Message-Id: <9310170604.AA15555@expresso.bunyip.com>
From: Peter Deutsch <peterd@bunyip.com>
Date: Sun, 17 Oct 1993 02:04:11 -0400
In-Reply-To: Simon E Spero's message as of Oct 17, 1:34
To: Simon E Spero <ses@tipper.oit.unc.edu>
Subject: Re: The URN: wrapper and URLs...
[ Simon wrote: ]
> I rise to a point of order; the question has not yet be put; therefore the
> member's vote is not in order. Furthermore, may I request that he remove his
> puppies from this sandbox to prevent the spread of ringworms.
So I assume we have to suspend the counting of the votes
until the chair recognizes the motion. Never having read
Robert's I don't know how email voting is to be handled in
this case but I am reasonably sure a member is not allowed
to manage the vote. That's why we have chairs (also, they
keep us from falling to the ground when we sit).
And we don't have a puppy, we have a cat...
> The problem that the member for Montreal seems to be addressing is that of
> identifying URIs within arbitrary text. This is a valid concern, and one
> that I am particularly glad to see the member raise in the light of statements
> made at the previous meeting of this body.
To give recognition where due, this example was raised by
Erik Ostrom, I merely endorsed it as an example of the
as-yet unrealized kind of thing we can expect in the near
future once we stabilize things. And what statements? It
was a long week...
> However, although the concerns of the member are valid, I put it to the house
> that they do not pertain to the definitions of URLs, URNs, or other URIs per se,
> but are instead an separate issue to be dealt with in a separate motion.
> In a previous meeting, a proposal was made that URLs be demarcated in
> plain text by the use of the characters '<' and '>', this demarcation
> being a convention agreed upon outside the scope of the specification.
I'm not sure that the demarcation issue should be treated
as outside the scope of the URL spec. Seems to be an
integral part of things. I'm wondering why it was decided
to not mention the "<..>" convention in the URL spec?
Also, I don't recall the actual discussion (I think it
might have happened during the session I missed in
Amsterdam, or maybe I'm just getting old) but I thought
that the "<..>" convention was adopted to allow us to
recognize the start and stop of a UR*. It seems that we'd
still need the prefix to determine what we are demarking
(is that a word?) when we go to process it.
Again, apologies for missing/forgetting this particular
thread. In any event a summary for the list does seem
appropriate at this point.
> Since it might be argued that this syntax is less than ideal, an amendment
> by the member for Montreal might be favourably recieved.
_Any_ amendment? This might be the chance I was waiting for! :-)
First, I suggest that someone with a more functional
nueral net than mine summarize the "<..>" convention to
the list.
Then, assuming that this is convention proves acceptable
to the list I move that the summary find its way into the
URL document, and I further move that we adopt the
convention of prefixes for all cases of URLs and URNs. At
that point, we're probably done...
(I just know this wont work, at least not just yet, but I
can dream...)
- peterd
--