|
» Subscribe » Favorite Links » What is S60? » Freeware & Trials » S60 devices » Hints and tips » About this blog |
» Announcements (20) » Feedback (13) » Fun (1) » General (5) » Tips and tricks (12) |
|
Main links » Mymobilesite.net » Discussions » Feedback For developers » Forum Nokia start page » Wiki » Discussions |
Research » Research project » Open Source project » Wiki » SourceForge Related technology links » Apache HTTP server » Mod_python » Python for S60 |
|
Subscribe RSS 2.0 feed |
Subscribe Atom feed If you wish to receive email notification, please here » |
|
» Mobile Web Server, spreading the love » Your input needed: Mobile Web Server feeds » Your input needed: adding access rights » Fully MWS Compatible PAMP » Mobile Web Server 1.4: out now! |
|
» September 2008 » August 2008 » July 2008 » June 2008 » May 2008 » April 2008 » March 2008 » February 2008 » January 2008 » December 2007 » November 2007 » October 2007 |
|

« Use your phone via browser | Main | Hot news: PHP and MySQL coming to S60! »
![]()
If you go to my mobile site and check out Presence, you will notice we've added a Connection type entry in there (2G, 3G or WLAN). So this entry will definitely be available in our next beta release towards the end of the year.
Now, what would be your ideas for new items to have in there? I know some are pretty obvious choices such as location, but then what to do with that more specifically?
As always, any other ideas welcome too!
-Jukka
Comments
I think it would be cool to have the location/activity as a more generic category, e.g. a house icon when someone's at home, a work icon.
Those could be set automatically by GPS if desired according to user-defined location areas, and could also be set manually, e.g. when you're out somewhere shopping or whatever. Personally, I'm not interested in Jaiku style knowing whether someone is currently sitting on the loo, but I would find it great to know where they are generally or what they're up to. It allows to interpret the availability of someone much better.
Some other info would be the song currently on playback, or the last n played songs, or the current destination if currently navigating with Maps (I know, this is very specific stuff...).
Posted by: jennifr | November 14, 2007 02:26 PMHello Jukka,
Posted by: nbelousov | November 14, 2007 04:10 PMi think a good idea could be to represent the current coordinates (from GPS, if any connected) of an user. The coordinates should be displayed as hyperlink to Google Maps for example. If no GPS connected - mobile carrier's network ID and cell ID. If in WiFi medium the name of access point in use.
Hello Jukka,
Elisa had a service where you could follow a mobile on the map. Not any more. I am guessing there is not enough info in the phone to calculate that location, but...! You are Nokia and you could implement on the network side that the location info could be sent to the phone. Than that info could be awailable for the web server...
How about that?
Posted by: Aron | November 14, 2007 10:15 PMT.:Aron
@jennifr: Those are very good ideas! I don't think the coordinates themselves are not that interesting or could have privacy issues but something higher level like work or home are interesting.
@Aron: The problem is that Elisa knows where some specific cell tower is located, but we don't know that (at least not yet). But we could still use the cellid, for example 'has been in the same place for 30 minutes'.
How about listening to accelerometer sensor present in N95 and some other models? What could we do with that?
Posted by: Jukka Eklund | November 15, 2007 04:31 PM@Jukka: Acceleration sensors are useful for detecting if the phone is more or less attached to the user (on hand, in pocket, or carried in a bag) or in a car with the user (...or if it is left on table, or if user is practically dead). They're pretty useless for detecting being in the move in general, since long-distance trains and planes have very little acceleration most of the time.
Augmented with GPS or cell info, they could provide some extra semantic information, though. Even without that, one could probably train software to detect things like being on a passive car cradle, or detecting it being on pocket while driving (reasonably constant-direction gravitational acceleration being close to the normal of other acceleration vectors), or while walking in contrast to running, which cause obviously recognizable acceleration swings. One can see interesting statistical stuff even from good-quality walking GPS tracks...
Obviously there has been quite a bit of work figuring out various typical-situation contexts, and it would be nice to have presence application that could make educated guesses. Having GPS and accelerometer combined in, the guesses could probably make plenty of sense.
Posted by: kirma | November 19, 2007 09:15 AM