Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Thu, Nov 20, 2003 at 09:34:01AM +0100, Arnoud Engelfriet wrote: Branden Robinson wrote: I'd sure like to know what Eben Moglen thinks about this issue. He submitted comments on behalf of the FSF on November 14. See: http://nagoya.apache.org/eyebrowse/[EMAIL PROTECTED]msgId=1127301 Quote: FSF believes that broad automatic termination provisions like that contained in the first sentence of section 5 are fraught with potentially serious unintended consequences, and are not an appropriate vehicle for protecting the freedom of free software against the serious threat posed by software patent litigation. Thanks a lot for bringing this to my attention. -- G. Branden Robinson| Debian GNU/Linux | Extra territorium jus dicenti [EMAIL PROTECTED] | impune non paretur. http://people.debian.org/~branden/ | signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Fri, Nov 14, 2003 at 07:43:01PM -0500, Brian T. Sniffen wrote: There is also no way to be sure that the next minor upstream Emacs release will still be entirely free software, and Debian has been bitten by this before. So why not move everything to non-free which is not under a GPL, version 2 only license? Yes, that follows quite clearly. People don't share your opinion, so resort to hysterically absurd arguments. That the GNU FDL is not DFSG-free tells us nothing about the DFSG-freeness of *any* other license. -- G. Branden Robinson| I suspect Linus wrote that in a Debian GNU/Linux | complicated way only to be able to [EMAIL PROTECTED] | have that comment in there. http://people.debian.org/~branden/ | -- Lars Wirzenius signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Sun, Nov 16, 2003 at 02:46:37AM -0500, Anthony DeRobertis wrote: I think you must look at the entire picture --- not just the copyright one --- to determine if software is free. I don't think its free if the copyright holder decides to use patents, instead of copyright, to limit your freedoms. In short: * If party A releases software, and in some maner prevents you from excercising your DFSG-freedoms, then that software is not DFSG-free. It doesn't matter if that manner is copyright, patent, or the mafia. * If party A releases software, and allows you all your DFSG-freedoms but an unrelated party B does not allow you to modify it to, e.g., play DVDs, party A's software is still free. To put it yet another way, you can't use patents as an end-run around freeness. I entirely agree. It's not always easy to perform this sort of analysis, and we may be tricked from time to time through secret conspiracies between separate organizations, one of whom holds a copyright and the other who holds a patent implemented by the copyrighted code. It would not have been fair to penalize the authors of the GIMP for the hostile and malicious actions of Unisys regarding the LZW patent, so we did not. The case where the same organization holds a copyright and an applicable patent is a no-brainer, though. Both the patent and the copyright licenses must satisfy the DFSG, and furthermore, any subsequent transfer of either or both the copyright or patent to other parties must be closely scrunitized. As a rule of thumb, I'd say any such transfer should not be regarded as materially affecting the situation. A hypothetical: If Microsoft had a patent (free for use in web browsers only) on HTTP and had licensed Internet Explorer under the GNU GPL, IE would have to go in non-free. If they then sold their patent to IBM, the situation would not change unless IBM then went and made the patent license fully DFSG-free. After all, we really have no way of knowing whether would Microsoft offered the HTTP patent to IBM for a steeply discounted price if IBM would agree to never relax the license terms on the patent. From what I recall of the Microsoft federal antitrust suit in the U.S., this sort of arrangment would be entirely within Microsoft's character. If a single organization (or multiple organizations when there is reason to believe collusion has taken place) asserts non-free copyright or patent licenses applicable to the same work, then the copyrighted implementation must be regarded as permanently tainted until both the copyright and patent license are made unambiguously DFSG-free. Organizations need to pay more than just lip service to freedom for us to recognize their participation in the Free Software community. Freedom which you cannot exercise is not freedom at all, and those responsible for preventing the exercise of freedom must be held to account. (Yes, I've been reading Chomsky lately. Does it show? :) ) -- G. Branden Robinson|Optimists believe we live in the Debian GNU/Linux |best of all possible worlds. [EMAIL PROTECTED] |Pessimists are afraid the optimists http://people.debian.org/~branden/ |are right about that. signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Fri, Nov 14, 2003 at 07:45:04PM -0500, Brian T. Sniffen wrote: In the current patent-litigation context, a large stable of patents to cross-license is considered a vitally important corporate defense strategy. *shrug* That's not our problem. President Bush considers a missile defense shield a vitally important military defense strategy. That doesn't mean he's right, or that he deserves our support. -- G. Branden Robinson|I have a truly elegant proof of the Debian GNU/Linux |above, but it is too long to fit [EMAIL PROTECTED] |into this .signature file. http://people.debian.org/~branden/ | signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 17, 2003 at 10:16:43AM -0600, John Goerzen wrote: On Mon, Nov 17, 2003 at 10:43:01AM -0500, Glenn Maynard wrote: However, this is essentially what the reciprocal patent clause is requiring. As part of the Apache license, you must agree not to sue any contributor for any of your software patents, for as long as you continue to use Apache. The only problem I see here is return fire: if I'm holding patents as a defense strategy, I want to be able to use them to return fire if an Apache contributor decides to attack me with his own patents, unrelated to Apache. This is only useful if you do not have a valid defense for the problem already. In other words, it is only useful as a strong-arm tactic to let your own company effectively ignore patents of others. After all, if the lawsuit filed against you has no merit, you don't need a patent portfolio to defend against it. So, its only real purpose is to let the patent holders thwart the patent law. I don't like that one bit. Well, I think it's probably intended more as a technique to force the dispute out of the courts and into settlement. Maybe that's what you mean. Given the general impression of the USPTO over the past decade or so, it may be that a significant number of patents in the U.S. would be invalidated if they were ever brought to litigation. In which case, the reciprocity clause actually strengthens the existing broken patent regime by preventing bullshit software patents[1] from being litigated at all, rather than promoting the development of a pool of patents available to the Free Software community (or any other supposedly benevolent purpose). I'd sure like to know what Eben Moglen thinks about this issue. [1] This phrase may be redundant. -- G. Branden Robinson| Reality is what refuses to go away Debian GNU/Linux | when I stop believing in it. [EMAIL PROTECTED] | -- Philip K. Dick http://people.debian.org/~branden/ | signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Scripsit Branden Robinson [EMAIL PROTECTED] On Fri, Nov 14, 2003 at 07:43:01PM -0500, Brian T. Sniffen wrote: There is also no way to be sure that the next minor upstream Emacs release will still be entirely free software, and Debian has been bitten by this before. So why not move everything to non-free which is not under a GPL, version 2 only license? That the GNU FDL is not DFSG-free tells us nothing about the DFSG-freeness of *any* other license. Um, the GFDL was not a part of that debate at all. Brian was responding to some opinions I had about Apache's apparent intent to knowingly include patent-encumbered algorithms in their product. He was saying, by a fairly usual reductio-ad-absurdum argument, that he did not find my reasoning convincing. Even though I still think my point was valind, I don't find his counterargument hysterically absurd. -- Henning Makholm Den nyttige hjemmedatamat er og forbliver en myte. Generelt kan der ikke peges på databehandlingsopgaver af en sådan størrelsesorden og af en karaktér, som berettiger forestillingerne om den nye hjemme- og husholdningsteknologi.
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Branden Robinson [EMAIL PROTECTED] writes: On Fri, Nov 14, 2003 at 07:45:04PM -0500, Brian T. Sniffen wrote: In the current patent-litigation context, a large stable of patents to cross-license is considered a vitally important corporate defense strategy. *shrug* That's not our problem. President Bush considers a missile defense shield a vitally important military defense strategy. That doesn't mean he's right, or that he deserves our support. There's a difference between lack of support, which I endorse, and active opposition. A license which has a cost to anyone holding a software patent, as the currently proposed Apache license, is non-free. -Brian -- Brian T. Sniffen[EMAIL PROTECTED] http://www.evenmere.org/~bts/
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Henning Makholm [EMAIL PROTECTED] writes: Scripsit Branden Robinson [EMAIL PROTECTED] On Fri, Nov 14, 2003 at 07:43:01PM -0500, Brian T. Sniffen wrote: There is also no way to be sure that the next minor upstream Emacs release will still be entirely free software, and Debian has been bitten by this before. So why not move everything to non-free which is not under a GPL, version 2 only license? That the GNU FDL is not DFSG-free tells us nothing about the DFSG-freeness of *any* other license. Um, the GFDL was not a part of that debate at all. Brian was responding to some opinions I had about Apache's apparent intent to knowingly include patent-encumbered algorithms in their product. He was saying, by a fairly usual reductio-ad-absurdum argument, that he did not find my reasoning convincing. Even though I still think my point was valind, I don't find his counterargument hysterically absurd. I try to be only hysterical *or* absurd, and never both at once. Fire hose. My original intent was to express this opinion: that software should not be put into main or non-free based on its potential future freeness, but on its freeness today. If that state changes, it can be moved -- though this is unlikely, since most free licenses cannot suddenly become non-free licenses (patent grants justify that most). Aardvark. By retaining absurdity, I hope to avoid hysterics. v.42bis High-Security Streaming Pants. -Brian -- Brian T. Sniffen[EMAIL PROTECTED] http://www.evenmere.org/~bts/
(OT) Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Ken Arromdee said on Mon, Nov 17, 2003 at 04:20:27PM -0800,: by which you could create it. I find it highly unlikely that patent lawyers cost appreciably more than software developers) (snip) But that's not cheap. Going to law school costs a lot of money. Becoming a software developer only requires buying a computer, which costs a lot less. There is at least one case of somebody from IBM going on to be a patent attorney. See:- http://www.fenwick.com/attorneys/4.2.2.asp?aid=435 Googling consequent to the discovery that the link mentioned in the Halloween IX docs on SCO revealed the above link. (see the comments about para 111, the part about IBM not releasing code without an IP audit.) -- +~+ Mahesh T. Pai, LL.M., 'NANDINI', S. R. M. Road, Ernakulam, Cochin-682018, Kerala, India. http://in.geocities.com/paivakil +~+
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Nov 17, 2003, at 11:16, John Goerzen wrote: This is only useful if you do not have a valid defense for the problem already. In other words, it is only useful as a strong-arm tactic to let your own company effectively ignore patents of others. After all, if the lawsuit filed against you has no merit, you don't need a patent portfolio to defend against it. I have seen estimates of the cost of litigating patents in the rang of one and a half to two million dollars (quick google search citations below). That's why people use patents defensively --- because it costs way to much to litigate! http://www.cafezine.com/index_article.asp?Id=627deptid=4 $2 million through trial http://www.atkinson-burrington.com/pressreleases/ cost_patent_litigation.htm £100,000-£1,000,000 through trial http://www.hunton.com/pdfs/article/Risk_Reward.pdf $1 million for discovery alone. $1.5 million through trial. Includes footnotes with explanations and citation
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Nov 17, 2003, at 13:35, Andrew Suffield wrote: On Mon, Nov 17, 2003 at 10:15:40AM -0800, Ken Arromdee wrote: I'm a software developer. So the services of one may, under some circumstances, cost me nothing at all (except my spare time). I don't think patent lawyers can get cheaper than this. You could go to law school and become a patent lawyer. Possibly, he has morals ;-)
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Sat, Nov 15, 2003 at 12:19:35AM +0100, Henning Makholm wrote: The argument proposed was attempting to say No company is ever going to grant free patent licenses; I pointed out the argument applies equally to software And I point out that it doesn't. If the company patent their invention at all, it must be because they intend to restrict people from using it (or at least keep an option open for using the patent to restrict what people do). If they do not intend that, why would they apply for a patent at all in the first place? Are you attempting to suggest that because companies choose to spend money on gaining patent licenses, they should be permitted to restrict people from using it? If not, then I can't see what the relevance is. They can either avoid taking out a patent, or they can grant a free license for its use. I don't care which. There's no excuse for them taking out a patent and then only granting a non-free license for its use. Finally, it is totally unacceptable to tie this into a software copyright license, such that accepting the license affects the status of your own patents. That's non-free however you look at it. (And this still applies just as much to software licenses. It is *hard* to gain a copyright license; you have to create the work. To gain a software patent, you merely have to describe the general method by which you could create it. I find it highly unlikely that patent lawyers cost appreciably more than software developers) -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 17, 2003 at 06:02:12AM +, Andrew Suffield wrote: gain a software patent, you merely have to describe the general method by which you could create it. I find it highly unlikely that patent lawyers cost appreciably more than software developers) While I agree with your general point about the bollocks nature of patents and software licences that try and control them, your reasoning on the cost of patent attorneys is way out of the ballpark. Having been involved (ephemerally) with patent application, I can tell you that patent attorneys are *incredibly* expensive. I'd love to be paid 1/4 of what they are for software development. - Matt
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 17, 2003 at 06:02:12AM +, Andrew Suffield wrote: Finally, it is totally unacceptable to tie this into a software copyright license, such that accepting the license affects the status of your own patents. That's non-free however you look at it. Your own patents are only affected if you contribute code that uses them. If I distribute modifications to a GPL work, the status of my own copyright is affected, too. Also, note 3B and 3C. Modifications being marked a contribution, and thus having patent licenses attached, is completely optional. I can't find any requirement in section 7 that all distributed modifications be contributions. Hmm. It seems that if someone forks Apache, the Apache team could not integrate anything from the fork; if someone uses Apache's code and improves it, they couldn't use it, at least without discussion with copyright holders, since the changes wouldn't be contributions. I wonder if that's intentional. (Well, there's no real guarantee in in the license that all code in Apache will be contributions, but that would defeat the point.) (And this still applies just as much to software licenses. It is *hard* to gain a copyright license; you have to create the work. The GPL requires that any distributed modifications be freely (according to the GPL) licensed to whoever receives it. I think the general notion here is to have a similar requirement for patents that affect the code. I'm inclined to think of it as if you contribute code, we want a license to use it under *both* copyright and patent laws, not just copyright. I'm undecided about reciprocity for something we don't require to begin with (patent licenses). -- Glenn Maynard
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Glenn Maynard [EMAIL PROTECTED] writes: On Mon, Nov 17, 2003 at 06:02:12AM +, Andrew Suffield wrote: Finally, it is totally unacceptable to tie this into a software copyright license, such that accepting the license affects the status of your own patents. That's non-free however you look at it. Your own patents are only affected if you contribute code that uses them. If I distribute modifications to a GPL work, the status of my own copyright is affected, too. That first sentence is not true. Specifically, the candidate Apache license says: 5. Reciprocity. If You institute patent litigation against a Contributor with respect to a patent applicable to software (including a cross-claim or counterclaim in a lawsuit), then any patent licenses granted by that Contributor to You under this License shall terminate as of the date such litigation is filed. [snipped] If I use Apache, and have a significant cost to switch my operations away from Apache, then I dare not sue any Apache contributor who infringes on my patents: if I do so, I lose my licenses to use Apache. If there were a list of relevant patents, this might be at least a *little* more reasonable. But given that, for example, IBM has contributed to Apache, I cannot sue IBM for patent infringement without losing my license to use Apache. -Brian -- Brian T. Sniffen[EMAIL PROTECTED] http://www.evenmere.org/~bts/
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 17, 2003 at 08:19:01AM -0500, Joe Moore wrote: Here's a bit from a hypothetical software license: In addition, by using this software, you grant to the Original Author a non-exclusive right to use, modify, and/or distribute any work of which you own copyright, for as long as you use or distribute The Program. Clearly, no one would argue that this license is a Free Software license. It requires a significant cost (all of your copyrights) to use the software. However, this is essentially what the reciprocal patent clause is requiring. As part of the Apache license, you must agree not to sue any contributor for any of your software patents, for as long as you continue to use Apache. The only problem I see here is return fire: if I'm holding patents as a defense strategy, I want to be able to use them to return fire if an Apache contributor decides to attack me with his own patents, unrelated to Apache. I can't decide if that makes it non-free, or is just an ugly loophole in the license. (I don't care about preserving people's freedom to use patents to restrict freedom, but I do sympathise with allowing them to be used defensely.) If you do, then all of your Apache use is unlicensed. All you lose is a license to the patents granted by that Contributor (under section 4b). Section 5 does not revoke licenses granted under section 4a, only 4b; you don't lose your copyright license to Apache, and you don't lose patent licenses granted by other Contributors. -- Glenn Maynard
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 17, 2003 at 10:43:01AM -0500, Glenn Maynard wrote: However, this is essentially what the reciprocal patent clause is requiring. As part of the Apache license, you must agree not to sue any contributor for any of your software patents, for as long as you continue to use Apache. The only problem I see here is return fire: if I'm holding patents as a defense strategy, I want to be able to use them to return fire if an Apache contributor decides to attack me with his own patents, unrelated to Apache. This is only useful if you do not have a valid defense for the problem already. In other words, it is only useful as a strong-arm tactic to let your own company effectively ignore patents of others. After all, if the lawsuit filed against you has no merit, you don't need a patent portfolio to defend against it. So, its only real purpose is to let the patent holders thwart the patent law. I don't like that one bit.
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Glenn Maynard [EMAIL PROTECTED] writes: On Mon, Nov 17, 2003 at 08:19:01AM -0500, Joe Moore wrote: Here's a bit from a hypothetical software license: In addition, by using this software, you grant to the Original Author a non-exclusive right to use, modify, and/or distribute any work of which you own copyright, for as long as you use or distribute The Program. Clearly, no one would argue that this license is a Free Software license. It requires a significant cost (all of your copyrights) to use the software. However, this is essentially what the reciprocal patent clause is requiring. As part of the Apache license, you must agree not to sue any contributor for any of your software patents, for as long as you continue to use Apache. The only problem I see here is return fire: if I'm holding patents as a defense strategy, I want to be able to use them to return fire if an Apache contributor decides to attack me with his own patents, unrelated to Apache. I can't decide if that makes it non-free, or is just an ugly loophole in the license. So an Apache contributor who owns patents on parts of Apache can force Apache users to either license him their unrelated patents at no cost, or give up their right to use (his patents in) Apache. The two paths provided, then, are payment or arbitrary revocation of the license. That's non-free. -Brian -- Brian T. Sniffen[EMAIL PROTECTED] http://www.evenmere.org/~bts/
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
John Goerzen [EMAIL PROTECTED] writes: On Mon, Nov 17, 2003 at 10:43:01AM -0500, Glenn Maynard wrote: However, this is essentially what the reciprocal patent clause is requiring. As part of the Apache license, you must agree not to sue any contributor for any of your software patents, for as long as you continue to use Apache. The only problem I see here is return fire: if I'm holding patents as a defense strategy, I want to be able to use them to return fire if an Apache contributor decides to attack me with his own patents, unrelated to Apache. This is only useful if you do not have a valid defense for the problem already. In other words, it is only useful as a strong-arm tactic to let your own company effectively ignore patents of others. After all, if the lawsuit filed against you has no merit, you don't need a patent portfolio to defend against it. So, its only real purpose is to let the patent holders thwart the patent law. I don't like that one bit. If the lawsuit filed against you has *no* merit, that's true. But in practice, given the current broken state of the American patent law system, it's much, much cheaper to countersue and work out a quick settlement -- even if both patents on both sides are bullshit -- than to slog through the courts. This isn't nice, it isn't good, it isn't right -- but it isn't Debian's fight, or Apache's, and this isn't the right way to solve it. -Brian -- Brian T. Sniffen[EMAIL PROTECTED] http://www.evenmere.org/~bts/
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 17, 2003 at 11:36:10AM -0500, Brian T. Sniffen wrote: If the lawsuit filed against you has *no* merit, that's true. But in practice, given the current broken state of the American patent law system, it's much, much cheaper to countersue and work out a quick settlement -- even if both patents on both sides are bullshit -- than to slog through the courts. This isn't nice, it isn't good, it isn't right -- but it isn't Debian's fight, or Apache's, and this isn't the right way to solve it. But it is. This is a real problem. Let's say I own a manufacturing company and I am looking for a solution to deploy online shopping services -- this is going to be critical to my business in the future, and a webserver will be a critical part of it. As a manufacturer, I own patents on various manufacturing processes that let me maintain a chance of competing against much larger companies. Now, let's say that somebody that contributed to Apache (with this type of patent grant) decides to violate my patent on mouse trap assembly. I am now stuck between a rock and a hard place: I can either see my product illegally copied by someone else, or defend my patent and see my e-commerce suite come crashing down. Either way, I'm screwed. I don't see how you, or anyone else, can possibly claim that this is acceptable even for proprietary software, much less Free Software. It directly violates DFSG's no discrimination clause, in that people that exercise their legal rights are discriminated against. Now, s/mouse trap/software/ and you have exactly the situation comptemplated by the license proposal. I agree that software should not be patentable at all. But I disagree that people should be prevented from exercising the same rights as everyone else just because they run Apache. If the proposed Apache license becomes DFSG-free, people will no longer be able to trust that Debian is a Free operating system. They will now have to review every legal action taken against any company against all software they use from Debian (which could be in the thousands) to see if this will terminate some rights somewhere. That is ludicrous. -- John
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Brian T. Sniffen wrote: 5. Reciprocity. If You institute patent litigation against a Contributor with respect to a patent applicable to software (including a cross-claim or counterclaim in a lawsuit), then any patent licenses granted by that Contributor to You under this License shall terminate as of the date such litigation is filed. [snipped] If I use Apache, and have a significant cost to switch my operations away from Apache, then I dare not sue any Apache contributor who infringes on my patents: if I do so, I lose my licenses to use Apache. Actually only the patent license would be terminated, not the copyright license. If you use Apache in a country where the patent is not in force, you are not affected by this clause. Or maybe you have a cross-license with the patent holder. Anyway, it's my understanding that this part of clause 5 has be deleted in the next draft. Arnoud -- Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Added license@apache.org to this subthread, since my final question is directed to them. Please CC debian-legal on replies. On Mon, Nov 17, 2003 at 11:36:10AM -0500, Brian T. Sniffen wrote: This isn't nice, it isn't good, it isn't right -- but it isn't Debian's fight, or Apache's, and this isn't the right way to solve it. Which fight are we talking about here? The fight against patents is certainly Apache's fight. Their strategy (require a patent grant for all contributions) seems like a potentially useful way to fight back. The patent grant (4b) seems to be the key part of this strategy. Other than the mixing of patent and copyright, it seems few people have issues with it. I'm not sure if there's a separate fight behind the reciprocity clause (#5). Is it there as another defense mechanism, or is it there to make 4b more palatable to patent holders? -- Glenn Maynard
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, 17 Nov 2003, Andrew Suffield wrote: (And this still applies just as much to software licenses. It is *hard* to gain a copyright license; you have to create the work. To gain a software patent, you merely have to describe the general method by which you could create it. I find it highly unlikely that patent lawyers cost appreciably more than software developers) I'm a software developer. So the services of one may, under some circumstances, cost me nothing at all (except my spare time). I don't think patent lawyers can get cheaper than this.
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Glenn Maynard [EMAIL PROTECTED] writes: Added license@apache.org to this subthread, since my final question is directed to them. Please CC debian-legal on replies. On Mon, Nov 17, 2003 at 11:36:10AM -0500, Brian T. Sniffen wrote: This isn't nice, it isn't good, it isn't right -- but it isn't Debian's fight, or Apache's, and this isn't the right way to solve it. Which fight are we talking about here? The fight against patents is certainly Apache's fight. Their strategy (require a patent grant for all contributions) seems like a potentially useful way to fight back. The patent grant (4b) seems to be the key part of this strategy. Other than the mixing of patent and copyright, it seems few people have issues with it. I'm not sure if there's a separate fight behind the reciprocity clause (#5). Is it there as another defense mechanism, or is it there to make 4b more palatable to patent holders? The fact that the license can be revoked over unrelated squabbles between users and authors appears to be an attempt to make software patents impractical and useless. If it only made software patents *on Apache* useless (the second clause of S5), I'd think it reasonable. That would parallel what the GNU GPL does for copyrights for example. What's currently there attempts to use the usefulness of Apache to buy non-enforcement of software patents elsewhere, which I believe is inappropriate for Free Software. -Brian -- Brian T. Sniffen[EMAIL PROTECTED] http://www.evenmere.org/~bts/
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 17, 2003 at 10:15:40AM -0800, Ken Arromdee wrote: On Mon, 17 Nov 2003, Andrew Suffield wrote: (And this still applies just as much to software licenses. It is *hard* to gain a copyright license; you have to create the work. To gain a software patent, you merely have to describe the general method by which you could create it. I find it highly unlikely that patent lawyers cost appreciably more than software developers) I'm a software developer. So the services of one may, under some circumstances, cost me nothing at all (except my spare time). I don't think patent lawyers can get cheaper than this. You could go to law school and become a patent lawyer. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 17, 2003 at 01:17:23PM -0500, Brian T. Sniffen wrote: What's currently there attempts to use the usefulness of Apache to buy non-enforcement of software patents elsewhere, which I believe is inappropriate for Free Software. If that's all it did, I'd be fine with it. However, I don't like the side-effect of making Apache users defenseless against unrelated software patent attacks from Apache contributors. I believe Arnoud Engelfriet mentioned that this clause (#5) has been removed from the draft. I havn't checked. If so, that's good; this is clearly the most problematic clause. -- Glenn Maynard
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
John Goerzen [EMAIL PROTECTED] writes: On Mon, Nov 17, 2003 at 11:36:10AM -0500, Brian T. Sniffen wrote: If the lawsuit filed against you has *no* merit, that's true. But in practice, given the current broken state of the American patent law system, it's much, much cheaper to countersue and work out a quick settlement -- even if both patents on both sides are bullshit -- than to slog through the courts. This isn't nice, it isn't good, it isn't right -- but it isn't Debian's fight, or Apache's, and this isn't the right way to solve it. But it is. This is a real problem. Let's say I own a manufacturing company and I am looking for a solution to deploy online shopping services -- this is going to be critical to my business in the future, and a webserver will be a critical part of it. As a manufacturer, I own patents on various manufacturing processes that let me maintain a chance of competing against much larger companies. Now, let's say that somebody that contributed to Apache (with this type of patent grant) decides to violate my patent on mouse trap assembly. I am now stuck between a rock and a hard place: I can either see my product illegally copied by someone else, or defend my patent and see my e-commerce suite come crashing down. Either way, I'm screwed. Exatly: that's what I've been saying: the clause of the candidate Apache license seeking to make software patents unusable is not Free, and not even a good idea. The this in my last sentence was referring to that clause. I don't see how you, or anyone else, can possibly claim that this is acceptable even for proprietary software, much less Free Software. It directly violates DFSG's no discrimination clause, in that people that exercise their legal rights are discriminated against. Now, s/mouse trap/software/ and you have exactly the situation comptemplated by the license proposal. We appear to be in violent agreement. I agree that software should not be patentable at all. But I disagree that people should be prevented from exercising the same rights as everyone else just because they run Apache. If the proposed Apache license becomes DFSG-free, people will no longer be able to trust that Debian is a Free operating system. They will now have to review every legal action taken against any company against all software they use from Debian (which could be in the thousands) to see if this will terminate some rights somewhere. That is ludicrous. -Brian -- Brian T. Sniffen[EMAIL PROTECTED] http://www.evenmere.org/~bts/
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Scripsit [EMAIL PROTECTED] (Brian T. Sniffen) 5. Reciprocity. If You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that a Contribution and/or the Work, without modification (other than modifications that are Contribution(s)), constitutes direct or contributory patent infringement, then any patent licenses granted to You under this License for that Contribution or such Work shall terminate as of the date such litigation is filed. That's certainly better. It still has a problem in the following scenario: 1. I start using Apache. 2. I develop a new process -- let's say an encryption algorithm, like RSA -- and patent it. 3. Somebody contributes an implementation of my algorithm to Apache. This somebody has patents on critical parts of Apache. Now I'm screwed: I can't sue Apache for illegally using my work without my permission, or I'll lose my license to their code. I don't see that. If is only the grants under this License *for* that Contribution or such Work that terminate. If you does not use the version of Apache with your work in it, then your license to the version you do use does not self-destruct as a consequence of your suit. You may be screwed if you only discover the violation after you yourself have converted your website to use an Apache version that itself contains the violation. In that case you will need to backport the new features you need to an older Apache that does not contain your patent (and which thus has a license that will not self-destruct). -- Henning Makholm Also, the letters are printed. That makes the task of identifying the handwriting much more difficult.
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Henning Makholm [EMAIL PROTECTED] writes: From: Henning Makholm [EMAIL PROTECTED] Subject: Re: [EMAIL PROTECTED]: Review of proposed Apache License, version 2.0] To: debian-legal@lists.debian.org Date: 17 Nov 2003 23:01:38 + Resent-From: debian-legal@lists.debian.org Scripsit [EMAIL PROTECTED] (Brian T. Sniffen) 5. Reciprocity. If You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that a Contribution and/or the Work, without modification (other than modifications that are Contribution(s)), constitutes direct or contributory patent infringement, then any patent licenses granted to You under this License for that Contribution or such Work shall terminate as of the date such litigation is filed. That's certainly better. It still has a problem in the following scenario: 1. I start using Apache. 2. I develop a new process -- let's say an encryption algorithm, like RSA -- and patent it. 3. Somebody contributes an implementation of my algorithm to Apache. This somebody has patents on critical parts of Apache. Now I'm screwed: I can't sue Apache for illegally using my work without my permission, or I'll lose my license to their code. I don't see that. If is only the grants under this License *for* that Contribution or such Work that terminate. If you does not use the version of Apache with your work in it, then your license to the version you do use does not self-destruct as a consequence of your suit. You may be screwed if you only discover the violation after you yourself have converted your website to use an Apache version that itself contains the violation. In that case you will need to backport the new features you need to an older Apache that does not contain your patent (and which thus has a license that will not self-destruct). Whoah. You're right, I missed that. OK, that might actually be Free. I'm not sure, and I'll need to think about it hard. It also seems to be a fine enough point that it invites situations akin to Pine: a malicious or just confused copyright^H^H^H^H^H^H^H^H^H patent holder might interpret it differently. -Brian
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, 17 Nov 2003, Andrew Suffield wrote: On Mon, Nov 17, 2003 at 10:15:40AM -0800, Ken Arromdee wrote: On Mon, 17 Nov 2003, Andrew Suffield wrote: (And this still applies just as much to software licenses. It is *hard* to gain a copyright license; you have to create the work. To gain a software patent, you merely have to describe the general method by which you could create it. I find it highly unlikely that patent lawyers cost appreciably more than software developers) I'm a software developer. So the services of one may, under some circumstances, cost me nothing at all (except my spare time). I don't think patent lawyers can get cheaper than this. You could go to law school and become a patent lawyer. But that's not cheap. Going to law school costs a lot of money. Becoming a software developer only requires buying a computer, which costs a lot less. Besides, researching and filing patents still has some costs even if you're already a lawyer and don't need to hire one.
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Jennifer Machovec, who's drafting the license, posted a new version to license@apache.org on November 13. You can read it at http://nagoya.apache.org/eyebrowse/ [EMAIL PROTECTED]msgNo=24 Correction: Jennifer Machovec is not drafting the license. She is an attorney at IBM who submitted comments on behalf of IBM. I am drafting the license on behalf of the ASF. I will most likely include the changes that IBM suggested, assuming that there aren't any more objections. We've already gone through several iterations on the patent sentences, mostly to deal with derivative work, so it isn't too surprising that we need to scale it back again. FYI, prior to a couple months ago, Drew Wright was representing IBM and made many suggestions to improve the license. I have also had discussions with folks from Apple, Sun, Mozilla, OSI, FSF, and a few independent attorneys. The hope is that the license will be reviewed by all of the companies that use Apache software, since that is less expensive than us hiring a couple thousand IP lawyers. Roy
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Sat, 2003-11-15 at 11:24, Brian T. Sniffen wrote: The patent prevents you from solving the covered problem, no matter how you come to that solution. So the unlawfullness of integrating the patented method into the parsing of your favorite text editor has nothing to do with the web server. Saying that the webserver's code is thus unfree -- attributing the unlawfulness to the method by which you became aware of the unlawful option -- seems wrong to me. I disagree, depending on circumstances. For example, when the LZW patents were still valid, I could not (legally) modify Gimp to produce normal GIFs. That didn't make Gimp non-free, it made the US and various other countries non-free. However, if, for example, a company releases under the MIT X11 license a program that does A, but also patents doing A --- and enforces that patent --- then A is not free software. If they grant a limitless license to the patent, it becomes free again. I think you must look at the entire picture --- not just the copyright one --- to determine if software is free. I don't think its free if the copyright holder decides to use patents, instead of copyright, to limit your freedoms. In short: * If party A releases software, and in some maner prevents you from excercising your DFSG-freedoms, then that software is not DFSG-free. It doesn't matter if that manner is copyright, patent, or the mafia. * If party A releases software, and allows you all your DFSG-freedoms but an unrelated party B does not allow you to modify it to, e.g., play DVDs, party A's software is still free. To put it yet another way, you can't use patents as an end-run around freeness. signature.asc Description: This is a digitally signed message part
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Thu, Nov 13, 2003 at 11:12:04PM +0100, Henning Makholm wrote: There's the difference that it takes explicit action and quite a bit of money to acquire and keep holding a patent. Going through that trouble just to grant the public a perpetual, non-exclusive, worldwide, fully paid-up and royalty free patent licence without a reciprocity clause would be senseless. An easier way to achieve the same legal result would be not to take out a patent at all. Exactly; all the more reason not to do it. Companies with patents are like countries with nuclear weapons; once you have them, it's very difficult to resist the temptation to play with them, and intimidate one's neighbors. This is not a temptation that ordinary people of ordinary means have to wrestle with. The Free Software and Open Source Movements are a community of rough equals. There is no society among giants. -- G. Branden Robinson| The more ridiculous a belief Debian GNU/Linux | system, the higher the probability [EMAIL PROTECTED] | of its success. http://people.debian.org/~branden/ | -- Wayne R. Bartz signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
[EMAIL PROTECTED] (Brian T. Sniffen) wrote: Walter Landry [EMAIL PROTECTED] writes: [EMAIL PROTECTED] (Brian T. Sniffen) wrote: Henning Makholm [EMAIL PROTECTED] writes: Scripsit [EMAIL PROTECTED] (Brian T. Sniffen) And, as it happens, companies do grant free patent licenses: it's common practice when working on a standard which must be approved by a standards body with a RF policy: typically, the patent is licensed for any use which implements that standard. A patent license that applies only to implementations of a specific standard is not free (as in free speech). Can you explain this to me? I see free software, and some external limits on how you may use certain modifications of it. You can't modify the code in the webserver to improve the parsing of your favorite editor. But you couldn't add those features to your favorite editor *anyway*, because they're patented. Unlike copyright, chain of custody and derivation is irrelevant for patents. I was answering the question of why software with a narrow patent grant is not free. You are not free to modify the code. But this only comes up if there is an actual patent that applies to the software. Regards, Walter Landry [EMAIL PROTECTED]
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Thu, Nov 13, 2003 at 11:12:04PM +0100, Henning Makholm wrote: Scripsit Andrew Suffield [EMAIL PROTECTED] On Sun, Nov 09, 2003 at 02:55:56PM +1300, Adam Warner wrote: No sane company will ever grant a perpetual, non-exclusive, worldwide, fully paid-up and royalty free patent licence without a reciprocity clause. No sane company will ever grant a perpetual, non-exclusive, worldwide, fully paid-up and royalty free copyright license without a reciprocity clause. There's the difference that it takes explicit action and quite a bit of money to acquire and keep holding a patent. Going through that trouble just to grant the public a perpetual, non-exclusive, worldwide, fully paid-up and royalty free patent licence without a reciprocity clause would be senseless. An easier way to achieve the same legal result would be not to take out a patent at all. On the other hand, copyright springs into being automatically. It makes sense for somebody who have accidentally become bestowed with a copyright to explicitly license it to the general public under free terms, if he wants to create a situation reminiscent of the one where he does not have a copyright. Are you attempting to suggest that a sane company doesn't want to hold copyright in the software they create? I don't find that likely. Your argument doesn't seem relevant to the subject at hand. The argument proposed was attempting to say No company is ever going to grant free patent licenses; I pointed out the argument applies equally to software (it's the same one that proprietary software advocates have been making for about 20 years, claiming that free software can't work), and companies *do* grant free software licenses. They can grant free patent licenses for the same reasons. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Andrew Suffield [EMAIL PROTECTED] writes: The argument proposed was attempting to say No company is ever going to grant free patent licenses; I pointed out the argument applies equally to software (it's the same one that proprietary software advocates have been making for about 20 years, claiming that free software can't work), and companies *do* grant free software licenses. They can grant free patent licenses for the same reasons. And, as it happens, companies do grant free patent licenses: it's common practice when working on a standard which must be approved by a standards body with a RF policy: typically, the patent is licensed for any use which implements that standard. This is an interesting restriction on modification-and-use: you can modify the program to use the patented technology outside the scope of the standard, but you can't compile and use that code without infringing on the patent. I *think* the result can be Free Software, but I'm not entirely convinced: it seems that the standard is included by reference in the patent spec. This is made even worse in cases where the standard isn't freely available, so you don't even have the text of the patent license unless you pay for the standard. That's probably not Free Software. But for the case of Apache, for example, it's enough for every contributing company to grant a universal license to their patented technology for use in web browsers, and for the ASF to refuse contributions to the mainline Apache from anyone who doesn't agree. Yes, this means unscrupulous or even just secretive companies can fork Apache and integrate their proprietary, patented technology. That would be unfortunate. But the freedom to do that is a necessary freedom. -Brian -- Brian T. Sniffen[EMAIL PROTECTED] http://www.evenmere.org/~bts/
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Scripsit Andrew Suffield [EMAIL PROTECTED] On Thu, Nov 13, 2003 at 11:12:04PM +0100, Henning Makholm wrote: On the other hand, copyright springs into being automatically. It makes sense for somebody who have accidentally become bestowed with a copyright to explicitly license it to the general public under free terms, if he wants to create a situation reminiscent of the one where he does not have a copyright. Are you attempting to suggest that a sane company doesn't want to hold copyright in the software they create? It is apparent (many concrete examples in Debian) that some companies create software without wanting very much of the protection that copyright gives by default. The argument proposed was attempting to say No company is ever going to grant free patent licenses; I pointed out the argument applies equally to software And I point out that it doesn't. If the company patent their invention at all, it must be because they intend to restrict people from using it (or at least keep an option open for using the patent to restrict what people do). If they do not intend that, why would they apply for a patent at all in the first place? (it's the same one that proprietary software advocates have been making for about 20 years, claiming that free software can't work), No, it has nothing to do with that argument. -- Henning Makholm `Update' isn't a bad word; in the right setting it is useful. In the wrong setting, though, it is destructive...
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Scripsit [EMAIL PROTECTED] (Brian T. Sniffen) And, as it happens, companies do grant free patent licenses: it's common practice when working on a standard which must be approved by a standards body with a RF policy: typically, the patent is licensed for any use which implements that standard. A patent license that applies only to implementations of a specific standard is not free (as in free speech). This is an interesting restriction on modification-and-use: you can modify the program to use the patented technology outside the scope of the standard, but you can't compile and use that code without infringing on the patent. I *think* the result can be Free Software, I think it is clear that it is not. But for the case of Apache, for example, it's enough for every contributing company to grant a universal license to their patented technology for use in web browsers, and for the ASF to refuse contributions to the mainline Apache from anyone who doesn't agree. If ASF makes public an intention to include in upstream Apache patented code that cannot be used in X servers, desktop calculators or Forth compilers, then we can never be sure that the next minor upstream release will still be free software. Of course the Debian maintainer for apache *may* choose to audit each new upstream release to see if non-free patents have crept in, but I would advise moving it to non-free right away to avoid future trouble. -- Henning Makholm ... and that Greek, Thucydides
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Henning Makholm [EMAIL PROTECTED] writes: Scripsit [EMAIL PROTECTED] (Brian T. Sniffen) And, as it happens, companies do grant free patent licenses: it's common practice when working on a standard which must be approved by a standards body with a RF policy: typically, the patent is licensed for any use which implements that standard. A patent license that applies only to implementations of a specific standard is not free (as in free speech). Can you explain this to me? I see free software, and some external limits on how you may use certain modifications of it. This is an interesting restriction on modification-and-use: you can modify the program to use the patented technology outside the scope of the standard, but you can't compile and use that code without infringing on the patent. I *think* the result can be Free Software, I think it is clear that it is not. Well, it's certainly not *clear*, or we wouldn't be having this discussion. But I'm entirely convinceable -- go ahead and try? But for the case of Apache, for example, it's enough for every contributing company to grant a universal license to their patented technology for use in web browsers, and for the ASF to refuse contributions to the mainline Apache from anyone who doesn't agree. If ASF makes public an intention to include in upstream Apache patented code that cannot be used in X servers, desktop calculators or Forth compilers, then we can never be sure that the next minor upstream release will still be free software. Of course the Debian maintainer for apache *may* choose to audit each new upstream release to see if non-free patents have crept in, but I would advise moving it to non-free right away to avoid future trouble. There is also no way to be sure that the next minor upstream Emacs release will still be entirely free software, and Debian has been bitten by this before. So why not move everything to non-free which is not under a GPL, version 2 only license? -Brian
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Henning Makholm [EMAIL PROTECTED] writes: The argument proposed was attempting to say No company is ever going to grant free patent licenses; I pointed out the argument applies equally to software And I point out that it doesn't. If the company patent their invention at all, it must be because they intend to restrict people from using it (or at least keep an option open for using the patent to restrict what people do). If they do not intend that, why would they apply for a patent at all in the first place? In the current patent-litigation context, a large stable of patents to cross-license is considered a vitally important corporate defense strategy. (it's the same one that proprietary software advocates have been making for about 20 years, claiming that free software can't work), No, it has nothing to do with that argument.
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Scripsit Brian T. Sniffen Henning Makholm [EMAIL PROTECTED] writes: And I point out that it doesn't. If the company patent their invention at all, it must be because they intend to restrict people from using it (or at least keep an option open for using the patent to restrict what people do). If they do not intend that, why would they apply for a patent at all in the first place? In the current patent-litigation context, a large stable of patents to cross-license is considered a vitally important corporate defense strategy. Yes, but a patent could not be part of such a portfolio if if were licensed freely to the general public. -- Henning MakholmNu kommer han. Kan du ikke høre knallerten?
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Sat, Nov 15, 2003 at 12:58:39AM +, Henning Makholm wrote: In the current patent-litigation context, a large stable of patents to cross-license is considered a vitally important corporate defense strategy. Yes, but a patent could not be part of such a portfolio if if were licensed freely to the general public. ... unless it's licensed with a condition that if you sue them, the patent grant is withdrawn. That seems to be the purpose of the reciprocity clause. It seems the intent is to require a patent license (under 4b), while still allowing those patents to be used defensively (against other patents). At least on its face, it seems like a useful compromise: companies often legitimately won't want to give out unrecovable patent licenses, since they need them to defend against other, hostile patent holders. Still undecided. I can sympathise both with attempts to find defenses against patents (of which free software has scarce few), and to do so in a way that doesn't force others to weaken their own patent defenses. -- Glenn Maynard
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Henning Makholm [EMAIL PROTECTED] writes: Scripsit Brian T. Sniffen Henning Makholm [EMAIL PROTECTED] writes: And I point out that it doesn't. If the company patent their invention at all, it must be because they intend to restrict people from using it (or at least keep an option open for using the patent to restrict what people do). If they do not intend that, why would they apply for a patent at all in the first place? In the current patent-litigation context, a large stable of patents to cross-license is considered a vitally important corporate defense strategy. Yes, but a patent could not be part of such a portfolio if if were licensed freely to the general public. But it could be part of such a portfolio if it were licensed for use in otherwise-free software only, or for use in implementing specifications with RF policies. -Brian
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Glenn Maynard [EMAIL PROTECTED] writes: On Sat, Nov 15, 2003 at 12:58:39AM +, Henning Makholm wrote: In the current patent-litigation context, a large stable of patents to cross-license is considered a vitally important corporate defense strategy. Yes, but a patent could not be part of such a portfolio if if were licensed freely to the general public. ... unless it's licensed with a condition that if you sue them, the patent grant is withdrawn. That seems to be the purpose of the reciprocity clause. It seems the intent is to require a patent license (under 4b), while still allowing those patents to be used defensively (against other patents). At least on its face, it seems like a useful compromise: companies often legitimately won't want to give out unrecovable patent licenses, since they need them to defend against other, hostile patent holders. Still undecided. I can sympathise both with attempts to find defenses against patents (of which free software has scarce few), and to do so in a way that doesn't force others to weaken their own patent defenses. My employer just hosted a lawyer to tell us all about the Dangers of F/OSS (Free or Open Source Software). His talk was largely FUD, but one of the few pieces which found purchase with management was Patent Litigation Fear: that if we were using Mozilla (the MPL has a similar clause) anywhere in the company, or even worse had standardized on it, and got into a patent lawsuit with any Mozilla contributor, we could lose our license to use Mozilla, or to distribute code which derived from Mozilla. That's just too scary to risk: if somebody else really does violate one of our (non-software, even) patents, we have no recourse without first switching to some other code base. Yech. That pretty much seems like a usage restriction: it restricts us from doing things in private, based on our attempts to exercise *unrelated* legal rights. -Brian
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Scripsit Brian T. Sniffen Henning Makholm [EMAIL PROTECTED] writes: Yes, but a patent could not be part of such a portfolio if if were licensed freely to the general public. But it could be part of such a portfolio if it were licensed for use in otherwise-free software only, OK, granted. -- Henning Makholm Nemo enim fere saltat sobrius, nisi forte insanit.
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
[EMAIL PROTECTED] (Brian T. Sniffen) wrote: Henning Makholm [EMAIL PROTECTED] writes: Scripsit [EMAIL PROTECTED] (Brian T. Sniffen) And, as it happens, companies do grant free patent licenses: it's common practice when working on a standard which must be approved by a standards body with a RF policy: typically, the patent is licensed for any use which implements that standard. A patent license that applies only to implementations of a specific standard is not free (as in free speech). Can you explain this to me? I see free software, and some external limits on how you may use certain modifications of it. You can't modify the code in the webserver to improve the parsing of your favorite editor. Regards, Walter Landry [EMAIL PROTECTED]
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Scripsit Andrew Suffield [EMAIL PROTECTED] On Sun, Nov 09, 2003 at 02:55:56PM +1300, Adam Warner wrote: No sane company will ever grant a perpetual, non-exclusive, worldwide, fully paid-up and royalty free patent licence without a reciprocity clause. No sane company will ever grant a perpetual, non-exclusive, worldwide, fully paid-up and royalty free copyright license without a reciprocity clause. There's the difference that it takes explicit action and quite a bit of money to acquire and keep holding a patent. Going through that trouble just to grant the public a perpetual, non-exclusive, worldwide, fully paid-up and royalty free patent licence without a reciprocity clause would be senseless. An easier way to achieve the same legal result would be not to take out a patent at all. On the other hand, copyright springs into being automatically. It makes sense for somebody who have accidentally become bestowed with a copyright to explicitly license it to the general public under free terms, if he wants to create a situation reminiscent of the one where he does not have a copyright. -- Henning MakholmNej, hvor er vi altså heldige! Længe leve vor Buxgører Sansibar Bastelvel!
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
Brian M. Carlson said on Sat, Nov 08, 2003 at 10:39:29AM +,: I'm not sure that this is even legal, at least in the US. Will you please clarify why?? -- +~+ Mahesh T. Pai, LL.M., 'NANDINI', S. R. M. Road, Ernakulam, Cochin-682018, Kerala, India. http://in.geocities.com/paivakil +~+
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, Nov 10, 2003 at 03:22:39PM +0530, Mahesh T. Pai wrote: Brian M. Carlson said on Sat, Nov 08, 2003 at 10:39:29AM +,: I'm not sure that this is even legal, at least in the US. Will you please clarify why?? I'm assuming you meant the copyright assignment statement, and certainly, I will clarify. According to David Turner, IIRC, it requires written paperwork for copyright assignment. Debian, though, usually accepts emails as well, but not licenses that have default assignments. This was a big deal with ReiserFS (search the archives for more info). -- Brian M. Carlson [EMAIL PROTECTED] 0x560553e7 signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Mon, 2003-11-10 at 14:39, Brian M. Carlson wrote: On Mon, Nov 10, 2003 at 03:22:39PM +0530, Mahesh T. Pai wrote: Brian M. Carlson said on Sat, Nov 08, 2003 at 10:39:29AM +,: I'm not sure that this is even legal, at least in the US. Will you please clarify why?? I'm assuming you meant the copyright assignment statement, and certainly, I will clarify. According to David Turner, IIRC, it requires written paperwork for copyright assignment. Debian, though, usually accepts emails as well, but not licenses that have default assignments. This was a big deal with ReiserFS (search the archives for more info). The proposed Apache licensed doesn't do copyright assignment -- it does licensing. That's an entirely different matter. I'm not jumping into this discussion, just pointing out that what I might have said might have been confusing :) -- -Dave Turner GPL Compliance Engineer Support my work: http://svcs.affero.net/rm.php?r=novalisp=FSF
[fielding@apache.org: Review of proposed Apache License, version 2.0]
- Forwarded message from Roy T. Fielding [EMAIL PROTECTED] - From: Roy T. Fielding [EMAIL PROTECTED] Subject: Review of proposed Apache License, version 2.0 Date: Sat, 8 Nov 2003 00:33:17 -0800 To: announce@apache.org X-Mailer: Apple Mail (2.552) The Apache Software Foundation is considering the adoption of a new set of licenses for our open source projects. We invite you to review and comment on the proposed 2.0 license and related material, which can be found at http://www.apache.org/licenses/proposed/ A mailing list has been set up for public review and discussion of the license prior to the ASF members meeting at ApacheCon [*]. The address is license AT apache.org and you can join it by sending a blank message to license-subscribe AT apache.org. The goals of this license revision have been to reduce the number of frequently asked questions, to allow the license to be reusable without modification by any project (including non-ASF projects), to allow the license to be included by reference instead of listed in every file, to clarify the license on submission of contributions, to require a patent license on contributions that necessarily infringe the contributor's own patents, and to move comments regarding specific Apache trademarks and attribution notices to a location outside the license terms (the NOTICE file). The result should be a license that is compatible with other open source licenses, such as the GPL, and yet still remains true to the original goals of the Apache Group and supportive of collaborative development across both nonprofit and commercial organizations. At least that's the idea -- if you note a potential problem with the new license, please let us know so that we can work on fixing it now rather than after it has been applied to our projects. Cheers, Roy T. Fielding, director, The Apache Software Foundation (fielding AT apache.org) http://roy.gbiv.com/ [*] ApacheCon is in Las Vegas, Nov 16-19 http://www.apachecon.com/ - End forwarded message - -- Martin Michlmayr [EMAIL PROTECTED]
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
I am including the licenses inline. I will immediately follow up with comments, so that it is apparent which comments are mine and which are not. = == DO NOT PANIC! This is a draft for discussion purposes only. == == It has not yet been approved. It does not yet apply to any == == software distributed by the Apache Software Foundation.== == == == This long version combines a contributor license with the== == copyright license, thus allowing us to talk about patents, == == providing disclaimers for all of the contributors rather than == == just the Licensor, and creating a community condition that == == eliminates the need for CLAs from every contributor. == ==+== Apache License Version 2.0, October 2003 http://www.apache.org/licenses/ Copyright (C) 2003 The Apache Software Foundation. Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. TERMS AND CONDITIONS FOR USE, REPRODUCTION, DISTRIBUTION, AND MODIFICATION 1. Licensing the Work. These terms and conditions for use, reproduction, distribution, and modification (the License) apply to any original work of authorship (the Work) containing a notice placed by the copyright owner (the Licensor) indicating that the Work is licensed under the Apache License, Version 2.0. 2. You. You or Your means an individual or legal entity exercising permissions granted by this License. By exercising any of the permissions granted to You in Sections 4 through 8 herein, You indicate Your acceptance of this License and all of its terms and conditions. 3. Contributors and Contributions. A. The Licensor and any individual or legal entity that voluntarily submits to the Licensor a Contribution to the Work are collectively addressed herein as Contributors. For legal entities, the entity making a Contribution and all other entities that control, are controlled by, or are under common control with that entity are considered to be a single Contributor. For the purposes of this definition, control means (i) the power, direct or indirect, to cause the direction or management of such entity, whether by contract or otherwise, or (ii) ownership of fifty percent (50%) or more of the outstanding shares, or (iii) beneficial ownership of such entity. B. A Contribution is the original version of the Work and any modification or addition to the Work that has been submitted for inclusion in the Work, where such modifications and/or additions to the Work originate from that particular Contributor, or from some entity acting on behalf of that Contributor. C. A Contribution is submitted when any form of electronic, verbal, or written communication is sent to the Licensor, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Licensor for the purpose of discussing and improving the Work, but excluding communication that is conspicuously marked or otherwise designated in writing by the Contributor as Not a Contribution. D. Any Contribution submitted by You to the Licensor shall be under the terms and conditions of this License, without any additional terms or conditions, unless You explicitly state otherwise in the submission. 4. Contributor Grant of License. Subject to the terms and conditions of this License, each Contributor hereby grants to You: (a) a perpetual, non-exclusive, worldwide, fully paid-up, royalty free, irrevocable copyright license under its licensable copyrights in the Work to reproduce, prepare derivative works of, publicly display, publicly perform, sublicense, and distribute the Work and such derivative works; and, (b) a perpetual, non-exclusive, worldwide, fully paid-up, royalty free, irrevocable (subject to Section 5) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work and derivative works thereof, where such license applies only to those patent claims licensable by such Contributor that are necessarily infringed by their Contribution alone or by combination of their Contribution with the Work to which such Contribution was submitted by the Contributor. No
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
BIG NOTICE: None of these licenses are official. They are all drafts. On Sat, Nov 08, 2003 at 10:03:55AM +, Brian M. Carlson wrote: I am including the licenses inline. I will immediately follow up with comments, so that it is apparent which comments are mine and which are not. 3. Contributors and Contributions. C. A Contribution is submitted when any form of electronic, verbal, or written communication is sent to the Licensor, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Licensor for the purpose of discussing and improving the Work, but excluding communication that is conspicuously marked or otherwise designated in writing by the Contributor as Not a Contribution. D. Any Contribution submitted by You to the Licensor shall be under the terms and conditions of this License, without any additional terms or conditions, unless You explicitly state otherwise in the submission. I'm not sure that this is even legal, at least in the US. 5. Reciprocity. If You institute patent litigation against a Contributor with respect to a patent applicable to software (including a cross-claim or counterclaim in a lawsuit), then any patent licenses granted by that Contributor to You under this License shall terminate as of the date such litigation is filed. In addition, if You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Work itself (excluding combinations of the Work with other software or hardware) infringes Your patent(s), then any patent licenses granted to You under this License for that Work shall terminate as of the date such litigation is filed. I think that we have prohibited such litigation-termination licenses as non-free. 7. Redistribution with Modification. You may modify Your copy or copies of the Work or any portion of it, thus forming another work product based on the Work (a Derivative Work), and reproduce and distribute such modifications or the Derivative Work, provided that You also meet the following conditions: (a) You must give any other recipients of the Derivative Work a copy of this License along with the Derivative Work. (b) You must retain, in the source code of any Derivative Work that You distribute, all copyright, patent, or trademark notices from the source code of the Work, excluding those notices that only pertain to portions of the Work that have been excluded from the Derivative Work. If the Work includes a NOTICE file as part of its source code distribution, the Derivative Work must include a readable copy of the notices contained within that NOTICE file, excluding those notices that only pertain to portions of the Work that have been excluded from the Derivative Work, in at least one of the following places: within a NOTICE file distributed as part of the Derivative Work; within the source code or documentation, if provided along with the Derivative Work; or, within a display generated by the Derivative Work, if and wherever such third-party notices normally appear. You may add Your own notices alongside or as an addendum to the original NOTICE information. The contents of the NOTICE file are for informational purposes only and do not modify the terms and conditions of this License. Others might wish to comment on this section. My problem is that if NOTICES contains advertisement notices (like in this case), the license is probably not GPL-compatible. (c) You must cause any modified files to carry prominent notices stating that You changed the files. You may add Your own copyright statement to such modifications and may provide (sublicense) additional or different license terms and conditions for use, reproduction, distribution or further modification of Your modifications, or for the Derivative Work as a whole, provided that the sublicense complies with the conditions stated in this License. 8. Redistribution with Additional Terms. You may choose to offer, and to charge a fee for, warranty, support, indemnity, or liability obligations and/or other rights consistent with the scope of the license granted herein (Additional Terms). However, You may do so only on Your own behalf and as Your sole responsibility, not on behalf of any other Contributor, and only if You agree to indemnify, defend, and hold every Contributor harmless for any liability
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
How Apache went from a rather decent 5 clause license to the proposed 11 clause license is a mystery to me. I strongly suggest the license be gone over with a fine toothed comb and searched for areas where it can be made more general and less specific. On Sat, 08 Nov 2003, Brian M. Carlson wrote: D. Any Contribution submitted by You to the Licensor shall be under the terms and conditions of this License, without any additional terms or conditions, unless You explicitly state otherwise in the submission. I can see no way that this term can ever be enforced without active agreement on the part of the contributor. Mere transfer of IP is not enough to establish an affirmation of the terms of the license for contributions. Furthermore, the entire section 3 of this license is totally useless and belongs somewhere else besides a software license. 5. Reciprocity. If You institute patent litigation against a Contributor with respect to a patent applicable to software (including a cross-claim or counterclaim in a lawsuit), then any patent licenses granted by that Contributor to You under this License shall terminate as of the date such litigation is filed. In addition, if You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Work itself (excluding combinations of the Work with other software or hardware) infringes Your patent(s), then any patent licenses granted to You under this License for that Work shall terminate as of the date such litigation is filed. This is not DFSG free. While software patents are generally held to be bad, it is not the purpose of a Free Software license to discriminate against who can use the software. [This also has the wierd side effect of effectively granting to each Contributor a license to use without royalty any of the Licensee's patents applicable to software, no matter how legitimate those patents are.] We should be dealing with invalid and/or improper software patents through legislation and/or litigation rather than adding usage restrictions to our formerly free software licenses. 7 (b) [...] The contents of the NOTICE file are for informational purposes only and do not modify the terms and conditions of this License. Either NOTICE is copyright and warranty information, and it can't be removed, or its not, and it can... this section seems to want it both ways. I'm tempted to consider the requirement to keep the NOTICES file compatible with the DFSG as it appears to contain copyright notices, but if it doesn't actually fullfill that role (ie, is purely informational) it needs to be removeable to comply with the DFSG. 11. Limitation of Liability. Under no circumstances and under no [...] This limitation of liability shall not apply to liability for ^ death or personal injury resulting from Licensor's negligence ^ to the extent applicable law prohibits such limitation. Some ^^^ The limitation phrase in this warranty is suspicious. Most likely what they mean is we are not liable to the maximum extent possible under applicable law rather than we are liable to the extent law prohibts such limitation. [I'm not up on my warranty law, but someone really ought to get a second opinion on this clause, as it definetly doesn't jive with what I'm used to reading.] When I get a chance, I'll look at the other licenses, but I see that some of the same troubling clauses from the Apache License 2.0 show up there as well. [Feel free to forward these comments anywhere appropriate and/or draw attention to them. I haven't done so because I'm not set in stone yet on my opinions.] Don Armstrong -- When I was a kid I used to pray every night for a new bicycle. Then I realised that the Lord doesn't work that way so I stole one and asked Him to forgive me. -- Emo Philips. http://www.donarmstrong.com http://www.anylevel.com http://rzlab.ucr.edu signature.asc Description: Digital signature
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Sun, 2003-11-09 at 01:25, Don Armstrong wrote: 5. Reciprocity. If You institute patent litigation against a Contributor with respect to a patent applicable to software (including a cross-claim or counterclaim in a lawsuit), then any patent licenses granted by that Contributor to You under this License shall terminate as of the date such litigation is filed. In addition, if You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Work itself (excluding combinations of the Work with other software or hardware) infringes Your patent(s), then any patent licenses granted to You under this License for that Work shall terminate as of the date such litigation is filed. This is not DFSG free. While software patents are generally held to be bad, it is not the purpose of a Free Software license to discriminate against who can use the software. [This also has the wierd side effect of effectively granting to each Contributor a license to use without royalty any of the Licensee's patents applicable to software, no matter how legitimate those patents are.] We should be dealing with invalid and/or improper software patents through legislation and/or litigation rather than adding usage restrictions to our formerly free software licenses. I'm unable to get into a long term discussion about this right now. Just be on notice that this kind of opinion is senseless and utterly counterproductive to free software development. So you want companies to grant perpetual, non-exclusive, worldwide, fully paid-up and royalty free patent licenses that are completely irrevocable even when another company is using their software and suing them for software patent infringement? Get a clue people. If a patent licence does not terminate under such circumstances then a free software friendly company may have no useful negotiating position whatsoever. It can't counterclaim with its own claims of patent infringement because it gave them up. The company needs to be able to say Yes, I may be infringing upon your software patent X but if you initiate legal action against me you may be infringing upon software patent Y. Care to negotiate? If it can't do that it is a sitting duck for any opportunistic litigant. While reciprocity cannot stem the rise of companies that are just patent licensing/litigation shells (that are unlikely to infringe upon patents because they don't produce anything) it may still just provide a slight disincentive for them to target free software if their business runs upon quality software products like Apache. No sane company will ever grant a perpetual, non-exclusive, worldwide, fully paid-up and royalty free patent licence without a reciprocity clause. Debian should not be indirectly legitimising the business endeavour of engaging in litigation against free software developers and destroying free software itself via licensing royalties. Debian should not be attempting to kneecap the potential negotiating positions and counterclaims of free software friendly companies in order for their software to be included in Debian. Regards, Adam
Re: [fielding@apache.org: Review of proposed Apache License, version 2.0]
On Sun, 09 Nov 2003, Adam Warner wrote: So you want companies to grant perpetual, non-exclusive, worldwide, fully paid-up and royalty free patent licenses that are completely irrevocable even when another company is using their software and suing them for software patent infringement? What I'd like to see is perpetual, non-exclsive, worldwide, fully paid-up and royalty free patent licenses that are completely irrevocable when included in a work of Free Software or a derivative of such a work. Useage restrictions like this are, plain and simple, not free. Consider the following: Entity Alfred is using a piece of software which is essential to Alfred's business model that Entity Betty has developed. The software is licensed under a license with a reciprocity clause. Alfred also develops Free and proprietary softare, and has numerous Software Patents, some of which Betty is possibly infringing upon. Betty sues Alfred for software patent violations unrelated to tne pieces of software under discussion. Alfred cannot counter sue Betty without loosing the ability to use the piece of software developed by Betty, without which, Alfred would go out of business (or have to replicate at prohibitive costs.) If a patent licence does not terminate under such circumstances then a free software friendly company may have no useful negotiating position whatsoever. We currently don't allow Free Software licenses to use copyright as a negotiating stick to beat upon their litigation opponents. If we were to allow such a clause for software patents, we might as well allow a similar reciprocity clause for software copyrights. Software patents are a serious problem, but restricting the usage of Free Software is not the solution. Don Armstrong -- Three little words. (In decending order of importance.) I love you -- hugh macleod http://www.gapingvoid.com/graphics/batch35.php http://www.donarmstrong.com http://www.anylevel.com http://rzlab.ucr.edu signature.asc Description: Digital signature