Autor: Gregorio Kus (Grego_at_RMnet.it)
Data: Wed 26 Feb 1997 - 14:42:38 MET
From: http://www.pcweek.com/opinion/opinion.html
February 17, 1997
PC At Work The ABM (Anyone But Microsoft) mentality
By Peter Coffee
The other day I found myself using, not for the first time, the
initialism "ABM." This used to stand for "anti-ballistic missile," the
ground-based part of the "Star Wars" program. But I was referring to its
more current meaning: "Anyone But Microsoft."
ABM isn't a position that I hold myself, though I've found it a tempting
doctrine from time to time. It does have many loud-voiced adherents on
both the supply and the demand sides of workgroup computing.
And I can't resist the analogy that results.
ABM systems of the missile type were supposed to employ the latest
technology to overcome a massive nuclear attack. Big, dumb rockets, in
theory, were no match for more nimble weapons with the latest in target
detection and tracking.
ABM systems of the computer type are supposed to use the latest
technology to overcome the mediocre quality and obsolete philosophy of
Microsoft's platforms and applications. Platforms such as Windows 95,
hobbled by backward compatibility with a decade of dirty DOSware;
applications like Office 97, inferior to alternatives like ClarisWorks in
its integration of the tasks that users do most often.
But the ABMs of the missile type couldn't make an effective case against
sheer numbers of incoming warheads, and against the impossibility of
testing the complex system under realistic conditions.
Likewise, the ABMs of the computer type can't make an effective case
against the huge installed base of Microsoft platforms and
applications--whose actual behavior is merely hinted at by their formal
specs, but whose quirks are studied, documented and worked around by an
entire industry that finds it profitable to do so.
Portable technologies such as Java find it hard to compete against the
resulting performance and--paradoxically--the overall reliability of the
code base that results from this truly bizarre situation.
The future isn't here. There are many times when I've thought the
internal contradictions of the Wintel architecture had finally reached a
point of no return. Microsoft would find that it had to abandon Intel's
X86 design, with its genetic roots in a traffic-light controller rather
than a VAX; Intel would have to go to a new approach to remain
competitive with chips designed from scratch for the needs, and the
fabrication processes, of the next century (such as Digital's Alpha).
I've publicly said, "The Wintel approach is a Model A Ford. It's a lot
more comfortable than a Model T, like the Apple II, but it's not even a
1940s Buick--let alone a Taurus SHO. We have much better things to come."
But I first said this at least three years ago, and the time of
transition seems as far away as ever. The worst predictions I've ever
made were that the Pentium would prove that you could only push the X86
approach so far, and that Windows 95 wouldn't stand up to OS/2. Now
Pentium derivatives have trounced the more expensive Alphas in parallel
processing price/performance, and Windows 95 is doing what people bought
it for.
I hope I don't sound satisfied with this situation. Apple's suicide,
IBM's surrender and Sun's adolescent arrogance are mysteries to me. But I
won't let them hold up the work that needs doing this year. And I don't
think you will, either.
Is ABM your doctrine? If so, whom do you prefer? Tell me at
peter_coffee_at_zd.com.
-- /----------------------------------------------------------------- Gregorio Kus Grego_at_RMnet.it Grego_at_cyberspace.org ROMA, Italy 2ndAdmin_at_iName.com Grego_at_FreeNet.hut.fi Anonymous Mail Service - http://free.rmnet.it/~grego/AnonMail.html
To archiwum zostało wygenerowane przez hypermail 2.1.7 : Tue 18 May 2004 - 15:56:26 MET DST