Python becomes the most popular language

[jfiresto]

... I have witnessed senile drivers get completely mentally stuck in a roundabout, suddenly braking to a complete stop to "give way" for someone entering the roundabout, has happened more than once....

For some German roundabouts – specifically, any that do not have yield signs at the entrances – you are supposed to yield in the roundabout. A local supermarket has such an unsigned roundabout in their parking lot and not everyone remembers the law, still, is yield to the right.

Roundabouts are more naturally suited to driving on the left side of road.

[brucehoult]

For some German roundabouts – specifically, any that do not have yield signs at the entrances – you are supposed to yield in the roundabout.

In roundabouts in Vanuatu vehicles in the roundabout have to give way to vehicles entering the roundabout. This seems exceptionally weird.

Roundabouts are more naturally suited to driving on the left side of road.

I don't understand that. If all rules are mirror-imaged then everything works exactly as well (or not).

[Cerebus]

You also need to trust in only having to look for traffic on the right (on the left in USA/Europe) -- and also of course that traffic in front of you isn't stopped. Still, much much easier than trying to get onto a busy road from a side road or driveway and spot gaps arriving from both left and right at the same time.

That "trust" can be a pitfall, you can't trust other people to be following the rules, as this little video demonstrates:



Although that's an extreme example it's illustrative of a common problem, if you're not prepared for people making poor emerges in front of you, and prepared to hold back if necessary to avoid them, you'll have issues. However, if you pick the correct lane, take good observations on your approach, adjust your speed to fit into oncoming gaps on the roundabout where possible, and remain prepared to deal with people with poor emerge and lane skills while on the roundabout, you can flow over roundabouts and never stop moving; they are in fact easier than junctions and keep the traffic flowing better once you've got the correct technique.

[brucehoult]

I've had something like that happen to me.

I was also on a motorcycle, on a multiple lane roundabout. I was already on the roundabout, in the innermost lane, when a car came around the corner in my lane going the wrong way. I was leaned well over and going ... probably above the speed limit. Somehow I managed to jump a pretty substantial kerb onto a footpath in the center of the roundabout. Not even quite sure how I did that. I think I fundamentally managed to pretty much stop but it was apparent that the car wasn't going to, so I got out of its way.

[nctnico]

... I have witnessed senile drivers get completely mentally stuck in a roundabout, suddenly braking to a complete stop to "give way" for someone entering the roundabout, has happened more than once....

For some German roundabouts – specifically, any that do not have yield signs at the entrances – you are supposed to yield in the roundabout. A local supermarket has such an unsigned roundabout in their parking lot and not everyone remembers the law, still, is yield to the right.

That is a problem with roundabouts indeed. At the basis a roundabout is a one way street laid out in a circle and who needs to give way is up to who designs the roundabout but without signage, the basic traffic rules apply. But there are a lot of roundabouts that overrule that so you need to keep your eyes open for the signage.

BTW: In the NL you can also encounter roudabouts which have a bi-directional cycling lane around them to make matters more interesting.

Roundabouts are more naturally suited to driving on the left side of road.

That is nonsense; it doesn't matter. Whether clockwise or counter-clockwise, driving in a circle is the same.

You also need to trust in only having to look for traffic on the right (on the left in USA/Europe) -- and also of course that traffic in front of you isn't stopped. Still, much much easier than trying to get onto a busy road from a side road or driveway and spot gaps arriving from both left and right at the same time.

That "trust" can be a pitfall, you can't trust other people to be following the rules, as this little video demonstrates:

Although that's an extreme example it's illustrative of a common problem, if you're not prepared for people making poor emerges in front of you, and prepared to hold back if necessary to avoid them, you'll have issues.

That is very true. Participating in traffic means being able to deal well with mistakes that other people make and hope other people can correct for the mistakes you make. Never think you have seen all the weird crap that can happen in traffic; there are always people creative enough to surprise you.

[Nominal Animal]

Thus: As a traffic planner, how much weight would you place on public preference versus traffic flow and accidents, when designing new roads and intersections?

I think it is well known that humans overall are creatures of habit and such, and not very rational or logical in their personal preferences.

Which, in my understanding, means that "popularity" is only useful metric for entertainment and such, not for important stuff like traffic and programming languages.

[Cerebus]

I can speak to British traffic planners following "received wisdom" rather than common sense. It is an article of faith for British traffic planners that if you deliberately reduce visibility at the approach to a hazard drivers will both be more cautious and reduce their speeds on approach. Five minutes spent observing in the real world would show anyone that this is not true. So we get junctions engineered to be, in fact, less safe in the name of making them more safe. I guess the programming language designer's version of this is to reduce the incidence of type errors by doing away with types.

[Nominal Animal]

In Finland, all politics is data-driven.  As in, if statistics or gathered data shows that a political decision was less than desirable, the data gathering will be immediately stopped and statistics un-gathered.  If a third party later shows that the decision was less than desirable, it is straightforward and effective to claim that "nobody could have predicted this", because those who did predict it, had been labeled "conspiracy theorists" and "foil hatters" already.
The most free news in the world, indeed: we're just the best closing our eyes from observable facts.

This is to say, human idiocy is infinite, and rational and logical thought requires effort.  Even when a logical, rational, proven better options exist, it is usually still an uphill battle to try and change the inertia of past decisions and established social "norms": this is what it is to be human.

I do find it interesting (and somewhat disconcerting) how many of the "best" software projects are driven by benevolent dictators, instead of any kind of democratic processes.  There are enough important counterexamples (C, Debian, Apache) that I don't think it is anything inherent, but more a combination of issues we haven't dealt with yet.

Considering all that, I'm not sure how I feel about "Python becoming the most popular language".  There is something definitely sinister in that to me, even though I write Python code very often; if not daily, nearly so anyway.

[SiliconWizard]

In Finland, all politics is data-driven.  As in, if statistics or gathered data shows that a political decision was less than desirable, the data gathering will be immediately stopped and statistics un-gathered.

So, uh... data-driven backwards.
It's not much better elsewhere...

Considering all that, I'm not sure how I feel about "Python becoming the most popular language".  There is something definitely sinister in that to me, even though I write Python code very often; if not daily, nearly so anyway.

Fortunately for me, I don't use Python. But I find the trend sinister as well.

But it's somewhat as sinister as the trend that is making machine learning the alpha and omega of computing these days, as though there was nothing else anymore, no other approach worth mentioning, and no problem that couldn't be solved with it.

But is it all a sign of human "idiocy", or ultimately just a sign of greed?

[Picuino]

I don't find it sinister that a particular language is the most popular. In fact it is inevitable, some language MUST be the first in popularity. Until recently it was C and Java. Now it's Python.

Given that I particularly like the Python language, I think this popularity is an advantage and can be used to improve its libraries, frameworks and other tools.

As for changes in the language itself, I also miss less version changes in the Python language. C has very few changes and you can program today perfectly well in the C99 standard of 23 years ago and there is nothing wrong with that.

[brucehoult]

But it's somewhat as sinister as the trend that is making machine learning the alpha and omega of computing these days, as though there was nothing else anymore, no other approach worth mentioning, and no problem that couldn't be solved with it.

The worst thing about the current "ML" fad is that it is totally infeasible to understand why the computer came up with any particular answer. The algorithm was not designed by anyone, it was "trained" until it seemed to work.

"Computer says 'NO'" becomes totally unassailable.

I guess some regard this as the best thing about it.

[nctnico]

As for changes in the language itself, I also miss less version changes in the Python language. C has very few changes and you can program today perfectly well in the C99 standard of 23 years ago and there is nothing wrong with that.

Maybe that is what is keeping C alive for all this time. PHP is also a language that severly breaks backwards compatibility with new versions. I get the reasoning (must be suitable for idiots to make safe web applications) but it would be so much nicer if it was possible to keep existing code running. More professional languages like C, VHDL, Verilog, etc have the possibility to tell the compiler which language version to use so old code can still be compiled succesfully.

[cfbsoftware]

In fact it is inevitable, some language MUST be the first in popularity. Until recently it was C and Java. Now it's Python.

FWIW the most popular video on YouTube is Baby Shark Dance. Go figure.

[SiliconWizard]

In fact it is inevitable, some language MUST be the first in popularity. Until recently it was C and Java. Now it's Python.

FWIW the most popular video on YouTube is Baby Shark Dance. Go figure.

 

Evaluation of popularity is ridden with pitfalls and biases anyway. But behind the popularity index, there's a very real fact: there's an increasing number of "developers" knowing Python on the market, and an alarmingly decreasing number of developers knowing other languages, so for any work for which Python is not appropriate (and yes, there's still a lot of stuff that can't be done with Python, despite what some would like to think), good luck finding candidates these days...

[Picuino]

In the end it doesn't matter whether a language is the most popular or is actually in second or third place.
What is important in this case is the undisputed rise in popularity of Python (an interpreted language) to the top ranks of popularity.
The fact that Python is slower than other languages is another reason that makes its rise in popularity even more surprising.

In fact, I believe that if it had not been Python it would have been another interpreted language that would have reached the top positions. And that's a major change in the way software is developed.
The advantages of interpreted software make it much more effective in some cases. Web development is one such paradigmatic case.

[Picuino]

https://insights.stackoverflow.com/survey/2021#section-most-popular-technologies-programming-scripting-and-markup-languages

at Stack Overflow joins programming languages with markup languages such as HTML and the popularity of the different languages looks like this:

[tszaboo]

You also need to trust in only having to look for traffic on the right (on the left in USA/Europe) -- and also of course that traffic in front of you isn't stopped. Still, much much easier than trying to get onto a busy road from a side road or driveway and spot gaps arriving from both left and right at the same time.

That "trust" can be a pitfall, you can't trust other people to be following the rules, as this little video demonstrates:

I said this before, and I say it again:
If someone decides to buy a Peugeot, you take their drivers license and throw it in the shredder. They are clearly not interested in cars, driving, safety or reliability, and they lack common sense and shouldn't be allowed on the road.

[Nominal Animal]

I say this once, and once only.

If someone decides to use Windows and pay Microsoft for a license, you need to crush their fingers so that they won't be able to touch any computer, ever.
That decision shows they are not interested in the long-term interests of the human race, and instead are driven by instinct, habit, social pressure, and advertisements only.

See, how superlative language becomes horrible, when you switch things you don't like with things you do like?  Food for thought, innit.

[PlainName]

The advantages of interpreted software

What do you see those as? To me, a major issue is that the first time to code gets to be parsed is when it's about to be run for real. With compiled languages, the code may be crap but all of it has to compile before you can run it.

[brucehoult]

If someone decides to buy a Peugeot, you take their drivers license and throw it in the shredder. They are clearly not interested in cars, driving, safety or reliability, and they lack common sense and shouldn't be allowed on the road.

Got any other brand opinions?

I endorse the contents of this video, even if (as I do) you live in a slightly different country to JOhn (and eevblog):

[tszaboo]

I say this once, and once only.

If someone decides to use Windows and pay Microsoft for a license, you need to crush their fingers so that they won't be able to touch any computer, ever.
That decision shows they are not interested in the long-term interests of the human race, and instead are driven by instinct, habit, social pressure, and advertisements only.

See, how superlative language becomes horrible, when you switch things you don't like with things you do like?  Food for thought, innit.

Nah, what you are saying is bullshit.

[Picuino]

The advantages of interpreted software

What do you see those as? To me, a major issue is that the first time to code gets to be parsed is when it's about to be run for real. With compiled languages, the code may be crap but all of it has to compile before you can run it.

For me, the main advantage of interpreted languages is their speed of development. You can make changes and test them immediately, without waiting for the compilation to complete.
Also you can run command line commands to test how the program works without the need for compilations.

In the microcontroller realm, in addition to having to wait for the compilation to complete you have the problem of having to load the compiled hexadecimal into the micro. When you want to make small changes and test immediately that causes quite a time expenditure. With an interpreted language like LUA, Basic or FORTH, command line testing is immediate.

Once you have tested the algorithm and when you see that it works well, you can transfer it to C language, which is faster in execution.

[madires]

What is important in this case is the undisputed rise in popularity of Python (an interpreted language) to the top ranks of popularity.
The fact that Python is slower than other languages is another reason that makes its rise in popularity even more surprising.

I'm not surprised. It's way many humans tick. And at the same time many of the developers coding in Python support a green future.

The advantages of interpreted software make it much more effective in some cases. Web development is one such paradigmatic case.

They are great for automation and one-off tasks. But if you need to scale up things for many users or tons of data they suck badly because of their poor performance.

[Cerebus]

They are great for automation and one-off tasks. But if you need to scale up things for many users or tons of data they suck badly because of their poor performance.

Although performance can be an issue, the suckiest thing (if there is such a word) in that context is usually maintainability. Scripting friendly languages tend to attract coders and coding styles that are poor in maintainability terms. This has been true of 95% of "other people's code" that I've had to deal with over the years that has been written in one of the scripting friendly languages whereas it's probably only 70% for 'mainstream' programming languages.

[Siwastaja]

I said this before, and I say it again:
If someone decides to buy a Peugeot, you take their drivers license and throw it in the shredder. They are clearly not interested in cars, driving, safety or reliability, and they lack common sense and shouldn't be allowed on the road.

It's interesting how culturally dependent such ideas are; here, no one would ever label a Peugeot driver like that. In fact, I'd say getting a Peugeot would be the most neutral action imaginable, communicating boringly modest amount of interest in cars, driving, safety, reliability, and common sense.

Here, the symbol for not being able to drive at all, and not having any interest in fixing that, seems to be a Nissan Qashqai, specifically white color. And then, I'm sure this doesn't apply where you live, at all, so I don't make a post like yours about it.