Re: Public identifiers for SVG 1.1 Tiny and Mobile

On Oct 9, 2007, at 22:43 , Robin Berjon wrote:
>> I trust you know the landscape better than I do, so I'll trust you  
>> on this. Note that the markup validator (a validating parser  
>> itself) is going to make it much easier to validate standalone  
>> SVG, with or without doctype. Test cases (from the specs): http:// 
>> qa-dev.w3.org/wmvs/HEAD/dev/tests/#valid_doctypeless
>
> Is there any chance of support RNG and NVDL there? That could go a  
> long way :-)

RNG is close on the radar, yes, since SVG 1.2 uses that as a schema  
language, among others. Good implementations of that and NVDL seem to  
be mostly in java though, maybe as a perl XML hacker you have  
suggestions on how to tackle this in perl (as this is what the  
validator is built in) or if this is a bad idea.

(I'm not sure if this part of the discussion still belongs in here,  
feel free to move to e.g www-validator)

> Oh I'm not saying that the WG shouldn't publish an erratum on the  
> very real issue you uncovered, I just think that said erratum  
> should be "forget about doctype declarations entirely, they were  
> invented before XML was finished" rather than "use this declaration  
> and not the contradictory ones we had before".

Your point makes sense, I'm just not sure if opinions (or indeed  
suggestions) should be in errata. Others more familiar with authoring  
errata would know better. I guess something like "the declaration is  
optional and unnecessary, and if you really want to use one use this  
one" would probably do the trick. Up to the WG at this point, really.

-- 
olivier

Received on Tuesday, 9 October 2007 14:43:02 UTC