Date: Mon, 15 Aug 94 07:31:00 PDT
From: Fisher Mark <FisherM@is3.indy.tce.com>
Subject: Re: [Dienst, A Protocol for a Distributed Digital Document Library]
To: "'URI'" <uri@bunyip.com>
Message-Id: <2E4F7DB7@MSMAIL.INDY.TCE.COM>
Chris Weider wrote in <9408112321.AA04990@mocha.bunyip.com>:
>Mark:
> You say that:
>> I see WWW as either the overall architecture or as the "glue" that joins
>> other architectures together (which may be saying the same thing). With
>WWW
>> and HTML you achieve a uniform UI over many different "schemes" (http,
>> gopher, ftp, wais, etc.). WWW and HTML provide the capability for a
>> universal Internet client -- something we will need at Thomson, as our
>> Internet user community will be mainly engineers who want to do their
jobs
>> (design TVs) rather than fiddle with the differing command interfaces for
>> each Internet protocol.
>
>I think that this is mistaking the train for the train/track combination.
>One of the big reasons why the IETF is pushing the use of URLs and URNs
>is that these allow a uniform 'api', if you will, that *any* information
>tool can use. As Peter Deutsch and I pointed out, these identifiers
>provide a defacto integration of information services, always assuming
>the URL client library can handle all the schemes.
>
>The fact that Tim and company invented URLs, and used them widely in
>the Web, is a good thing. But they have much broader applications.
I agree. I should have added a qualification that WWW and HTML are
presently the best universal Internet client (IMHO). I suspect in 10 years
(or sooner!) there will arise new architectures that make WWW look as old
hat as FTP and TELNET do now. I am eagerly looking forward to the next wave
of research on networked information access.
======================================================================
Mark Fisher Thomson Consumer Electronics
fisherm@tcemail.indy.tce.com Indianapolis, IN
"Just as you should not underestimate the bandwidth of a station wagon
traveling 65 mph filled with 8mm tapes, you should not overestimate
the bandwidth of FTP by mail."