The FSF objects (again) to the Google book settlement [LWN.net]

Here's a release from the Free Software Foundation detailing more objections to the proposed Google book search settlement. Essentially, they see it as a way for Google to bypass the requirements of the Free Documentation License. "But under the proposed amended settlement, Google would generally have permission to display and distribute these works without abiding by the requirements to pass the freedoms guaranteed under the GFDL on to Google Books readers. Authors who wanted to use the GFDL or another free license would be required to designate that license in a Registry -- and the Registry would determine which licenses could and could not be chosen"

to post comments

The FSF objects (again) to the Google book settlement

Posted Feb 3, 2010 5:26 UTC (Wed) by Requiem (guest, #51519) [Link] (1 responses)

Or they could just tell Google not to display their works at all if passing the GFDL on isn't done. That is an option.

can't work

Posted Feb 3, 2010 6:36 UTC (Wed) by coriordan (guest, #7544) [Link]

Just this once, just for Google, just for FSF's books, yes, that would work.

But if we want GFDL'd books to always come with the freedoms their licence grants, then we need a general solution - one that will work for all Google's book programs, and for the book programs of all redistributors, and for all GFDL'd books (not just FSF's, not just those of whoever hears about a particular program of distribution of their book and had the time to do an explicit opt-out).

Missing the big picture

Posted Feb 3, 2010 9:09 UTC (Wed) by djao (guest, #4263) [Link] (8 responses)

I usually agree with the FSF, and am skeptical of Google, but in this case, my sympathies are reversed. The FSF is really missing the forest for the trees.

The FSF's ultimate goal is to give all computer users the freedom to share software and other creative works such as books, music, movies etc. without undue restriction. Google Books contributes positively, not negatively, towards that goal. Using Google Books, we have a situation where for the first time in human history, any member of the general public can efficiently and quickly search all written works ever published, in full text, for free, on demand at any time. Administered properly, it's like giving each internet user free, continuous, and unrestricted access to the Library of Congress. Unfortunately, the effectiveness of Google Books is limited by uncooperative publishers, who employ copyright restrictions to prevent certain works and certain pages of works from being represented in searches. This settlement lessens the impact of such opposition, by subjecting all publishers to the same uniform terms, terms which seem to be fairly reasonable.

It is the nature of such settlements that many of the individual parties affected by the settlement will be affected negatively. In this case, the FSF, and FDL users in general, happen to be among the parties who are affected negatively. Clearly, however, traditional publishing houses have a lot more to lose from this than the FSF. In this case, the FSF should subscribe to the theory that "the enemy of my enemy is my friend," because this settlement, and Google Books in general, represents such a dramatic and positive change in the copyright regime that it deserves support despite the concerns about free licenses.

Imagine if all copyright restrictions on all software were suddenly abolished, and the FSF objected to the change because of its effect on GPL software. To a lesser extent, that's exactly what's going on here.

Missing the big picture

Posted Feb 4, 2010 17:38 UTC (Thu) by rsidd (subscriber, #2582) [Link] (7 responses)

Imagine if all copyright restrictions on all software were suddenly abolished, and the FSF objected to the change because of its effect on GPL software.

The situation may be imaginary but the FSF's reaction to such a scenario isn't. It is here. Short summary: the FSF is concerned with availability of source code with certain user rights, not with removing copyright. A copyright-free regime would defeat the FSF's goals and, in particular, make the GPL unenforceable.

Missing the big picture

Posted Feb 5, 2010 2:01 UTC (Fri) by djao (guest, #4263) [Link] (6 responses)

Indeed, the FSF's response in the link you pointed out suffers from the same flaw, although the issue there is confounded by the presence of EULAs. One thing the FSF failed to mention is that in practice, EULAs depend on copyright for their existence -- without copyright, a EULA would have no legal enforcement mechanism, just as EULAs on hardware like the CueCat proved to have no force in law.

I find also that much of the force of the FSF's moral argument disappears once you start requiring publishers to take active steps such as releasing source code. The argument against copyright on computers was always that copying files is so easy that to prohibit it requires oppressive, draconian, and totalitarian measures. In practice, removing these restrictions would accomplish about 95% of the good that the FSF is trying to bring about. To oppose such a change on the grounds of the remaining 5% is shortsighted -- far better to accept the 95% improvement immediately, and then fight for the remaining 5% later.

Finally, in the context of books and printed matter, which is what we are discussing here, there is much less of a difference between object code and source code. "Decompiling" books in most cases is nothing more than OCR, which is a tractable technical problem on which great progress has been made over the years. So the FSF's concerns about software freedom from the point of view of the distinction between object code and source code are much less relevant here.

Missing the big picture

Posted Feb 5, 2010 3:12 UTC (Fri) by rsidd (subscriber, #2582) [Link] (3 responses)

The argument against copyright on computers was always that copying files is so easy that to prohibit it requires oppressive, draconian, and totalitarian measures.

Some people have made such an argument against copyright, but not, as far as I know, the FSF. In fact I don't think the FSF has ever argued against copyright. Their only cause has been availability, modifiability and redistributability of source code to the user. The original motivation was Stallman's being confronted with a buggy binary-only printer driver, back in 1981 or so.

Missing the big picture

Posted Feb 5, 2010 8:55 UTC (Fri) by djao (guest, #4263) [Link] (2 responses)

RMS has definitely argued against large portions of the current copyright system, including the prohibitions on private copying, both in person and in print. I consider RMS to be philosophically the same entity as the FSF.

From "Free Software, Free Society: Selected Essays of Richard M. Stallman," p. 144:

By allowing the noncommercial verbatim copying, it means the copyright no longer has to intrude into everybodys home. It becomes an industrial regulation again, easy to enforce and painless, no longer requiring draconian punishments and informers for the sake of its enforcement. So we get most of the benefit, and avoid most of the horror, of the current system.

and again on p. 153:

Im convinced now that people should have the right to publicly, noncommercially, redistribute verbatim copies of everything.

What is interesting especially about the latter quote is that it is directly relevant to the debate that we are having right now. Suppose that I want to noncommercially redistribute verbatim binary copies of GPL software without also giving out the source code. (Assume GPLv2 for concreteness.) According to a literal reading of the above quote, RMS is stating that this should be allowed. However, the GPL does not normally allow this. There is a slight exception in Section 3(c), but even that exception contains an obligation to transmit certain pieces of accompanying information along with the binaries -- and suppose that I do not do that. We now have a scenario that corresponds almost exactly to what Google is doing, namely transmitting binary-only copies of GFDL works without transmitting essential accompanying license information, except that Google of course is a commercial entity.

Even by the FSF's uncompromising standards, this topic lies in a gray area and involves conflicting principles. It's definitely not clear that the FSF's position is the one that most furthers their own goals.

Missing the big picture

Posted Feb 5, 2010 9:24 UTC (Fri) by rsidd (subscriber, #2582) [Link] (1 responses)

Interesting: I didn't know he had said that. Certainly, as you say, such a
stance undermines the GPL and also contradicts what he said in the Pirate
Party case.

But one should note that this was in a question-answer session after a talk,
so it may have been a spur-of-the-moment response and not well thought out.
(Though he sounds categorical enough, as always.)

Stallman and copyright

Posted Feb 5, 2010 15:30 UTC (Fri) by pboddie (guest, #50784) [Link]

My impression, after attending a talk by Stallman last year, was that he's taking copyleft and formulating a broader system as a reform of copyright - sort of like going from special relativity (copyleft) to general relativity (copyleft for programs and content), or something. Thus, his arguments touch upon some of the things that arise in the spectrum of Creative Commons licensing: there may be things such as scientific papers where you want to impose restrictions on modification or redistribution (somewhat like CC-BY-ND) in order to prevent cherry-picking of regions of the text to support or suggest the opposite of the meaning of the text; other works might be freely shared provided you aren't profiting from that sharing (CC-BY-NC); software would naturally observe a system much like copyleft is today.

Missing the big picture

Posted Feb 9, 2010 11:40 UTC (Tue) by nye (guest, #51576) [Link] (1 responses)

>EULAs depend on copyright for their existence -- without copyright, a EULA would have no legal enforcement mechanism

IANAL, but I believe this is not true.

Copyright has no control at all over usage. Once the copyright owner has sold or otherwise conveyed a copyrighted work to an and user, that user is free to use it in any way they wish without requiring permission, subject to the restriction that they do not copy/broadcast/redistribute it without permission from the copyright holder. Even in jurisdictions where a EULA is considered enforceable, there is no legal requirement to accept it in order to use the software[0]. However, if you *do* click through the EULA, then it is often considered to be equivalent to having signed a contract, to which you are now legally bound. This would be the same regardless of whether or not copyright applies.

[0] Some claims have been made that loading software into memory in order to run it counts as copying and thus requires a license, and there may be some jurisdictions which uphold that repugnantly deliberate misinterpretation of the law. In this case a EULA would indeed be required to use software, but this interpretation is absurd - eg. it would probably mean that you can only run a GPL program if the source is loaded into memory alongside the binary.

Missing the big picture

Posted Feb 9, 2010 22:12 UTC (Tue) by djao (guest, #4263) [Link]

Even in jurisdictions where a EULA is considered enforceable, there is no legal requirement to accept it in order to use the software

True.

However, if you *do* click through the EULA, then it is often considered to be equivalent to having signed a contract, to which you are now legally bound.

Debatable, but for the sake of argument, let's assume it's true.

Now put the two together.

Without copyright, it would be legal to distribute modified versions of software. One could, for example, modify a piece of software to remove the click-through dialog for the EULA, and then distribute it. Since clicking the EULA was never a requirement for using the software in the first place, the result would be that people could use the software without clicking on the EULA. This means that, in practice, companies would have no way to force users to click through a EULA. A company could still impose a EULA using a paper contract, as is true in any industry, but the widespread abuse of EULAs would cease if click-throughs were abolished. It is in this sense that EULAs (as they exist today) depend on copyright law.

Even if the EULA itself forbids modifying the software, it is technically possible for people to modify software without clicking through the EULA, or even installing the software.

The FSF objects (again) to the Google book settlement

Posted Feb 3, 2010 11:52 UTC (Wed) by trasz (guest, #45786) [Link] (5 responses)

So basically, FSF wants to prevent Google from giving people easy access to the books, because that would reduce the users' freedom?

The FSF objects (again) to the Google book settlement

Posted Feb 3, 2010 12:15 UTC (Wed) by coriordan (guest, #7544) [Link]

Basically, FSF wants Google to comply with the licence, like everyone else does.

The FSF objects (again) to the Google book settlement

Posted Feb 3, 2010 12:30 UTC (Wed) by coriordan (guest, #7544) [Link]

FSF's first objection is more descriptive of the problem:

http://www.fsf.org/licensing/google-book-search-objection...

For example, Google's system displays large parts of the books without necessarily displaying the licence. So they're distributing a partial copy and chopping off the licence text (by negligence, not by malice).

To comply, Google should display the full licence whenever they display GFDL'd works. That's one example problem.

The FSF objects (again) to the Google book settlement

Posted Feb 3, 2010 15:23 UTC (Wed) by Trelane (guest, #56877) [Link] (2 responses)

http://www.fsf.org/news/2010-02-gbs-objection

"The GFDL is a copyright license that authors use for their works when they believe others should have the freedom to share and improve those works. It was designed primarily for use with technical documentation, but has been used for many different kinds of written works -- from print biographies to Wikipedia articles. Whereas copyright is normally used to prohibit others from distributing works, the GFDL encourages sharing, with the requirement that any such redistribution must also be under its pro-sharing terms.

"But under the proposed amended settlement, Google would generally have permission to display and distribute these works without abiding by the requirements to pass the freedoms guaranteed under the GFDL on to Google Books readers. Authors who wanted to use the GFDL or another free license would be required to designate that license in a Registry -- and the Registry would determine which licenses could and could not be chosen."

Perhaps not surprisingly (since they come from the same organization, it's like GPL. The point is to use copyright to preserve the end-user's freedom, by removing the intermediaries' freedom to remove the end-users' freedom. Is the freedom to remove someone else's freedom a freedom we wish to preserve?

You're using the BSD argument that it's the intermediaries' freedom which is the most important; the FSF believes that it's the end-users'.

The FSF objects (again) to the Google book settlement

Posted Feb 4, 2010 3:27 UTC (Thu) by mheily (guest, #27123) [Link] (1 responses)

The GFDL with Invariants is hardly a "free licence"; ask Debian, or Wikipedia.

> The point is to use copyright to preserve the end-user's freedom, by removing the intermediaries' freedom to remove the end-users' freedom. Is the freedom to remove someone else's freedom a freedom we wish to preserve?

Would you like some freedom fries with that? The word "freedom" is so overused as to be practically meaningless. Please be more specific.

I fail to see how anyone's freedom is lost when Google displays an excerpt of a book. The original work is not harmed or modified. End users are always free to access the original work in it's entirety. Without the assistance of Google's search and indexing system, the end user might never find the document at all.

> You're using the BSD argument that it's the intermediaries' freedom which is the most important; the FSF believes that it's the end-users'.

Not true. The BSD license does not discriminate between end-users and intermediaries. All recipients of the code are treated equally.

The FSF objects (again) to the Google book settlement

Posted Feb 4, 2010 12:50 UTC (Thu) by pboddie (guest, #50784) [Link]

I fail to see how anyone's freedom is lost when Google displays an excerpt of a book.

I'm sure the problem lies in the subsequent use of such excerpts. If people treat them as "public domain" stuff as in "I found this on the Google Internet - it would look great in my book/thesis/homework" and any notion of authorship is removed, someone's rights or privileges (if not exactly freedoms) are being curtailed.

The BSD license does not discriminate between end-users and intermediaries. All recipients of the code are treated equally.

Sure, in the sense that someone who gets a modified piece of BSD-licensed software as a binary with only the copyright and permissions statements attached has the same access (in theory, depending on the chain of distribution) to the original project's source code as the people who gave them the binary, but we all know that the distinction between copyleft and permissive licensing is about the end-user being able to use/modify/redistribute the actual source code for the entire system (or a specific part of that system).

In fact, what should have been asserted was that you're defining freedom in terms of what has been given rather than what is received.