Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
New W3C Patents Policy

By bizcoach in Internet
Thu Apr 03, 2003 at 06:53:16 AM EST
Tags: Freedom (all tags)
Freedom

Because the internet experience of most computer users is primarily through a "web browser", the World Wide Web Consortium (W3C), where new standard technologies for the Web are developed, is probably the organization with the greatest influence on the future of the internet.

Now the W3C is planning to change their rules so that patented technologies will be accepted as standards for the Web, subject to "royalty-free license" requirements. The W3C says in their press release that these requirements are "consistent with Open Source/Free Software Terms". Unfortunately the royalty-free patent license might allow only to implement what is required by the standard. This has led the DotGNU project to submit a comment that the planned new rules would lead to a "standards war".


What precisely are the problems with the new patents policy?

First of all, the claim that the "royalty-free license" requirements are "consistent with Open Source/Free Software Terms" is simply not true. As the Free Software Foundation points out on their website, these "W3C Royalty-Free license requirements" allow standards that cannot be implemented in Free Software because of patent issues. More precisely, when the royalty-free patent license allows only to implement what is required by the standard, that is called a "field of use" restriction. Any such restriction is in conflict with the user's freedom to modify the program, which is an essential part of both the Free Software Definition and the Open Source Definition.

However it is true that the Royalty-Free Patent Policy draft is a big improvement over earlier policy proposals.

Programs may appear to be Free Software / Open Source even when in reality, they aren't

When a standard is affected by this problem (that essential technology is covered by a patent with a royalty-free license which however has a "field of use" restriction), programmers are allowed to implement the standard and make the resulting software available for download. What happens if the programmer tries to make the program Free Software / Open Source?

Both in the Open Source movement and in the Free Software movement, the most generally recognized license is the GNU General Public License (GNU GPL). This is a "copyleft" license, which means that redistribution of the program is allowed only if the program will be Free Software for the recipient. That is impossible in this situation, therefore any attempt to use a copyleft license will result in a situation where only the copyright holder truly has the right to distribute the program.

The effects of using a non-copyleft license are less dramatic. While the program doesn't become Free Software / Open Source, at least within the constraints of the "field of use" restriction, people are allowed to redistribute modified and unmodified copies of the program.

Why we cannot simply use a non-copyleft license

A few years ago, I would have considered this to be a minor issue, something that concerns only those for whom Free Software and Open Source are matters of religion. I want to view computers and software merely as tools for getting things done, and like most computer users I wasn't interested in paying much attention to subtle details of software licensing.

My perspective changed dramatically when I realised how much Microsoft's plans with .NET were an attempt to extend their effective monopoly on desktop systems to the internet and e-commerce. Encouraged by the success of the GNU/Linux system, the objective of the DotGNU project is to create a competitor to the .NET framework, a competing system which protects the freedoms of internet users. Here the term "freedoms" refers to the four freedoms which are enumerated in the Free Software Definition as well as (and most importantly) privacy-related freedoms: Some software licensing schemes which have been proposed publicly would make "big brother" type surveillance possible; anyone who is capable of wiretapping the internet would be able to see who uses which programs when.

It is easy to understand that when we want to create a successful Free Software competitor to .NET, we certainly need to use a "copyleft" license. Otherwise, we'd put a lot of work into building an alternative system, and when parts of it become so good that many computer users want to use them, Microsoft and other proprietary software companies could simply insert the code (on which we worked so hard) into their system, without giving computer users any access to the program's source code or the freedom to change it. Our system would thereby lose its competitive advantage, and with almost unlimited financial resources for adding other nice features and for marketing, Microsoft could still build the effective monopoly they want.

In other words, when the goal is to prevent Microsoft from gaining monopolistic control of the software that makes the internet work, Free Software development with non-copyleft licenses cannot possibly help much. It may help a little, and it generally won't hurt, but it certainly cannot bring us any decisive victories. As Professor Bryan Pfaffenberger wrote in his article "Why is Microsoft Attacking the GPL?", if GNU/Linux used a non-copyleft license, "Microsoft would have probably already released a version of Linux, Linux++ or Linux# or L-Nux".

Therefore, any Free Software project which like DotGNU seriously wants to compete with Microsoft needs to use a copyleft license.

My proposal for a resolution of the issue

I propose to change point 3 in section 5 of the policy draft from "may be limited to implementations of the Recommendation, and to what is required by the Recommendation" to something like "may be limited to implementations of the Recommendation, and to what is required by the Recommendation, if at the same time a separate royalty-free license allows use of the patent in all Free Software which is licensed under the terms of the GNU General Public License".

This change would allow all W3C recommendations to be implemented in GPL'd Free Software without directly harming the commercial interests of any proprietary software company.

The W3C still invites the public to comment on the matter of the Royalty-Free Patent Policy via the email address www-patentpolicy-comment@w3.org until April 30, 2003. So far only a few people have commented. Does this mean that the Free Software and Open Source community is not sufficiently aware of this matter? Please contribute to making these issues more widely known.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o World Wide Web Consortium
o press release
o DotGNU
o comment that the planned new rules would lead to a "standards war".
o Free Software Foundation points out on their website
o Free Software Definition
o Open Source Definition
o Royalty-Fr ee Patent Policy
o GNU General Public License
o "copyleft"
o Professor Bryan Pfaffenberger
o "Why is Microsoft Attacking the GPL?"
o the policy draft
o Also by bizcoach


Display: Sort:
New W3C Patents Policy | 47 comments (39 topical, 8 editorial, 0 hidden)
New W3C policy is good enough for me (4.00 / 3) (#7)
by FlipFlop on Wed Apr 02, 2003 at 07:09:50 PM EST

So far only a few people have commented. Does this mean that the Free Software and Open Source community is not sufficiently aware of this threat?

Slashdot posted a story on the topic in December. They posted another story on the the revised policy two weeks ago. The first comment (rated above -1) to that story linked to the FSF's views on the topic. The Free Software Foundation's front page mentions their position on the issue. I suspect the Free Software and Open Source communities are well aware of the issue. Most likely, people just don't find it all that troubling.

If the Open Source and Free Software community whines about the current policy, we will just come off looking like a bunch of extremists loonies. The W3C received an overwhelmingly negative response to the original RAND policy (and rightfully so). The W3C did the right thing by extending the comment deadline. They did the right thing again when they changed to a royalty-free policy. I applaud the W3C for addressing the concerns of Open Source and Free Software developers. The latest policy may not be perfect for Open Source or Free Software. It isn't perfect for commercial interests either. I personally see the latest policy as a good balance between commercial and open source interests.

The FSF should draft a Patent LGPL (PLGPL) which grants the usual rights to the code, but only limited rights to patents. Developers should be able to take code from PLGPL software and use it in LGPL software as long as they don't violate the patents.

AdTI - The think tank that didn't

Free software and patents (4.66 / 3) (#8)
by enterfornone on Wed Apr 02, 2003 at 08:32:43 PM EST

Problem is that a royalty free patent licence still requires an explicit licence. Which means that everyone who wanted to make a fork of the code would need to obtain a licence from the patent holder. It's ridiculous to expect any free software author to do this.

--
efn 26/m/syd
Will sponsor new accounts for porn.
[ Parent ]
Why is that ridiculous? (2.00 / 1) (#9)
by jjayson on Wed Apr 02, 2003 at 09:07:19 PM EST

It seems like a very small burden for using the IP of someone else.
_______
Smile =)
* bt krav magas kitten THE FUCK UP
<bt> Eat Kung Jew, bitch.

[ Parent ]
Well the point of free software... (none / 0) (#10)
by enterfornone on Wed Apr 02, 2003 at 09:09:09 PM EST

...is that you shouldn't have that burden.

--
efn 26/m/syd
Will sponsor new accounts for porn.
[ Parent ]
The licence could be written... (4.00 / 1) (#19)
by Cloaked User on Thu Apr 03, 2003 at 04:51:49 AM EST

...so as to not require each and every author to individually seek permission, just as the GPL is.

Something along the lines of "If you're writing software to be released under an approved Open Source licence, go ahead, consider this to be permission to use our patented stuff. Everyone else, we invite to discuss terms with our legal department", couched in suitable legal terms, should be sufficient, I would have thought. Of course, I'm no lawyer.
--
"What the fuck do you mean 'Are you inspired to come to work'? Of course I'm not 'inspired'. It's a job for God's sake! The money's enough and the work's not so crap that I leave."
[ Parent ]

Amusingly... (2.00 / 1) (#25)
by igor on Thu Apr 03, 2003 at 11:34:41 AM EST

Having a patent license like that would turn BSD/X11-type copyright licences into GPL-type copyright licenses, as it would implement the 'viral' feature of pulling all derivative works into Free Softwareland. The point of the BSD/X11 licenses is that you can use the code for anything (except lawsuits), not just to develop more free software. So, while such a license would fly nicely in GNUland, it'd restrict other Open Source licenses. -jeff

[ Parent ]
It WOULDN'T fly nicely in GNU-land (none / 0) (#41)
by greenrd on Fri Apr 04, 2003 at 08:09:34 AM EST

What people seem to be forgetting is that the GPL explicitly forbids adding additional restrictive terms and conditions. It doesn't matter if those terms and conditions are written by the FSF, and blessed by His Holiness Saint RMS Himself - they would still be GPL-incompatible.


"Capitalism is the absurd belief that the worst of men, for the worst of reasons, will somehow work for the benefit of us all." -- John Maynard Keynes
[ Parent ]

This is enough for GPL software now (5.00 / 1) (#26)
by MfA on Thu Apr 03, 2003 at 12:38:57 PM EST

The problem is that the W3C patent policy allows the license (implicit or explicit) to be restricted to implementations of the W3C standards ... and that is incompatible with the GPL, for good reason.

[ Parent ]
The trouble is (none / 0) (#40)
by greenrd on Fri Apr 04, 2003 at 08:06:17 AM EST

Most software patents are partially or completely bogus and illegitimate. Take the example of the Aspect-Oriented Programming patent (the very concept of which is highly dubious, let alone its implementation - what would you think of a patent on OOP??) - the strategy seems to be to take ridiculously overbroad claims and then progressively narrow them down point by point. I can understand this from a legal point of view - but it's totally unethical and should be smacked down by any competent patent office, which the US patent office clearly is not.


"Capitalism is the absurd belief that the worst of men, for the worst of reasons, will somehow work for the benefit of us all." -- John Maynard Keynes
[ Parent ]

-1 GNU Hippie Whinging BULLSHIT (1.63 / 22) (#12)
by Talez on Wed Apr 02, 2003 at 09:26:05 PM EST

Before I start, a note to the GNU activists out there. If you think I'm wrong, feel free to comment. If you're feeling cowardly, feel free to moderate me a 1 and move on.

Now all I hear from the FSF is:

"If we can't have EVERYTHING FREE then we're going to pack up our ball and go home!"

Another example of GNU's inability to coexist with anything that isn't 100% socialist.

"Field of Use", as the FSF software calls it, is required to ensure that people who use these patents don't go off and use them to enhance their product's value while not paying royalties they might otherwise have to pay.

Someone invents (at great cost) a new way of compressing images losslessly down to 1/100th the size of anything available right now. In the spirit of helping humanity, this person releases the technique so that it can be used in web browsers to decode images.

Now, without that "field of use", Adobe can come along and say "We'll just use the WC3 patent to implement decoding these images. Sure we won't be able to save in them but we can let people open them and then save them in something else like JPEG".

Adobe Photoshop would then increase its value without paying any royalties to the inventor. In the capitalist world we consider this to be unfair.

Back to FSF's example. You can't release the code because the license would have a field of use condition. Boo hoo. Take out the offending piece of code, make the offending piece of code into a seperate non-free project and link it against the original piece of code.

Problem Solved.

Si in Googlis non est, ergo non est

Utter bollocks. (3.71 / 7) (#14)
by pwhysall on Thu Apr 03, 2003 at 12:27:02 AM EST

But then, being misinformed about the FSF and the GPL is SOP around here.

Hey! You can SELL GPLed software! You can MAKE FILTHY LUCRE out of it! IT EVEN SAYS SO RIGHT THERE IN THE LICENSE!

But then, actually READING it would be a bit of a stretch, no?

Your example is crap, too. The inventor of said compression technique made a conscious decision to NOT profit from his invention. So Adobe use it. So they don't pay him. So what?

"In the spirit of helping humanity" and "capitalism" are opposing forces.

The Culture has a saying - "Money is a sign of poverty".
--
Peter
K5 Editors
I'm going to wager that the story keeps getting dumped because it is a steaming pile of badly formatted fool-meme.
CheeseBurgerBrown
[ Parent ]

Selling GPL software (4.00 / 2) (#15)
by Cloaked User on Thu Apr 03, 2003 at 02:08:41 AM EST

That's true, you can do that. However, there's still no restrictions placed on redistribution - the first person you sell it to is free to give it away to anyone who wants it (or sell it for a lower price, etc).

Therefore, whilst it's true that you can sell GPLed stuff, I wouldn't count on trying to make much money that way. The only way I can see to make money from GPLed software is in charging for support and bespoke development work - updates, patches, etc. I don't see either of those things having that large a market, though - certainly, home users are not going to want to pay for either (or the software itself, for that matter, if they can get it for free elsewhere).
--
"What the fuck do you mean 'Are you inspired to come to work'? Of course I'm not 'inspired'. It's a job for God's sake! The money's enough and the work's not so crap that I leave."
[ Parent ]

I get paid for that all the time (5.00 / 1) (#31)
by hardburn on Thu Apr 03, 2003 at 01:45:26 PM EST

I sometimes do consulting work. I get paid to support Free Software all the time. I know I'm not alone. Every GNU/Linux sysadmin on earth is getting paid to support Free Software.

My regular job is a web devloper (mostly writing Perl all day). I recently wrote the CGI::Search Perl module to simplify many of our flat-text database searches on our web site. We're not a software company, and releasing CGI::Search publically wasn't going to hurt us. It could very well benifit us if other people use it and find bugs, add new features, etc. So I got paid to write Free Software.

What is so easy to forget is that the ammount of software on the shelf at Best Buy makes up a very tiny precentage of the actual ammount of software produced. Most of the software is written internally by companies. Much of this software is specific to the buisiness rules of the company, and would therefore be of no use anywhere else. In that case, there is no point in either selling it or making a Free Software release. However, some of that software is more generic (like my CGI::Search example), but the company doesn't want to support it as a purchased product. In that case, a Free Software release makes sense.


----
while($story = K5::Story->new()) { $story->vote(-1) if($story->section() == $POLITICS); }


[ Parent ]
Oh bullshit (1.00 / 4) (#16)
by Talez on Thu Apr 03, 2003 at 02:32:57 AM EST

Hey! You can SELL GPLed software! You can MAKE FILTHY LUCRE out of it! IT EVEN SAYS SO RIGHT THERE IN THE LICENSE!

But then, actually READING it would be a bit of a stretch, no?

Peter fakes with the "I've got something interesting to say" and results in a "Let's spout random bullshit about other people that has nothing to do with the argument".

Your example is crap, too. The inventor of said compression technique made a conscious decision to NOT profit from his invention. So Adobe use it. So they don't pay him. So what?

Right into a "see the world in black and white" combo!

Tonight he's on a hot streak!

"In the spirit of helping humanity" and "capitalism" are opposing forces.

From downtown the "you can't profit form something helpful" argument.

The Culture has a saying - "Money is a sign of poverty".

It's going in! That obscure quote seem to put his post right on course with the basket!

But seriously.

You added absolutely nothing to my post. You simply spouted some bullshit about common mistakes and propsed that I didn't read the license 50,000 times.

Then you proceed to say "well if he gives up the right to make money in THIS area, he has to give up making money in ALL areas" which is just totally unreasonable it makes me sick just thinking about it.

Not to mention you fail to address and problems with my solution involving linking against the patented component.

You're typically full of sensible debate filled with ideas based on reason rather than outlandish remarks; are you on crack or something?

Si in Googlis non est, ergo non est
[ Parent ]

Re (4.00 / 2) (#17)
by djotto on Thu Apr 03, 2003 at 03:07:20 AM EST

Someone invents (at great cost) a new way of compressing images losslessly down to 1/100th the size of anything available right now.

In the "capitalist world", we don't patent maths. Or at least, we didn't used to.

Now, without that "field of use", Adobe can come along and say "We'll just use the WC3 patent to implement decoding these images. Sure we won't be able to save in them but we can let people open them and then save them in something else like JPEG".

Adobe can't incorporate OSS code without opening the rest of Photoshop.



[ Parent ]
Re: Re (5.00 / 1) (#18)
by Talez on Thu Apr 03, 2003 at 04:36:58 AM EST

In the "capitalist world", we don't patent maths. Or at least, we didn't used to.

If the algorithm satifies the Freeman-Walter-Abele test its patentable. While it may be unlikely, lets assume for the sake of argument it is.

Adobe can't incorporate OSS code without opening the rest of Photoshop.

You seem to have copyright and patents confused. This example has nothing to do with OSS. If a patent holder is forced to license something freely with the W3C then anyone can get at it and implement it. Whether or not it has anything to do with the W3C is irrelevant.

In the case of Adobe, using a decompression algorithm to provide support for a new format has nothing to do with the W3C and enhances Photoshop's value while not returning anything to the patent holders.

This is the kind of hijacking crap "field of use" tries to stop.

Si in Googlis non est, ergo non est
[ Parent ]

Re. Re: Re (none / 0) (#23)
by bizcoach on Thu Apr 03, 2003 at 08:12:46 AM EST

I think djotto's comments are in the conext of my suggestion to amend the clause which allows "field of use" restrictions with the condition that then there needs to be, in addition, a separate patent license which allows to freely implement the technology in GPL'd programs. This would allow to prevent what you call "hijacking crap" (precisely because Abobe can't incorporate GPL'd code into Photoshop without opening the rest of Photoshop) without excluding DotGNU from implementing W3C recommendations.

[ Parent ]
But (none / 0) (#24)
by Talez on Thu Apr 03, 2003 at 08:37:15 AM EST

What you have to remember is that the application of "field of use" restrictions (at least in my understanding) is only where the patent in question has been invalidated by the working group.

In essence the W3C is saying "We've fucked up. We've got no other alternative. We'll do anything you want so long as you keep letting us use this stuff for free."

What then? Is the W3C is supposed to dump technology because its simply not compatible with the GPL?

Yes it's possible to add special cases to licenses to cover the GPL but its not appropriate for W3C policy to dictate how companies license patents that are already infringed upon. You really have to take up that kind of stuff with the owners of the patents.

Si in Googlis non est, ergo non est
[ Parent ]

Huh? (none / 0) (#30)
by MfA on Thu Apr 03, 2003 at 01:44:02 PM EST

Companies dont have to give out patent licenses for anything. This is meant for contributors to get their patents in the standards. If a patent is infringed by the standard and the owner does not want to license it under the W3C policy then the standard will have to be retracted (same situation as ISO is in with JPEG).

What that policy is is completely arbitrary in that case, there might be a slightly higher chance that if such patents are found they might be licensed under this new scheme ... but then again, it is also a stick to beat free software with.

As you say yourself "its not appropriate for W3C policy to dictate how companies license patents that are already infringed upon". Logic left you for a moment when you implied a policy requiring GPL compatible licensing to be any more forced than RF :) (Or even RAND for that matter.)

[ Parent ]

Ok (none / 0) (#34)
by djotto on Thu Apr 03, 2003 at 05:39:55 PM EST

On the 2nd point - I understand what you're saying now. That without these "field of use" clauses, W3C would essentially be asking the patent holder to donate their patent to the public domain. That's fair enough, though I suspect there are ways to work within the letter of the law... build a webdav client into Photoshop and say "hey, it's a browser now!"

On the 1st point (and drifting from the topic here) - I still believe that math is/should be unpatentable... business process be damned. The whole underpinning of the patent system is broken - a patent is granted nationally but enforced internationally, which inevitably leads to governments allowing any old crap to be patented, on the off-chance it might pull in some money some day. Bad system in need of overhaul.

[ Parent ]

The only thing worse that a dirty GNU hippie... (2.66 / 6) (#20)
by Theranthrope on Thu Apr 03, 2003 at 05:31:12 AM EST

...is a dirty anti-GNU troll whining loudly about dirty GNU hippies.

hmmm... It smells like... hummis!
"Turmeric applied as a suppository will increase intelligence." -- HidingMyName
[
Parent ]
Patents and Inventions (5.00 / 1) (#21)
by Argon on Thu Apr 03, 2003 at 05:34:07 AM EST

I agree that some merit and retribution should be given to the ones that invent something.
However I still have some doubts on the entire system.
One big issue is that no invention is created on vacuum. So, what should we do? Pay tribute to the original inventors?

For instance, if I invented math would I receive tribute for Physics? Or for any engineered project that used my invention?

This is a complex issue, and simple capitalist rules may no apply. Knowledge and information is not something that can be weighted and measured the same way any physical substance.

So in some way, the FSF quest makes sense, even if it is against your capitalist view of the world. Please remember that Capitalism is not the answer for all problems, and is not the ultimate system. Keep your mind open for new systems.

[ Parent ]

GPL doesn't allow linking with non-free code (4.50 / 2) (#22)
by bizcoach on Thu Apr 03, 2003 at 08:03:19 AM EST

A strong copyleft license like the GPL doesn't allow linking with non-free code like Talez suggests. In my article I don't go into the difference between strong (i.e. GPL-style) copyleft and weak (i.e. LGPL-style) copyleft, but the argument which I present really shows that in order to accomplish its objective, DotGNU needs to use strong copyleft licensing.

[ Parent ]
Interoperability (4.50 / 2) (#29)
by hardburn on Thu Apr 03, 2003 at 01:34:48 PM EST

Take out the offending piece of code . . .

This would be great, except that you lose interoperability. In your image compression example, people will no doubt spread such images everywhere on the web. As soon as that happens, Free Software cannot read a good chunk of the images on the web due to patent restrictions. You are forced into using non-Free Software if you want to view them.

This is where problems with patents and copyright on various networking technologies happen. If you want to use a patented system on your computer and never have it leave your hard drive, that's your choice. But as soon as that system has to be used outside your computer, you've now forced someone to use that patented system.


----
while($story = K5::Story->new()) { $story->vote(-1) if($story->section() == $POLITICS); }


[ Parent ]
Cannot link to non GPL software. (3.00 / 1) (#44)
by Alhazred on Fri Apr 04, 2003 at 02:53:58 PM EST

GPLed code cannot link to non-GPL code dude. If it could then I could simply make 99% of my program closed source, link a GPL shell with it, and call it 'Open Source Software', and that would give me the right to incorporate other people's TRUELY open source code into my project? While my real contribution remains hidden? Doesn't work that way.

Thus your 'problem solved' is 'un-solved'.

And that brings up the real point. My use of the GPL is intended to mean 'I will be open IF YOU WILL BE OPEN TOO'. Even capitalism understands reciprocity. Trying to keep your part to yourself is just plain cheating.
That is not dead which may eternal lie And with strange aeons death itself may die.
[ Parent ]

GPL code can link to proprietary code (4.00 / 1) (#46)
by werner on Mon Apr 07, 2003 at 05:02:50 AM EST

but with restrictions. Your proprietary code should only link to the documented API and it should also have considerable functionality of its own, i.e. not be wholly dependent on the GPL code.

That wouldn't really apply here, as I think we are talking more about other-licenced code running as a plugin or extension to a GPL'ed app. The GPL'd app is using proprietary code, not vice versa. I don't see how this is any different to installing a Flash plugin in Mozilla.

[ Parent ]

Some might even question that... (none / 0) (#47)
by Alhazred on Tue Apr 08, 2003 at 02:06:56 PM EST

I think the basic criteria is "is the program useable without the proprietary part?". If it is, then by writing it I've made, within the open part, a useful contribution of code, which probably warrants being allowed under the GPL.

Another criteria might be "how useful is the proprietary part on its own?" The aformentioned Flash plugin for example does work in mozilla, but it also works in Netscape and other non-free browsers, like Opera, as well as in other free ones, like Konqueror. That also tends, IMHO, to indicate that the author's purpose in putting functionality into a plugin was to make modular software, not mearly to subvert the GPL.

And finally of course note that the flash plugin was not written by the Mozilla developers. In fact even if they abhorred its existence they could hardly stop someone from making and distributing a proprietary plugin that adheres to their APIs. They might well create license restrictions theoretically preventing users from using such a plugin, but I hardly think it would be effective.

What I was arguing was that the GPL MUST not allow people to 'shroud' most of their functionality inside a non-free part. In point of fact if someone were to make 1/2 of their program non-free, it could still be questionable if the free part did something useful but was obviously crafted such that the non-free part could have been better built directly in as additional free functionality. It is a judgement call, but ultimately should be judged by what is fair and reciprocal. Thats what the GPL is about. Unfortunately in today's world thats NOT what the legal system is about.
That is not dead which may eternal lie And with strange aeons death itself may die.
[ Parent ]

But then, you wouldn't need a patent (none / 0) (#27)
by hardburn on Thu Apr 03, 2003 at 01:28:03 PM EST

If you're going to allow anybody to make use of your invention, you don't need a patent to apply copyleft principals to it.

I know how some people think on this. You have an invention, and you want make sure it stays free for anyone to modify and use without the worry that somebody else will patent the same device at a later time. So you patent it and give it away, saying "anyone is allowed to use and modify this device".

Well, you can go about it that way, but you actually don't have to bother with filling out the patent application. You only need to document your device and release the documents publically. If you want a legal assurance, you talk to a lawyer who can keep dead-tree versions of the documents on file, signed and dated by you (the price for doing this is probably a lot less than paying the same lawyer to walk you through the patent application). If anybody tries to patent the same device, you have evidiance that you were there first, and can thus invalidate their patent claim.

Of course, IANAL.


----
while($story = K5::Story->new()) { $story->vote(-1) if($story->section() == $POLITICS); }


Say: 'GPL is ok, but for proprietary you pay!' (none / 0) (#43)
by bizcoach on Fri Apr 04, 2003 at 08:28:39 AM EST

You can can patent a technology and then make it publicly available royalty-free under a license that allows implementing the technology in any GPL'd program.

Implementing the technology in proprietary software would be allowed only under a different patent license (one which is not royalty-free.)

[ Parent ]

Free Software is more flexible than you think (2.00 / 1) (#28)
by lordcorusa on Thu Apr 03, 2003 at 01:31:34 PM EST

Both the original poster and some of the commenters seem to think that Free Software is much less flexible than it really is. In reality, you can get 99.9% of the benefits of GPL software with only a single little change in the license. Forgive me if someone has pointed this  out deep in one of the threads, but I think it deserves greater emphasis.

If you want to implement a GPLed browser rendering engine, for example, and there is a W3C royalty-free patent in the standard, all you have to do is include in your project and code files the following comment:

/***********************************************
This code is released under the terms of the GNU GPL version 2 or later, with the following exception:

W3C standard number ABC contains a patented method XYZ which is released under a royalty-free license only for the purposes of implementing standard number ABC. Code for this method may not be used for any purpose other than implementing relevant W3C standards.
************************************************/

That's all. It would also be appropriate to restate this warning immediately before the specific code dealing with the patent. Such exceptions to the vanilla GPL are used routinely in other Free Software projects. GNU Classpath is one example, in that it allows non-GPL software to link (in any way) to it.

Of course, the most zealous of GPL advocates will not accept this as free software. However, it will protect your code under terms essentially identical to the GPL, and it will still grant others leeway to modify your code. The only thing they will not be allowed to do is to copy one particular piece of your code (the patent implementation) into an unrelated program.

Most reasonable people should be satisfied with this. You should still be able to link other GPLed code in with yours.

I am not saying that I disagree with the original poster's proposed draft changes. I agree and am preparing to submit a comment to W3C saying much the same thing. However, if the W3C keeps the draft as it currently is, I have shown that this is not the end of the world for Free Software.

Good luck getting a change (5.00 / 1) (#32)
by hardburn on Thu Apr 03, 2003 at 01:53:42 PM EST

Take the Mozilla project, for example (this is an excelent example, since they are a high-profile Free Software project that is quite likely to be effected by whatever the W3C decides). Mozilla has had legions of programmers working on it over the years. I'm sure there are some programmers that have been there since Netscape first released the code, and there are probably other programmers that dropped in a single bug patch a few years ago and were never seen again.

For your license addition to be legally acceptable, you would have to contact every single programmer who has ever contributed code, be it a single-line bugfix up to the main writers of the rendering engine, and get their permission for the addition to be applied. This task is so large that you might as well give up now.

Changes in licensing aren't a big deal for newer or smaller projects, but for something as big as Mozilla, forget it.


----
while($story = K5::Story->new()) { $story->vote(-1) if($story->section() == $POLITICS); }


[ Parent ]
Wrong (none / 0) (#38)
by greenrd on Fri Apr 04, 2003 at 05:45:09 AM EST

Such exceptions to the vanilla GPL are used routinely in other Free Software projects. GNU Classpath is one example, in that it allows non-GPL software to link (in any way) to it.

The big difference is, those give you more rights, not less.

You should still be able to link other GPLed code in with yours.

No, you won't be able to. The GPL mandates that no further conditions are imposed on the recipient beyond its own conditions. It wouldn't do its job if it didn't!

However, this is a bit academic since patent infringements are still patent infringements whether or not they were "allowed" by the license.


"Capitalism is the absurd belief that the worst of men, for the worst of reasons, will somehow work for the benefit of us all." -- John Maynard Keynes
[ Parent ]

Not totally wrong, but not really helpful (none / 0) (#42)
by bizcoach on Fri Apr 04, 2003 at 08:11:16 AM EST

While the precise exception text which lordcorusa suggests wouldn't work (that text is not GPL-compatible and hence it would result in licensing that does not allow anyone besides the copyright holder to make copies and distribute them), in principle something like what he suggests could be done.

The language to use would be something like "as a special exception, the field-of-use restriction of the royalty-free patent license XYZ for US patent no. 798654 shall not be considered as something that restricts you from copying and redistributing this program".

Other GPL'd projects can link to that code if and only if they add similar language. This could easily create huge practical problems for everyone who wants to re-use copylefted code for a purpose which is different from the purpose for which the code was originally written.

Those practical issues are not the only problem with this idea.

The biggest problem is that this approach would be that it would leave the Free Software movement very vulnerable to embrace-and-entend attacks on the standards from proprietary software companies. Suppose that a W3C standard gets widely adopted for which essential technology is patented, and that there is a royalty-free patent license with a field-of-use restriction. The field-of-use restriction on the royalty-free license means that extensions of the standard will be implemented only in proprietary software. Let's not forget that the company which currently present the greatest threat is a master at embrace and extend tactics. Currently they're in a phase of embracing the notion of publicly-available standards, but at the same time they would like to have exclusive rights to extend these standards later.

So, yes, in a way it is true that it is possible to work around the "cannot implement because we're using the GPL" problem, but should we do this when this means giving the proprietary software companies exclusive rights to extend those standards later?

No!

[ Parent ]

interesting (none / 0) (#33)
by lordcorusa on Thu Apr 03, 2003 at 03:03:22 PM EST

You bring up an interesting point. I suppose the only thing for new projects to do to protect themselves is to demand that the central organization or person in control of the project be assigned "ownership" of all of the patches submitted to be included in the the project proper. Operating under whatever charter the organization uses, license changes could be made without such a difficult and time consuming process.

For example, lets say that hypothetically, the Mozilla Organization manages to gain copyright ownership over all code in its tree. We'll assume that the organization's charter has some rules about what the organization can do with that code, ie: require a vote of actively contributing members to make a licensing change. At that point, making a change as I suggested would require only a vote, which could be conducted in a few weeks. As long as a quorum of active members voted, and assuming it passed, the change would be made immediately.

Now this will not help existing projects like Mozilla, which will have to undertake another long process to get this centralized control approved by all patch owners, but we should learn fron the mistakes of others when creating new projects.

Now that you bring up the subject of Mozilla license changes, I do remember that fiasco (which I believe is still technically not fully resolved). While I and (I'm sure) many others cringe at the risks of centralized ownership of major projects, the kind of inflexibility shown by the Mozilla example could be disasterous should something like this patent policy go through. Remember, all systems in the universe, be they natural or artificial, must be able to adapt to new conditions, or they run the risk of dying out.

Perhaps this patent threat will give us all a new parameter to be used when we evaluate the way we manage control over our Free Software projects.

Flexibility for later license changes (none / 0) (#39)
by bizcoach on Fri Apr 04, 2003 at 06:05:34 AM EST

lordcorusa wrote: "Remember, all systems in the universe, be they natural or artificial, must be able to adapt to new conditions, or they run the risk of dying out."

I agree.

What we do in DotGNU is that all code is licensed by with the words "you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version."

This means that for example, if a royalty-free patent license becomes popular which in principle is not incompatible with Free Software in general, but which happens to be incompatible with the GPL, then that incompatibility can be resolved by releasing a new version of the GPL.

"Field of use" restrictions cannot be addressed in this way, because that would defeat the whole point of the GPL as a copyleft license.

[ Parent ]

The GPL (4.00 / 2) (#35)
by Anonymous Brave on Thu Apr 03, 2003 at 08:06:55 PM EST

I usually prefer BSD over GPL precisely because no requirement of just using the code in open-source projects is made.

In here this requirement is presented as an advantage: "we'd put a lot of work into building an alternative system, and when parts of it become so good that many computer users want to use them, Microsoft and other proprietary software companies could simply insert the code [...] into their system, without giving computer users any access to the program's source code or the freedom to change it".

My question is "so, what's the big problem?"

I just consider due credit should be given, and BSD requires that.

correspondente.net - reflectir e discutir em portuguÍs

The freedom to change it (5.00 / 1) (#36)
by hardburn on Thu Apr 03, 2003 at 11:37:24 PM EST

I want complete access to the source code of everything running on my system. I want to make sure any code I write will always stay free to change it, because otherwise I wouldn't bother making it into Free Software (I'd close up the source and charge for it).

If you don't personally care about the licensing status of software on your system, or about what happens to your own code later on, that is up to you. Personally, I like the inheirant flexibility of having access to the source, and I want to keep it that way.

Further, I don't want to be forced into using non-free software just because somebody else sends me something in a closed format (Word docs being the typical example). The W3C's patent policies work against this principle. The W3C is all about the web, and the web is just a dream unless you have more than one computer working together in a network. Therefore, anything the W3C does in regards to patents will directly affect anyone who uses the web.


----
while($story = K5::Story->new()) { $story->vote(-1) if($story->section() == $POLITICS); }


[ Parent ]
Towards a viable Free Spoftware industry (5.00 / 1) (#37)
by bizcoach on Fri Apr 04, 2003 at 05:42:23 AM EST

Anonymous Brave asks "so, what's the big problem?"

The DotGNU project was born out of the vision to create a viable Free Software industry. Of course, besides Free Software businesses, in the Free Software community there will always also be room for hobbyists and for prophets and for fanatics. (The difference between a "prophet" and a "fanatic" is that the message of the prophet makes sense, while the fanatic doesn't know what he's talking about.)

Now I'll freely admit that theoretically, it would be possible for such a Free Software industry to use a BSD license. The problem is the lack of defenses against embrace and extend attacks from proprietary software companies with monopolistic ambitions. Suppose there was a successful community of Free Software businesses around the FreeBSD project. Now imagine that this BSD-based Free Software industry becomes so successful that it seriously threatens Microsoft's business position. Then Microsoft could easily take this successful system, make a few minor improvements, and turn it into a proprietary product. Whenever improvements are made to the BSD-licensed Free codebase, Microsoft will be able to easily incorporate them into their system. At the same time, Microsoft can make improvements to the proprietary system which the Free Software businesses would find difficult to duplicate. Besides patent-related obstacles, the Free Software companies would be seriously disadvantaged because they have to re-implement every feature, while Microsoft has the right to simply copy-and-paste their code.

For this reason, I think that a Free Software industry needs to be based on copyleft licensing in order to be viable in the long run.

Hence I claim that it's important to have the right to implement in GPL'd Free Software everything that is considered to be a standard.

[ Parent ]

The big problem is (none / 0) (#45)
by werner on Mon Apr 07, 2003 at 04:50:10 AM EST

many developers fail to see why they should let companies like Microsoft appropriate and market something which they have done for free.

If I spend many hundreds of hours writing a program which I decide to give away for free, I am hardly going to be enraptured if a company decides they want to start selling my code as their own at virtually no cost to them.

Imagine how much easier Microsoft's dirty little strategies would be to implement if all they had to do was take Gecko and extend it a bit and break it a bit rather than having to build or buy their own broken rendering libraries. I fail to see the point in helping a company like that walk all over everyone.

[ Parent ]

New W3C Patents Policy | 47 comments (39 topical, 8 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest © 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!