Re: SVG12: #text traits

Just to reinforce what Robin says regarding "lengthy process", as I 
remember there were somewhere in the range of 20-40 joint teleconferences 
between the JSR-266 EG and the SVG WG over 4-6 months. In fact, during a 
particular 3 month period, there were probably 2x or 3x as many joint 
teleconferences with JSR-266 as there were regular SVG WG teleconferences. 
My point is that many people invested many hours in the coordination 
activity. I mentioned previously that coordination activity was sometimes 
difficult and painful. Although no one got their way on every single issue 
(e.g., #text), despite the huge time commitment and the painfulness of the 
process, most people felt that there was a very positive result where J2ME 
and SVG-t are not only integrated from a specification point of view and in 
solid technical manner, but also that there are major commercial efforts 
around that specification. I hate to use the word "compromise" because most 
of the time the technical decisions were based on technical arguments to 
find the best technical solution, but in a couple of cases there was a 
little bit of give and take on technical issues.

Jon

At 07:31 PM 5/23/2005, Robin Berjon wrote:

>Bjoern Hoehrmann wrote:
>>org.w3c.dom.svg is owned by W3C and formally defined in W3C Technical
>>Reports. JSR-226's extensions to org.w3c.dom.svg are proprietary until
>>there is W3C consensus about these extensions.
>
>They are not extensions as they have been ratified by the SVG WG after a 
>lengthy process of discussion and cooperation with the JSR 226 EG.
>
>>It's also clear that the expert group and the SVG WG
>>did not coordinate very well as JSR-226 is not a subset of and not
>>binary compatible with the current Last Call Working Draft.
>
>An editorial oversight has caused a small backwards compatibility issue 
>which is being fixed, but to infer from that that there was poor 
>coordination between the groups is very much mistaken, bordering on bad faith.
>
>>Now, depending on whether and how the draft is changed to resolve all
>>these problems it might make sense to keep the #text trait as depre-
>>cated "backwards"-compatibility feature that must not be implemented
>>for anything but <text>, where it behaves exactly as defined in JSR-
>>226 (and I do not think this is exactly like textContent).
>
>As someone who fought long and hard against #text in many of the joint SVG 
>WG/JSR 226 EG telcons (alas, without result) I can only support the 
>limitation of its inclusion in SVG Tiny. However it would certainly not 
>make any sense to remove it and break compatibility with JSR 226, which is 
>already a finalized specification and already has shipping implementations.
>
>>I would like to point out though that I've registered my concerns
>>regarding #text and legacy interfaces in the SVG DOM subset long ago,
>>   http://www.w3.org/mid/41e46161.181259968@smtp.bjoern.hoehrmann.de
>>   http://www.w3.org/mid/4216ac75.23952484@smtp.bjoern.hoehrmann.de
>>It seems, by formally addressing my comments and regular publication
>>of Working Drafts such incompatibilities could have been avoided.
>
>Given that they were discussed in the WG and had an impact on the 
>specification I doubt that formally replying to them would have made any 
>difference. Also, at the time you made your comments JSR-226 had already a 
>frozen specification for over half a year.
>
>--
>Robin Berjon
>   Research Scientist
>   Expway, http://expway.com/
>

Received on Tuesday, 24 May 2005 15:58:16 UTC