Date: Thu, 15 Sep 1994 20:36:06 -0400 (EDT)
From: "Steven D. Majewski" <sdm7g@virginia.edu>
To: Michael Mealling <Michael.Mealling@oit.gatech.edu>
Subject: Re: The <URL: Wrapper take 2
In-Reply-To: <199409152156.RAA08878@oit.gatech.edu>
Message-Id: <Pine.A32.3.90.940915201354.14609A-100000@elvis.med.Virginia.EDU>
On Thu, 15 Sep 1994, Michael Mealling wrote:
> AH!!... I think I see the confusion. In the case of a URIs that aren't
> locators it would NOT be 'URL:urn' but 'URN:'. For example, we now have
> signigicant work being done with LIFNs (Location Independent File Names)
> that aren't URNs nor are they URLs but they LOOK like URNs as far as
> syntax is concerned. Therefore we have to have something to differentiate
> twixt the two: LIFN:<value> and URN:<value>. This is why we need to be
> able to know without funky regular expressions that everything after
> the colon is a URL or a URN or a LIFN or <insert your favorite UR* here>.
_IF_ there *is* a need to prefix URL's with URL: in plain-text to
distinguish them from URN's or LIFN's, _THEN_ the same need exists
(or will exists, when URN and other non URL locators are common)
to distinguish them in other places, link in HTML docs.
Again: I think URL: is redundant, but if I'm missing something here -
if there is a real NEED for URL:, then it's needed everywhere, not
just in plain text, and so ought to be put back into the syntax and
not just part of the plaintext "wrapper" .
(all the way) In or (all the way) Out.
"In" is (IMHO) ugly and redundant and unnecessary - but I can live
with it. ( But if it's "in", I'ld like to reopen the question of
whether schemes are legitamately "chainable" in a hierarchy, and
whether some global syntax is required to distinguish them. Specifically,
with respect to message id's, which, again, are not locators.
We need to be able to map transformations that map the same object
( message-id ) via different protocols. )
- Steve Majewski (804-982-0831) <sdm7g@Virginia.EDU>
- UVA Department of Molecular Physiology and Biological Physics