Re: SVG 1.2 Comment: Detailed last call comments (all chapters)

At 12:28 PM 11/11/2004, Boris Zbarsky wrote:
>Jon Ferraiolo wrote:
>>People are wasting their breadth talking about the potential conflict 
>>between the 'filter' property name in SVG and Microsoft's proprietary 
>>'filter' property.
>
>Of course.  "filter" was just an example pointing out that name conflicts 
>_do_ arise, sent in response to a claim that concerns over name conflicts 
>arising were spurious.  Let's not get hung up on "filter" and keep the 
>focus on the general issue that was raised.

Boris,

Yes, name conflicts might arise. I agree that it is legitimate to ask about 
whether the property name 'overlay', which could be used in various common 
W3C scenarios outside of SVG, should be used up to allow for SVG's 
particular notion of overlays. It is legitimate to ask whether SVG should 
use an alternative name for the given feature to leave open the name 
'overflow' for some future use. My personal view is on the other side. I do 
not think we should worry in this particular case about potential future 
collision with some potential future enhancements to other languages. The 
upcoming month of May will be the seven year anniversary for CSS2. If 
CSS2.1 or CSS3 or any other W3C grammar hasn't found a need for an 
'overlay' property by late 2004, I think the statute of limitations has run 
out, so to speak, and it is OK for SVG to use the name 'overlay'. If anyone 
else needs to do something that corresponds somehow to a notion of 
overlays, they will have to find a different name. Also, I believe that the 
SVG 'overlay' feature makes sense in a Compound Document scenario as 
something that would work in a mixed-namespace, HTML+SVG context also, 
thereby allowing renderings of mixed-namespace content to be overlayed on 
top of mixed-namespaced content, such as for rich tooltips.


>>Anyways, there is no conflict. MS's 'filter' property only applies to 
>>elements within the "Microsoft HTML" language. SVG's 'filter' property 
>>only applies to elements within W3C's SVG language.
>
>Ian's employer (like other desktop browser vendors to one extent or 
>another) is likely to want to support both languages, possibly within 
>mixed-language environments.  In which case it becomes an issue.

In that case, it is an issue for Opera, which they can discuss with 
themselves within their own internal forums, but it isn't an issue for the W3C.

Jon


>-Boris

Received on Thursday, 11 November 2004 21:21:24 UTC