[OT] GPL

[Jim Witte]

To hook onto a *very* old thread:
>> I don't believe he [Stallman] is contradictory.  His position is 
>> quite subtle.
> Section 2.b of the GPL sez:
>    You must cause any work that you distribute or publish,
> <snip>
> The widely held interpretation of this is that you can indeed sell your
> program for any price you like, provided you also give it away for free
> under the GPL.

   What I've never understood about the GPL is seciton 3b (GPL 2.0): 
"Accompany it with a written offer, valid *for at least three years* 
[..]source distribution"

   It seems to me that this is backward (in a way).  At the beginning, 
you as an author *want* to have protection for a work (to be able to 
recoup the cost of making it, the same original intent of the U.S. 
patent-system..)  It's only *after* some period of the that a 
*requirement* of source release becomes important, but not quite from 
the POV of Stallman - from the POV of "abandon-ware".

   What happens if I make a really neat OS (oh, let's call it NewtonOS 
or BeOS) and then 3 years later it folds (because of another, much more 
ruthless OS vendor we'll call, oh, MS..)  The code is locked up (be a 
company we'll call, Oh, Apple or Palm), and never seen again.  But is 
it worthless?  Perhaps, from a pragmatic point of view, but then, so 
are lots of things, like the designs for the internal combustion 
engine.  Anyone who *really* wants to make an ICE can do so *from 
scratch* (without any help from an engineering textbook), and anyway, 
there aretens oif not hundreds of companies out there that make ICE's 
commercially from which I can buy one.  So just because the ICE has 
been around for about 70 years or more doesn't necessarily mean that 
the principles for the design of one should be found in an introductory 
engineering textbook..

   Yes, I know, this is a *very* bad analogy, and there are some 
horrendous holes in it.  But that section of the GPL to me says that 
"it" (and/or the people who promulgate it) give little thought to the 
problem of "abandonware" and to solving it, and much more thought to 
their own little idea of "software commuism/socialism" (when it is true 
that even programmers have to eat.)

   Another implicit implication I see in the statements that "you can 
charge as much as you want or can for a piece of GPL software, as long 
as the source if Free", is that the only people who are going to pay 
for the software when the source is already available (neglecting 
"add-ons" like tech-support, on-site installation, etc) are people who 
*don't know how to build it from source*.  Which, if you're software is 
the next MS Office that no Tom, Dick, or Harry can live without, is 
most likely true.  But is the typical OSS developer going to bet on 
that fact, and be "inspired" (as MS so smugly says in it's recent 
commercials) to create really good OSS software?

   I don't know..  I doubt it, especially when it comes to really good, 
bullet-proof software for making UNIX installation and maintenance 
easier (read, as easy as Mac, or at least Windows).  Which is why I 
don't think Linux will hit the desktop in a big way for a long time 
(unless perhaps the EU does something very drastic to MS..).  Because 
for it to do so, I think that the UNIX model that Linux inherited 
(three-character directory names, when we've had long-file-names for 
years; thousands of files required for the OS to run, when many of them 
could be hidden in logically-named bundles like MacOSX does; preference 
'rc' and 'profile' files scattered here there and everywhere [some in 
etc/, some in other places]; somewhat illogical naming conventions - 
why is there 'bin' and 'sbin' for instance) will have to change and 
simplify.  The Linux developers are not going to particularly want to 
do that - they already understand all this complexity, so they don't 
see it as a hindrance (to the contrary - it's flexibility to them - and 
it is, in reality, but that doesn't change the fact that it *looks* 
like complexity).

    The only people who would do that are outside forces, like Nautilus, 
which folded - I'm not exactly sure why, but I'd bet (a small amount 
under $20) that it had something to do with money and ROI.

My (very late) 0.0245 euros (due to the weakening dollar..)
Jim Witte
jswitte at bloomington.in.us
Indiana University CS