Subject: Re: why Linux is popular (was Re: the walls have ears)
Date: 1 Jun 1999 16:48:45 -0000

>Craig Burley wrote:
>> As I said in private email to Johnathan Shapiro, it appears to me that
>> Linux is getting lots of attention these days, not because it's UNIX
>> per se
>I started using Linux because it was Unix per se.  This meant
>Walled forks and possible forks were not on my mind.
>I think Linux is getting lots of attention these days because of all
>the people who started using it in 1993-1996 writing great software for
>it, most of which is free.

I'm referring generally to the *media* attention Linux is getting.

In all the media reports I've seen, not a single one started out saying
"Finally, a really great UNIX system!".

Instead, what they've focused on is the fact that people are finding
Linux to be a robust alternative to WNT on servers (mainly).

They don't care that it's UNIX.  In fact, they'd probably rather it
not be UNIX.  Further, if Linux was, instead, a GPL'ed clone of
*WNT* that was crash-proof, the "hype machine" over it would probably
have started one year *sooner*, and today be at least 10x higher
than it is at the moment.  (After all, Linux *is* a different *platform*,
but would not be perceived that way if it was a WNT clone, since
WNT-based applications would run, without modification, on it.
Linux is finally doing what no combination of other UNIX variants
was likely ever going to be able to achieve: convince high-end vendors
like, if not specifically, Oracle, to port their server software to
UNIX, not *just* WNT.  Oracle might have already been doing that,
but lots of vendors weren't, until Linux hit.)

The people who *do* care that Linux is UNIX probably wouldn't be so
inclined to recommend Linux, because, as people often point out,
Linux is "merely" a GPL'ed UNIX, another example of GPL'ers simply
cloning existing technologies, rather than inventing anything really
new.  (Said somewhat sarcastically, but it's not far off-base in this

That is, anybody really interested in just the UNIX component has long
had reasonable (sometimes superior) alternatives, like NetBSD, to
choose.  I doubt Linux really made enough of a difference in *that*
group to explain the media hit Linux has become.

So, AFAICT, all Linux has to offer is the *promise*, if not the present
reality, that all those technologies worth cloning will *someday* end
up in (libre) Linux, when compared to what other UNIX systems offer
people choosing platforms for their businesses.  Linux therefore offers
probably the greatest *credible* promise *ever*, in the history of
computing on this planet, of a platform with long-term stability,
portability, and upgradability.  (There are just too many people
who thought that of systems ranging from IBM MVS to Digital VAX/VMS to
MS-DOS to MS-WNT, and learned otherwise.  There's not a single example
I can name of a widely used, but dead/dying, platform under the GPL,
though of course the GPL is comparatively new.)

One thing for sure: those same technologies are of far less value when
they're made available only in separate, walled-off forks, or camps,
of UNIX variants, again, when viewed from the perspective of people
choosing platforms for their businesses.

        tq vm, (burley)