Re: ISSUE-55: Re-enable @profile in HTML5 (draft 1)

On Mon, 12 Oct 2009, Manu Sporny wrote:
>
> Here's my definition of a backwards incompatible change as it applies to 
> this discussion:
> 
> "Any change in HTML5 that, given an HTML5-conformant User Agent, will 
> cause a previously conformant HTML4 document to no longer be a 
> conformant HTML5 document."

I don't think that definition makes sense. What does the user agent have 
to do with the conformance of a document? Why would an HTML4 document ever 
be a conforming HTML5 document?


> Now, if one were to have a large number of articles written in HTML4, 
> migrating those articles to HTML5 isn't as simple as removing the 
> DOCTYPE at the top of the document.

Why would one need to migrate legacy documents to HTML5? The whole point 
of the backwards-compatibility design is that HTML4 documents will 
continue to render the same in HTML5-conforming user agents as they did in 
HTML4-contemporary user agents.


> HTML6 may want to make several backwards-incompatible spec changes based 
> on authoring use in the field such that default parsing behavior matches 
> real-world usage, in which case a version specifier is needed to ensure 
> that previously conforming documents continue to be conforming after the 
> change.

If HTML6 needs to do that, then there was an error in HTML5, and we should 
change HTML5.


> How will you know if somebody omits the @version tag on an HTML6 
> document if you should parse as HTML5 or parse as HTML6?

Assuming HTML6 is designed the same way that HTML5 was, there would be no 
difference. HTML6's parser would be a superset of deployed 
HTML5-contemporary parsers, just like HTML5's parser is a superset of the 
HTML4-contemporary parsers.


> If we create a rule now that says anything that doesn't have a @version 
> attribute will be parsed using the latest rules known to the User Agent, 
> we're covered.

That way leads to there being multiple parsers. That's a terrible place to 
be, as it leads to spiralling QA costs, engineering costs, and authoring 
costs. It's imperative that we not end up in that situation.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Monday, 12 October 2009 05:54:31 UTC