On Thursday, July 13, "Steve Arnold" wrote:
> > Hmm, saying that the FSF brought the GNU/Linux OS into
> > existance, is like saying that churchill and his croonies created
> > the atom bomb. You (and they) may have created a position or
> > environment where the item being created flourished in. It is
> > not a very sound argument to assume that the item being created
> > would have never made it without these conditions.
>
> Boy for being around this long (a real 'net old-timer :) you sure seem
> to have missed the boat on a few things. How can you say that FSF, and
> the GNU tools, weren't instrumental in bringing GNU/Linux into being?
No, you are making the assumption that without such tools, linux would have
never been written. This is an erronous argument. Yes, linux may have been
written a lot later, in assembly, or even in pascal, but it *may* still have
been written. In other words, "it's nice to step on others shoulders, but
certainly not necessary in order to reach a specific goal". I will gladly
credit the FSF/GNU with creating an environment, in which certain types of
activity flourished. It's quite a strech (IMHO) to credit an inanimate
piece of paper (the GPL), or a political group (the FSF) with the creation
of linux (or other software), unless they actually coded the piece.
> > You talk about freedom, about right and wrong. I fail to see
> > where the freedom is in the GPL.
>
> Say what? The GPL is all about maintaining freedom (it explicitly
> forbids you or anyone else from taking freedom away). I guess you
> haven't read much of Richard Stallman's definitions, explanations, and
> other philosophical stuff.
Ok, I want the freedom to distribute some GPL code in a binary only form.
If (and only if) you can give me that, then I will conceede that the GPL
does not restrict my freedom.
> > Did their original licensing terms for the BSD code not follow in
> > a "truer" sense of the word of freedom? What about the original
> > MIT project athena license for X?
>
> In short, no, and it sucks. RS created the GPL precisely because the BSD
> license was too restrictive. And the XFree86 guys use the GPL because
> they didn't like the X license (again, too restrictive).
Huh!?! Have you read the original BSD license? The MIT athena X license?
Other than to provide credit (credit where credit is due), they don't
restrict the use of code under said license in any way. In particular, you
can distribute a binary-only package, including modifications you've made
to the source, without having to disclose your source.
> > Would it not be better to try and see why the opensource movement
> > was "created" in 1998? Since (as a lot of people seem to state)
> > the two movements have roughly the same goals in mind, does the
> > creation of the opensource movement not point out a deficiency in
> > the FSF?
>
> Not at all. As I said, I think the genesis of this more recent
> "movement" was all about not alienating corporate/business types (ie,
> suits). The core RS/GNU philosophy is just way to socialist for real
> suits. Although I don't understand why they fail to see the viability
> of the "give the code away but charge for your expertise" open source
> business model.
Here I believe you are being closed minded. Why not consider the possibility
that the FSF/GPL state of afairs has a deficiency? There are reasons for the
opensource movement, as there are for the FSF/GPL movement. I merely pointed
out that the two seemed to have similar (if not the same) goals, and that
there may be a reason that split occured. Also, if business "types" are
telling the FSF/GPL there is a problem, then listening to them may actually
be a good thing.
Also, not every business is built on "sell your expertise". I know of many
software vendors that operate on "sell product". For them it is not an
option to disclose the source to *every* product they sell. Youre failure
to see the viability in the FSF/GPL model leads me to believe that you live
in the proverbial ivory-tower...
> OTOH, the GNU tools, Linux, XFree86,
> Enlightenment, etc, are like a dream come true. Hey, they actually
> behave the same way each time you run them (unless you change the
> configuration)! What a concept! It's almost like mainframe stability on
> PC hardware.
Ahh, ever try to port a linux application to another unix machine? One of
my pet-peaves is that 90% of all linux "centric" applications are very hard
to port. In many ways the coders of these pieces of software are just as
bad (if not worse) then a lot of commercial software out there. They make
use of obscure and usually undocumented systemcall semantics, etc...
I truly believe that the only way to create quality software, is to write
software that works in many places. An exercise that will quickly force
you to write architected and engineered software, or loose the game all
together.
> Granted, most free software projects don't exactly follow good SE
> practices either (but at least they do *some* things right, like control
> code via cvs, etc) but the nature of open source lets as many people
> eyeball/test the code as are interested, and actually produces some of
> the highest quality software (ie, fewest defects, highly stable, etc) we
> have today.
It also produces some of the worst code I've seen. And I have seen both
commercial and "free" code. You can play both sides of this coin. Ultimatly
it rarely matters what license the source-code is drapped with. It matters
more what the coders feel is an acceptable level of profesionalism in their
care and feeding of their code.
--Toby.
