Message-Id: <9403071620.AA02071@expresso.bunyip.com>
From: Peter Deutsch <peterd@bunyip.com>
Date: Mon, 7 Mar 1994 11:20:05 -0500
In-Reply-To: John Curran's message as of Mar 5, 22:23
To: John Curran <jcurran@nic.near.net>,
"Rob Raisch, The Internet Company" <raisch@internet.com>
Subject: Consensus alert! Consensus alert! :-)
It may, just may, be that we're working towards consensus
on this?
[ John wrote: ]
. . .
> ] We really need to return to first principles.
> ]
> ] METHOD://LOCATION/INSTRUCITONS_FOR_RETRIEVAL
> ]
> ] If we do not, all retrieval engines will need to understand the
> ] differences between the various ways in which systems store information.
> ]
> ] And this would break the one untransgressable law: Thou shalt scale.
>
> I am in strong agreement with Rob, Mark, et. al. with respect to requiring
> a clear algorithmic mapping from each URL to the intructions for retrieval,
> even if it results in a more complicated encoding.
Dare I say "me, too"?? I don't think it gets any simpler
than this. We need clear instructions for all supported
cases and that means clear algorithmic steps are required
in all cases. If it means we have something that's more
script-like than subroutine call-like, that's fine. In
fact, I personally think it would be a win.
- peterd
--
---------------------------------------------------------------------------
"The future belongs to neither the conduit or content players, but
those who control the filtering, searching and sense-making tools
we will rely on to navigate through the expanses of cyberspace."
- Paul Saffo, (_Wired_: March,1994)
---------------------------------------------------------------------------