Re: Lotus Notes -- Too much Hype !!!

Brian Behlendorf (brian@wired.com)
Fri, 9 Sep 1994 09:47:14 +0200

Okay, this is my last post on this, I swear.

On Fri, 9 Sep 1994, Ravi Kalakota wrote:
> Peer-to-peer interaction is very useful in environments such as engineering,
> R&D, problem-solving essentially, where a dialogue is essential and
> exchange is more like rapid-fire. But it creates anarchy in terms of
> organizational communications as there is not much structure in the process.

A problem that has social and technical solutions. A lack of structure
is a good starting place, because one can build tools to manage it.

> The producer-consumer model is an all pervasive model and it's one
> organizations want to see (at least the ones I interacted with) in
> domains such as technical documentation, sales and marketing information
> and transactions with clients and other collaborators (EDI) because it
> gives them control.

Because they're used to mediums which don't talk back.

> You must remember that CSCW while very popular in terms of research
> with respect to things like collaborative editing etc., has not made
> any inroads in terms of organizational applications as there is often
> no control or filter on the information posted. The issue is not at a
> technological one but more of a social one.

Filtering filtering filtering.... keep repeating it... filtering
filtering filtering.... this is a big gaping problem just waiting to be
solved. Information filtering is *very* rudimentary currently - you rely
on a friend to forward you anything interesting from News of the Weird,
for example, while he trusts you to forward anything interesting from
comp.unix.admin. There are whole mailing lists based on people trusting
other people's opinions - like the David Farber's "interesting-people"
list, or Peter Neumann's comp.risks. We are going to see a *torrent* of
information spilling out onto the web over the next year, and resource
discovery algorithms based solely on the statistics of the words in
documents is only one part of the solution.

We need a *generalized*, *secure* way of enabling all internet users to
become information filters for other users for any given WWW resource.

This is a *technological and social* problem, which will lead to a
combined technological and social solution.

'Nuff said.

Check out http://www.xerox.com/PARC/dlbx/filtering.html for some more
thoughts on this issue.

> My point is that we have not even scratched the surface of the
> network publishing model in terms of applications. We can effectively
> subsume the on-demand publishing (video on-demand, games on-demand etc.)
> applications very easily in this model. And, that is one of the reasons
> why it is so popular.

Video on demand is dead. 500 channels isn't it. It's one big
asynchronous pipe. The media doesn't understand asynchronicity. Yet.

Brian