Reconsidering the GPL licensing of Maxima


Jay Belanger wrote:

>Richard Fateman  writes:
>...
>  
>
>>>>In particular, while he asked DOE for permission to release
>>>>THEIR code (actually MIT's, his, mine, ...) under GPL, he
>>>>didn't say that his additional code was also restricted by GPL.
>>>>        
>>>>
>...
>  
>
>>(Jay...)
>>    
>>
>>>It seems to me that he did.
>>>The only indication of copyright in the old sources is the GPL, and
>>>the included spec file indicates it's GPLed.
>>>      
>>>
>>I don't know what sources you checked,
>>    
>>
>
>I checked the maxima-5.6 sources, which I thought were the latest
>sources released by Bill Schelter.
>
I agree that the later sources include the GPL files COPYING  ( and also 
COPYING1, which
has to do with not distributing Maxima to certain countries.)

>
>  
>
>>I looked at the old source for Maxima-5.2a-beta which is in the
>>file
>>http://www.ma.utexas.edu/users/wfs/pub/maxima/maxima-5.2a-beta.tgz
>>This version is significant because it is dated 16-September-1998.
>>
>>That is, it is the last version just before Bill got a letter (3 weeks
>>later) from the DOE, on 6-October-1998.  In that version 5.2a-beta,
>>there is no mention of GPL.
>>    
>>
>
>I would think that sources released after he got the letter would be
>relevant, not sources from before he got the letter.
>
Here I do not agree.  Until Oct, 1998,
Bill distributed his own code and additions without the GPL. He posted 
them and informed people
of their existence. Though not too loudly because he didn't want DOE to
get upset. That's why he was relieved to know that DOE really didn't care.
But looking at the code from version 5.2a-beta, he did not restrict 
access to,
or distribution of, his own code, so far as I can tell.  So it 
presumably represents
his intentions as of the date on that tarfile, Sept. 16, 1998.  Can one
give away code and then later impose a restriction on it? Does that 
constitute closing the
barn door after the horses have escaped?  Now it is also the case that 
Bill placed
copyright notices on everything, possibly confusing the issue.  But he 
also added
to version 5.2a-beta and earlier releases code that I wrote. It did not 
occur to
me to ask him to add my name to the copyright.

>
>  
>
>>If Bill had meant for his own code to be "protected" by GPL, he
>>wrote all that code from about 1984 to 1998 and never mentioned
>>it.
>>    
>>
>
>But he included a copy of the GPL with the code after he got
>permission.
>
Well, now the permission from DOE is posted on the ESTSC web site. 
Copying/Unlimited
is what it says.  So none of the code he or I submitted to DOE is really 
restricted by GPL.
It could be considered "dual licensed" to anyone who cared to take it 
out of the ESTSC
library.

>
>  
>
>>>Since the GPL provides much fewer restrictions than some of the other
>>>stuff he wrote himself, I don't see why this point would be argued.
>>>      
>>>
>>Actually, I don't see GPL to be consistent with Bill's intentions, 
>>    
>>
>
>I'm going by his actions.  It does seem somewhat disrespectful to
>assume that he didn't know what he was doing.
>
Maybe. I think that the other side of the coin is that by interpreting 
his words  literally
you may be making his contributions relatively unavailable for re-use, 
and thus
inconsistent with his other actions.  I think it is arguably more 
respectful to
look at his contributions as advancing the public access to computer 
algebra, and
how to continue to honor those intentions.   But certainly things could be
clearer than they are now.

>>As I said, I think he believed he was being respectful to the DOE.
>>    
>>
>
>And he did what he did, which was use the GPL.
>
>  
>
>>Several people have suggested that Bill's heirs inherited his
>>copyright, and could say what they would like to do with material
>>specifically authored and copyrighted by Bill.
>>
>>I've started making an inquiry along these lines.
>>    
>>
>
>Which is, of course, a different issue that what the license currently
>is.
>  
>
yes.

Regards

RJF