OSHW 1970's Edition

[ehughes]

My take on what OSHW means:

https://circuitmaker.com/blog/#Blogs/OSHW-1970sEdition

[donotdespisethesnake]

My take on what OSHW means:

https://circuitmaker.com/blog/#Blogs/OSHW-1970sEdition

An interesting history, but you miss many points along the way. The article in PE is copyrighted material, and the user by virtue of buying the magazine was granted a license to use the material.  If you started copying the article and giving it to people, you would be infringing copyright. Now that magazines publish electronically, you will find they protect their IP, and the terms of use make it clear users are not free to re-distribute material.

The point of open source is not merely to publish a complete design for personal and private use, it is to make the design free for use, distribution and modification by others. What PE did (and others still do) is completely not Open Source. So if you think it is, you really have no clue what Open Source is about.

ETA: It's ironic, but not unexpected, that a company targetting the "open source community" seems to have virtually no clue what Open Source means. Invariably these companies are uninterested in the principles - but are very interested in ways to make money out of it. "Open Source" is just tacked on as a marketing checkbox. But anyway I followed the link to your website, so I guess your ad campaign worked

[EEVblog]

An interesting history, but you miss many points along the way. The article in PE is copyrighted material, and the user by virtue of buying the magazine was granted a license to use the material.

There is no licence involved. You are simply paying for access to read the magazine in convenient form that was not free to produce.
If you wanted to you could go to the library, newsagent, or electronics store and read it for free.
And you are allowed to photocopy the article for personal use.

  If you started copying the article and giving it to people, you would be infringing copyright.

Just like today when you can't just go copy someone web site describing the project.

The point of open source is not merely to publish a complete design for personal and private use, it is to make the design free for use, distribution and modification by others. What PE did (and others still do) is completely not Open Source. So if you think it is, you really have no clue what Open Source is about.

It is free and open. It's just the format description in the article that is the copyright of the magazine.
Kind of like you have to pay to access the internet in order to get the free open source information today.
Sure you can go to a library or wherever and get free internet access, but so you could also do the same with magazines.
And just like copyright on a magazine article meaning you can't reproduce it as-is, you also can't usually go copy or reproduce someones entire web page with their open source article. That web page "layout" is just like the magazine articles of old.

What Eli was getting at is the circuit design etc is as "open source" as any open source project today.
Open hardware is about the project design and describing it fully so other can modify or reproduce the design, it's not about the format of the magazine article or the website equivalent today.

[donotdespisethesnake]

Totally and completely missed the point of Open Source. It is not just about publishing a design

If you look at the PE magazine, it clearly says "Copyright .. All rights reserved".

Sorry Dave, you are quite clueless too.

[EEVblog]

Totally and completely missed the point of Open Source. It is not just about publishing a design
If you look at the PE magazine, it clearly says "Copyright .. All rights reserved".
Sorry Dave, you are quite clueless too.

I'm clueless?
I have had projects published in magazines before the whole "open hardware" movement came along, have you?
I have had my magazine projects copied and people made money from it, have you?
I had a (very small) hand in culturing the ethos of magazine project "open hardware" back in the day, did you? Do you even know what the magazine project ethos was back in the day?

You clearly have no idea what copyright is. Every author automatically gets "Copyright" on their published work, even today with OSHW, you still get "Copyright Blah Blah" on the schematic.
The article in question actually has no copyright message attached to it if you actually read it, not that it matters because it's not the point. Of course the magazine holds copyright over the layout of the article as published in the magazine!, just like you or I have natural copyright on our web page layout that describe our OSHW projects. Same thing!
Take the prime example of OSHW today, the Arduino. Go and copy their webpages as-is (the modern day equivalent of photocopying that magazine article and republishing it) and see how you get on:
https://www.arduino.cc/en/Main/CopyrightNotice

[helius]

Actually, prior to the adoption of the Berne Convention, copyright protection was not automatic and required a statement of copyright on the work plus registration. The Night Of The Living Dead was famously never protected because its distributor failed to do this.

[ehughes]

What PE did (and others still do) is completely not Open Source.

I think you are still conflating hardware and software.      If getting a schematic,  board layout , a theory of operation, mechanical specs does not satisfy you, nothing will.   Hardware is not free to reproduce and the design tools are *not* the design.   It is theoretically impossible to meet the definition of open source by your standards.   

In several instances,  people have taken the the wah-wah *design* from PE and built companies around it.     You are hung up on the idea of the how the schematic was shown in the article. 

Here is the reality.   The copyright "schematic" is not what anyone wants.  It is the *Design* that is interesting.      A good hardware designer will recreate and tweak in the tools of their choice.  Only software people care about the "direct clone"  as s it much easier to step and repeat.  The items I listed are all that are needed to achieve this goal.   

My original position stands.  If you are hung up on the legalistic software application of open source as it applies to software,  you have failed.   How about spending time writing a theory of operation on your design?   That will make it more open and free than the GPL or gear logo ever will.   The theory of the design allows me to really take the concept and move on.   I could care less about the tools you used to generate the schematic and layout.    Those will come and go.   The design lives beyond the tools and implementation.

This goes back to the "teach a man to fish" parable.     I could  care less about the fish you may give me for free.  I want to know *how* to fish*.   At that point I can build own fishing pole.


All that said...   The software people do not get to define what "open".    Especially to hardware people who have been sharing work for many years. :-) 

Invariably these companies are uninterested in the principles - but are very interested in ways to make money out of it

Yes, it costs money to keep the lights on,   pay employees and develop a product.     Even in the software world,  there is profit motive behind the software development.   IBM, et. al.  contribute with their resources to open source projects as it helps their bottom line.     Open source principles havelittle effect.    Profit motive is a very good thing for open source.  That is the reason there are some very good open source projects.   The projects that are fueled purely by principle tend to be the most awful in design.

[EEVblog]

My original position stands.  If you are hung up on the legalistic software application of open source as it applies to software,  you have failed.   How about spending time writing a theory of operation on your design?   That will make it more open and free than the GPL or gear logo ever will.

+1
And this is why the old magazine articles are often far superior "open hardware" projects than modern ones who think all they need to do if whack the design files up on github and slap the gear logo on, and job done.
That's ok or course, you are still sharing your design and making it open, great, well done, but it's a far cry from a nicely written article and theory of operation with block diagrams, timing diagrams, wiring diagrams and other niceties. To poo-poo someone's magazine design and say it's not "open hardware" is just ludicrous. Modern open hardware designs should live up to the standards done back then.

And there are those that argue that magazine projects were not open hardware because they didn't have any modern licencing associated with them, and therefore you shouldn't touch it and it was hence worthless as an open hardware project.
What a load of bollocks. Sorry, but the world worked fine with people building companies around and resusing these magazine article projects since Wireless Weekly in the 1920's.
It was like getting research published in a peer reviewed journal. You got the project published so that:
a) You could make a name for yourself
b) People could critique it and give you feedback.
c) Share designs and idea with other people and advance the industry
d) Encourage people to recreate or expand and build upon your ideas
It was an honor to get your project published and see people build upon it.

[rrinker]

 I miss the classic mags - in the US the biggies, at least to me, where Popular Electronics and Radio-Electronics. I learned an awful lot about electronics reading those magazines as a kid, because the best articles always explained how something worked and why it was built the way it was. Dis the ones in other countries have those reader service cards  where you could circle a number (printed at the bottom of an ad) to send away for additional info on various products? I got all sorts of databooks and so forth that way. And I would read them over and over until I understood what was going on.

 Proper documentation of a project is an absolute must if anyone is to learn from your design. You are spot on. A copy of the schematic, Gerber files for a PCB, and a gear logo basically say "here, just build this, it's cool." It's very hard to learn from such things. Add in the how and why documentation and you have a winner. Young players and veterans alike can always learn something from a new and interesting project.

[free_electron]

i still find this whole 'open sauce' licence formats a big pile of drivel.

Wanna make something really 'open'? release it without any copyrights.
make it 'public domain'.

anyone can do what he wants with it. in the early days of the internet that was what people did. Stuff was released as public domain. No need for complicated legalese.
people were nice enough to automatically attribute a design to the original owner or put 'based on' in the files.

[helius]

As I explained above, the rules were different before the Berne Convention. If you did not claim copyright, it was in the public domain.
The Berne Convention works differently and doesn't have a public domain concept. The best you can do is to explicitly derogate your authorship rights, which is not even possible in some countries.

[EEVblog]

i still find this whole 'open sauce' licence formats a big pile of drivel.
Wanna make something really 'open'? release it without any copyrights.
make it 'public domain'.
anyone can do what he wants with it. in the early days of the internet that was what people did. Stuff was released as public domain. No need for complicated legalese.
people were nice enough to automatically attribute a design to the original owner or put 'based on' in the files.

That's exactly how the magazine article worked, your project effectively became "public domain", there was no license attached to it, which I agree is the best form of "open".

[EEVblog]

As I explained above, the rules were different before the Berne Convention. If you did not claim copyright, it was in the public domain.
The Berne Convention works differently and doesn't have a public domain concept. The best you can do is to explicitly derogate your authorship rights, which is not even possible in some countries.

You do know the Berne Convention was in 1886, right?

[coppice]

As I explained above, the rules were different before the Berne Convention. If you did not claim copyright, it was in the public domain.
The Berne Convention works differently and doesn't have a public domain concept. The best you can do is to explicitly derogate your authorship rights, which is not even possible in some countries.

Before the Berne Convention means back in the 19th century, and of no relevance to anything being discussed here. I think you might be referring to the US joining the Berne Convention a century after everyone else.

At the time of the Berne Convention the US had laws denying copyright to foreigners, and allowing Americans to rip off anyone not American. They certainly were not going to sign up at that point. Look up "Gilbert and Sullivan" for an interesting read on America's weak attitude to copyright when it suited them, versus now when they think the whole world should pay eternal homage to Mickey Mouse.

[helius]

I think you might be referring to the US joining the Berne Convention a century after everyone else.

When adopting a supercilious and condescending attitude, it is a good idea to ensure you are right; otherwise you merely appear as a fool. I wonder what the members here from South and Central America think about your implication that they simply don't matter.

[coppice]

I think you might be referring to the US joining the Berne Convention a century after everyone else.

When adopting a supercilious and condescending attitude, it is a good idea to ensure you are right; otherwise you merely appear as a fool. I wonder what the members here from South and Central America think about your implication that they simply don't matter.

Many countries were not asked to be signatories to the convention in 1889, but I believe the US was the only country to refuse to sign when asked. It was also the country which had created the most friction over copyright issues, by refusing to respect the rights of foreigners. Most of Europe already had piecemeal bi-party agreements respecting the rights of foreigners in a largely compatible manner. Most other countries signed up at an appropriate point in their history. In some cases that lead to them signing up very recently.

Its worth saying that colonialism played a big part in the lists of countries who signed and didn't sign in the 1880s. A lot of countries were colonies at that time, and so didn't sign for themselves. Much of South America had only stopped being colonies 50 or 60 years before, so they were probably not especially eager to sign mutual agreements with their old colonial masters, even if they were asked.

[apelly]

i still find this whole 'open sauce' licence formats a big pile of drivel.

Wanna make something really 'open'? release it without any copyrights.
make it 'public domain'.

Open source is not public domain.

Your constant bashing is boring. You are welcome to maintain your ignorance and attitude. It'd be nice if you contained your childish whining though.

[EEVblog]

Open source is not public domain.

No, but public domain is a valid way to open source a design.
Let's take the most recognised definition we have:
http://freedomdefined.org/OSHW

Open Source Hardware (OSHW) is a term for tangible artifacts -- machines, devices, or other physical things -- whose design has been released to the public in such a way that anyone can make, modify, distribute, and use those things.

Public domain certainly fits that requirement.

[apelly]

I didn't think there was a need to state the obvious. An apple is a fruit. An orange is a fruit. An apple is not an orange.

There are many ways to skin the open source cat. It's up to the original developer to decide what suits them. Not what suits you.

[apelly]

My feeling is simply that it's nice that people work together for the fun of it and just give us the fruits of their labour. No need to shit on them for it.

[aon]

Without delving too much into the specifics here, isn't the gist of this pretty much the same argument that Dan Bernstein has made for software - basically that when you acquire software that has no licensing terms that waive the copyright holder's rights, you're still allowed to do pretty much whatever you want with it, as long as you don't distribute the modified copies (http://cr.yp.to/softwarelaw.html, https://en.wikipedia.org/wiki/License-free_software)?

While there's nothing wrong with that view as such, what happens if you want to make changes to the original design files and share them? If the original author doesn't want to integrate them, the only way to make them public is by publishing patches, which is a pretty cumbersome way to deal with things in the age of distributed version control. Speaking of which, trying to submit your improvements by making a public forked repository and sending a pull request to the author, as is commonly done, would mean publishing a derivative work and violating the copyright.

The way I see it, open source is more about the ease of collaboration, not what an individual person gets or doesn't get to do when they acquire the resulting work. Public domain as a concept doesn't exist everywhere, so licensing seems to be the commonly agreed-upon way to do things.

[rsjsouza]

Free as in freedom exists since the magazines of yore, but the difference with the new "open source" concept is that the new deal is more restrictive: everything you do with the code must be kept open as well. I am not sure if this was done this way to force recognition to the previous contributors or to prevent companies reaping the benefits of free code (free as in free beer). Perhaps that is what the OSHW guys are trying to do? Or to prevent a company that copies an OSHW design and scratches the codes of all the ICs? I am not sure (I haven't read their manifesto).

Despite this, the OSHW examples mentioned here certainly seem to be "open source": all design files and details are released, but with no explanation why things were done in a certain way - as if they were
SW without comments or an API guide.

To me the excellent HW projects of the old electronics magazines were not "open source" but "open design" instead. The equivalent in the SW realm would be a book like "Numerical Recipes in C": it has all the theoretical details about the algorithms and also show its implementation in C language. Also similarly to a HW project in a magazine, you are allowed to replicate the algorithms there "as-is" for free (by typing yourself) or modify them to fit your own purpose. Even better, you can purchase a copy of the SW in electronic form - similarly to a HW kit purchased from the magazine store. 
 

[aon]

Free as in freedom exists since the magazines of yore, but the difference with the new "open source" concept is that the new deal is more restrictive: everything you do with the code must be kept open as well.

Not quite. To quote OSHWA's best practices document:

There are two main classes of open-source or free-software licenses: copyleft (or viral) licenses which require that derivatives be licensed under the same terms; and permissive licenses, which allow others to make modifications without releasing them as open-source hardware. Note that the definition of open-source hardware specifies that you must allow modification and commercial re-use of your design, so do not use licenses with a no-derivatives or non-commercial clause.

Neither OSHWA's open hardware definition nor OSI's open source definition mandate copyleft licenses.

What I tried to point out is that the main difference between "free as in a magazine article" and having an open source license is that in the latter case you don't have to ask the original author for permission to make your modified design files public (or to just redistribute the original). Same goes for software - I don't think I've seen anyone argue that open source software is unnecessary because computer magazines in the 80's had BASIC listings on them

Despite this, the OSHW examples mentioned here certainly seem to be "open source": all design files and details are released, but with no explanation why things were done in a certain way - as if they were
SW without comments or an API guide.

To me the excellent HW projects of the old electronics magazines were not "open source" but "open design" instead.

I think this is just because there's no barrier of entry to putting something on the Internet, whereas a magazine article has to be something people actually want to read to get published. When you can get something out there without doing the boring parts, why would anyone do it when you could just move on to the next thing (see: CADT)?

Seems to me that a similar thing is sort of happening with OSHW, though - the projects with better documentation tend to get picked up by large blogs and such. On the other hand, I don't remember there ever being a shortage of badly documented circuits on the Internet, even way before anyone talked about OSHW.

[rsjsouza]

Free as in freedom exists since the magazines of yore, but the difference with the new "open source" concept is that the new deal is more restrictive: everything you do with the code must be kept open as well.

Not quite. To quote OSHWA's best practices document:

There are two main classes of open-source or free-software licenses: copyleft (or viral) licenses which require that derivatives be licensed under the same terms; and permissive licenses, which allow others to make modifications without releasing them as open-source hardware. Note that the definition of open-source hardware specifies that you must allow modification and commercial re-use of your design, so do not use licenses with a no-derivatives or non-commercial clause.

Neither OSHWA's open hardware definition nor OSI's open source definition mandate copyleft licenses.

You are correct regarding OSHWA; I didn't read their terms. I was skewed by GPLv3, BSD (IIRC) and a few others that mandate copyleft.

What I tried to point out is that the main difference between "free as in a magazine article" and having an open source license is that in the latter case you don't have to ask the original author for permission to make your modified design files public (or to just redistribute the original). Same goes for software - I don't think I've seen anyone argue that open source software is unnecessary because computer magazines in the 80's had BASIC listings on them

In lieu of your highlight, I think the difference between the magazine and OSHWA is that you can't photocopy the magazine article without permission (a restriction when compared to OSHWA, which grants redistribution), but you could transfer the code or the design by your own means (typing source code or using a copper clad board, overhead projector marker and iron perchlorate). Both allowed publishing a new article if you built upon the original.

BTW, thanks for reminding me of the old computer magazines with their source code listings, which I think also fall into the concept of "open design", as frequently they were accompanied with a thorough explanation of the code an methods used.

Despite this, the OSHW examples mentioned here certainly seem to be "open source": all design files and details are released, but with no explanation why things were done in a certain way - as if they were
SW without comments or an API guide.

To me the excellent HW projects of the old electronics magazines were not "open source" but "open design" instead.

I think this is just because there's no barrier of entry to putting something on the Internet, whereas a magazine article has to be something people actually want to read to get published. When you can get something out there without doing the boring parts, why would anyone do it when you could just move on to the next thing (see: CADT)?

Seems to me that a similar thing is sort of happening with OSHW, though - the projects with better documentation tend to get picked up by large blogs and such. On the other hand, I don't remember there ever being a shortage of badly documented circuits on the Internet, even way before anyone talked about OSHW.

Yes, I think you are on the mark here. A design to be approved in a magazine required a more thorough process of investigation and testing - only in the latter years the quality started to fall in certain magazines and several untested designs made to the pages (or were protected in some way when microprocessors started to gain popularity).

[Cerebus]

Actually, prior to the adoption of the Berne Convention, copyright protection was not automatic and required a statement of copyright on the work plus registration. The Night Of The Living Dead was famously never protected because its distributor failed to do this.

What you're missing there is a heavy sprinking of "in the United States".

The Berne Convention dates from 1886 when it was adopted by: Belgium, France, Germany, Haiti, Italy, Liberia, Spain, Switzerland, Tunisia, and the United Kingdom. Many other countries adopted it over the years, including Australia in 1928. The US was rather late to the game, adopting the convention in 1989. That is one the reasons you'll see a lot of older books that say "Not for sale in the United States" in the frontispiece.