Re: Gap analysis: SCXML (was Re: Gap analysis: SMIL)

Been stuck in a project that needed a bit more immediate attention,
and this reply has been rotting in my draft box for the last couple
weeks.

...apologies.

On 12/11/12 10:07 AM, Futomi Hatano wrote:
> On Mon, 10 Dec 2012 16:55:04 +0900
> Kazuyuki Ashimura <ashimura@w3.org> wrote:
>
>> BTW, I think my point is:
>>
>> - We should not restrict the possibilities of any existing
>>     approaches/mechanisms including SCXML, CSS, SVG and SMIL, at this
>>     stage, i.e., Gap Analysis.
>>
>> - We should not jump into technical details at this stage, though
>>     detailed discussions on possible implementations/services should
>>     follow.
>>
>> - We should involve all the stakeholders and get their opinions as
>>     well.  For example, we should ask all the participants in the June
>>     workshop [1] to provide their views.

I do agree that we should not put too hard of constraints on the what
kind of approaches that the group will take to tackle the use cases,
but we *do* have to be realistic.

Unless we can get commitment from at least two browser vendors that a
certain specification that we bring to the table will be implemented,
putting that into a best practice document would make the document
itself a lost cause.

As for SCXML, at least Opera has no plans to _natively_ support it.

I cannot speak for any other browser vendor, but it does not seem like
there is any native implementation in any browser at the moment from a
quick investigation, and I couldn't find any backlog of such work being
tracked either.

I _did_ find this: https://github.com/jbeard4/SCION

I believe the largest problem is the amount of complexity that SCXML
brings to the table versus the amount of use cases it solves.

If there is a *absolute* need for a declarative model for timed
elements that cannot be covered with a imperative development model
(Javascript), we should try to find a pragmatic and simple solution
that is readily available and allows the developers to write content
immediately, rather than rely on a standard where the implementation
status of browsers or learning curve may slow them down.

I have yet to take a look at web animations, that could be a possibility
- although I'm still trying to get my head around the bit if there is a
absolute necessity for the above-mentioned declarative model.

Sangwhan

Received on Tuesday, 25 December 2012 08:28:53 UTC