Re: mid and cid URLs

Harald.T.Alvestrand@uninett.no
Sun, 26 Nov 1995 22:52:15 +0100


Message-Id: <199511262152.WAA07904@dale.uninett.no>
From: Harald.T.Alvestrand@uninett.no
To: asg@severn.wash.inmet.com (Al Gilman)
Cc: uri@bunyip.com
Subject: Re: mid and cid URLs
In-Reply-To: Your message of "Fri, 24 Nov 1995 17:06:04 EST." <9511242206.AA29205@severn.wash.inmet.com>
Date: Sun, 26 Nov 1995 22:52:15 +0100

Al,
(note: I've taken ietf-types off the recipient list; I hate multilist
debates, and I think this has drifted far enough away from ietf-types'
main domain by now)

thanks for presenting your thoughts on this subject!
My immediate reactions include:

- we need an Internet information architecture, and Chris Weider has
  started forming an Internet Research Task Force to work on this.
  Karen Sollins is involved. Make sure this stuff gets there.

- I think the mixing of "precise reference" with "jumble of material that
  might be useful for finding the referenced object" is a Bad Thing.

- I think that the DRUMS group may be a nice sounding board for evaluating
  your suggested re-restrictions of header fields, while you are just in the
  right place for discussing your suggested URI extensions.

Specific to the last point:

>     mailto: URLs able to nominate but not dictate header values
>          for resulting RFC 822 message
May agree, but have security worries.
>
>     mid: Citations in HTML and MIME headers able to quote existing
>          resource header values for retrieval assistance
Disagree, because I think it mixes referring with searching.
The current in-reply-to syntax is, IMHO, a hindrance to interoperable
referrals, not a help.
>
>     common syntax for header-in-URI embedding for the above two.
No problem, IF it is a good idea.
>
>     In-reply-to, References, etc. headers refined, not extended.
>          is: *( msg_id | phrase )
>          to_be: *( msg_id | cite | subject-phrase)
>               ; where "cite" is the URI-embedding syntax
This MIGHT be a backwards-compatible change, if the chosen syntax
for "cite" is legal under 822 rules.
There's always the theory that you could end up with old values being
treated as if they were URIs, but I don't see it as a big problem.
The proper arena for suggesting backwards-compatible changes to
822 is either the MAILEXT list or the DRUMS list, I think.

Sounds as if we're not that far apart on this one....

     Harald