On Wed, Aug 20, 2008 at 2:39 AM, Brendan Taylor <[EMAIL PROTECTED]> wrote:
> On Tue, Aug 19, 2008 at 11:12:40PM +0200, Remko Tronçon wrote:
>> > Numbers of human activities very big. And impossible shove them all in
>> > one document. And impossible draw icons for all them too.
>>
>> That's why we have <text>. If you have the time to implement a
>> gazillion of state icons, you have time to implement something that
>> parses a piece of text for a string to match the icons.
>
> Microparsing isn't a very good extensibility mechanism.
>
> Something like this would be better:
>
>    <activity xmlns='http://jabber.org/protocol/activity'>
>      <other>http://example.org/activities/phrenology-lessons</other>
>    </activitiy>
>
> It's extensible, you only need an extra icon for "other" and the
> contents of <text> are still human-readable.
>
> I think extensibility is important, it would be nice if my robotic lawn
> mower could send me an event like:
>
>    <activity xmlns='http://jabber.org/protocol/activity'>
>      <other>http://robotic-lawn-mower.example.org/recharging</other>
>    </activity
>
> It seems silly to try to specify a taxonomy of human (and robot)
> activity.
>

As another use case for this, I'm working on a media device which
*will* want to publish its activity, be it playing audio, a movie, or
recording something. I'm not sure it would do to overload any of the
existing activities for that.

At the same time as having nice semantics I want stock clients to be
able to display something meaningful also. Dilemma :)

How about:

   <activity xmlns='http://jabber.org/protocol/activity'>
     <other>
           <charging xmlns='http://robotic-lawn-mower.example.org/' />
     </other>
   </activity>

...which fits more closely with the current spec.

Matthew.

Reply via email to