Re: [cgmo-webcgm] Required changes to IDL

[Lofton Henderson <lofton@rockynet.com>]

I favor option #1 also.

-Lofton.

At 05:15 PM 8/23/2004 -0400, Benoit Bezaire wrote:
>Hi all,
>
>   I'm trying to generate a single complete IDL document (webcgm.idl).
>   This exercise raises a few questions...  The IDL was obviously based
>   on the W3C DOM and W3C SVG IDLs but we have some inconsistencies to
>   resolve...
>
>   We have the same definition as the W3C DOM Level 3 for DOMString and
>   NodeList; but quite a number of differences for several
>   interfaces such as Node, Attr, DOMStringList, Event etc...
>
>   I'm wondering what to do about that?  I don't think it's a good idea
>   for us to have a Node interface that is different from the W3C DOM
>   Node interface while using the same name.  I think the group will get
>   criticized for that.
>
>   I believe we have three options:
>   1) removing the name confusion by prefixing all datatypes and
>   interfaces with "WebCGM"?  (ie; WebCGMString, WebCGMNode etc...)
>
>   2) using DOM interfaces.
>
>   3) ignoring the problem.
>
>   My vote goes for 1) for the following reasons:
>   - option 2 is unfortunately not really an option since we have
>   demonstrated that the exact DOM approach is inappropriate for
>   WebCGM.
>   - ignoring the problem may be a possibility but creating a
>   dependency on the W3C DOM by 'using' some of their datatypes and
>   interfaces may be risky in the long run.
>   - prefixing all datatypes and interfaces with "WebCGM" requires
>   little effort from vendors in this early stage of development and in
>   my opinion reduces confusion regarding the relationship of our DOM
>   and other W3C DOM specifications.
>
>   Comments on these changes would be appreciated.  Note: the complete
>   IDL document (with the above changes) is attached to this email.
>
>--
>  Benoit                 mailto:benoit@itedo.com