Re: [apps-discuss] FW: New Version Notification for draft-thaler-uri-scheme-reg-ps-00.txt

[Larry Masinter <[e-mail]>]

I like Dave's draft, and I think it will help set the agenda for reviewing
an updated IRIREG process. I didn't think we were making progress
on irireg precisely because we didn't have an agreed on problem statement.

Of course, what originally led the discussion on irireg was to clarify
the issue of whether a URI scheme was also an IRI scheme, whether
all URIs could be internationalized or did they need processing one
at a time, and whether the scheme definition could be (should be)
in terms of  Unicode code points rather than ASCII characters.

Some other questions that arose during registration process review
included:

Did the scheme define fragment identifiers (no), allowable
"methods" (can you GET, PUT, POST to a URI of various schemes, and
did the scheme definition have to tell you how.

The changes were extensive: 
http://tools.ietf.org/rfcdiff?url1=rfc4395&url2=draft-ietf-iri-4395bis-irireg-04.txt
but worth (I think) re-reviewing.

The issue of dynamic allocation was raised as a discussion of
"registerProtocolHandler" in HTML5. Originally the discussion was
about registering "web+" prefix but I think the main issue is: if web sites 
can cause browsers and OSes to recognize new URI schemes dynamically,
why bother with registering at all, ever?

http://www.ietf.org/mail-archive/web/apps-discuss/current/msg06774.html
and I raised this http://www.ietf.org/proceedings/85/slides/slides-85-appsawg-8

some other links:

http://tools.ietf.org/id/draft-castellani-core-advanced-http-mapping-02.txt
http://lists.w3.org/Archives/Public/public-ietf-w3c/2012Sep/0004.html
http://lists.w3.org/Archives/Public/public-html/2012Aug/0115.html
https://www.w3.org/Bugs/Public/show_bug.cgi?id=13904
http://lists.w3.org/Archives/Public/public-ietf-w3c/2012Sep/0035.html