Weekly Edition
Return to the Announcements page
| |||||||||||||
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw)
Groklaw has a
transcript of Eben Moglen's keynote at LibrePlanet. "Freedom
will from here out be endangered. Freedom will be attacked, freedom will be
undermined, freedom will be evaded in various ways -- some of them clever
some of them stupid -- but from here on out, the relationship
between technological sophistication, agility, reliability, adaptability
and low cost, means that freedom has acquired an extraordinary set of
unintentional allies. They may not care about freedom at all, but they no
longer have a choice but to further freedom's interests." Video
is also available.
(Log in to post comments)
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 8, 2010 11:12 UTC (Sun) by FlorianMueller (guest, #32048) [Link] Eben Moglen's claims in that speech concerning some large corporations is questionable. He often tries to portray Microsoft as a unique archenemy of open source. In reality, other large players such as IBM, Oracle, Google and Apple also have closed/proprietary core businesses (execution of mainframe legacy workloads; RDBMS; search etc.) and will do anything to defend those against open innovation. The "enemy's enemy" principle is therefore not unique to Microsoft's competitors. However, allegiance to IBM is a consistent element all throughout Eben Moglen's biography. Concerning Oracle, it's worthwhile reading the European Commission's clearance decision on the Sun acquisition and search for the name "Moglen". One can see that Eben Moglen provided Oracle with a paper that became an annex to an important Oracle filing in the process. What's more important is that the European Commission makes several references to Eben Moglen's paper in its decision to point out that he was off-base on some key issues and also off-topic (in terms of what's relevant to a competition case): footnote 431, item 721, footnote 451 (makes it particularly clear that he just made a wild, baseless assertion concerning MySQL's business model), footnote 456 (on Eben Moglen making a claim about patent law without citing a single court decision, which doesn't suggest a very clean approach for a law professor...), footnote 457 (on a purposeful misrepresentation of the GPL's content), footnote 459, footnote 468 (again a claim without citing an actual court decision, and totally speculative on a business issue). The European Commission's intelligent rebuttals of the central claims made by Eben Moglen in that paper he wrote for Oracle call into question his credibility whenever large companies and their business models (quite apparently he's never run a business) and business interests are concerned.
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 8, 2010 17:15 UTC (Sun) by njs (guest, #40338) [Link]
Those are very interesting claims; thanks for looking through that, and providing the link. However, after skimming it for myself, I want to quibble some...
> footnote 451 (makes it particularly clear that he just made a wild, baseless assertion concerning MySQL's business model)
Moglen's "wild, baseless assertion" is an argument that MySQL has done okay with the dual GPL/proprietary model, but would do even better if it went GPL-only. I'm not sure I believe that -- and I'd like to at least see what Moglen's argument is before judging, never mind deciding it was "wild" and "baseless" -- but more to the point, this sounds *much* more like the sort of argument you'd expect from a strong FOSS advocate than from a corporate patsy?
> footnote 456 (on Eben Moglen making a claim about patent law without citing a single court decision, which doesn't suggest a very clean approach for a law professor...)
This is a very strange characterization of this footnote. The footnote doesn't say that he didn't cite a single court decision; it says that he didn't cite any court decisions that had ruled on the *particular issue under discussion*, and furthermore, they express doubt that any such decisions exist. Plus, it sounds like his statement was much weaker than your wording ("a claim about patent law") implies -- he said he "is 'doubtful' [...] as regards the main text's analysis in relation to the hypothetical claim...". I don't see a problem with a law professor, in the absence of relevant case law, making their best guess and expressing it in cautious terms, which seems to be what happened here.
> footnote 457 (on a purposeful misrepresentation of the GPL's content),
Suppose EvilDBCompany bought the ownership of MySQL's copyrights, and immediately ceased distribution under the GPL. Moglen's "purposeful misprepresentation", AFAICT, is to say that if this happens, then we can still legally get copies for MySQL from anyone who has them under the GPL's terms. The EC, on the other hand, suggests that as soon as EvilDBCompany purchases the MySQL copyrights, all extant copies suddenly become impossible to legally redistribute -- the GPL just sort of evaporates.
I don't know who's right here -- the EC has an interesting argument based on the exact wording of the GPLv2 -- but Moglen's "purposeful misrepresentation" is basically the same thing that's believed by everyone who has ever distributed GPL'ed code, and is clearly the intent of the license, modulo any subtle bugs in the wording. So he could be wrong, but if so then the GPLv2 is deeply broken and we have worse problems.
So overall, sure, I'm not sure I'd hire Eben Moglen as a business strategy consultant, since that's not his area of expertise. And maybe his analysis of the marketplace's strategic battlefield is not quite spot on, that's an interesting and important discussion to be having. But, what I'm getting from your comment is that you think we should go into that discussion expecting him to be wrong because he's at best clueless and at worst in the pocket of large corporations. And I just don't see how that follows from the evidence given.
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 8, 2010 17:37 UTC (Sun) by FlorianMueller (guest, #32048) [Link] I'd like to at least see what Moglen's argument is before judging He published his written submission to the EC on the Web. I call the business model stuff wild and baseless because MySQL AB was founded in 2001 and sold in early 2008 for $1 billion, and it had one of the most successful venture capital funds in history (Benchmark, which also invested in Red Hat and other open source businesses) on board. The assumption that they all got the business model wrong and failed to see an economically more attractive route (while Eben Moglen with apparently zero business expertise believes he knows better) is really unbelievable. more to the point, this sounds *much* more like the sort of argument you'd expect from a strong FOSS advocate than from a corporate patsy? I'd call it an argument that a true FOSS advocate might make, but so might a corporate patsy trying to preserve a reputation and masquerading as a FOSS advocate... The footnote doesn't say that he didn't cite a single court decision; it says that he didn't cite any court decisions that had ruled on the *particular issue under discussion* Yes, that related just to the particular issue, although his paper as a whole didn't cite any court decision either. I don't see a problem with a law professor, in the absence of relevant case law, making their best guess and expressing it in cautious terms, which seems to be what happened here. There's enough patent litigation going on all the time, in the US and also in Europe, that there's no scarcity of applicable case law if one wants to find it; but presumably a scarcity of case law that would support the point he tried to make in order to help Oracle. I don't know who's right here -- the EC has an interesting argument based on the exact wording of the GPLv2 -- I said purposeful because I can't imagine he doesn't know what the GPL says, and if he makes reference to it in a written submission to a regulatory agency, he could quote what's actually stated in there as opposed to talking about something that isn't in its text. basically the same thing that's believed by everyone who has ever distributed GPL'ed code, and is clearly the intent of the license, modulo any subtle bugs in the wording. So he could be wrong, but if so then the GPLv2 is deeply broken and we have worse problems. It's also always been my understanding that GPL'ing is an irrevocable decision for the code in question, although it doesn't constitute any obligation on the copyright holder to also GPL derived works. But, what I'm getting from your comment is that you think we should go into that discussion expecting him to be wrong because he's at best clueless and at worst in the pocket of large corporations. And I just don't see how that follows from the evidence given. I didn't say he'd always have to be expected to be wrong. What I said is that a great deal of skepticism is needed when business issues and business interests of large corporations are involved. I didn't say he's clueless or in the pocket of large corporations. I'm just very cautious when I hear him say those things. This has been the case since the first time I met him (June 2004), when he tried to discourage MySQL AB from lobbying against software patents and instead suggested to support his patent busting efforts concerning the FAT patents (which never got anywhere). Fortunately, MySQL AB didn't adopt his advice.
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 8, 2010 21:16 UTC (Sun) by nix (subscriber, #2304) [Link] I'd call it an argument that a true FOSS advocate might make, but so might a corporate patsy trying to preserve a reputation and masquerading as a FOSS advocate...Isn't this sort of thing pretty much the definition of an ad hominem attack? More to the point, it's also impossible to defend against. (And this series of comments of yours leaves a bad taste in my mouth.)
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 8, 2010 21:18 UTC (Sun) by nix (subscriber, #2304) [Link] I didn't say he's clueless or in the pocket of large corporations.So when you said However, allegiance to IBM is a consistent element all throughout Eben Moglen's biography.you were just throwing it out at random without trying to suggest any sort of connection at all? Please. A three-year-old child wouldn't be fooled by this. You did strongly imply he was in the pocket of a large corporation, to wit IBM.
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 9, 2010 5:38 UTC (Mon) by drag (subscriber, #31333) [Link]
> Eben Moglen's claims in that speech concerning some large corporations is questionable.
Your right. The guy is a lawyer, professor and academic; Not a business expert and not a economist. There is no proof that he could even run a 7-11*, much less give pertinent economic advice to a entire sector of business.
*(although he is certainly smart enough that it is likely he would manage just fine)
So like advice received from anybody else you have to use your own judgement.
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 9, 2010 14:29 UTC (Mon) by pboddie (subscriber, #50784) [Link] What's more important is that the European Commission makes several references to Eben Moglen's paper in its decision to point out that he was off-base on some key issues Please define "off-base". It appears that you might mean "unable to address specific points adequately, instead addressing related but distinct points", but in my experience "off-base" is a sort of cover-all business word that people use when they cannot or will not articulate their objection to what someone else has said. and also off-topic (in terms of what's relevant to a competition case): footnote 431, item 721, footnote 451 (makes it particularly clear that he just made a wild, baseless assertion concerning MySQL's business model), ... This last footnote is interesting, but it's far from a "baseless assertion" that MySQL AB could have made money the Red Hat way. Indeed, the criticism in the footnote is arguably uninformed: Red Hat - if one of the parties tangentially referenced in the decision - doesn't merely make money from shipping and supporting the "Linux kernel". And from what information one can readily dig up about MySQL AB's revenues (see this blog post, for example), with the bulk of the revenues coming from selling proprietary licences, one can wonder whether selling support was really a priority of the company or whether it was an area of missed opportunity that the company could have better exploited. ...footnote 456 (on Eben Moglen making a claim about patent law without citing a single court decision, which doesn't suggest a very clean approach for a law professor...), footnote 457 (on a purposeful misrepresentation of the GPL's content), footnote 459, footnote 468 (again a claim without citing an actual court decision, and totally speculative on a business issue). This last footnote is amusing: expecting a lawyer to give specific business advice in a competition investigation. Don't people pay good money for that kind of thing? The European Commission's intelligent rebuttals of the central claims made by Eben Moglen in that paper he wrote for Oracle call into question his credibility whenever large companies and their business models (quite apparently he's never run a business) and business interests are concerned. As for Moglen's credibility, he seems to have advised numerous companies over the years, big and small, and has managed to point out various peculiarities in the Statement of Objections (such as the "threat" to storage vendors from the copyright holder actually turning out to arise from a specific vendor getting into a contractual dispute with MySQL AB). Maybe you meant "motivations" instead of "credibility", but even then, I don't see much more than the message that Free Software projects are no more at risk from an Oracle-owned MySQL than if someone else had bought the technology. We all know that much of the wailing on the matter of this particular acquisition comes from proprietary software vendors (some of whom wanting to have a second shot at selling something they once owned), but that shouldn't mean that any response from the Free Software community should in any way defend these people's particular interests. I'll accept that people should open their eyes about the behaviour of corporations other than Microsoft (the "yay, IBM!" crowd, for example), but seeing nothing spectacularly bad about an acquisition isn't the same as doing a corporation's bidding.
Patent royalties are the lesser evil, exclusionary use is the real threat Posted Aug 8, 2010 11:35 UTC (Sun) by FlorianMueller (guest, #32048) [Link] In my previous comment I pointed out why I'm generally skeptical about Eben Moglen's advice concerning business models and large corporations. The passage quoted above and the relevant part of that speech refer to Microsoft collecting patent royalties. If it were up to me, there would be no software patents and therefore no such royalties. However, Eben Moglen focuses on the much lesser evil (a company making patent licenses available and demanding royalties for them). The real problem is exclusionary use: demanding the removal of featuers from products, or requiring vendors to take products out of the market. That's what really hurts innovation and can kill products or even entire companies. IBM has a tradition of using (or at least trying to use) patents that way in connection with the mainframe; Apple apparently never offered a license deal to HTC but simply wants HTC to drop multi-touch; Google doesn't make some of its patents (especially on search) available to anyone else. I don't know what the future will bring, but so far Microsoft has focused completely on license deals as opposed to requiring alleged "infringers" to remove features from their products or take products out of the market. Right now it would be a major improvement if all large corporations could at least do what Microsoft does. Eben Moglen, in my opinion, completely fails to address the most serious issue in connection with patents. Exclusionary strategic use of patents is even worse than trolls.
Always exclusionary to free software Posted Aug 8, 2010 11:48 UTC (Sun) by job (guest, #670) [Link]
This is only partly true. Most licensing deals exclude certain business models.
The right to freely distribute software is inherently incompatible with not only per-unit licensing but also one-time licenses which are either revocable or not redistributable. The very fact that you have a patent license that you are unable to relicense when distributing your software may very well lock you out of the free software ecosystem.
Always exclusionary to free software Posted Aug 8, 2010 11:57 UTC (Sun) by FlorianMueller (guest, #32048) [Link]
If the deal structure is that a given company must pay royalties on a given set of commercial products or services, then it's just a revenue-sharing arrangement.
I agree that per-unit patent royalties might restrict distribution. That is, by the way, why I believe MPEG LA's AVC/H.264 licensing terms with a royalty cap aren't too bad: major players like Mozilla and Opera could accept them, while any per-unit royalty, no matter how small, would make free distribution an incalculable economic risk. And that is one of the reasons why FOSS is particularly threatened by software patents. All of this really depends on the deal structures and their contractual implementation. The companies from which Microsoft appears to collect royalties don't seem to be locked out of the free software ecosystem. At least I'm not aware of any of them having ceased the distribution of anything. Maybe that's because those don't really give much back to the free software ecosystem anyway. Also, those tend to be, in general, companies who themselves want software patents and use them to their advantage (example: Amazon...).
Always exclusionary to free software Posted Aug 11, 2010 7:40 UTC (Wed) by job (guest, #670) [Link]
Even with a royalty cap your customers need to get a license from you. That is incompatible with free software. This would otherwise be a loophole where you could distribute GPL software, take a patent on it and use it to nullify the user's rights. The GPL is pretty harsh about distributing things that can not in turn be redistributed.
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 8, 2010 16:26 UTC (Sun) by gvy (guest, #11981) [Link]
It's a tad bit sad to hear so much on freedom from a country which is deeply buried in personal and national debt. One can do HH with free software on his free/liberated device -- but can't avoid loans.
Ah yes, and it might just be economically better to use free software to further enslave the very those who write it.
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 8, 2010 16:47 UTC (Sun) by JoeBuck (subscriber, #2330) [Link] You are not hearing from a country. Moglen isn't a country, he's a person, and clearly not a person whose views are mainstream in the US. You are dragging in an unrelated political discussion: the US debt has nothing to do with this.
Eben Moglen's LibrePlanet 2010 Keynote (Groklaw) Posted Aug 9, 2010 5:34 UTC (Mon) by drag (subscriber, #31333) [Link]
Yes:
USA != USA Government != USA individuals.
There are plenty of people floating around with zero debt. So far your still allowed to run your life how you see it and lots of people refuse to run their's into the ground.
The LLVM jab is amusing Posted Aug 8, 2010 18:47 UTC (Sun) by renox (subscriber, #23785) [Link]
The LLVM jab is amusing: sure Apple is behind LLVM for their own reason but there are quite a few contributors which are just fedup with the FSF's use of GCC as a 'political' tool:
I'm thinking about some developers which tried (and failed) to make a D frontend for GCC, so now other are targetting LLVM.
Sure the FSF has authorised now external contributions, but this is quite late and LLVM's momentum is impressive..
The LLVM jab is amusing Posted Aug 9, 2010 1:21 UTC (Mon) by Zack (guest, #37335) [Link]
I'm not sure the jab is at LLVM or its developers. LLVM is Free Software after all.
I think it's more directed at apple throwing their weight behind LLVM for reasons that can only be related to not wanting to comply with the gplv3.
Seeing how apple has been abusing patents and how the main perceived difference between the gplv2 and gplv3 is the explicitness of the patent language, it's a reasonable assumption that apple avoids the gplv3 specifically because of patent related reasons.
You may be correct that there are contributors to LLVM that are simply fed up with the FSF's use of gcc as a polical tool, and it's hard to blame them. But apple's support "for their own reason" is likely not to be a personal dislike for the FSF but for their own "political reasons", which are likely to be patent abuse related.
The LLVM jab is amusing Posted Aug 9, 2010 1:47 UTC (Mon) by foom (subscriber, #14868) [Link]
If everyone who wanted to abuse patents first went and wrote (or, at least, massively contributed to!) such a fabulously useful piece of Free Software, the world would certainly be a better place.
Maybe Apple's reasons are not 100% pure, but, really, LLVM is *awesome*. A set of compiler components with competitive optimizations, available in library form, with the sub-pieces able to be invoked separately? About time!
As much as I hate what Apple is doing with the iPhone, I really don't see how you can say their contribution to LLVM is a bad thing. It's providing significant brand new functionality to the Free Software world, not just a replacement of GCC.
The LLVM jab is amusing Posted Aug 9, 2010 4:59 UTC (Mon) by gmaxwell (subscriber, #30048) [Link]
LLVM is conceptually pretty neat, but its actual delivered results, in terms of the quality and performance of the resulting binaries compared to GCC, are not yet impressive. LLVM's fans do it no service to overstate its qualities prematurely.
While LLVM remains arguably inferior to GCC in some regards it does an opponent of GCC (or, rather, GCC's copyleft licensing) little harm to publicly collaborate on the improvement of LLVM towards matching GCC: Giving the free software world two GCC level compilers doesn't reduce your ability to offer something superior to what the free software world has. But after meeting that point there is no reason to believe why the most valuable Apple's further contributions won't take the form of licensing restricted binary-only software.
We've already seen this happen with LLVM: Adobe's Alchemy is binary-only licensing encumbered LLVM to FlashVM byte code converter as well as some binary only LLVM bytecode optimization passes.
GCC's development model and technical design have historically inhibited people from creating and distributing enhancements. LLVM's licensing model allows for economic incentives against the same thing.
In both cases people who care primarily about software freedom loseĀ though they lose more in the LLVM case because absent LLVM the proprietary world also went without these improvements except for the few companies with the resources available to allow them to build a complete GCC competitive compiler suite on their own.
The LLVM jab is amusing Posted Aug 9, 2010 19:50 UTC (Mon) by foom (subscriber, #14868) [Link]
> LLVM is conceptually pretty neat, but its actual delivered results, in terms of the quality and performance of the resulting binaries compared to GCC, are not yet impressive. LLVM's fans do it no service to overstate its qualities prematurely.
I don't really care about it as a fixed binary compiler to replace GCC. We already have GCC for that, and it works pretty well at that.
I like all the *other* things you can do with LLVM, such as call it from within a program to do JIT compilation, or use the C++ frontend to parse the code, but not compile it. Those are things that you just can't do with GCC, and yet are incredibly valuable features. LLVM is infinitely better than GCC at those. :)
That it also compiles to code that has competitive performance (not *better* than GCC, sure, but pretty good!) is great too.
>We've already seen this happen with LLVM: Adobe's Alchemy is binary-only licensing encumbered LLVM to FlashVM byte code converter as well as some binary only LLVM bytecode optimization passes.
Sure, but we've also seen things like LightSpark, Unladen Swallow (Python backend targetting LLVM), the IcedTea JVM JIT support for non-x86/sparc CPUs, Gallium3D shader compilation...and those are just the ones I know of off the top of my head...LLVM enables amazing Free Software to be built that probably could not have been feasibly built before, due to the high cost of building an optimizing compiler.
So, you can build proprietary software with it too...shrug, oh well. The Free Software projects people are already building far outweigh any "harm" done by allowing proprietary software to also be built.
The LLVM jab is amusing Posted Aug 15, 2010 11:50 UTC (Sun) by nakano (guest, #69567) [Link]
Well Apple still seem to be trying to hamper the LLVM project by patenting the side effects of producing various frontends to LLVM with patents like this recent one:
http://www.faqs.org/patents/app/20100153929
Any JavaScript frontend for generating IR would violate this patent (they even mention LLVM IR explicitly in this patent).
The patent has only recently been published and can still be opposed. The patent harms anyone considering using LLVM as an intermediate between JavaScript and back-end machine specific executables.
Smells bogus to me i.e. prior art:
So much for Apple not abusing its contribution to LLVM.
The LLVM jab is amusing Posted Aug 15, 2010 23:52 UTC (Sun) by bronson (subscriber, #4806) [Link]
Great, Apple just patented the UCSD p-System.
The LLVM jab is amusing Posted Aug 19, 2010 9:26 UTC (Thu) by renox (subscriber, #23785) [Link]
Thanks for posting this.
I wonder why this was not widely discussed?
If I were a LLVM contributor, I would be very annoyed|worried..
The LLVM jab is amusing Posted Aug 19, 2010 14:10 UTC (Thu) by foom (subscriber, #14868) [Link]
With the way the system works today, it's not particularly reasonable to say that someone is doing something wrong by filing or owning a patent. Almost everything in software is patented by someone-or-other, those patents just aren't being asserted against you. Really it's only the use of the patent in an aggressive manner which is the bad thing.
Of course, if there is prior art, it's certainly best to get the patent invalidated before it's issued; that'd avoid any possibility of aggressive use in the future. But in this case, I bet you *won't* be able to find prior art for every one of the 34 claims, and you'd have to fall back on the more difficult to prove "obviousness" defense to knock out some of them.
Of course, if apple has actually pledged the patents to the LLVM project already, they're doing it a *favor* by getting the patents filed, since the patent office only seems to look for prior art in the patent database by default, and not in wider literature.
The LLVM jab is amusing Posted Aug 19, 2010 11:18 UTC (Thu) by coriordan (guest, #7544) [Link]
Another Apple patent covering LLVM is this one: http://www.patentstorm.us/patents/6263421/claims.html
LLVM's contributor policy requires contributors to give free access to necessary patents. http://llvm.org/docs/DeveloperPolicy.html#patents
Anyone know if Apple has done this? Is there a list of declared patents anywhere?
I'm documenting this here:
Another new Eben video Posted Aug 9, 2010 19:50 UTC (Mon) by coriordan (guest, #7544) [Link]
Very interesting. It's a more developed version of his "freedom in the cloud" talk from a few months ago.
Another new Eben video Posted Aug 9, 2010 20:53 UTC (Mon) by kragil (subscriber, #34373) [Link]
You're welcome.
I love the idea of the freedombox and I really hope the Debian guys will put a "Pure Blend" together once the diaspora guys release their stuff next month (OK, I know I am being very optimistic here, but I think something like this is really needed. I would run it.)
I am in the moot to brainstorm what would be needed.
-a DHT where all the freedomboxen publish their IPs (Retroshare and Emule use those)
Not easy, but still not something the FOSS community couldn't put together.
Another new Eben video Posted Aug 9, 2010 20:58 UTC (Mon) by dlang (✭ supporter ✭, #313) [Link]
take a look at roundcube http://roundcube.net/ it seems to be pretty decent (at least on a normal browser, it fails miserably on portable devices)
there are really good IMAP mail servers out there (I use Cyrus). the problem is in the spam filter (which is a separate component)
Another new Eben video Posted Aug 13, 2010 15:22 UTC (Fri) by stevem (subscriber, #1512) [Link]
See
http://wiki.debian.org/FreedomBox
for further discussion and ideas prompted by Eben's inspiring speech at DebConf.
|
|||||||||||||