Reconsidering the GPL licensing of Maxima

--- Albert Reiner  wrote:

> Just two comments:
> 
> - Much of the proposal revolves around the possibility of making a
>   case that Bill Schelter didn't understand the license he chose, and
>   implying that he would have accepted any other of the licenses
>   mentioned just as well.
> 
>   Not only do I find this rather distasteful and arbitrary, it also
>   makes me wonder whether there is reason to believe that any court
>   in this world would follow this line of argument.  (The purported
>   relevance of the last mail quoted completely eludes me, BTW.)

I raised this point off list, and it does concern me.  The significant
questions are:

a)  How many changes did Bill really make from DOE-Macsyma, or more
specifically how critical are those changes to our subsequent changes? 
I'm guessing the only way to know this is to request a copy from the
DOE, and find out.  If they are such that duplicating the relevant and
important changes is not too difficult or can be done as part of a
broad scale cleanup and modernization, then we can presumably start
with DOE-Macsyma, map contributions to Maxima from those willing to
relicense onto that code base (updating as advisable to make things
work), and proceed from that point as LLGPL (or whatever).  (Maybe we
will get the docs that originally came with DOE-Macsyma released as
well - drool.)  I can't help thinking that for industrial commercial
purposes, by today's standards, DOE-Macsyma needs a major overhaul to
be fixable and maintainable in a robust manner.  I get the impression
Bill was mainly focused on xmaxima, netmath, and affine along with
making the codebase work better on more modern lisps, but I don't know
enough to be able to say how correct that impression is.  Anybody have
more details?

b)  Who actually does have the copyright to Bill's changes now, and can
he/she be contacted?

>   As far as I understand, GPL is not in the least concerned with
>   *running* a program, but with the question of its *distribution*.
>   Certainly, Maxima is routinely run in commercial lisps now.

Right - but I think the subset of users who want to do major work on
Maxima and not distribute it is pretty small, so it's still an
important point.

Cheers,
CY


		
____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs