From: burley@geech.gnu.ai.mit.edu (Craig Burley) Subject: Re: At last: the media woke up Date: 15 Oct 1992 17:47:56 GMT
In article <1992Oct15.020946.5978@ee.eng.ohio-state.edu> coil@sample-o.eng.ohio-state.edu (Stacy D. Coil) writes:
In my there is two major problems with heralding Linux as the OS of the future
(Please forgive me Linus). Linux is not very stable (at the moment), and it
is public domain.
I think you're wrong on the first count, and you're definitely wrong on
the second.
Linux seems to me to be more stable, and less prone to viruses, than Windows
3.anything, and probably will be at least as stable as Windows NT when they
are at the same level of maturity.
Linux is not public domain, it is licensed under terms of the GNU Public
License.
But your point was concerning support, wasn't it? There, Linux can easily
outperform Windows NT, because anybody can hire anyone to support Linux,
but to support Windows NT, you can hire only Microsoft. Given the fact
that there are already more companies providing paid-for support for other
GPL-protected software like GNU C, GNU EMACS, and so on, than there are for
similar proprietary products (where there's only one company per product),
it is clear that Linux, should it become popular, would enjoy the
advantage of having a variety of companies competing to provide support for
it, while Windows NT will probably always have the effective monoply on
support by Microsoft.
No, the main reasons Linux might not supercede Windows include things like
this:
1. Linux, being a UNIX clone, isn't targeted at the same market
as Windows.
2. Linux doesn't have smaller semi-compatible versions that support
pre-386 processors and systems with <2MB RAM.
3. Linux doesn't have a built-in low-overhead windowing system.
There are other lesser concerns that are easily addressed (e.g. documentation,
putting to bed the so-called "risk" of distributing GPL-protected software
that so many companies have been willing to take, having an easy installation
procedure, having a high-quality newsletter/magazine for the user base,
running apps meeting various API/AIBs like DOS, Windows, and SVR4).
These three points, however, are ones that the people should think about
knocking off _first_ before asserting that Linux is a more viable alternative
to a typical future Windows user.
I see no reason why an organization like, say, the FSF (which created GNU C
and GNU EMACS and Project GNU itself, along with the GPL to protect the
stuff, but not Linux), couldn't design and build an OS that has the
advantages of Linux and easily beats out Windows. But Linux isn't that
OS, at least not yet, though it could be a great starting-point. (In
case you're wondering, I'm considering whether it'd be useful to start
a new FSF-like organization that'd do things just like this, i.e. non-GNU
free software. But I'm still too busy with GNU Fortran to do anything
other than dream. :-)
Curiously, the industry may finally be learning the value of a free base
OS complete with source code. I've seen items in WSJ (Wall Street Journal)
about app developers for Windows complaining that Microsoft isn't
always documenting Windows features that it's applications group uses
for their (competing) apps. It's hard to do that if you provide source code
to the entire OS!