[2173] in Commercialization & Privatization of the Internet
thinwire (was: Trickle Down? How About Bubble Up?)
daemon@ATHENA.MIT.EDU (Bob Sutterfield)
Wed Jan 29 10:26:08 1992
Date: Wed, 29 Jan 92 10:23:05 -0500
From: Bob Sutterfield <bob@MorningStar.Com>
To: mfidelma@diamond.bbn.com
Cc: com-priv@psi.com
In-Reply-To: "Miles R. Fidelman"'s message of Mon, 27 Jan 92 19:48:46 EDT <9201280048.AA02852@diamond.bbn.com>
From: "Miles R. Fidelman" <mfidelma@diamond.bbn.com>
Date: Mon, 27 Jan 92 19:48:46 EDT
It seems to me that dial-up IP misses the point for small users...
Rather it is more effective to connect to a more powerful machine,
that has high bandwidth connectivity, and use it as an
agent/filter... a more general purpose version of something like
Compuserve's Navigator, or Applelink... to split functionality
between client and server... Any body want to step up to developing
a prototype?
That sounds a lot like RFC-914. I think the reason the idea hasn't
yet caught on in the wider Internet (despite its success in special
cases like Applelink) is because of the additional trouble of
developing new application protocols to work on new transport layers.
It's much easier to wait until modem bps/$ numbers increase, then
continue using the same old tried-and-true protocols that have served
well on the mainstream network. (VJ TCP header compression
accelerated the process of making fatter protocols work on thinner
wires.) Thinwire client/server applications tend to isolate
themselves behind a gateway that provides a subset of the functions of
the general purpose protocols, at a snapshot in time.
Maybe, now that the market is growing wider in the low-speed end of
the spectrum, we'll see more offerings like PSI's new service, aimed
at people who don't need the full rich thick capability (sounds like a
chocolate commercial, doesn't it? :-), but can drink skim white milk.
