For Approval: Open Source Software Alliance License
Rick Moen
rick at linuxmafia.com
Thu Sep 25 21:55:34 UTC 2003
Quoting Sean Chittenden (sean at chittenden.org):
> Strictly speaking from a legal sense, it is not needed, however for
> interpretation's sake, you're probably right. I'll add an FAQ section
> to the license page to handle this case.
I strongly recommend that misleading ambiguities in the licence text be
remedied _in_ the licence text.
[My clarification that GPLv2's source-access provision does _not_ get
triggered by "linking" (contrary to OSSAL's wording), but rather by
redistribution, snipped.]
> I know, which is why the GPL is problematic. If you're writing an in
> house application, you have within your right the ability to link
> against the GPL. If you try sell a product, the OSSAL prevents you
> from shipping that product if it uses GPL'ed code, which is what I
> want. Means that freshmeat.net or other avenues for software
> announcements will be venerable gold mines for businesses in terms of
> software that they can use in products.
Why or whether particular licences are problematic is completely
irrelevant to this discussion. (Moreover, licence advocacy is
infamously tedious, and I try to have nothing to do with it.)
My overall point is that your concern (if I understand you correctly)
about some other codebase's licence attaching itself to a OSSAL-covered
work used with it is misplaced: It can't happen. Basic copyright law
guarantees that it cannot.
So, I conclude that your licence (like Darren Reed's) attempts to cure
(in that particular) a nonexistent problem.
[Snip my attempt to explain why your assumption that "free software ==
GPL" isn't correct, and that free software is just a synonym for open
source, with a different marketing program.]
> Bah, this is exactly why I avoid this kind of semantic propaganda
> silliness.
On the contrary: I thought it evident that that classification distinction
(dividing free aka open-source software into copyleft vs. non-copyleft
categories) is devoid of advocacy, and is an aid to clarity.
I strongly recommend you heed what is meant by those terms, as otherwise
you'll lose a great deal of time dealing with misunderstandings.
(As an aside, I'm really boggling at your conclusion that my category
explanation was "propaganda". That's one huge chip on your shoulder,
Sean: It seems to be partially blocking your hearing.)
> Fair enough.
Cool.
> copyleft + product == !possible; non-copyleft + product == possible;
Plainly outside the scope of licence-evaluation discussion. (Moreover,
tedious.)
> This doesn't mean it hasn't happened, however. Having it explicitly
> stated doesn't hurt anyone, esp since this isn't the 1st time this has
> happened.
>
> http://slashdot.org/bsd/01/09/24/1432223.shtml
As Ian Lance Taylor said -- and as I was saying, for that matter --
that's basic copyright violation, and would be so regardless of the
code's licence. That's a tort. When someone violates your copyright,
send them a stiffly worded demand letter, and expect immediate
correction preferably accompanied by an abject apology. If sufficiently
incensed, and if you think you can get any, sue for damages.
Thus my point that the OSSAL provision discussed -- like Darren Reed's
-- is a no-op. Basic copyright law renders it irrelevant: It is
flat-out unlawful to "relicense" someone else's copyrighted property.
Persons other than the copyright owner lack title to do so. Therefore,
actions purporting to do so have zero effect -- except in committing
torts. That's what title _means_. And note that Søren's copyright
notice had been simply lopped off: Having had that copyright notice say
OSSAL instead of BSD would (obviously) have had zero effect on the
situation.
People like Darren write such licence provisions because they
misapprehend how basic copyright law works. You _do_ know how it works,
and so needn't repeat his mistake.
--
Cheers,
Rick Moen ROMANI, ITE DOMVM!
rick at linuxmafia.com
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
More information about the License-discuss
mailing list