FSF Releases Draft of Version 1.2 of the GNU FDL

bkuhn writes: "The FSF has released a draft version 1.2 of the the GNU Free Documentation License for comment by the Free Software community. Comments should be directed to <fdl-comments@fsf.org>."

Comments

Now if we only knew...

[tfurrows]

... what precentage of those in opposition were slashdotters. I know it was this slashdot post that got me on my toes.

Can it be that slashdot is effectively preserving our rights and liberties? It's nice to know that right is slowly winning out in all of this.

Re:Now if we only knew...

[tfurrows]

Mod me down for being a moron! I posted this comment while reading a freakin' Register article! DOH!

If anyone is interested in WHAT THE HECK I AM TALKING ABOUT please check out this article at The Register

It's going to be a long day....

Re:Now if we only knew...

[tfurrows]

Wow.... I didn't think they'd actually mod me down, heck I was hoping to get modded up for being funny in my reply...

Don't you moderators have a sense of humor? It was an honest mistake- honest!

(P.S. Now you mod me up for having the nerve or sheer stupidity to post AGAIN!)

Problem

[Anonymous Coward]

If I cite or quote a portion of a FDL text in my non-FDL text, that portion of my text remains mutable.

For it to remain mutable requires that my copyright be non-applicable in certain areas. It also requires that someone else be able to reproduce my work verbatim with a modified FDL quote. Thus, either the FDL is invalid once a text is cited, or it invalidates general copyright once a text is cited.

Now invalidating non-FDL-licensed works may sound like a good idea to some (screw them thar IP hoarders!). However, such a stupid loophole or trojan horse (however you want to look at it) is simply asking for a lawsuit down the road.

Also, the copyright date should be 2002.

The license is non-free?

[OiBoy]

From the FDL:

"Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed."

I see a bit of a dichotomy here.

Re:The license is non-free?

[bcrowell]

The GFDL is a GPL-style license, which means it's viral. You can't have a viral license if people are allowed to change the license after the fact. If you don't like GPL-style licenses, you can make your work public domain, or choose a different license.

The whole point of a GPL-style license is to take away people's freedom. For example, there might be a hacked Perl interpreter linked into the software that runs your PDA, and you might never know it, because Perl's license gives the PDA manufacturer the freedom to change Perl, use the changed version, and not publish the changes. A GPL-style license doesn't give you the same freedom.

Of course, "free" isn't necessarily the same as "good," as implicitly assumed by Stallman's rhetoric. I sometimes use the GFDL, because I want to take away certain freedoms from people who use use my books. For instance, I may not want them to have the freedom, which they would have under a BSD-style license, to use excerpts from my book in their own book, without making their book free-as-in-anything.

Re:The license is non-free?

[Anonymous Coward]

I may not want them to have the freedom, which they would have under a BSD-style license, to use excerpts from my book in their own book, without making their book free-as-in-anything

I think this may violate general copyright's concept of fair use. Now, if that is your goal, then so be it, but I think it is one of the more insiduous portions of the FDL.

Plagiarism is already illegal, so I don't see why this restriction is necessary except to make FDL'd texts off-limits to publishers.

Example: O'Reilly wants to publish a book about Language X that is under GPL using information from the manual which is licensed under FDL. Under normal copyright, the manual could be excerpted and expounded upon by the O'Reilly author without O'Reilly losing their own copyright. However, under the FDL, the manual could not be used as a source without O'Reilly having to also release their book under the FDL or "freer" license.

The FDL is bad for publishers and authors. It isn't even very good for original text writers as they don't gain very much from the license.

side-by-side diff

[bcrowell]

Here is a side-by-side diff. The main change seems to be that they've added some explicit discussion of no-warranty clauses. Not a momentous change, really -- does anyone really get sued because they gave away a book for free, and the book had a mistake in it???

Re:side-by-side diff

[Pembers]

Another difference appears to be that they've added restrictions on the lengths of Front-Cover and Back-Cover Texts (5 and 25 words, respectively).

These limits seem a little arbitrary. Also, they might be inappropriate for languages other than English. Most European languages would be OK, but I gather that something like Chinese uses many short words. At the other extreme are languages that say as much in one long word as English says in a sentence (Inuit, I think, is one example). Even Latin gets by with about half the number of words in a sentence that English normally does.

Could this cause problems, or am I just picking nits here?

Re:side-by-side diff

[Arandir]

No, it could very well be a problem. I can understand GNU wanting to limit the front and back texts, otherwise you could start including stuff that shouldn't be there (like an index). But five words is much too short for the front matter, and 25 words is much to short for the back.

In my opinion, I think GNU is going overboard on their copyleft crusade. It fits works okay for technical documentation associated with copyleft software, but it falls down for a lot of other stuff.

In my never humble opinion, just use the same license for the documentation as for the software. Make it all one package and license it as a whole. If it's good enough for the software it's good enough for its documentation.

free use? bad for authors?

[bcrowell]

I think this may violate general copyright's concept of fair use.
Well, the GFDL can't trump fair use. The only reason a reader would need to agree to the GFDL is to gain permission to copy the text. If they're copying under the fair use doctrine, they don't need permission from the copyright owner, and they don't need to agree to the GFDL.

Plagiarism is already illegal,...
No, it's not.

...so I don't see why this restriction is necessary...
From the FSF's point of view, I think the strategy is to built a worldwide corpus of free information. By forbidding the mixing of GFDL'd information with GFDL-incompatible information, they want to arm-twist people into GFDL-ing their information. It's the same strategy that was such a big success for the GPL and Linux.

From my point of view, the point is that if someone else wants to use one of my physics homework problems, verbatim and with credit, in their own GFDL'd physics textbook, they can. On they other hand, they can't just publish my whole book as a not-free-as-in-anything book, without dealing with me. Yes, I want to prevent this, since I sell printed copies myself.

...except to make FDL'd texts off-limits to publishers.
There are three perfectly good ways around such a problem: (1) the author can choose a BSD-style license; (2) the author can offer the book to the publisher under some separate contract, rather than under the GFDL; (3) the publisher can GFDL the stuff with which they're mixing the author's GFDL's work.

[The GFDL] isn't even very good for original text writers as they don't gain very much from the license.
Would you complain that the GPL isn't good for programmers because they don't gain much from the license? There are ways to make money from GPL'd code, but for most people it's a hobby; they have some other way to pay the rent.

Re:free use? bad for authors?

[Anonymous Coward]

Plagiarism is already illegal,...
No, it's not.

But suit can be brought against the plagiarizer. Perhaps because plagiarism is a violation of copyright? Hence, illegal.

if someone else wants to use one of my physics homework problems, verbatim and with credit, in their own GFDL'd physics textbook, they can...they can't just publish my whole book as a not-free-as-in-anything book, without dealing with me.

But quoting you is, as defined by the GFDL Section 4, releasing a Modified Version of your document. There is no concept of fair use in the GFDL at all allowing selective quoting.

There are three perfectly good ways around such a problem: (1) the author can choose a BSD-style license; (2) the author can offer the book to the publisher under some separate contract, rather than under the GFDL; (3) the publisher can GFDL the stuff with which they're mixing the author's GFDL's work.

Under a general "unfree" copyright, none of that rigamarole is necessary. Simply select the quotes and attribute them. The GFDL goes far beyond that, requiring such extreme measures on the part of publishers.

Would you complain that the GPL isn't good for programmers because they don't gain much from the license?

No. A nicely working GPL system results in a feedback loop of improvements back to the original author. These improvements are not necessarily easily implemented, and may be beyond the abilities of the software author.

A manual or other text is easily edited and the author is able to solicit feedback without having to give up redistribution rights to his own writings.

I don't see any reason an author wouldn't simply retain general copyright.

Re:free use? bad for authors?

[Arandir]

It's the same strategy that was such a big success for the GPL and Linux.

That's still open for debate. Without a suitable control group, there's no way to know. You can't look to BSD for guidance, since they were under a bogus lawsuit at the time Linux got started. You can't look at GNU as-a-whole, since it never took off until Linux.

But you can look at a several smaller projects. Here we see myriad GPLd projects that bit the dust and myriad MIT/BSD-style projects that took off. In fact, of the top four most successful Open Source projects (Apache, Linux, Perl, XFree86), four are not under a copyleft license.

Call me a heretic, but from my vantage point, the success of a project has a lot more to do with the project management, structure and culture than with its licensing.

Re:free use? bad for authors?

[Anonymous Coward]

That's funny, my /usr/src/linux/COPYING looks a lot like the GPL (version two). And Perl (which doesn't really enable anything you can't also do with a less deranged language) is hardly in the same class as, say, GCC.