Re: ISSUE-41/ACTION-97 decentralized-extensibility

On Mon, Oct 5, 2009 at 6:09 AM, Henri Sivonen <hsivonen@iki.fi> wrote:
> On Oct 5, 2009, at 15:53, Leif Halvard Silli wrote:
>
>> Henri Sivonen On 09-10-05 14.09:
>>
>>> On Oct 3, 2009, at 02:45, Shelley Powers wrote:
>>>>
>>>> I can speak for an SVG editor, Inkscape, which uses namespaced  elements
>>>> and attributes to record information about the SVG it  produces. And before
>>>> you mention using HTML comments, you should  spend time with an Inkscape SVG
>>>> file, to see how extensive the use  of namespaced elements and attributes
>>>> are in an Inkscape managed SVG  file.
>>>
>>> Inkscape is indeed a good example.
>>> It isn't clear to me why Inkscape couldn't serialize its state using a
>>>  proprietary key-value syntax inside comment nodes instead of using
>>>  attributes as the key-value syntax.
>>
>>
>> Another version of IE Conditional Comments, you mean? Can even be used to
>> hack a "namespace" solution ... HTML as we are forced to speak it ...
>> Doesn't sound lovely.
>
>
> Not at all. I'm not suggesting that comments change tree building in any
> way. I'm suggesting that authoring tools could locate their own comments in
> the parsed tree and parse the contents of the comments.
>
> (This would of course be highly inappropriate for any other kind of
> application except authoring tools storing data strictly for themselves. For
> example, I think putting RDF/XML in comments for Trackback is a bad design
> pattern.)

HTML5 already has a much better mechanism; microdata. I don't see why
the problems that inkscape is trying to solve couldn't be solved using
one or more microdata syntaxes.

However I don't think that inkscape needs to immediately changed.
Inkscape can keep doing what they are currently doing. We don't need
to solve every problem right now. If we can save a lot of complexity
(which I think XML Namespace-like syntax is) and the cost is not
supporting inkscape metadata right now, then I think in the long run
that is the way to go.

/ Jonas

Received on Tuesday, 6 October 2009 04:42:49 UTC