Re: your mail

[Kendall Clark <kclark@superfly.ntlug.org>]

All,

What Matthew puts in great and accurate detail below, can be summed up
with a few pithy rules:

	# never trust _anything_ anyone at MS says about _any other
	computing technology_

	# When Gates talks about open v. closed or proprietary standards,
	he is engaging in classic Orwellian doublespeak, and is not to 
	be trusted.

	# Unix on the x86 platform is a real challenge to MS hegemony,
	especially the free Unixes

	# It is by definition impossible to be more open and less
	proprietary than truly free software (software which comes 
	with all source code by terms of its license)

I don't want people on this list to think I'm beating anyone up with
Linux. I use it and I wouldn't use anything else, but I'm not even a
programmer. You don't have to be to use Linux, especially now.

I people really want open standards computing, however, they don't have to
use Linux or another free Unix. The most important thing to do is to
resist MS in some significant way.

MS poses the greatest risk of domination of a whole economy by
proprietary, closed standards than any other technology or company.

Bill Gates and MS lies and obfuscates about open standards. That is a
fact.

Thanks, Matthew, for such a well-informed and accurate post.

Best,

Kendall Clark

On Wed, 7 Jan 1998, Matthew Benjamin wrote:

> 
> Several statements have been made recently about the "failure" of open
> systems, and what caused "it".
> I think this is important, because, as the recent news.com interview
> with Gates shows, claiming "open systems failed" is a key
> rhetorical/marketing move by Microsoft.  Are they _really_ dead?  Well,
> in the Newsweek interview (June? July? 1997) Gates and Ballmer had an
> exchange on this topic.  Newsweek asked what could make MS domination
> plans run afoul.  Ballmer suggested, UNIX-but then said, but we think
> "it's too late for that."  Gates chimed in, "It is too late."
> The reality is, I think, that the most prominent UNIX vendors (IBM, SUN,
> Digital, HP) did balkanize the market-but the balkanization is in the
> _hardware_, not the UNIX interfaces.  To be fair though, these UNIX
> offerings allowed their vendors to _compete_, to sell the best
> _hardware_ for high end computing.  And there was, indeed, much
> competition, much back-and-forth movement and innovation in the market.
> Furthermore, it must be said, AT&T UNIX has run on Intel microcomputers
> for a _very_ long time-since before there was a Windows from Microsoft.
> There is, in many cases, _binary compatibility_ between these Intel AT&T
> UNIXes (SCO, MicroPort, AT&T/UnixWare, etc).
> However, Gates lies when he says that "UNIX locks you in more than
> anything," based on these facts about RISC UNIX.  First, and most
> important, UNIX is defined not by what runs on any one hardware
> platform-especially a hardware platform.  UNIX really does have
> comprehensive software interface standards, and the proof of their
> effectiveness is shown in the portability of UNIX source code:  software
> which runs on any one UNIX, typically can be compiled to run on others.
> It has even been possible for an _entire operating system_ (GNU/Linux)
> to be developed, solely from the specifications and open reference
> material-which is source code compatible with modern UNIX.  Because the
> standards are fully open, Linux can be distributed freely, in source
> code form, used for any purpose whatever, by anyone.  It is true that
> the POSIX and other specifications took time to develop-mainly because
> generations of UNIX had been developed separately, but in parallel,
> first by AT&T and the University of California, Berkeley, and then by
> many independent vendors.  The claim that standards violation balkanized
> UNIX simply doesn't hold water, however.
> I submit to you, if open systems standards were not in the consumer, and
> in the public interest, why would Microsoft be so interested in
> undermining them?  In fact the opposite is true:  destruction of open
> standards in the computing industry allows the wholesale plunder of
> standard art by the dominant interface vendors.  Attempts to redefine
> computing in terms of highly abstract interfaces-driven by proprietary
> middleware, such as Microsoft COM/DCOM/OLE-are being strategically used
> to close computing interfaces opened by (partly) by UNIX.  To a man,
> every one of these "programming models is a replacement for older
> technologies which were rich, efficient, and did similar jobs-but
> weren't owned by the promoting company.  
> My point is, I do _not_ believe that all outcomes are equal-that the
> only possible outcome is an computing industry dominated by one
> "interface vendor" or another.  
> That wasn't true in the last decade.  Only misconception and Microsoft
> marketing is making it true in this one.
> 
> Jamie love wrote:
> 
> >The term "open systems" was coined by  Unix vendors in the 80s when
> they saw interoperability among their systems as a way to >fight against
> losing market share to WinTel.  Their model was the IBM PC  which by
> publicly open standards (either released by >developers or set by
> consensus). Typically this was done by industry definition of common
> interfaces. 
> >The initiative failed because UNIX vendors had every reason to "cheat",
> and create proprietary extensions because that was the only >way they
> could recoup their  investment in new technologies (else they would have
> to give them up for free to the other vendors).
>