It is interesting Stallman thinks it is not necessary for hardware to be open source, because you cannot change it. If he would have been born a bit earlier and had started thinking about freedom at the time of the introduction of the integrated cirquit, he would have been against closed source integrated cirquits. Before ICs, all four freedoms were there for hardware: you could use it, study and adapt it, clone it and improve it. ICs made "reading" the hardware directly very difficult, disabling three of the four freedoms. With ICs you need the blueprints for those freedoms.

And what if Stallman had been born later, in a future in which "trusted computing" is already reality? Would he still want software to be free/open source? Reading and modifying software would be just as impossible as reading and modifying ICs then!

Please, mr Stallman, demand that not only as much software must be freed, but also hardware. Not just the BIOS, it should be possible to build a whole "open" computer. Not only will that completely stop the "trusted computing" nonsense, it is a necessity. Thanks to technology like FPGAs, it is practically possible for everyone to design their own hardware today. Maybe you can help building a free hardware community?

Seems he already wrote something about it.

http://features.linuxtoday.com/news_story.php3?ltsn=1999-06-22-005-05-NW-LF

But copying hardware is NOT hard! Just use a hardware prototyping device such as an FPGA. True, these easily rewritable chips are not as fast as "hardwired" hardware, but they can still be very useful. For example the developers of GIMP could create a design for an accelerator for some heavy image processing operations. The early adopters would load it into an FPGA, and when the design becomes stable, some hardware manufacturer takes it and "prints" it on single-purpose chips, making an even faster accelerator available to everyone.
There is one huge problem: as far as I know there is only expensive proprietary software for loading designs into FPGAs, possibly because the manufacturers do not publish the interface specifications.

Well... maybe the big names in hardware are not going to like freed hardware, and as a revenge they will not show their interface specifications to the f/oss community anymore... That would be bad for the near future of open source, but in the end having open hardware is very good for open software. Both with software and hardware, it is useful to be able to read more than the interface specification while writing hardware/software that interfaces with that hardware/software.

I must say i completely agree with you PK and i'm not completely sure where exactly does RMS stands in regard to the free hardware idea and in what sense.

What he described in your quotes is probably valid. We can't say that hardware is the same as software. It is still harder to copy it. But what exactly does copying hardware means? It is definitely not putting it in a replicator and ordering a duplicate. :quoi: (like in star trek).

What copying hardware essentially means is reproducing it according to the same exact blueprints and specifications that tell you how to produce it, provided the nessessery raw (or other) materials.
And this comes down to information again. To be able to copy hardware you gotta have a free and open information on how to do it.

And this is very well the same issue as with software. That information CAN be easily copied and it SHOULD be free as in freedom. I don't think RMS would disagree about that and if he would than i wouldn't agree with him in that.

I believe i will have to make an inquire to himself about that. :-)

This free information for hardware reproduction is what counts under free information and knowledge of production we are writing about in our cooperative article: Expanding FOSS Model to Other Areas of Economy (areas of material production - computer hardware involved).

There i believe it doesn't actually matter if the product is or isn't easily replicable because the information essential as a precondition to be able to do that reproduction is always easy to copy and share and it should thus be free. Software is different from other information only in that it is the blueprint and a product of that blueprint at the same time.

And of course, as for the free BIOS campaign i fully support it and intend to sign myself as a supporter.. No question about it.

Thank you
Daniel

As Danijel already said, this is exaclty the idea we developed in our [url=]Cooperative Article[/url]. See, please the Section 4, we describe such a free material production community (it would be nice if you could comment on that! :yes: )

Thanks,

Rijik.

No, not the same... I guess I should explain what an FPGA is :-D . An FPGA (field programmable gate array) is a programmable chip. Essentially an FPGA is to hardware blueprints what a computer is to computer programs: you can load a blueprint onto it, then use the FPGA as if it is the chip specified by the blueprint, then load a completely different blueprint onto it, etc...
This means that if you have an FPGA and the software and hardware to (re)write it, hardware designs have become just like software, and exactly the same arguments about freedom will apply to both! In fact hardware blueprints ARE software for a very untraditional processor called an FPGA, so integrated cirquits are not an "other area of economy" at all.

As for my comments on the article: you know I don't like the "self preserving" part in the GPL, not for software, and not for anything else. But applying that to other things than software is the whole point of the article... Open knowledge without the self preserving thing is nothing new, in fact I think most knowledge is available freely to everyone.

Something that would be very interesting is how to make the self-organizing way the FOSS community works possible for other types of communities (not software). Just a GPL-like license is not enough, there is more to a community than a license. How does the communication work? How are decisions made? How are standards created? Etc etc...

Yes! This is the chalenge for the possible cooperative communities of other areas of economy. We are not there, we are not TVs producers or material good developers. In fact, It's up to them to resolve how they would work cooperatively. We deal with this issue in the Section 6 of our cooperative article.

I only guess that it might be similar to the FOSS community way, but how exaclty, we can't know yet. We don't know even if they would take on the FOKP model! It's still just an idea.

Here in my Computer Science Departament we have some programable hardwares, but I didn't know that they were already so flexible. It's really amazing!

As far the 'self preserving' part of GPL license, you remind me that guy who loves to be in a nice beach under the sunshade, but doesn't like much that piece of wood that stands up the sunshade...

Am I being unfair with you? :-D

Anyway, many thanks for your nice comments,

Rijik.

Thank you for the explanation. I agree that sounds more like a software then hardware. The FPGA in itself is hardware (as a chip in itself), but it's functioning is dictated by "software" which are both blueprints and a products of those blueprints. :-)

Wise words Rijik.. :-) (sorry PK) ;-)

Thanks
Dan

Sorry if I disagree with the majority here, but I think people misunderstand when Stallman talks about Free Software. There cannot be free hardware the same way software can be free. For example, when Stallman thinks about the right to copy, he's not talking about taking some raw physical material, building it and practically "copying" the hardware model.

Whichever the economic model you choose (capitalism, social-democracy, marxist-leninist socialism, anarcho-sindicalism, coperative-based economies, etc. etc. etc.), the creation of artificial physical objects imply investment of a collective human labor to create an object which corresponds to a certain original model, and which is accounted in the economic system per-se. Stallman does advocate for a just distribution of capital, and in fact he doesn't oppose at all that people should share blueprints and information on that task. But that's not the point he's trying to make with Free Software.

The Physics of software is very different from the Physics of hardware. For example, I can copy the GIMP (to make an example), and I alone can make thousands of copies with it, with very little or no effort whatsoever. I can copy it on CD's, or provide copies in my site, then everyone can have their own copy, and everyone can copy that software, and redistribute it, etc.

To clarify a little bit more about the absurdity of talking of Free Hardware being in the same sense as Free Software. The only way we can talk about Free Hardware in the same sense as Free Software is that if we have something very similar to Star Trek's replicator. The replicator can actually make copies of physical objects just as we can make copies of software. So, if everyone has a replicator, there would be little or no human investment in replicating whatever we want.

When we talk about copying a software, we are not talking about copying "blueprints" of software. We are talking about copying the program itself. That cannot be possible in the case of hardware. Freedom of software is freedom of information, freedom of cooperation, freedom of having a community that benefits itself. Hardware is not information in the way software is. Blueprints, and information about hardware should be available to everyone, but hardware cannot be made free in the sense that software is! That's RMS's point!

And about changing the hardware ... well, actually no-one prevents me from changing the hardware. In fact, I can take a motherboard, and with an artistic mind I can burn it, or cut a heart-shape with it, and place it on the wall... I can make other kinds of changes.... whether it will work afterwards, it's another story!!! But you can change hardware. No one will actually stop you from doing it. If you buy a $160.00 CPU then bend the pins and make a hole... and hang it in the visor of the car... Well... that's $160.00 stupidly wasted... but no one stops you from doing those kinds of changes!!!!

However ... AND THIS IS STALLMAN'S ISSUE ... the software that comes in the BIOS to manage those chips are not Free Software. The programs provided for that task are only provided as executables. Which once again, prevents everyone from changing it, from enhancing it, from sharing it with a community of hackers and programmers. Besides, you never know (it doesn't matter what the companies claim) that such BIOS have treacherous programs, specially designed to spy on people's activities. That is the issue with the so-called "trusted computing". Hence a FreeBIOS is necessary. That situation of IBM's denial of providing the specifications of the BIOS is no different as when in 1978 he asked Xerox for a copy of the drivers so he could make the MIT printer work reliably.

I hope I have clarified a little bit what Stallman has in mind.

Agreed Prosario.

Hardware itself is just a product while information and knowledge (blueprints) to produce it are something else. Hardware cannot be copied in the same sense as information indeed, but the information on how to reproduce it exactly as it was can be easily copied and should be free and open.

The difference is this:

Software is both the product and a blueprint of itself.

Hardware is a product made with the blueprints of itself which are naturally different.

Hmm.. i kinda feel like i just repeated myself :-)

Anyway.. the only thing we're actually talking about here is *information* (whichever sort of it). Software is information, blueprints are information and information should be free as in freedom.

Material objects being produced "out of" that information is something else and different ethical and practical rules apply there.

Thank you
Daniel

Very good points, Prosario. Indeed, very clarifying. :-) Perhaps these blueprints could be "registered" under some kind of GPL license? Does this make sense?

Hmmm.... I think I am getting sleepy here... hehe... :-)

Thank you,
Andr�