On Wed, May 11, 2005 at 01:43:45PM -0600, Warren Young wrote:
> Chris Frey wrote:
> >Since the new documentation is based on the old documentation, it is under
> >the LGPL as well.
> The old documentation files themselves do not have copyright and license
> notice. There is a copy of the LGPL within the old documentation, but
> from the text leading up to it, it seems that this is meant to apply to
> the code itself.
I don't understand it that way. Going back to 1.7.9, there was an LGPL
file at the root, which generally applies to the whole tree. The docs
were under this tree, so the LGPL applies.
Not all the headers have copyright notices either, but that doesn't
exempt them from the LGPL.
> I'm not saying that the documentation is under no license. If that were
> true, we'd have no right to use the documentation at all, as it would
> fall under standard copyright. I'm just saying that it seems that there
> was no particular thought put into the topic. This isn't surprising
> because the various documentation licenses we have now were either
> nonexistent or just barely coming into existence while MySQL++ was under
> significant development under the previous regime. Since LGPL is a poor
> match for documentation, I believe they would have chosen one of these
> if they'd known of their existence, and cared.
It's probably true that no particular thought went into it, but changing
it now is counterproductive, in my opinion. There is nothing to be gained
from changing the license for documentation, as far as I can see. It only
> As a result, I don't think it's a practical problem to assign a
> different license explicitly, as long as it has the same effective terms
> as we assumed it had before. Specifically, it should be possible to use
> it commercially, so long as you share your changes and not revoke any
> privileges you recieved when distributing it.
I don't understand how these things are prevented by the current situation,
and I'm not sure what you mean by "commercially". Do you want to allow
people to write a book based on the mysql++ docs? I don't think the
LGPL prevents that at all, and any changes would need to be returned,
as the docs are part of the package.
> >Plus, I don't think comments can be separated from code, license-wise,
> Sure they can. A license is a contract, and you can write a contract
> with any terms you wish. Since the comments are completely separable
> from the source code proper (the same executable code is generated
> independent of their presence or content) I don't see a problem treating
> them separately.
Technically they can be separated out into other files and licensed
differently by the original author. But practically, it is all code,
comments and all. It is part of the "preferred form" of the source code
mentioned by the LGPL. I've never heard of souce code comments being
under a different license than the code they were a part of.
> I'm not trying to hijack the documentation here, just put it under a
> license suited to documentation.
I'm just not sure what advantage there is to changing it. I see only
> >There is some distaste for the new documentation licenses as well,
> All of them, or just some of them?
> And, is this distaste in the form "we think this could be better", or
> "we won't add this to the distribution unless you change the license"?
> If the former, I'm not inclined to worry about it unless they come up
> with that better thing. If they do have a better thing, let's use that.
> As I see it, both the GNU FDL and the Creative Commons Attribution
> ShareAlike licenses are compatible with the intentions of the primary
> creators of MySQL++. The artifacts they left behind for us to guess
> these intentions are the fact that it was LGPL licensed, and this
> statement from the original docs: "The intent of [licensing under LGPL]
> is allow developers to use my library to develop commercial programs and
> to allow it be distributed with commercial databases."
I haven't been following the whole documentation license issue particularly
closely, but even so, the fuss over the GNU FDL has reached me. :-)
Debian has declared it a non-free license:
In my view, technical documentation doesn't really need any special
treatment when it comes to licensing, since it is only explaining
the software that is free anyway, and we want to encourage people to
contribute and fix it. It is when you get into novels, editorials,
and works of art that the GPL really makes little sense, and preserving
the document's and author's integrity needs to be embodied in the license.
Technical docs should be considered as much a part of the software as the
software itself, since without it, there won't be very many users.
Especially for a package as complex as mysql++. :-)
> Tell you what, I'll ask Sinisa, and see if I can track down Kevin
> Atkinson and ask him as well.
Cool. License disputes are a drag, but can really affect how people use
software and contribute to it. I look forward to hearing their comments.