W3C home > Mailing lists > Public > public-iri@w3.org > October 2010

Oct 19th 2010 iri wg interim meeting notes

From: Marc Blanchet <marc.blanchet@viagenie.ca>
Date: Wed, 27 Oct 2010 00:38:47 +0100
Message-ID: <4CC76687.5090806@viagenie.ca>
To: "public-iri@w3.org" <public-iri@w3.org>
CC: proceedings@ietf.org
Title: IETF IRI wg interim meeting
Date: Tuesday, October 19, 2010
Time: 8:00 pm, Eastern Daylight Time (New York, GMT-04:00)

Presences:
Larry Masinter
Adam Barth
Tim Berners-Lee
Noah Mendelsohn
Dan Appelquist
Martin Durst
Marc Blanchet

1. Administrativia
1a. Scribe: Marc Blanchet

2. Adam Barth document
http://github.com/abarth/url-spec/blob/master/drafts/url.txt

Adam: Webkit: two different url parsing librairies. they differ in 
processing. some Adam: browsers use one, some others.
Adam: have the two libraries to do the same thing.
Adam: write a specification they should be doing.
Adam: then fixing the implementation.
Adam: roughly two camps: IE and Firefox. IE does more canonicalisation 
than others. Adam: Firefox does not like the interop issues, so want to 
move to interop.
Adam: current status of doc: mostly url parsing.
Question: why not put test cases in the document?
Adam: Not written IETF or W3C style. might be fixed.
Larry: fine to have this document. also need a normative document. What 
senders are sending and what receivers should expect.
Adam: agree
Larry: there are use cases of IRI are not used in web context.
Martin: issue with document is people reading the document don't get the 
context. Might want to add some context.
Adam: currently, doc is more notes for myself than standard doc.
Larry: could be informative document on how Webkit parses URI.
Adam: I care most that Webkit agree with itself. Than I also care that 
all browsers agree with.
Larry: OS may do something else if not using DNS for name resolution. 
For example, IDNA may not be resolved in non-DNS contexts.
Larry: test suite: 
http://github.com/abarth/url-spec/blob/master/tests/gurl-results/by-browser.txt
Marc: Adam, would you make your document an Internet-draft? and if yes, 
when?
Adam: intended to add html section before publishing. can be done as soon
Larry: contacting browser url parsers implementors would be good.
Larry: trying to find flash url parsers to come.
Marc: please publish as soon as internet-draft submission window will be 
reopened
ACTION ITEM: Adam to publish individual internet-draft when submission 
window opens.

3. draft-ietf-iri-3987bis: opened issues to be discussed

Martin: Martin, Larry and Addison and others will have an editing 
session end of this week.
Larry: many tickets are editorial. only need other to review and then 
apply them to the doc.

   #18 rewrite Security Considerations section (move spoofing out)

Larry: UTF8 spoofing text has changed to reference to UTR-36.
Larry: spoofing rewording. trying to tell people that one cannot fully 
fix spoofing. you need the context. no way to get complete guidelines.
??: browsers can do various things: user interfaces such as mixed 
language warnings, colors or fonts coloring.
Larry: even exists in pure ascii: l and 1, 0 and O. with unicode, 
showing to the end-user is not going to be fully safe.
Larry: look at the ticket and comment text.
Larry: spoofing is not a IRI issue.
Martin: What Larry has done is good. document is closed to 15 years 
old... Removing text is more difficult, but sometimes good to do.
Received on Tuesday, 26 October 2010 23:39:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 30 April 2012 19:52:00 GMT