Re: SVG Semantics Re: SVG and MathML in text/html

On Tue, 30 Sep 2008 10:12:36 +1000, Maciej Stachowiak <>  

> On Sep 29, 2008, at 4:27 PM, Charles McCathieNevile wrote:

>> Our conclusion is that something more like the SVG-WG proposal and less  
>> like the original proposal incorporated in the spec (we're not  
>> convinced that it is perfect yet, but they are also waiting on Ian's  
>> feedback in particular) will be substantially easier to implement  
>> effectively in the actual browser.
> When you say "our" do you refer to SVG WG or to Opera, and is it Opera's  
> official position that the SVG WG proposal would be easier to implement?

I refer to Opera, and it is Opera's official position that implementing  
something like teh SVG WG proposal would be better than implementing  
something like the "commented-out" proposal. For what it is worth, it is  
also our position that these proposals are just that, and not some final  
position to defend to the death but potential bases from which to develop  
a standard.

>> It would, prima facie, seem relatively simple to implement in an  
>> authoring tool that already handles SVG, since there is no need to  
>> implement HTML parsing there, just slurp out the SVG bit and not touch  
>> the HTML (which is relatively simple text processing).
> Using simple text processing to extract content from HTML, instead of an  
> HTML parser, is acceptable only if you don't care about correctness.

That depends what the content you are trying to extract is. If you are  
planning to deal with all the HTML, of course you need a parser. If you  
are extracting particular things, then you can optimise and simplify...



Charles McCathieNevile  Opera Software, Standards Group
     je parle français -- hablo español -- jeg lærer norsk   Try Opera 9.5:

Received on Wednesday, 1 October 2008 00:23:46 UTC