location-oriented services

ewanm89 ewanm89 at googlemail.com
Mon Apr 7 12:25:38 CEST 2008


On Sat, 5 Apr 2008 12:31:29 +0100
john <jptmoore at gmail.com> wrote:

> Hi. I have been working on a D-Bus location-oriented service. I would
> like to get some feedback to help direct the development and possibly
> get some others involved. There are also some important non-technical
> issues I would like to address.
> 
> Firstly, the D-Bus service is a server type process to look-up the
> distance of other users of the system. It has a simple interface which
> has been designed to be self-clocking. What this means is you need to
> "ping" in location to update your local database of distance
> information. One of the ideas behind this was to encourage updates
> rather than be dominated by look-ups. A key feature of updating the
> database in this system is that transferring the data required is done
> in a very efficient manor. It actually "bit packs" the data. This
> makes no assumptions about the quality or cost of the underlying
> network.
> 
> So how would this be used? Well, I need some help with ideas and
> clients. A simple use case would be to match users to your contacts
> list and alert when they are within a certain distance. There are
> other D-Bus location-oriented projects which could be integrated or
> used in some way.
> 
> Are there any students working on GSOC projects in this area?
> 
> It is my intention to release all code including server. This brings
> me to the non-technical issues. Although I don't have any problems
> running server processes on my own hardware etc I think there are some
> issues which need to be addressed. I understand the privacy concerns
> of using these types of services and want to make this as open as
> possible. Releasing the server code is fine but it does not actually
> mean the "real" server is some other code doing something evil and
> logging your data etc. This is where a trusted third party needs to be
> involved. If these server processes were run from such a source it
> would be a step in the right direction. How do others feel about this?
> I know some people will never use these types of services full stop.
> That is fine. I know others are interested in using them. I personally
> am not interested in plotting on a map where I am. I don't mind
> knowing that I am approximately close to something or someone and
> vice-versa.
> 
> Anyway, I think D-Bus provides us with a good opportunity to develop
> some cool applications in this mobile space. Interested in your
> thoughts?
> 
> John (zedstar on IRC).
> 

The ad-hoc bluetooth/wifi transparent networking stuff may be useful?
The idea is to automatically initiate a ad-hoc mesh with nearby
phones, given a free IP network between phones, at which point we can
use bonjour and other similar zeroconf services.

-- 
Ewan Marshall (ewanm89/Cap_J_L_Picard on irc)

http://ewanm89.co.uk/
Geek by nature, Linux by choice.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.openmoko.org/pipermail/openmoko-devel/attachments/20080407/5a1cf5ab/signature.pgp


More information about the openmoko-devel mailing list