[tex-live] license question

[David Kastrup]

Jonathan Kew <jonathan at jfkew.plus.com> writes:

> On 10 Sep 2008, at 7:33 AM, David Kastrup wrote:
>
>> Stephan Hennig <mailing_list at arcor.de> writes:
>>
>>> Karl Berry schrieb:
>>>>    (ii) the underlying list of hyphenated words and scripts to
>>>>         generate patterns under a copyleft flavoured license.
>>>>
>>>> I very strongly suggest using the GPL(v2 or later).
>>>> There is no problem with using the GPL for anything for which
>>>> "source"
>>>> can be identified -- clearly not an issue here.
>>>
>>> Actually, our list of hyphenated words is partly based on another
>>> larger list of unhyphenated German words (plus lots of garbage)
>>> sorted by word frequency.  That list is /non-free/.  We have
>>> explicit permission to derive substantially different work and do
>>> anything we want with that.
>>
>> [...]
>>
>>> So I think we're on the save side.  But in fact, there is source for
>>> our work that cannot be distributed.  What does that mean for GPL?
>>> I'd think it doesn't matter as long as we're acting in compliance
>>> with our permissions.
>>
>> It doesn't matter to you.  You have all the relevant permissions.
>>
>>> For users of our work it's the same situation as if there were no
>>> sources.  Anything wrong with that?
>>
>> Yes.  Since the GPL mandates distributing the sources (the "preferred
>> form of the work for making modifications to it"), nobody can meet
>> the conditions.  As a result, the patterns are not redistributable
>> legally.
>
> I don't follow this. In the case of Stephan's patterns, the "preferred
> form for making modifications" would surely be the list of hyphenated
> words, which he is proposing to distribute as well. The fact that he
> made use of another (non-free) list of words -- I notice that it
> didn't even have hyphen positions -- while creating his hyphenated
> list is irrelevant; the patterns will come with perfectly acceptable
> source.

I have to admit I have been mislead by his use of the word "source".
The GPL has quite a clear definition of "source code" with regard to the
scope of the covered material.  I agree that it does not match what
Stephan describes with "But in fact, there is source for our work that
cannot be distributed." and "For users of our work it's the same
situation as if there were no sources."

So yes, the GPL would appear to be a suitable license for distribution
even though Stephan is not the sole copyright holder (the copyright for
the original word list and thus also part copyright for the derived work
rests with the original party).

-- 
David Kastrup