W3C

Results of Questionnaire ISSUE-56: Bring "URLs" section/definition and IRI specification in alignment - Straw Poll for Objections

The results of this questionnaire are available to anybody.

This questionnaire was open from 2011-03-04 to 2011-03-11.

4 answers have been received.

Jump to results for question:

  1. Objections to change the definition of URL to normatively reference to IRIBIS
  2. Objections to change the definition of URL to normatively reference IRI specification using a well-defined interface
  3. Objections to restore the removed text that explained how to translate input strings contained in text/html documents into URIs

1. Objections to change the definition of URL to normatively reference to IRIBIS

We have a Change Proposal to change the definition of URL to a normatively reference to IRIBIS. If you have strong objections to adopting this Change Proposal, please state your objections below.

Keep in mind, you must actually state an objection, not merely cite someone else. If you feel that your objection has already been adequately addressed by someone else, then it is not necessary to repeat it.

Details

Responder Objections to change the definition of URL to normatively reference to IRIBIS
Roy Fielding
Julian Reschke
Henri Sivonen I strongly object to relying on IRIBIS, because the IETF WG has failed to deliver on its promised schedule. (Note that the proposal linked to here is over a year old!)

Meanwhile, the absence of a reality-based normative reference chain from the HTML5 spec has at least once slowed down my actual implementation work. I think it was a bad call to remove the previous text from the HTML5 spec before a replacement was proven not to be vaporware.

At least as long as IRIBIS hasn't delivered what the HTML WG needs, I think the HTML WG should follow one of alternative proposals to avoid a dependency on IRIBIS.
Theresa O'Connor

2. Objections to change the definition of URL to normatively reference IRI specification using a well-defined interface

We have a Change Proposal to change the definition of URL to normatively reference the IRI specification using a well-defined interface. If you have strong objections to adopting this Change Proposal, please state your objections below.

Keep in mind, you must actually state an objection, not merely cite someone else. If you feel that your objection has already been adequately addressed by someone else, then it is not necessary to repeat it.

Details

Responder Objections to change the definition of URL to normatively reference IRI specification using a well-defined interface
Roy Fielding
Julian Reschke Note that the current HTML5 already refers to IRIBIS in this way.
Henri Sivonen
Theresa O'Connor This proposal relies on the IRI specification providing two algorithms (address parsing and address resolving), which are capable of operating on arbitrary strings. But in the year since this proposal was advanced, the IRIbis spec has failed to provide such interfaces. I see little reason to believe that, were this WG to accept this proposal, the IRIbis spec would be changed at that time to provide these interfaces. This would leave us without, as Henri put it, a reality-based normative reference chain that implementors could follow. (I believe Henri's comment on the first proposal applies to this proposal.)

3. Objections to restore the removed text that explained how to translate input strings contained in text/html documents into URIs

We have a Change Proposal to restore the removed text that explained how to translate input strings contained in text/html documents into URIs. If you have strong objections to adopting this Change Proposal, please state your objections below.

Keep in mind, you must actually state an objection, not merely cite someone else. If you feel that your objection has already been adequately addressed by someone else, then it is not necessary to repeat it.

Details

Responder Objections
Roy Fielding The removed text was a complete fantasy that does not reflect how any implementation processes hypertext reference. Restoring it does not solve anything.

The solution was given in my response to this CP on the mailing list. Why is the mailing list discussion being ignored by the decision process?

http://lists.w3.org/Archives/Public/public-html/2010Jul/0036.html
Julian Reschke As far as I can tell, the CP does not restore text that describes "how to translate input strings contained in text/html documents into URIs" -- it is a complete parsing/resolving algorithm, introduced with the claim that what RFC 3986 contains is non-sufficient/buggy.

As Roy has explained multiple times, the contents of a/@href is not supposed to be a URI/IRI (reference). What it is is something that can be *transformed* into one, such as by stripping leading/trailing whitespace, and handling non-ASCII characters (such as for the special case of non-ASCII in query parameters).

Roy's mail <http://lists.w3.org/Archives/Public/public-html/2010Jul/0036.html> proposes just that, and my impression was that Adam Bart actually supported that approach, In particular, in <http://lists.w3.org/Archives/Public/public-html/2010Jul/0037.html> he says:

"...That's fine with me. I don't know what the specific text should be. I was mostly suggesting reverting http://svn.whatwg.org/webapps@3245as a starting point, but the text you have above seems like a reasonable starting point as well...."

...so the proposed change was not meant to actually resolve the issue, but to get the spec into a better defined state before proceeding.

As such, I have to object to the CP as it's not even *meant* to resolve the issue.

To make progress, we should go back to the drawing board and turn Roy's proposed text into a concrete change proposal.
Henri Sivonen
Theresa O'Connor

More details on responses

  • Roy Fielding: last responded on 4, March 2011 at 21:42 (UTC)
  • Julian Reschke: last responded on 8, March 2011 at 10:09 (UTC)
  • Henri Sivonen: last responded on 11, March 2011 at 15:37 (UTC)
  • Theresa O'Connor: last responded on 11, March 2011 at 17:19 (UTC)

Everybody has responded to this questionnaire.


Compact view of the results / list of email addresses of the responders

WBS home / Questionnaires / WG questionnaires / Answer this questionnaire

Report issues on GitHub project w3c/wbs-design (preferred) or by mail to sysreq.