The polyglot specification should be suitable for normative reference

I am writing here as an individual TAG member, not as chair.

Henri Sivonen [1] and Lachlan Hunt [2] have offered opinions regarding the 
criteria that should be used to determine whether a document such as the 
HTML/XHTML Compatibility Authoring Guide [3] (the so-called polyglot 
specification) should be "normative".

I am convinced by the argument that Tim Berners-Lee made on our TAG 
teleconference last week that "normative" describes not the status of an 
individual document, but of a reference from one document to another. So 
the question is really "What is the requirement if any for a polyglot 
specification to be suitable as a normative reference from some other 
recommendation or specification, and what does that in turn say about 
whether polyglot itself should be a Recommendation, a Note, or something else?"

My first conclusion is that being the subject of such normative references 
is an important use case for the polyglot specification. Let's say some 
particular industry group wished to define their own standards for HTML to 
be exchanged among applications used by group members. Referencing a 
document like [3] is exactly what their specification should do to 
establish agreement that their documents should be servable on the Web as 
text/html, and also directly processable by XML tools and databases.

Having concluded that the polyglot specification should be suitable as the 
target of such normative references, I am more or less convinced that 
justifies making it a full W3C Recommendation.

Thank you.

Noah

[1] http://lists.w3.org/Archives/Public/public-html/2012Nov/0006.html
[2] http://lists.w3.org/Archives/Public/public-html/2012Nov/0021.html
[3] http://www.w3.org/TR/html-polyglot/

Received on Thursday, 6 December 2012 01:44:18 UTC