Which GPL applies?

[Laird J. Heal]

In article <2058 at sixhub.UUCP> davidsen at sixhub.UUCP (bill davidsen) writes:
>  This question came up in comp.unix.shells, but I think it is of more
>general interest.
 
>  There is talk of a change to the GPL to require only making the source
>of the FSF stuff available. 
> [ . . . . . . .]  But after talking to a lawyer(*), I think there is
>some question as to which GPL applies, and that in general it is the one
>in force when the software was released.
 
>(*) I don't just consult lawyers at the drop of a rumor, I was having a
>beer with one, and his specialty is patent, rather than copyright, law.
 
I do not really like the GNU Public License; also realize that they can,
if they choose, offer a private license to selected parties with any terms
whatsoever.  I shy away from any source subject to the GPL unless I only
want to compile and run it.
 
The issue of what should and should not be in the GPL merits general
discussion.  The question should be how to maximally benefit society as a
whole, or the users of the software in particular.  The biggest problem I
have with binary-only software is that a company can choose to withdraw
support (or have bankruptcy make that choice for them).  The GPL solves
that problem.  A less restrictive solution, namely to cause a vendor to
archive the source for release when the vendor stops supporting it would
also be acceptable.  The current Copyright law permits a vendor to submit
only a portion of the source code as a fingerprint of sorts, and this is a
big problem.
 
The GPL is a response to another problem as well.  We are practically in a
global village, running the same software on a wide variety of platforms.
The developers of the operating system, in an effort to maximize their own
return, both charge outrageous prices for the software and restrict access
to the source code of the operating system, while users as a body are now
accustomed to sending source code for their own programs so that they may
be compiled on the sundry configurations.  The developers, as a matter of
fact, send their lowering flocks of legal eagles to swoop on anything which
threatens the almighty rate of return.  The GPL here works a guerilla war
intended to undermine this very obnoxious practice.
 
In general, there are some things that should be public knowledge, and the
various OS incarnations are usually well-documented as to their workings;
any answer not in the documentation should be there, and if not the source
code should be provided.  People do have the right to get the market value
of their products, but selling a program without providing support for as
long as it is usable is like working a fraud upon the purchasers.  Society
at large will benefit more from protecting the needs of the purchasers
rather than maintaining excessive profit in the hands of the developers.
 
Furthermore, a situation can arise like the following.  I once worked for
a computer manufacturer; they are still in business but not so prominent
as then.  One of their products had some problems, so I wanted to peek at
the source.  Unfortunately they had been unable to find the source code
since one of the software engineers left; the best guess was that he had
used the binary debugger to meet a deadline and never gotten around to
putting the corresponding changes in the source code.
 
Thus, while I personally do not like the GPL, I think that as a whole it
strikes a much better balance than the normal end-user agreement.  Most
users never need source code, but without it they never have any guarantees
that they can ever obtain the support they may require.
 
What I have stated above are general goals.  Anyone should feel free to
propose detailed changes in the terms of the GPL or to state their own view
of what is highest in the public interest.

-- 
My .signature is on vacation ------------- like me!