At 04:56 PM 8/10/2008, Thomas Dalton wrote:
> > Are you saying that even if someone else published a work under
> > CC-BY, you couldn't publish a derivative work as a Knol article, even
> > if you attributed the base work to the author as required by CC-BY,
> > and published it under Knol's own CC-BY licensing option?
>
>That's my understanding, yes.
>
> > Would it be, not because you can't re-mix a CC-BY work into your own
> > CC-BY work (everyone says you can do this), but because you can't
> > grant *Google* the unlimited right to create "derivative works",
> > because that would include derivative works that don't include the
> > CC-BY attribution?
>
>I'm not entirely sure. Depending on jurisdiction, there may well be
>moral rights to attribution granted by law in addition to whatever
>rights are reserved when granting the license, so that might not even
>be an issue. I think it's simply a matter of you not being able to
>grant a license to someone else's work, regardless of what that
>license says. If it's a derivative work, then it might be ok, since
>you're the copyright holder of that work and it just contains, under
>license, someone else's work. What people have been doing, however, is
>just copying stuff straight from Wikipedia onto Knol and no modifying
>it at all, it that case it isn't a derivative work it's just copying
>the original work and you certainly can't grant Google a license to
>that.
I would assume that if user A grants everyone in the world a license
to do X, Y, and Z, then you're allowed to submit your work to company
Q which requires you to agree to terms that say "YOU give us
permission to do X, Y, and Z". Even though the permission is
technically not yours to give.
Because, logically, if you did interpret it this way, what could
possibly happen that anyone could sue for? If you grant company Q
the right to do X, Y, and Z and company Q actually does one of those
things, user A can't claim they were wronged, because they granted
the whole world the right to do X, Y and Z anyway.
-Bennett
bennett(a)peacefire.org http://www.peacefire.org
(425) 497 9002
> Are you saying that even if someone else published a work under
> CC-BY, you couldn't publish a derivative work as a Knol article, even
> if you attributed the base work to the author as required by CC-BY,
> and published it under Knol's own CC-BY licensing option?
That's my understanding, yes.
> Would it be, not because you can't re-mix a CC-BY work into your own
> CC-BY work (everyone says you can do this), but because you can't
> grant *Google* the unlimited right to create "derivative works",
> because that would include derivative works that don't include the
> CC-BY attribution?
I'm not entirely sure. Depending on jurisdiction, there may well be
moral rights to attribution granted by law in addition to whatever
rights are reserved when granting the license, so that might not even
be an issue. I think it's simply a matter of you not being able to
grant a license to someone else's work, regardless of what that
license says. If it's a derivative work, then it might be ok, since
you're the copyright holder of that work and it just contains, under
license, someone else's work. What people have been doing, however, is
just copying stuff straight from Wikipedia onto Knol and no modifying
it at all, it that case it isn't a derivative work it's just copying
the original work and you certainly can't grant Google a license to
that. I'm really not very sure how all this works - I need to go and
read CC-BY in detail, for a start! (I've read GFDL in plenty of
detail, but have never had a need to read the full CC licenses
properly, I've just scanned them.)
At 02:39 PM 8/10/2008, Thomas Dalton wrote:
> > That's interesting. Which part of the Knol TOS specifically, is it #8:
> > http://knol.google.com/k/-/-/si57lahl1w25/12#
> > which grants Google the right to "modify" and "create derivative
> > works based on" what you submit to them? Why is that incompatible,
> > is it because this grants Google the right to "create derivative
> > works" which take out the sharealike requirement, and you're not
> > allowed to do that?
>
>Yes, it's section 8. It's not anything unique to GFDL, you can't
>upload anything to Knol that you didn't write yourself (or you have
>explicit permission of the author) since you can't grant Google a
>license to something you don't hold the copyright for. It's worth
>noting that uploading someone else's work wouldn't be a copyright
>violation because of that, it would just be a violation of Google's
>terms of service, and Google would be committing a copyright violation
>if they tried to use the license you were meant to grant them but
>weren't legally able to.
Are you saying that even if someone else published a work under
CC-BY, you couldn't publish a derivative work as a Knol article, even
if you attributed the base work to the author as required by CC-BY,
and published it under Knol's own CC-BY licensing option?
Would it be, not because you can't re-mix a CC-BY work into your own
CC-BY work (everyone says you can do this), but because you can't
grant *Google* the unlimited right to create "derivative works",
because that would include derivative works that don't include the
CC-BY attribution?
-Bennett
bennett(a)peacefire.org http://www.peacefire.org
(425) 497 9002
> That's interesting. Which part of the Knol TOS specifically, is it #8:
> http://knol.google.com/k/-/-/si57lahl1w25/12#
> which grants Google the right to "modify" and "create derivative
> works based on" what you submit to them? Why is that incompatible,
> is it because this grants Google the right to "create derivative
> works" which take out the sharealike requirement, and you're not
> allowed to do that?
Yes, it's section 8. It's not anything unique to GFDL, you can't
upload anything to Knol that you didn't write yourself (or you have
explicit permission of the author) since you can't grant Google a
license to something you don't hold the copyright for. It's worth
noting that uploading someone else's work wouldn't be a copyright
violation because of that, it would just be a violation of Google's
terms of service, and Google would be committing a copyright violation
if they tried to use the license you were meant to grant them but
weren't legally able to.
(Again, IANAL, YMMV, etc., etc.)
(Disclaimer: IANAL)
> Thanks. So just to be clear, does that mean it's still technically
> illegal to copy an article from Wikipedia and republish it under
> CC-BY-SA?
Correct.
>But once an FDL version is released that's compatible with
> CC-BY-SA, it'll no longer be illegal?
I think that's the plan. We'll have to wait and see.
> (So, for people to start copying Wikipedia content to Knol,
> presumably two things would have to happen -- Knol would have to
> allow CC-BY-SA as a publishing option, and FDL would have to be
> revised to be CC-BY-SA-compatible.)
In addition to that, they would need to remove the requirement from
their terms of service that people give Google a much wider ranging
license.
> Since it was announced in December 2007 that they planned to make the
> FDL compatible with CC-BY-SA, does that mean they're still working on
> it? Are their pro bono lawyers just really careful with things like
> this, so they usually take a long time before coming out with a new
> version of FDL that meets a specific goal?
As far as I know, they're still working on it. I haven't heard
anything about it recently.
At 01:18 PM 8/10/2008, David Gerard wrote:
> > Since it was announced in December 2007 that they planned to make the
> > FDL compatible with CC-BY-SA, does that mean they're still working on
> > it? Are their pro bono lawyers just really careful with things like
> > this, so they usually take a long time before coming out with a new
> > version of FDL that meets a specific goal?
>
>
>They are slooooooow. Because, obviously, they have to be sure they've
>gotten it *exactly* right. Wikimedia is by far the largest collection
>of GFDL material, but it certainly isn't the only one.
Is there any reason why they couldn't just say: "Version 1.3 of the
GFDL is identical to CC-BY-SA"?
That would presumably meet the murky requirement that future versions
of the GFDL have to be "in the same spirit" as the present one.
-Bennett
2008/8/10 Bennett Haselton <bennett(a)peacefire.org>:
> Thanks. So just to be clear, does that mean it's still technically
> illegal to copy an article from Wikipedia and republish it under
> CC-BY-SA? But once an FDL version is released that's compatible with
> CC-BY-SA, it'll no longer be illegal?
That's the present situation, and that's the idea.
> (So, for people to start copying Wikipedia content to Knol,
> presumably two things would have to happen -- Knol would have to
> allow CC-BY-SA as a publishing option, and FDL would have to be
> revised to be CC-BY-SA-compatible.)
Yep.
> Since it was announced in December 2007 that they planned to make the
> FDL compatible with CC-BY-SA, does that mean they're still working on
> it? Are their pro bono lawyers just really careful with things like
> this, so they usually take a long time before coming out with a new
> version of FDL that meets a specific goal?
They are slooooooow. Because, obviously, they have to be sure they've
gotten it *exactly* right. Wikimedia is by far the largest collection
of GFDL material, but it certainly isn't the only one.
- d.
At 12:52 PM 8/10/2008, Thomas Dalton wrote:
> > 1) I thought that the GFDL was already compatible with CC-BY-SA 3.0,
> > since they both required derivative works to be published under the
> > same license. Is there a specific part where they're incompatible,
> > or is it just a case that there are ambiguities about compatibility,
> > and the FDL will be revised to remove all doubt?
>
>Indeed, they require both require new versions to be under the same
>license as the original, and GFDL isn't the same as CC-BY-SA 3.0, thus
>they are incompatible. In spirit, they're pretty similar, but they
>have to exactly the same license (up to version numbers, at least) for
>them to be interchangeable.
Thanks. So just to be clear, does that mean it's still technically
illegal to copy an article from Wikipedia and republish it under
CC-BY-SA? But once an FDL version is released that's compatible with
CC-BY-SA, it'll no longer be illegal?
(So, for people to start copying Wikipedia content to Knol,
presumably two things would have to happen -- Knol would have to
allow CC-BY-SA as a publishing option, and FDL would have to be
revised to be CC-BY-SA-compatible.)
Since it was announced in December 2007 that they planned to make the
FDL compatible with CC-BY-SA, does that mean they're still working on
it? Are their pro bono lawyers just really careful with things like
this, so they usually take a long time before coming out with a new
version of FDL that meets a specific goal?
-Bennett
bennett(a)peacefire.org http://www.peacefire.org
(425) 497 9002
> By the way, that means that if authors are submitting content to
> Wikipedia, with the intention that nobody would be able to create a
> derived work from their article and slap "all rights reserved" on it,
> those authors are putting a lot of trust in the Free Software
> Foundation, aren't they? Since the FSF might someday release a
> version of the FDL which allows third parties to create derivative
> works published under "all rights reserved", like CC-BY does. (Not
> that the FSF is ever likely to do that, obviously, but it's still
> unusual to have an agreement that one party can unilaterally change
> at any time in the future.)
The license says:
"The Free Software Foundation may publish new, revised versions of the
GNU Free Documentation License from time to time. Such new versions
will be similar in spirit to the present version, but may differ in
detail to address new problems or concerns."
Removing the requirement that derived works be released under a free
license would not be "similar in spirit".
2008/8/10 Bennett Haselton <bennett(a)peacefire.org>:
> By the way, that means that if authors are submitting content to
> Wikipedia, with the intention that nobody would be able to create a
> derived work from their article and slap "all rights reserved" on it,
> those authors are putting a lot of trust in the Free Software
> Foundation, aren't they?
Yep, much as developers are when they put software under "GPL v2 or
later". So far RMS has proven incorruptible.
- d.