W3C home > Mailing lists > Public > public-poiwg@w3.org > May 2011

Re: Fwd: ISSUE-17 (dynamic-pois): How are POIs with dynamic properties represented?

From: Rob Manson <roBman@mob-labs.com>
Date: Thu, 12 May 2011 08:48:07 +1000
To: "public-poiwg@w3.org" <public-poiwg@w3.org>
Message-ID: <1305154087.3037.331301.camel@robslapu>
> Do we need this within the POI itself?
> The data source distributing the POIs could give the dynamic
> properties itself - (in our case, we are streaming updates to clients,
> allowing POIs to move by their location co-ordinates been
> changed....actually a quite like a IRC bots work). Meanwhile if a
> webpage distributes a POI either PHP or Javascript could be used to
> create/manipulate it.

I agree with that...definitely wasn't referring to dynamic
location...more to do with animation and interaction.

If 2d and 3d model formats that support time can be linked to from/by
POIs then that deals with the animation.


> I think a POI should be seen as just a way to position data - not as a
> source of manipulation itself.

I generally agree with that statement.


> (ie, if we are thinking POIs should have "onclick" and such this,
> imho, is the wrong road - the item being positioned by the POI might
> have all manor of functionality, but the POI should really just be a
> data standard for the positioning).

Well that really all depends upon how thinly we scope this standard.  I
can see benefits for just having a skeleton POI that is simply a
reference point separate from presentation and interaction information.

However, the main mobile AR browsers now all include some form of
interaction definition at the POI level and I find that really useful on
a daily basis.

So it all really depends on whether we expect that presentation and
interaction context will solely be supplied by the consuming
application...or if we also expect that in "some cases" the POI creator
would also like to be able to specify (at least recommended) options for
this.

Personally, I think not allowing a POI to be linked to the equivalents
of styles and events in any way at all could quickly become very
limiting and will simply drive more network connections.


> Animation meanwhile is easy - the data your linking too should be animated ;)
> (ie, a animated 3d mesh format)

Yeah see comment above.


roBman


> On 11 May 2011 22:43, Points of Interest Working Group Issue Tracker
> <sysbot+tracker@w3.org> wrote:
> >
> > ISSUE-17 (dynamic-pois): How are POIs with dynamic properties represented?
> >
> > http://www.w3.org/2010/POI/track/issues/17
> >
> > Raised by: Matt Womer
> > On product:
> >
> > From: roBman@mob-labs.com
> > Date: Tue May 03 06:17:01 2011
> > Archived: http://lists.w3.org/Archives/Public/public-poiwg/2011May/0004.html
> >
> > [[
> > How would you see dynamic aspects like interaction, scripting and
> > animation being integrated into this data model?
> >
> > At the moment it's a completely static model with no scope I can see for
> > extension in this direction (except perhaps related to "time").
> > ]]
> >
> >
> >
> >
> 
> 
Received on Wednesday, 11 May 2011 22:48:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 11 May 2011 22:48:24 GMT