Re: Version information

On Fri, 06 Apr 2007 20:17:33 +0200, Chris Wilson  
<> wrote:
>> Also, I don't think that introducing versioning for HTML at this point
>> would be a good idea. It makes:
> We are not introducing it.  The HTML 4.01 doctype contains its version  
> number.  We use that to flip into standards mode, just as other browsers  
> do.

Well yes, standards mode and quirks mode is an unfortunate thing from the  
past we can no longer remove.

>> * authoring more complex;
> I don't see how.  You know what version of the spec you're writing to,  
> you put it in the DOCTYPE.  No?  It makes maintaining content easier,  
> because you can expect that the behavior will never change underneath  
> you.

Copy and pasting content from other sites won't work reliably. Version  
triggers in the browser you author against might not be the same as in  
other deployed browsers, etc. Browsers will have bugs in their  
implementation of how this is supposed to work which makes it even more  
tricky, etc.

>> * developing new user agents more complex
> Yes, unless you just presume that treating HTML < 4.01 the same as HTML  
> 4.01 is okay.

That is okay and that is exactly what we're doing today. The only  
"versioning" we have is a rendering mode switch which introduces a limited  
set of specific quirks which should probably be standardized somewhere.

> Doing that isn't going to be any worse in terms of compat problems than  
> having a continuously evolving spec for HTML.

>> * maintaining existing user agents more complex;
> Not really.  You finish HTML 5 support, you're done.  If the spec  
> "changes" in HTML 6, you don't have to change in that support.

You don't just finish support for a specification. Specifications are  
incrementally implemented, not as a whole thing.

>> * specifications more complex.
> Not for HTML5, because we are a new spec, not a delta spec.  Again, cf  
> Charter.  I don't think it makes the HTML 6 spec any more complex  
> either, as it just calls out deltas from HTML5.

If we don't define stuff before HTML5 we're not helping much here I think  
and I don't really see why that would be any better than XHTML2.

> [...]

Anne van Kesteren

Received on Saturday, 7 April 2007 01:23:36 UTC