[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Tim Bray on "Which Technologies Matter?"

• To: "'Mike Champion'" <mc@x...>,<xml-dev@l...>
• Subject: RE: Tim Bray on "Which Technologies Matter?"
• From: "Michael Kay" <michael.h.kay@n...>
• Date: Sat, 16 Mar 2002 16:24:28 -0000
• Importance: Normal
• In-reply-to: <72VPEC52C0626ZZ0VUE0QM2YOIIDC7.3c926d9d@MChamp>
• Reply-to: <michael.h.kay@n...>

> He then goes back and rates the technologies that matter,
> and didn't, on several criteria that might be expected to
> differentiate them...

There's been a lot of good academic analysis of the factors that affect
whether a new technology succeeds or not. But it's a long time since I read
this stuff so I can't point you to many references. The dominant factor
seems to be that people will adopt a technology if they believe that
everyone else is going to do so. The early adoption (which happened for all
the technologies you listed) comes from individual users for whom the
benefit exceeds the cost. The bulk adoption comes from people who get a
benefit (either a real benefit, or just a sense of security) simply from
doing the same as others are doing, or more importantly, from doing what
they *think* others will be doing: the "self-fulfilling prophecy".

Synergy benefits are critical: you adopt Unix (or Windows) not because you
think it is a good operating system, but because you believe that everyone
else will use it and therefore there will be lots of applications and low
prices. As a technology designed for interworking, XML obviously has very
high synergy potential so long as everyone believes in it.

4GL is an interesting counter-example. 4GLs were actually very widely
adopted in the 1980s and gave very substantial benefits to their users (and
as you point out, they are still giving these benefits today, it's just that
the users don't want to shout about it). But for some reason which I don't
think anyone has been able to explain, the market was incredibly fragmented,
there were hundreds of products out there, all incompatible, and none with
more than 2% market share. What stifled the whole market was the lack of a
clear market leader.

I don't think anyone can predict why the industry will sometimes decide to
agree on a single standard (or product), will sometimes split into two rival
camps, and will sometimes fragment into hundreds of camps, but there is no
doubt that the big successes come when for some reason all the pundits
decide to back the same horse.

Most of Tim's factors below:
>
> ?Management support ?Investor support ?Standardization process
> ?Technical elegance ?Apparent ROI ?80/20 point ?Compelling idea
> ?Happy programmers ?Good implementations ?Military backing ?

are only relevant to the extent that they encourage the community to believe
that a particular horse is going to win, and therefore to back it. All of
them can be thwarted if a rival technology has equal backing.

Michael Kay
Software AG
home: Michael.H.Kay@n...
work: Michael.Kay@s...

• References:
  • Tim Bray on "Which Technologies Matter?"
    • From: Mike Champion <mc@x...>

• Prev by Date: RE: XLinks
• Next by Date: Re: XLinks
• Previous by thread: Tim Bray on "Which Technologies Matter?"
• Next by thread: do leading indicators matter?
• Index(es):
  • Date
  • Thread