interesting-people message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Subject: [IP] more on -- rest in peace -- OS/2 Obituary

  • From: Dave Farber <>
  • To: ip <>
  • Date: Tue, 24 Dec 2002 16:22:42 -0500

------ Forwarded Message
From: "Bob Frankston" <>
Date: Tue, 24 Dec 2002 14:46:00 -0500
To: "Dave Farber" <>
Cc: "David Reed" <>
Subject: Re: [IP] rest in peace --  OS/2 Obituary

I have my own theory about an aspect of the failure of OS/2. I started using
it back in 1986 well before it was released and was big fan but I couldn't
use it as my only operating system because it simply didn't run critical DOS
apps and, worse, drivers.

As bad as Windows was by Windows/386 it had acquired a virtual machine
capability. With years of opportunity OS/2 didn't take advantage of the 386.
Based on conversations with key people it seems that IBM had promised its
customers that their 286 machines would continue to be supported. IBM also
had a license to fabricate 286 processors in their Burlington Vermont

Thus IBM chose to not advantage the 386. Microsoft had no such inhibitions.

Since anything involving Microsoft is a sensitive point, I want to emphasize
that supporting the 386 would not have guaranteed OS/2's success since there
were other problems with OS/2 (both in the design and in the development
process) and Microsoft soon saw little advantage in the relationship and a
large liability. But the inability to run OS/2 as my only operating system
made it difficult for it to gain traction during this critical period. I'm
fascinated by these seemingly minor issue that become major leverage points.
This is why I focus on the DNS and EIPV6 as the key Internet issues.

In the end IBM was selling to the IT departments and Microsoft was selling
to users. 

It's the same as ISDN vs. POTS. The Telcos focused on price skimming the
market. Had ISDN been priced like POTS and had the installation process been
as simple as plugging in a phone, then it would have simply taken over and
modem traffic would have been a light load since only the data bits would
have been carried. I originally intended HomePNA as a solution to the
installation problem but the ISDN part quickly became irrelevant.

A broadband agenda is the same thing -- letting the "natural" monopoly
define the product instead of allowing the marketplace. By adding
electronics along the wire DSL can run many times the current rates and with
no distance limits. Unlike OS/2 vs. Windows and ISDN vs. POTS, there is no
opportunity for a marketplace to provide an alternative.

The natural monopoly is, of course, nonsense at this point but without a
marketplace the real damage isn't obvious. If we look at ISDN deployment the
US lagged but it didn't matter because the marketplace triumphed and more
ISDN deployment in itself wasn't the issue -- the ability to create new
markets was the real issue and the US Telcos were less able to prevent that
back then because common carriers were limited in their ability frustrate
POTS innovation. Such inhibitions don't apply to meddling in IP traffic.

As long as people ask for solutions instead of opportunity will get what we
ask for and no more.

Though I'm focusing on telecom because it is such an egregious violation of
the marketplace the same applies to the current mania for "security" that
sees any disruption and innovation as a threat.

Bob Frankston

You are subscribed as
To unsubscribe or update your address, click

Archives at:

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Search: Match: Sort by:
Words: | Help


Powered by eList eXpress LLC