newton applied to a mass of 1 kilogram produces an acceleration of 1 meter per second per second.

A force of 1 newton acting over a distance of 1 metre does 1 joule of work.

A voltage of 1 volt driving a current of 1 amp for a period of 1 second also does 1 joule of work.

So the kilogram can be related to the volt.

[vk6zgo]

They redefined the "pint" in the late '50s so the rhyme ended up a little bit "out".

Only a tiny bit though. For practical purposes, a gallon of water at 60°F still weighs 10 lb the same as it always did.

Only real Gallons,though,--------  "Gallonettes" are only 8 lb!

[miguelvp]

No snarky remarks yet?
You trying to figure it out?

Actually no, I am not trying to figure out EXACTLY how YOU did it. But what gives it away is that you arrived at the exact result, while the values I gave in inch were rounded. I.e. you didn't work with the rounded values or you did some fortunate rounding in your favor during your calculation, probably because you used knowledge of what the result should be. Maybe even just starting with 5/2.54

Of course I took the original problem and did it in my mind. No reason to complicate things by converting before doing the math, that's just plain silly

Edit: no need to know the final result, here it is. hypotenuse is the square root of the sum of the squares so 5cm. now why divide 5 by 2.54? so 2.54/5 is the same as (2.54/10)*2 so double it and shift and invert the result so 1/0.508

for the fraction is even simpler 1/0.508 is the same as  1000 / 508 and the same as 250/127 (127 being prime) so  250 - 127 is 123 and therefore 1(127/127) + 123/127 inches no need for calculators or anything. But having to deal with weird conversions and supplying results as fractions in school allows you to think in different ways expanding our math tool set.

[lewis]

Here's an excellent video about the origin of the kilogram and why it is the only SI unit to use a metric prefix.

[miguelvp]

I remember watching that video.

That's the most expensive implant known to mankind.

Edit: The surprising thing is that a kilogram is no longer the weight of a liter of water since that will weight 999.975 grams, so 25 milligrams short of a kilogram, I guess they will have to redefine the liter and therefore the meter and of course the speed of light because the meter changed.

Edit2: more alarming is that if the meter changes then the Avogadro's constant will change since it's based in volume, so the kilogram is no longer a kilogram 
Hopefully this million Euro sphere and a thousand wo/man hours doesn't go through, they might even change the value of an inch.

[lewis]

I remember watching that video.

That's the most expensive implant known to mankind.

Edit: The surprising thing is that a kilogram is no longer the weight of a liter of water since that will weight 999.975 grams, so 25 milligrams short of a kilogram, I guess they will have to redefine the liter and therefore the meter and of course the speed of light because the meter changed.

Edit2: more alarming is that if the meter changes then the Avogadro's constant will change since it's based in volume, so the kilogram is no longer a kilogram 
Hopefully this million Euro sphere and a thousand wo/man hours doesn't go through, they might even change the value of an inch.

A litre is defined as the volume of a cube of 10cm sides, and the metre is defined as the length of the path travelled by light in vacuum during a time interval of 1/299792458th of a second (talk about an imperial fraction!). So redefining the kg will not affect the metre or litre. But it will affect the Candela, Mole and Ampere

[miguelvp]

Redefinitions are inevitable as our measuring systems get better. It isn't a problem, for most purposes the initial accuracy was fine. The key ones like length and force are now extremely precisely defined, and things like the weight of a litre of water will change with altitude anyway so anyone using that measurement will need to be aware that it will rarely be exactly 1Kg anyway.

But that is not the case, is it? They just tried to get something other than the definition (making it an approximation) and make it a standard. Not because we can measure better or because there was some error on thee definition, but because they want a new physical object that doesn't change over time (although we don't know if the silicon sphere will keep it's weight).

We've been able to measure 25 mg long time ago. Nothing to do with measuring things better.

[mamalala]

Edit2: more alarming is that if the meter changes then the Avogadro's constant will change since it's based in volume, so the kilogram is no longer a kilogram 
Hopefully this million Euro sphere and a thousand wo/man hours doesn't go through, they might even change the value of an inch.

Just that the project behind that sphere is all about Avogadro:

http://en.wikipedia.org/wiki/Kilogram#Avogadro_project

The whole idea of the planned redefinition of the kg is in fact to be able to derive it from a fundamental physical constant, like what happened with the meter, which is tied to the speed of light in a vacuum.

Greetings,

Chris

[miguelvp]

Edit2: more alarming is that if the meter changes then the Avogadro's constant will change since it's based in volume, so the kilogram is no longer a kilogram 
Hopefully this million Euro sphere and a thousand wo/man hours doesn't go through, they might even change the value of an inch.

Just that the project behind that sphere is all about Avogadro:

http://en.wikipedia.org/wiki/Kilogram#Avogadro_project

The whole idea of the planned redefinition of the kg is in fact to be able to derive it from a fundamental physical constant, like what happened with the meter, which is tied to the speed of light in a vacuum.

Greetings,

Chris

I do get what they are trying to do define the kilogram based on a constant derived from a 1 Mol which in turn is defined by 12 grams of C12, so they are going to change the kilo and the constant.

As far as light yeah it does define the meter now, but it was measured in meters, but that would change since the moles in a cubic decimeter of water won't hold the same amount of molecules.

And of course the Mole is not an SI unit so let's define the Kilo based on a non SI concept. Very strange.

What's next the centigrade, it was all based on water and hydrogen at the beginning now its all a big mess of weird constants and values.

For example the Mole, it was 1 gram of Hydrogen, but later 12 grams of Carbon 12, now it's probably not even based in mass since it's going to be Si the one that defines mass.

[IanB]

I do get what they are trying to do define the kilogram based on a constant derived from a 1 Mol which in turn is defined by 12 grams of C12, so they are going to change the kilo and the constant.

As far as light yeah it does define the meter now, but it was measured in meters, but that would change since the moles in a cubic decimeter of water won't hold the same amount of molecules.

And of course the Mole is not an SI unit so let's define the Kilo based on a non SI concept. Very strange.

What's next the centigrade, it was all based on water and hydrogen at the beginning now its all a big mess of weird constants and values.

For example the Mole, it was 1 gram of Hydrogen, but later 12 grams of Carbon 12, now it's probably not even based in mass since it's going to be Si the one that defines mass.

I don't think you are quite on base with some of your comments in this thread. Like for instance, yes the mol is an SI unit. It is a measure of "amount of substance" and it is independent of mass.

[lewis]

and it is independent of mass.

And volume!

[miguelvp]

Sorry, I meant not metric got my metric and SI mixed.

And not independent of mass, the mole is defined as exactly 12 grams of C12.
and the Avogadro's constant is based on the number of molecules in there.

[miguelvp]

and it is independent of mass.

And volume!

Not anymore if it does define the kilo base on the volume of a sphere.

It doesn't really matter anyways, the math will make things work out regardless of the use of arbitrary values that keep on changing.

[IanB]

And not independent of mass, the mole is defined as exactly 12 grams of C12.
and the Avogadro's constant is based on the number of molecules in there.

Not quite. The mole is defined as the same number of elementary particles of any substance as there carbon atoms in exactly 12 grams of the carbon-12 isotope of carbon. This makes the mole a number of particles, a measure by counting, where the actual number is Avogadro's number.

You can have, for example, a mole of electrons, which has a charge of approximately 96 485 coulombs.

The use of carbon-12 in the definition is to provide a reference standard, just the same as a mass of 1 kg is the mass of any substance that is exactly the same as the reference kilogram stored in Paris.

But where comparisons of masses involve weighing, to find out if this thing has the same weight as that thing, comparisons of moles involve counting, to find out if this quantity contains the same number of particles as that quantity.

[miguelvp]

I know wikipedia is hardly an authority but it's all good they are redefining the Avogadro's constant as well and break a lot of links. Actually they are attempting to redefine a lot of things as well.

http://en.wikipedia.org/wiki/New_SI_definitions#Mole

So yeah they are proposing to break the link between the Mole and the kilogram as well as the link between the kilogram and the meter. I have to throw away my Encyclopedia of Physics from 1990 by Lerner/Trigg since it's already obsolete and uses 6.0220 x 10 ^ 23, and of course the energy output for conversion from mass to energy changes because the kilogram changes but the speed of light doesn't.

Other units like planck's constant, elementary charge (bye maxwell) and boltzmann's constant are redefined as well.

Edit: btw the ampere and the kelvin will change as well

I guess they already changed things in 2008 and broke the link between the kilo and the meter and adjusted Avogadro's constant then, but they have to take another stab at it.

[max666]

Changing the definition of the kilogram to be the mass of a specific amount of silicon atoms, will not change the metre. While you need to measure the diameter of that sphere, the unit of length you are using for that is irrelevant, because you also need to measure the number of silicon atoms per unit of length to the power of three. So I could measure the diameter in 'EEVblogs' and measure the number density in 1/EEVbolgs³ and it doesn't matter what the EEVblog actually is.

At some point in the future I'm sure the second will be redefined to not being the duration of a certain number of oscillations of a microwave transition of the caesium-133 atom, but being based on a certain number of oscillations of an optical transition of some element. And the reason being that we will be able to measure more accurately and more stable on an optical transition, because some of them are sharper and have inherently less uncertainty.

[miguelvp]

You are right, the link between the meter and the kilogram was broken when the liter was no longer the definition for the kilogram.
Oh well, buy gold and sell it in 2015 for a small profit

[lewis]

The kilogram and metre were separated in 1889 when the definition for the kilogram became the 'mass of the international prototype kilogram (IPK)'. Just like the metre and a physical distance around earth were separated, definitions get more and more precise and stable with time. It's nothing to be scared of. Sure, the odd 'constant' gets changed here and there, but the IPK has been drifting in mass since it was created, so these constants have been changing without anyone noticing anyway. The world hasn't ended, everything is still fine.

[miguelvp]

Even if the definition changed back then, the math was still there so you could create a kilogram easily, because it was all connected to water.

You could derive all the units if you had access to water, mercury and glass, including pressure and distance, regardless if you had access to 1 atmosphere. Now you need a piece of raw material that costs a million Euros and extremely expensive pre calibrated measuring equipment and once you create that kilogram you can't use it to derive other units.

I know the world wont end, but the metric system now has totally isolated units as much as the imperial system.

I also find it funny they call their meetings revolutions. "we all want to change the world" comes to mind.

[jahonen]

It should be remembered that those guys are not making the definitions of an unit for a carpenter to use in his daily job but for the science called metrology. A thing called uncertainty is the key why they do these changes of the definitions.

Primary problem of the artifact definitions like IPK is that physical artifacts are impossible to transfer and compare without moving the physical artifact(s) around, which causes wear of the artifact thus increased uncertainty. So primary goal is to be able to express the kilogram using fundamental constants which can be determined with less uncertainty. One simply can't make a realization of a kilogram with some mercury, glass and water with small enough uncertainty for metrology purposes no matter how good the equipment is. Not so with the silicon ball.

So being able to say that 1 kilogram equals this many Si atoms, can be realized as long as elementary particles do not change their mass, anywhere in the observable universe (probably). Similarly, once there were a prototype of a meter, but then it became impractical when somebody asked "at which point of this marker line it should be measured". So it was necessary to improve the uncertainty. Same applies to the mercury column or some volume of water, it might be precise enough for a casual carpenter but not for modern day metrology scientists to be used as an reference as there are too many involved things which increase the uncertainty. Temperature being one of the affecting factors.

Also, another way to realize a kilogram is the watt-balance method, which is also proposed as an alternative way.

Regards,
Janne

[jahonen]

This Veritasium video explains the whole thing better than me:



Regards,
Janne

[miguelvp]

Was already embeded on the previous page.
At the end of the day the kilogram is going to lose 25 milligrams,  25 ppm might not be a lot but it's not a small number either.

[Shale]

my 2 cents.

Metric is a better system. Much simpler to do anything with, having said that. I was raised on imperial, and when you say something in imperial i know exactly what you are talking about. personally I am not going to put in all the effort to learn the metric system as well as i know imperial. Its just not worth the effort to me even though it is a better system. I will just use a calculator to do the conversion.

[HackedFridgeMagnet]

I had a moment today when some lady said a box was 3 inches by 3 inches.
My brain went foggy for a moment, what's she talking about? Oh now I get it, inches, those things.

[vk6zgo]

I had a moment today when some lady said a box was 3 inches by 3 inches.
My brain went foggy for a moment, what's she talking about? Oh now I get it, inches, those things.

Of course it's impossible for a box to be 3x3", because that would be a square. "Box" implies three dimensions.

We can hardly feel superior,when people on this forum often sloppily refer to a rectangular wave as a "square" wave,when that term  properly only applies to a 1:1  mark/space ratio.

[Rigby]

Well now you have to come up with a better name.

[miguelvp]

To nitpicking, the time dimension has nothing to do with the signal dimension. So it's neither

[HackedFridgeMagnet]

I had a moment today when some lady said a box was 3 inches by 3 inches.
My brain went foggy for a moment, what's she talking about? Oh now I get it, inches, those things.

Of course it's impossible for a box to be 3x3", because that would be a square. "Box" implies three dimensions.

Maybe that's what caused my brain-fade.
So it wasn't the inches, and it wasn't the early onset dementia either.

[vk6zgo]

To nitpicking, the time dimension has nothing to do with the signal dimension. So it's neither

Indeed,but like a lot of terms ,these names have been used for so long that I doubt anyone wants to change them.

There is a difference in the Mathematical Expression for the two waveshapes,with a square wave being a "special case" of a rectangular wave.

Please don't ask me to write out the Expressions,as it has been 30 odd years since I have had to do so.

In answer to Rigby:- the "better name" for a rectangular wave is "rectangular wave"!

[AlfBaz]

Surely pulse or periodic pulse

[Rigby]

It's a square wave.  It's not a "square" it's a "square wave" so the definition of "square" doesn't really apply.

[retrolefty]

It's a square wave.  It's not a "square" it's a "square wave" so the definition of "square" doesn't really apply.

 So a red wave wouldn't essentially be the color red by definition?

[Monkeh]

It's a square wave.  It's not a "square" it's a "square wave" so the definition of "square" doesn't really apply.

And kilobytes are 1024 bytes.


 

[SirNick]

Yeah, and you park on a driveway and drive on a parkway.  Life's a beach, get over it. 

[Rigby]

It's a square wave.  It's not a "square" it's a "square wave" so the definition of "square" doesn't really apply.

 So a red wave wouldn't essentially be the color red by definition?

not necessarily.  things get named weirdly sometimes.

"red menace" used to mean communism.  communism isn't any color, communism is an economic system.

[max666]

And kilobytes are 1024 bytes.

 

Yes, the kilobyte carries some historical baggage, but the International Electrotechnical Commission (IEC) enacted standards for binary prefixes back in 1998 already, specifying the use of kilobyte  to strictly denote 1000 bytes and kibibyte (1 KiB) to denote 1024 bytes. However it took until 2007, that the IEC Standard had been adopted by the IEEE, EU, and NIST.

[miguelvp]

And kilobytes are 1024 bytes.

 

Yes, the kilobyte carries some historical baggage, but the International Electrotechnical Commission (IEC) enacted standards for binary prefixes back in 1998 already, specifying the use of kilobyte  to strictly denote 1000 bytes and kibibyte (1 KiB) to denote 1024 bytes. However it took until 2007, that the IEC Standard had been adopted by the IEEE, EU, and NIST.

When will the madness stop?
a kibibyte?

At least when you type kilobyte on google it converts it to the right 1024 bytes,

Mathematically it's even worse, kilo is 10^3, so a kibi should be 2^3 or 8.

2^10 should probably be called a kibidecbyte. Or better yet a 10dBByte.
Edit, no dB since that implies decimal, 10bBByte as in ten binaryBell Bytes.


So when I buy a 32KB memory module are they going to disable the 32*24 bytes extra? or do they have to sell it to me as a 32.768KB (32,768KB for the non English speaking countries)
a 64KB module then it will be a 65.536KB memory module.

wow, way to f-things up standard committees!

[Rigby]

I think the extra 'b' characters are really gonna dog adoption.  The abbreviations are more easily pronounceable (Bib, Kib, Mib, Gib, Tib, BiB, KiB, MiB, GiB, TiB, etc.) but the whole system sucks.  Bollocks to all of it.  I measure data in nibbles, only. Also, words only and octets only.  Only!

[max666]

When will the madness stop?
a kibibyte?

At least when you type kilobyte on google it converts it to the right 1024 bytes,

...

wow, way to f-things up standard committees!

At least when you type kilobyte on wolframalpha it converts it to the right 1000 bytes,
http://www.wolframalpha.com/input/?i=kilobyte

And I disagree that the standard committees did "f-things up" here, they resolved an ambiguous use of the 'k' prefix, which is already reserved for the kilo.

[JoeN]

The only thing I don't get is that one of the most important measurements is still in nothing like decimal form, anywhere in the world.

One year is 365 days, except for every fourth year which is 366 days.  One day is 24 hours (except when a leap second is necessary - last one was June 30, 2012 at 23:59:60 GMT).  An hour is 60 minutes.  A minute is 60 seconds.  Let's metric-ize time!  How about the number of seconds since Newton was born and just keep it simple?

[miguelvp]

The only thing I don't get is that one of the most important measurements is still in nothing like decimal form, anywhere in the world.

One year is 365 days, except for every fourth year which is 366 days.  One day is 24 hours (except when a leap second is necessary - last one was June 30, 2012 at 23:59:60 GMT).  An hour is 60 minutes.  A minute is 60 seconds.  Let's metric-ize time!  How about the number of seconds since Newton was born and just keep it simple?

Buried on page 7:

Thank God the French republican calendar didn't go through and was abolished by Napoleon.

Imagine 12 months 30 days each, each month with 3 decades, rest was on the last day of the decade (1 day out of 10) at the end of the year having 5 days (6 on leap years) of revolution celebratory days that didn't belong to any month. and a 10h/day clock with 100 minutes per hour and one hundred seconds (probably not SI seconds).



http://en.wikipedia.org/wiki/French_Republican_Calendar#Decimal_time

[Jeroen3]

Yes, time isn't decimal. But that isn't the worst thing.
The months are, 30 days, 31 days, then 28/29 days.

Unfortunately this is because the moon decided to do one cycle every 27.3217 earth rotations.
Which happens to be 365.242187 rotations per orbit round the sun.
Don't forget the stability errors in the orbit and rotation.

And then there are timezones:

[jahonen]

Problem is that astronomical time is drifting like crazy: http://en.wikipedia.org/wiki/Leap_second

Keeping atomic time in sync with astronomical time is a PITA. They propose to stop keeping UTC in sync with the astronomical time, due to problems with various systems with leap second insertion.

Regards,
Janne

[max666]

And I disagree that the standard committees did "f-things up" here, they resolved an ambiguous use of the 'k' prefix, which is already reserved for the kilo.

The problem is they tried to change kilobyte, instead of making kibibyte = 1000.

To stifle this madness I always use kibibyte = 1000 and kilobyte = 1024. Hopefully the confusion this creates will stop the breaking change.

I take it change must be very hard for you? 

No worries, you go ahead an do what you want. I still think changes for the better will stick, while changes for the worse get abandoned in the long run. And if you think, doing things solely because we've always done them this way, is a good reason, then fine let's see what happens. I think doing things so that they are simpler and less confusing is a better reason for doing something.

[SirNick]

And then there are timezones:

And the grand-daddy of pointless nonsense:  Daylight saving.  The sole reason my bathroom clock is an hour off half the year.  And I never really remember which half that is.  Basically, I can only count on the fact that it's approximately so many minutes past whatever hour it is.

[retrolefty]

And then there are timezones:

And the grand-daddy of pointless nonsense:  Daylight saving.  The sole reason my bathroom clock is an hour off half the year.  And I never really remember which half that is.  Basically, I can only count on the fact that it's approximately so many minutes past whatever hour it is.

Well I'm happily retired so my minimum time slice is seasonal. 

[Monkeh]

And I disagree that the standard committees did "f-things up" here, they resolved an ambiguous use of the 'k' prefix, which is already reserved for the kilo.

The problem is they tried to change kilobyte, instead of making kibibyte = 1000.

To stifle this madness I always use kibibyte = 1000 and kilobyte = 1024. Hopefully the confusion this creates will stop the breaking change.

I will now sell sugar by the 960-gram kilogram. Because it's sugar, so it's okay for 'kilo' to mean something else.

[lewis]

I will now sell sugar by the 960-gram kilogram. Because it's sugar, so it's okay for 'kilo' to mean something else.

Kilograms Troy?

[mamalala]

I take it change must be very hard for you? 

Yes, actually. I have vast amounts of code, technical documents, books and other literature that uses kilobyte = 1024. Changing it all is a massive amount of work and in many cases impossible.

All they had to do was make kibibyte the new 1000 byte unit and most people would have been unaffected.

But here's the thing: code, technical documents, books and other literature are not static and will get updated all the time. That means that the mistake of using the kilo prefix to denote 1024 will slowly phase out and be corrected. It really makes sense to have prefixes always mean the same, otherwise they become useless, since everyone can make up arbitrary, different meanings for a given prefix.

Oh, and harddisk manufacturers are using the kilo = 1.000, mega = 1.000.000, etc. for many years by now.

Greetings,

Chris

[JoeN]

Oh, and harddisk manufacturers are using the kilo = 1.000, mega = 1.000.000, etc. for many years by now.

Works for me.  The binary-friendly prefixes only really make sense if the device causes you to ask yourself "how many address lines do I need for this sucker" - that it is a directly addressed thing with A0...Ann lines, i.e. it's a problem for electrical engineers and low-level programmers.  For everyone else, the decimal-friendly prefixes usually make more sense.  Even for engineers, since a 1 Mb/s data rate, for example, is probably, in the end, based on a crystal somewhere in the system operating at a integer multiple frequency of 1,000,000, not 1,048,576.

[mamalala]

Are you volunteering to update all my old documentation and firmware for free? How about editing and re-printing my old books?

Why should i? You are either going to update it yourself at some point, or being forced by your employer/customer to update it, or just live with the status quo.

It wasn't a mistake, it was a result of computers being binary. Having 1.024 kibibytes of memory would have be annoying so a more sensible, round unit (kilobyte) was picked.

Now you are confusing the concept of a unit prefix with what it represents. <edit for clarity> 1 something-like-kilo-bytes can be 2.349 bytes, depending on what one "something" represents <end edit>. But it can also be one KiB, which would mean 1024 bytes. The prefix "kilo" was in used way before the digital age, and always denoted 1000. Similar for mega and giga, etc. That it was used to denote 1024 for digital stuff was simply a mistake that should never have happened in the first place.

Again, either you have unit prefixes that have a specific meaning, or you don't. In the former case, no one has to guess what is meant. In the latter case, the use of standard prefixes is futile. After all, what's stopping me from saying "one dish of spaghetti has to be made of 1 kilospaghetties, meaning 1654 single spaghettis"?

Oh, and for what it's worth: I also think of 1024 when i hear "kilo" in connection with digital stuff. But that's simply a slowly dying remnant of how i was trained. When it comes to documenting new stuff, or reworking existing stuff, i will use the proper unit prefixes now. But i still can't be bothered to work through all my old stuff without someone requiring (and thus paying) for it.

Greetings,

Chris

[miguelvp]

Awesome, the sooner they forget our archaic ways the better

They will be trying to figure out how to address some memory by doing base conversion with natural logarithms, while we'll be running circles around them at our old age.

[Richard Crowley]

Hard drive manufacturers use the Kilo, Mega, and Giga confusion to their advantage.  Unlike everyone else in the digital/computing sector, when they say "1 GB"  they actually mean 1,000,000,000, not 1,073,741,824.  It makes the drives sound like they have "more capacity". Like pricing things at 999.99, it sounds less than 1000

[miguelvp]

Yes and we have known that for quite a while, same as the old pixel count for cameras where they will count each color sensor as a pixel, so pretty much scaling it by a factor of 3, and of course using 1,000,000 for a MegaPixel count.

So now they will have to figure out how to change the 4K format for TVs because 1920x1080 has no relation to 1K of anything so 3840x2160 is 4K 

The thing is that all those K definitions don't confuse me at all.

Edit: Although I recall someone talking about a datasheet where they used 1024 as the definition of an RTC clock frequency, so they stated the crystal was oscillating at 32KHz instead of  32.768KHz. So for people that are not aware of your standard RTC frequency I can see that throwing them off.

[Richard Crowley]

2K resolution is a generic term for display devices or content having horizontal resolution on the order of 2,000 pixels.[1]

In the movie projection industry, Digital Cinema Initiatives is the dominant standard for 2K output. In the digital film production chain, a resolution of 2048x1556 is often used for acquiring "open gate" or anamorphic input material, a resolution based on the historical resolution of scanned super 35mm film.[2]

In television, the top-end 1080p high-definition television format qualifies as 2K resolution, having a horizontal resolution of 1920 pixels, with a vertical resolution of 1080 pixels.
http://en.wikipedia.org/wiki/2K_resolution

[SirNick]

I recall someone talking about a datasheet where they used 1024 as the definition of an RTC clock frequency, so they stated the crystal was oscillating at 32KHz instead of  32.768KHz. So for people that are not aware of your standard RTC frequency I can see that throwing them off.

I was complaining about that with one of the LPC datasheets, as an example of how they're generally not as good as, e.g., AVR.  I actually searched the document for "768".  Not one single reference.  It was all "32kHz".  The omission was so consistent and blatant, I began to wonder if they were really using a divide-by-1000 scheme.  (After all, unless you're using oldskool logic ICs, does it really matter how many oscillator clock ticks are in one second?)  But no, it was just lazy technical writing, which should be a contradiction in terms.

[jahonen]

That genius who conveniently redefined on a whim a established scientific 10^3 kilo probably didn't think it enough and now we have to deal with the situation where one can't be always sure which one is implied. If one sees a computer program indicating a transfer rate in kB/s, I certainly can't say for sure whether it is 1024 or 1000-based. But if it says kiB/s, then it is clear which one is intended.

Along these lines of thoughts if we expand this thinking that one can redefine any units where convenient, should one use 2^-10 instead of 0.001 on a whim? Or 2^-20 instead of "micro"? So for example thou would be 1/1024 of an inch, so one could conveniently write 0.004 to indicate one thou in base-16 hexadecimal notation instead of that decimal rubbish?

Regards,
Janne

[JoeN]

That genius who conveniently redefined on a whim a established scientific 10^3 kilo probably didn't think it enough and now we have to deal with the situation where one can't be always sure which one is implied. If one sees a computer program indicating a transfer rate in kB/s, I certainly can't say for sure whether it is 1024 or 1000-based. But if it says kiB/s, then it is clear which one is intended.

Along these lines of thoughts if we expand this thinking that one can redefine any units where convenient, should one use 2^-10 instead of 0.001 on a whim? Or 2^-20 instead of "micro"? So for example thou would be 1/1024 of an inch, so one could conveniently write 0.004 to indicate one thou in base-16 hexadecimal notation instead of that decimal rubbish?

Regards,
Janne

This is SI now.  You are only seeing ambiguities because that is what you want to see.  But, there are no ambiguities, except in very old documents, and even there it takes very little work to resolve which numbering system they are referring to.

[Monkeh]

including the two most popular operating systems in the world (Windows and Linux) use kB = 1024.

No, no we don't. We use SI units. Compatibility options are available for the backwards. A few tools are still broken.

Code: [Select]

alex@terra ~/.stuff $ dd if=/dev/zero of=testfile3 bs=1000 count=100
100+0 records in
100+0 records out
100000 bytes (100 kB) copied, 0.000730059 s, 137 MB/s
alex@terra ~/.stuff $ dd if=/dev/zero of=testfile4 bs=1024 count=100
100+0 records in
100+0 records out
102400 bytes (102 kB) copied, 0.000892386 s, 115 MB/s

N  and  BYTES  may  be  followed  by  the following multiplicative suffixes: c =1, w =2, b =512, kB =1000, K =1024, MB =1000*1000, M =1024*1024, xM =M GB =1000*1000*1000, G =1024*1024*1024, and so on for T, P, E, Z, Y.

Code: [Select]

[    0.583420] ata1.00: ATA-8: WDC WD5000AAKX-083CA0, 15.01H15, max UDMA/100
[    0.583539] ata1.00: 976773168 sectors, multi 0: LBA48 NCQ (depth 31/32), AA
[    0.584791] ata1.00: configured for UDMA/100
[    0.585027] scsi 0:0:0:0: Direct-Access     ATA      WDC WD5000AAKX-0 15.0 PQ: 0 ANSI: 5
[    0.585462] sd 0:0:0:0: [sda] 976773168 512-byte logical blocks: (500 GB/465 GiB)
[    0.585544] sd 0:0:0:0: Attached scsi generic sg0 type 0
[    0.585878] sd 0:0:0:0: [sda] Write Protect is off
[    0.585995] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
[    0.586038] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[    0.611745]  sda: sda1 sda2 sda3 sda4 < sda5 >
[    0.612393] sd 0:0:0:0: [sda] Attached SCSI disk


[miguelvp]

That genius who conveniently redefined on a whim a established scientific 10^3 kilo probably didn't think it enough and now we have to deal with the situation where one can't be always sure which one is implied. If one sees a computer program indicating a transfer rate in kB/s, I certainly can't say for sure whether it is 1024 or 1000-based. But if it says kiB/s, then it is clear which one is intended.

Along these lines of thoughts if we expand this thinking that one can redefine any units where convenient, should one use 2^-10 instead of 0.001 on a whim? Or 2^-20 instead of "micro"? So for example thou would be 1/1024 of an inch, so one could conveniently write 0.004 to indicate one thou in base-16 hexadecimal notation instead of that decimal rubbish?

Regards,
Janne

This is SI now.  You are only seeing ambiguities because that is what you want to see.  But, there are no ambiguities, except in very old documents, and even there it takes very little work to resolve which numbering system they are referring to.

But the SI system doesn't get it. A kilo represents three orders of magnitude bigger than one unit, so a kilogram is three orders of magnitude bigger than one gram.
So a kilobit will be 2^3 or 8, meaning that b1000 is three orders of magnitude bigger than 1 bit (b0001).
A kilobyte will be 256^3 meaning that 01 00 00 00 base 256 is three orders of magnitude bigger than one byte (01 base 256).

If they are going to be all scientific and such at least make it so it's consistent mathematically.

1000 is only a kilo if applied to a base 10 number.

So what we call one Kb should be really b10 Gb or 2*2^9.
and one KB should be really (base 256)04 daB (decabytes) or 4*256^1.

Then the math will be scientifically sound and we could convert units easily

Edit: and the OP has to change his nick to end in 1000 not that 1024 rubbish

[max666]

But the SI system doesn't get it. A kilo represents three orders of magnitude bigger than one unit, so a kilogram is three orders of magnitude bigger than one gram.
So a kilobit will be 2^3 or 8, meaning that b1000 is three orders of magnitude bigger than 1 bit (b0001).
A kilobyte will be 256^3 meaning that 01 00 00 00 base 256 is three orders of magnitude bigger than one byte (01 base 256).

If they are going to be all scientific and such at least make it so it's consistent mathematically.

1000 is only a kilo if applied to a base 10 number.

So what we call one Kb should be really b10 Gb or 2*2^9.
and one KB should be really (base 256)04 daB (decabytes) or 4*256^1.

Then the math will be scientifically sound and we could convert units easily

Edit: and the OP has to change his nick to end in 1000 not that 1024 rubbish

I will let this speak for itself:

A metric prefix or SI prefix is a unit prefix that precedes a basic unit of measure to indicate a decadic multiple or fraction of the unit.

[Monkeh]

No, no we don't. We use SI units. Compatibility options are available for the backwards. A few tools are still broken.

Actually all the examples you gave are GNU tools. If you look at the Linux kernel all the source code the vast majority of it uses the correct kilobyte = 1024. Most boot messages, for example, are correct. It's only the GNU stuff that is broken.

.. no, one of the examples I gave is straight out of the kernel, and another is certainly not a GNU tool. I think you'll find they just haven't gone around and fixed everything yet.

[miguelvp]

I wonder what Linus Torvalds will have to say to attempts in changing the kilobyte definition to 1000, after all he is from Europe so probably an easy sell, right?

[GeoffS]

I wonder what Linus Torvalds will have to say to attempts in changing the kilobyte definition to 1000, after all he is from Europe so probably an easy sell, right?

Whoever suggests that to Linus will probably learn some new words 

[Monkeh]

I wonder what Linus Torvalds will have to say to attempts in changing the kilobyte definition to 1000, after all he is from Europe so probably an easy sell, right?

Whoever suggests that to Linus will probably learn some new words 

Well it's already done in the entire block subsystem.

[miguelvp]

Hmm next time we do our weekly meeting I'm going to use kibibytes, gibibytes and tebibytes.
When the rest of the programmers (including my boss) look at me weird, i'm going to tell them that they are idiots for no adopting.

Hmmm, maybe I'll pass and call things like they've always been (I like to keep my job), plus we never use bytes just Ks, Megs, Gigs. Terabyte might be the only one but is not like we use it often at all colloquially. Once it becomes more common maybe we'll call it Ts or T-Rex-es.

Now, who is going to fix the 4K, 2K, 8K TVs?

[gxti]

There's little reason to suggest power-of-two prefixes for anything that doesn't have actual, physical address lines (SRAM, DRAM, raw flash chips). Hard drives don't have parallel address lines, they have tracks holding a series of bits which may be any length at all. Network cards don't have address lines either and their clocks and transfer rates are powers of ten. Might as well use the same prefixes as every other type of unit. Plus when you start talking about transfer rates, the extra "per second" part throws out any added value of using powers of two. 1MiB/second becomes 17.0666KiB/min and 291.27111 bytes/hour -- if there were 64 seconds in a minute it would make sense, but there's not. Also, mixing decimal values with power-of-two prefixes means that changing prefixes from KB to MB to GB requires doing a divide-by-1024, whereas a power-of-ten prefix would require only moving the decimal point.

It's time to give up on this confusing 1024 vs. 1000 business. I suspect that the reason most people who still prefer the 1024-based units do so because they perceive it as a trick to short-change them for their storage devices. The time to make that argument has long since passed.

[miguelvp]

Where I work all the programmers talk in Ks as in 1024 bytes, Megs as in 1024*1024 and Gigs as in 1024^3 regardless if it's about memory or storage.
We don't sell drives and we are well aware the sizes of those drives.

Our programs do a lot of caching and so does the hard drives themselves, and those buffers are power of 2.
And there is no way i'm going to say the world Kibibyte or its variants to a colleague in a professional setting.

[vk6zgo]

The only reason for the kilobyte kerfuffle is the unfortunate propensity of "computer people" to misuse already well established Engineering terms--------hence "kilobyte"  for 1024 bytes,"bandwidth" for something other than the amount of frequency spectrum occupied by a signal,"CRT" for the whole Monitor instead of the display device,& so on....

Shades of the CB-ers & their "Airwaves","Swarrs",etc!-----"All the good numbers to you Good Buddy"!!!

[mamalala]

The only reason for the kilobyte kerfuffle is the unfortunate propensity of "computer people" to misuse already well established Engineering terms--------hence "kilobyte"  for 1024 bytes,"bandwidth" for something other than the amount of frequency spectrum occupied by a signal,"CRT" for the whole Monitor instead of the display device,& so on....

Shades of the CB-ers & their "Airwaves","Swarrs",etc!-----"All the good numbers to you Good Buddy"!!!

Indeed. And it starts again with "4k" displays, just in the opposite direction, i.e. 4k < 4000 (3840, in this case).

CB? Want some old AM/SSB radios that can easily be modded for 10 meter?

Greetings,

Chris

[JoeN]

It's time to give up on this confusing 1024 vs. 1000 business. I suspect that the reason most people who still prefer the 1024-based units do so because they perceive it as a trick to short-change them for their storage devices. The time to make that argument has long since passed.

I think of it as a trick to short-change us from using our fancy << and >> operations... 

[XOIIO]

Not sure if it will add much, but born in '94 here, I use them pretty much interchangeably I suppose, I don't pay a huge amount, whatever is being used I just go with.

Gallons, liters, miles, kilometers, doesn't make much difference to me (though when I started driving, I had to be sure my racing games were in km/h, I was wondering why real life felt so slow lol.

[mrflibble]

Smaller drives use 512 byte sectors, larger ones use 4k (4096) sectors. Same for SD cards and other types of flash memory. They all use kilobyte multiples for sector sizes. It would be stupid having to refer to them as 4.096 kibibytes.

Indeed. Referring to a 4096 byte sector as "a 4 kibibyte sector" would be far less stupid.

[mariush]

Technically there's no reason hard drives would still use 512 bytes or 4096 bytes as a sector size, I guess it's mostly for backwards compatibility and probably they assume operating systems and file systems have some internal optimizations (using bit masks for clusters or sectors and so on).
Hard drives already have CRC information embedded with a sector or series of sectors, so they already read from drive chunks of data larger than 512 bytes or 4KB:

See   Western Digital’s Advanced Format: The 4K Sector Transition Begins  http://www.anandtech.com/show/2888




SSD drives still use multiples of 2 though internally, but they use the 1000 byte notation to hide some of the space for use with caching, for moving data around, as spare sectors for when one sector becomes unwritable due to too many writes etc etc

There's 2 bits per cell for MLC, 1 bit per cell for SLC, (some of the newer types have 3 or 4 bits per cell) and then you have strings of 32 cells in series, then these are arranged in blocks of 4KB and then blocks are arranged in pages (128 KB per page for example) and that's the smallest erasable unit..

Anantech - The SSD Anthology http://www.anandtech.com/show/2738/5


They could use 500 byte sectors, 4000 bytes etc but I'd best the controller inside (some of them based on ARM) are already optimized and use 8bit, 16bit, 32bit variables and do bit shifts and so on to work and be fast.

[AlanR]

I don't think it will ever happen. Besides, we score consistently low on the international tests in STEM subjects and have not had much improvement at all.  Our secondary education system that deals with STEM careers are overwhelmingly majority filled with foreign students with Americans making up a small percentage of  the total students in these subjects.

Don't get me started on this Common Core garbage!! That will just further dumb us down. Hell, even James Milgram, a MATHEMATICIAN that did the math for NASA space launches, got booted from the Common Core committee responsible for overseeing the subjects in Common Core because he pointed out the severe flaws in the math section of Common Core! In short, America is doomed pretty much...

[max666]

As long as your hardware is binary based, there is nothing wrong with using structures in multiples of 2, because then you're most optimized for a binary system.
What you shouldn't do, and this has been said now many times here, is call something 'kilo' when you actually mean 1024. Instead it has been suggested to use the prefix 'kibi' when you mean 1024.

Now I haven't heard any arguments against the use of 'kibi' meaning 1024, other than:

• It sounds stupid
• I’m not gonna use that when I talk to my friends, I might embarrass me
• I have always used 'kilo' meaning 1024, except when I stand on the bathroom scale, drive around (except in the USA), pee on a high-voltage line, talking about energy content at my weight watchers meetings, buy a resistor, ...

What if we used 'kojo' = 1024, 'mojo' = 1024^2, 'gojo' = 1024^3? would that make you happy? (yes, I’m talking to you mojo-chan )

[miguelvp]

It's not about friends, they wouldn't know kibi from kilo or what binary is other than just 0 and 1.
At a professional level I've never heard kibi come out of anyone's mouth, I'm not going to be the first one since it will just confuse the rest.

As far as I'm concerned we (the work force) set the lingo and kilo, as in 1024, is here to stay as far as programmers go.

[magetoo]

Hmmm, maybe I'll pass and call things like they've always been (I like to keep my job), plus we never use bytes just Ks, Megs, Gigs. Terabyte might be the only one but is not like we use it often at all colloquially. Once it becomes more common maybe we'll call it Ts or T-Rex-es.

This is the exact same thing I came up with.  Nobody outside Wikipedia will ever use the ridiculous "kibi" prefixes.  Going with "kays" "megs" and "gigs" seems like the obvious choice.

[Monkeh]

Now I haven't heard any arguments against the use of 'kibi' meaning 1024, other than:

You forgot the vast amounts of existing documentation and common language use. How much time and effort has already been wasted converting software and texts over? Why try to change the entire world when they could just have done what I did and define kibibyte=1000.

Because what you did is stupid?

[Rigby]

My joke detector is broken, so lemme lay it out.

"kilo" = SI prefix meaning 1000.

"kibi" = binary prefix meaning 1024.

the 'bi' means binary, so you must have a number that is a power of 2 somewhere in there.  for kibi, it's 1024.  since 1024 is close to 1000, and "kilo" is the SI prefix for 1000, they chose "kibi" to mean something close to "a thousand binary" whatevers, in this case, bytes.  Kibibites.

Like I said earlier, one sounds a lot less stupid if they just pronounce the shortened version.  Kib.  That's what I do, and yes, I use it verbally at work.

[vk6zgo]

The only reason for the kilobyte kerfuffle is the unfortunate propensity of "computer people" to misuse already well established Engineering terms--------hence "kilobyte"  for 1024 bytes,"bandwidth" for something other than the amount of frequency spectrum occupied by a signal,"CRT" for the whole Monitor instead of the display device,& so on....

I agree with you about bandwidth, but calling a monitor a CRT is like calling a car a "motor" or a "diesel". Everyone knows it refers to just a component part, but it's the most important part. In the case of monitors it is short of "CRT monitor", to differentiate the older type from newer LCDs.

Only Brits call cars "motors".

I've never heard a car or truck called a "diesel"---locomotives,yes,but that is an exception,as I've never heard anyone call a locomotive a "steam" or "electric".

Things like calling a transistor radio a "transistor",or a CRT type monitor a "CRT" may be OK for the general public,but hardly for anyone with Technical pretensions.

One beginner on this forum was very concerned that a CRT tube was full of lead,after reading a website referring to how much lead there was in a "CRT".
The website was,of course, referring to the lead content in the solder used on the  monitor's PCBs

[SirNick]

Is it OK to call a newer, flat-panel monitor an "LCD"?  That's just the panel, after all...

This one actually bugs the crap outta me -- how about those "LED TVs"?  (To be fair, "LCD backlit with LED arrays" is a bit of a mouthful.)

[echen1024]

Is it OK to call a newer, flat-panel monitor an "LCD"?  That's just the panel, after all...

This one actually bugs the crap outta me -- how about those "LED TVs"?  (To be fair, "LCD backlit with LED arrays" is a bit of a mouthful.)

I've corrected the TV salesman multiple times when we were looking for a new TV. I told him that LED and LCD were the same, only LED is describing how the LCD is backlit. OTOH, CCFL lamps generally perform better than LEDs in terms of color reproduction, which drew me quite a few stares as well when I mentioned that.

[denelec]

Soon we will have OLED TVs.
People will confuse them with LED TVs.

[Rigby]

They ARE LED TVs, they're just organic LEDs.  The display design is ENTIRELY different, but I don't think it's unfair to call them LED TVs.  I think it's wrong to call LED backlit TVs "LED TVs", though.