[Edu-sig] linux essay

[Jaime E. Villate]

On Tue, Nov 04, 2003 at 11:05:41AM -0800, Kirby Urner wrote:
> > Another thing: I think you should mention that even for those who still
> > refuse to substitute Windows for a GNU/Linux system, most free
> > programs can also run on top of Windows. For instance, they can use
> 
> I don't know if I'd say "most" here -- but "some" for sure, Python being a
> premier example.

Notice that I said "can also run" and not "has already been compiled".

> > free browsers such as Mozilla and Phoenix instead of Explorer, free
> > Office suits such as OpenOffice, instead of MSOffice, Gimp instead of
> > Photoshop, etc.
> 
> It's often the case that closed source proprietary software has more
> features
In most cases I know, proprietary software does not really have more
features but it has a large advertising machine behind them which
makes them looked as if they were much more advanced than their free
counterparts.

> and the people who buy it make plenty of money with it -- enough to
> easily justify the cost of ownership.
I'm not talking about costs here. I'm sure that all the money I spend
everyday to advocate for free software is more than I would spend if I
bought proprietary software and used it quietly. But my freedom is
more valuable than all that money I spend.

> And so it works both ways:  lots of proprietary, closed source software is
> run on top of Linux in this world, not just on top of Windows.  We shouldn't
> spread the myth that, once you've moved to an open source operating system,
> that everything you do after that is open source.  Not every C++ program
> compiled with gcc has to be GPLed (that's not a requirement for
> using gcc).
I agree. That's exactly why I advocate free software and not just
Linux. At a recent public debate I had with representatives of Oracle
and Microsoft, the guy from Oracle said that he was absolutely
shocked to find out that his views were closer to Microsoft's and
opposed to those of the free software association I lead.
He mistakingly thought that since Oracle DBMS runs on Linux, we were on the
same train and opposed to Microsoft.

> Nor does executable Python have to have any source code accompanying it
> (plus a lot of Python stuff isn't meant for distribution -- as Eric Raymond
> points out, tons of real world working software is custom in-house stuff,
> not any kind of commercial "product").  I've spent many years as a
> programmer, writing thousands of lines of code, and yet nothing that I've
> coded has ever been sold to anyone else in a shrink-wrapped box.  This is
> very typical.
Sure, I'm well aware of that.

> I'm very much in favor of open source software and am always glad when it
> proves itself as capable or more so, than the closed source
> competition.
For me it is not a matter of technical competition. I would not like
my country to undergo a competition between slavery and democracy to
find out which one leads to a stronger economy. I think slavery is
ethically wrong and the issue of whether it leads to a stronger economy
or not is irrelevant.

> But I don't expect all software to ever be open, not even all operating
> systems (many of which run on devices other than desktops or
> laptops).  
I don't either. I just expect the software that is imposed to me (by
my employer or government) to be open.

> Again, the analogy with schooling works:  whereas algebra is in the public
> domain and anyone is free to teach it or learn it, sometimes a new algorithm
> is kept under wraps for proprietary advantage, e.g. the patent on RSA only
> expired recently.  
I think that allowing a patent for the RSA algorithm is wrong. I
prefer the European policy that equations and algorithms cannot be
patented (I keep my fingers crossed that this policy is not changed at
the prime minister's meeting on next Monday).

> IBM, a heavy investor in Linux, still expects to be a major player in the
> proprietary market.  The commercial success of Linux has everything to do
> with big companies realizing that open source does not force them to
> surrender their right to keep secrets (only some loser companies like SCO
> couldn't see how to turn a profit without selfishly and erroneously claiming
> exclusive private ownership over ideas that belonged to the Unix
> community).
Exactly. And I do not see anything wrong with the desire to keep
secrecy as long as they do not claim ownership of any ideas.

> Finally, in practice what happens to a lot of GPL code is it gets forked
> internally to a company and moved into proprietary service.  It's not shared
> outside the company, so this is not strictly speaking against the license
> (when you play with GPLed code in on your own computer, you're not compelled
> to share these changes -- which may have broken it or made it worse).  And I
> agree with Jon Udell who writes about this phenomenon in 'Open Source
> Citizenship' in Infoworld (10.27.03), that this is often a shortsighted
> approach, that giving back to the community is about enlightened self
> interest.
I do not try to force people to give back to the community. If some
company wants to profit from my GPLed program and use it for their own
good without sharing any improvements, I do not feel bad; I'm glad
they could profit from my contribution. What I expect is that
commercial interests do not forbid me from giving back to the
community or sharing if I want to.

Regards,
Jaime