The following text is copyright 1996 by Network World, permission is hearby given for reproduction, as long as attribution is given and this notice is included.

Running in Karnak mode

Johnny Carson used do a comedy routine on the Tonight Show called the Great Karnak. In this bit Carson, resplendent in floppy turban, would be presented with a sealed envelope said to contain a question unknown to all. After a bit of deep thought he would tell the audience the answer. He would then open the envelope and read the question to the audience.

Devising the answer before you can clearly see the question seems to be a popular pastime in the data networking community. This is most evident in the trade press but does show up quite a bit in the announcements that we see coming out of companies that should know better. It is not a new phenomena and I wonder why it is such a hard habit to break.

Just as finding a common enemy in politics can distract people from their real life problems finding a common technical solution seems to stop people from thinking about just what their technical requirements actually are.

I've previously written about two instances of this phenomenon, ATM and the World Wide Web technology have been seen as the answer (now tell me your question) for networks and user interfaces respectively. The current common solution is the Internet (or its spelling challenged alter ego the corporate intranet).

It seems like there is a new front page story every week in this and other trade journals in which a corporation either announces its new Internet-centric strategy or is criticized for not having one. IBM, Microsoft, Digital, Apple and many others have decided that the only future is one that depends on them understanding and dominating the Internet. (That's the way that it gets reported anyway.)

I've been a strong supporter and proponent of the Internet and TCP/IP for quite a while. I've argued that the idea that the highly touted National Information Infrastructure and the mystic Global Information Infrastructure would evolve out of the Internet rather than be new creations that would replace the Internet as the pundits of a few years ago were telling us. Time and Newsweek (yes I know they should not be confused as technical journals) painted the network of the future as a global video store where the customer could view Terminator 17 and Rocky 34 whenever they had the whim. The Internet has always been much more and different than that (to me) apocalyptic vision, but it is not the only answer.

I don't see any more reason to expect that the Internet will be the solution to all needs any more than ATM is, or than ISDN, the ISO protocols, SNA, APPN, mainframes, PCs, etc. were. The Internet will not have a billion nodes by the year 2000. (But neither will Novell.) Technologies should be used where they should be used. Don't substute the common solution for the thought process.

Note that, in Carson's routine, part of the shtick seemed to be the worse the match between the answer and the revealed question, the better the audience reaction. We have not reached that point in the pundits' Karnak impersonatons but who knows what the future may bring.

disclaimer: Common solutions are not easy to come by in an aggressively decentralized place like Harvard, even imagined ones, in any case the above are my musings.