W3C home > Mailing lists > Public > uri@w3.org > June 2003

Re: draft-fielding-uri-rfc2396bis-03, section 6

From: Martin Duerst <duerst@w3.org>
Date: Wed, 18 Jun 2003 12:26:46 -0400
Message-Id: <>
To: "Roy T. Fielding" <fielding@apache.org>, uri@w3.org

For general comments, please see my comments about section 7.

This is about section 6.

This section is a very valuable addition to the document,
and I hope it will avoid many future discussions.

6.1, second paragraph (starting: "Even thought it is possible..."):
      Please add an example here, or at least a clarifying sentence
      such as:
      For example, an owner of two different domain names could
      decide to serve the same resource from two completely
      different URIs.

6.2 "distinguished by the amount of processing required and the
      degree to which the probability of false negatives is reduced"
     Please add "the degree to which a method can be defined
     exactly and reproducibly, not relying on scheme-specific
     knowledge". For applications such as xml namespaces and
     RDF, this is an overranging consideration, even more
     important than the often cited processing requirement.

6.2 "starting with those that": I think this would be easier
     to read if you changed to "starting with those practices that"

6.2.1 After "Java String object" add "(UTF-16)". Not everybody
     is familiar with the fact that Java Strings do not use ASCII.

6.2.1 "Thus, in principle, it is better to speak": Please
     remove "in principle", and change the relative "it is better
     to speak" to something absolute such as "the correct term is".

6.2.1, last paragraph: "Unicode defines a character as ...":
     Please remove this paragraph. It is more confusing than
     helpful in the ASCII-only context of this document and
     this section.

6.2.2: It may be better to move (escape normalization)
     before (case normalization). The main reason is that
     this would make it clearer that %a3 <-> %A3 is escape
     normalization, not case normalization. Case Normalization: it is strange to have a (subsub)section
     entitled 'normalization', but not saying which way is the
     normalized one. Please either copy/move some text from 6.3
     up, or add a pointer refering to 6.3 for details.

     Also, please add a note here explicitly saying that not all
     parts of an URI are case-insensitive. People too much used
     to a certain brand of operating system may otherwise tend
     to forget.

6.2.3 Scheme-based Normalization: Again, this only speaks about
     equivalence, not saying that if possible, the port number
     should be omitted. And this is not even mentioned in 6.3.
     Please add it at least in one place.
     Maybe we should also say that definitions of URI schemes
     should give advice on which way to normalize if there are
     several alternatives.

6.3 Canonical Form: I think this title raises the wrong expectations,
     namely that there is always a single canonical form. The section
     however just gives advice. So a title such as "Best Practice"
     seems more appropriate.

Regards,   Martin.

At 19:07 03/06/06 -0700, Roy T. Fielding wrote:

>I have just submitted draft 03, which can also be obtained via
>the issues list at
>    http://www.apache.org/~fielding/uri/rev-2002/issues.html
>Please note that all issues have been fixed or closed.  If you'd
>like to raise a new issue or reopen an old one, please do so
>within the next two weeks.
>Roy T. Fielding, Chief Scientist, Day Software
>                  2 Corporate Plaza, Suite 150
>                  Newport Beach, CA 92660-7929   fax:+1.949.644.5064
>                  (roy.fielding@day.com) <http://www.day.com/>
>                  Co-founder, The Apache Software Foundation
>                  (fielding@apache.org)  <http://www.apache.org/>
Received on Wednesday, 18 June 2003 12:31:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:06 UTC