Re: An HTML language specification

Henri Sivonen wrote:
> I think the discussion of whether these three HTML files should be be 
> in a separate "spec" is mostly about organizational aesthetics--it's 
> not about the implementability, and it's not about getting the 
> relevant text as separately addressable Web resources.

I agree that pagination is not the issue. I don't think the scare quote 
is appropriate, and I disagree that such a separation is of no benefit 
to implementors, since I would have been glad to get such guidance for 
Mozilla work in the past.

There are several use cases for HTML parsing that don't include script 
execution or any other browser API involvement. In fact, it is easy for 
me to recall sandboxing discussions where it was claimed that servers 
should "just use HTML parsers" instead of their current, insufficient 
filtering. I agree with that prescription, but I don't think the advice 
coming from the HTML5 crowd included script interpretation on the server 
side.

>   ... It's not horribly intertwined but there are some dependencies ...
 
I agree. That's why I don't think splitting parsing *and* vocabulary 
into a separate document is unreasonable on its face.


Ian Hickson wrote:
> html5lib doesn't implement document.write, either; are you saying we 
> should have a different parser spec for UAs supporting scripting and UAs 
> not supporting scripting?
>   
No.

- Rob

Received on Friday, 21 November 2008 08:05:48 UTC