Re: Web IDL spec.

Hi, Thierry-

Thierry Michel wrote (on 9/2/09 9:18 AM):
> Doug,
>
> The Web Applications Working Group, has published on 19 December 2008 a
> Working draft "Web IDL"
>
> The Media Annotations Working Group is looking for an interface
> definition language for a client-side API for cross-community data
> integration of information related to media objects on the Web and
> providing partial mappings between the existing formats.
> This API would provide methods for client side processing.
>
> Would Web IDL be the proper language for this purpose ?

Certainly.  It has three chief advantages over other IDL languages (such 
as he OMG IDL traditionally used in W3C specs):

1) it provides ECMAscript (Javascript) and Java bindings as a single 
declaration
2) it is easier to write
3) it is more precise

For this reason, it's being used in HTML5 and several other recent 
specifications.  I am leaning toward using it for DOM3 Events, as well 
(though that would mean a rewrite of parts of DOM3 Events, I think it's 
worth it).


> What is the status of "Web IDL", and when do you plan to go to Last
> Call. Would it be mature enough for the Media Annotations WG to rely on
> it ?

It is a bit of an unusual beast... since it is not intended to be 
implemented directly as software, but rather to be used in other 
specifications, the CR criteria are not as clear.  Until the other 
specifications that use Web IDL go to Rec, Web IDL won't be completely 
stable, because there may be details that need to change upon further 
examination.  However, it is relatively stable now, and we believe it 
will go to LC in a month or two.


> Thanks for your hints,

Thanks for looking into this.  We hope you will decide to use Web IDL, 
since it will make the implementation more consistent with other 
technologies that use it.

Regards-
-Doug Schepers
W3C Team Contact, SVG and WebApps WGs

Received on Wednesday, 2 September 2009 13:48:53 UTC