The following text is copyright 1998 by Network World, permission is hearby given for reproduction, as long as attribution is given and this notice is included.

Unfulfillable expectations

By Scott Bradner

Why do so many good technical people prefer to plan the future based on their wishful thinking rather than finding out what the facts are?

The particular thing that got me wondering about this was yet another instance where some people asserted that in the near future there was going to be an explosion of cheap Internet bandwidth for individual users. Not quite repeating the claim that bandwidth was going to be too cheap to measure which is popular with some people, but getting close. The identity of these specific myth spreaders is unimportant only because they are far from alone in their belief. It would be easy to do a little checking and figure out just how much money that the fiber people, like Quest, MFS, and Project Oxygen (, and the ISPs, like MCI, GTE/BBN, and UUNET, are planning on spending. The spending plans are far in excess of the current $3.3 Billion per year total annual revenue for ISPs in the US. This would seem to indicate that unless these companies are suddenly about to become somewhat more altruistic than their previous histories would indicate that cheap bandwidth is a pipe dream. An attractive pipe dream but a pipe dream none the less.

But this is far from the only example of this sort of reality disconnect. A few years ago we saw this in the assumption that ATM was going to replace all other networking technology. We see it today in the apparent belief that most Internet users already have some level of high speed access, judging that belief by the amount of graphics that all too many web page designers put on their pages.

Even a bit of thinking would have indicated that any wholesale replacement of network infrastructure with ATM was not in the cards, there are just too many alternatives, too much installed base and besides that, the network technologists were not about to stop thinking when ATM was designed--new technologies were bound to be developed.

It would seem to me to be very broken thinking to design web pages that take many minutes to download, with much of that time taken in transferring complex advertisements, and expect that Internet users will be all that happy to deal with you and come back for more abuse.

What is it that causes this type of inability or unwillingness to recognize reality?

Clearly some of the problem comes from a need to attract venture capital and sell products. Clear impressive stories are easier to sell than ones with qualifiers. Some also comes from the surface knowledge of the issues that many network managers and technology writers possess. On the web side, showing your boss a web page that is designed for the connectivity that the real user has is not all that rewarding an experience.

But the basic issue may be just that complexity is a pain and it takes time and makes people's heads hurt to actually think through the implications of reality.

disclaimer: Harvard's organizational structure can make my head hurt but the above are my musings.