- From: Roy T. Fielding <fielding@gbiv.com>
- Date: Wed, 27 Apr 2011 22:12:29 -0700
- To: Adam Barth <ietf@adambarth.com>
- Cc: public-iri@w3.org
On Apr 27, 2011, at 6:52 PM, Adam Barth wrote: > A lack of interoperability is not a license to make a free choice. > There is a "best" behavior we can choose, and it's not what RFC 3986 > says. RFC3986 defines how to parse valid URI references and transform them to absolute URI form when given a valid base URI. The IETF defines interoperability in terms of rough consensus and running code among all implementations, including non-browsers, and the Internet Standard is based on what has been implemented in fact by hundreds of independent vendors and deployed in thousands of applications. If you can prove that a majority of implementations differ in observable result from what RFC3986 defines, then we can go though the entire standards process again or publish errata. However, such a thing is not even remotely in scope for IRI, nor is it necessary for HTML5. As you well know, what HTML5 needs is a definition for parsing arbitrary attribute values in document encoding. Those attribute values are not URLs. They aren't even URI references. They are one or more space-separated or space-ignoring strings in an HTML attribute encoding, and each reference needs to be extracted and transcoded before the definitions in 3986 are even applicable. However, for the subset of possible references that do happen to match what are called valid URI references by RFC3986, then we have already tested consensus and deployed many implementations that conform exactly to the results given in RFC3986. If you find a difference between that and a single browser's behavior, then that browser has a bug and should be fixed. Rough consensus is determined by testing as many implementations as possible, including those listed at http://curl.haxx.se/docs/comparison-table.html http://curl.haxx.se/libcurl/competitors.html http://en.wikipedia.org/wiki/Comparison_of_web_browsers http://en.wikipedia.org/wiki/List_of_content_management_systems http://xmlsoft.org/html/libxml-uri.html#xmlURIPtr http://lwp.interglacial.com/ch04_01.htm http://hackage.haskell.org/packages/archive/network/2.2.1.3/doc/html/Network-URI.html https://github.com/ajrisi/fsm/blob/master/examples/uri-rfc3986.c ... or feel free to check out some of the 3.3 million hits for http://www.google.com/search?q=3986+parse ....Roy
Received on Thursday, 28 April 2011 05:12:57 UTC