-
Position paper for W3C/Wap Forum Workshop on "Position
dependent information services".
Philippe Caloud
Cegetel/Inovatel
Introduction
Inovatel is a fully owned Cegetel R&D division, with the mission to
specify and prototype new and innovative service infrastructures. One of
the projects currently owned by Inovatel is a project of specification
and prototyping of a generic infrastructure for mobile location services.
The mobile location services we target are services which bring to
a mobile user an information which is dependent on his geographical position,
and which can typically be viewed on a map. These services will address
both the enterprise (fleet management, etc.) and consumer (information
services like yellow pages, weather forecasts, routing and traffic services,
etc.).
Project drivers
We believe that these services could be deployed in a rather short time
frame, due to the following factors...
-
The GIS technologies are progressing very well. Vectorized map representations
can now be compressed to a level allowing to transfer such type of information
by air connections, even at the current GSM data bandwidth. For example,
a vectorized map representation of a local area around a mobile user current
position can take as little as a few tens of Kbytes. Another advantage
of this technology is that the service information can come as a layer
on top of map layers. Only the dynamic service information has to be updated.
-
The PDAs have an adequate form factor and enough memory and CPU power to
store, process and display vectorized maps.
Project summary
A typical service user will be able to download position dependent information
from a content provider site, this information coming as a layer on top
of a map which has already been downloaded to the PDA.
Our goal is to participate to the standardization of XML position dependent
information packages, together with the associated extensions to WML in
the first place, so that content providers can design their services independently
from the protocol and mobile device used by the user.
The primitives we are currently defining will allow a generic control
of the creation, and display of the map related information. For example,
it will be possible to specify in an XML document what information has
to be encapsulated in a map format, and also on the user agent side, how
it is possible from a textual representation of the information to switch
to a map representation and vice versa. For example, in the context of
a WML browser, if textual information about a point of interest is
displayed as a card, a reference URI displayed as a bitmap symbol allows,
when selected by the user, to show this very point of interest on a
map. It is therefore possible to navigate between the textual and map visual
information in an intuitive manner. In the WAP context, we therefore plan
to propose position dependent extensions, a la WTAI, where generic functions
are made available to the service layer.
General expectations on the workshop outputs
We are specifically interested in the data formats for position relevant
data, metadata aspects of location information, and interoperability aspects.
Our current work share some aspects with the NVML initiative, but we
plan to include more operative primitives (XML elements) so that a service
designer can have a way to specify how the information is best processed
depending on the
target device features (in a similar way as WTAI makes the telephony
call control functions available at the service layer).
One interesting output for us from the workshop would be to have comments
on this position, and to have a better understanding of the current initiatives
like NVML.