Open GoogleCodeExporter opened 8 years ago
However, the positioning data from cell tower triangulation is only
approximate, and it's not always possible to receive a GPS fix.
I would use this feature if it were reliable, but given the above issues and
the complexity of intelligently working around them, I'm wondering if this
wouldn't have the potential to degrade Timeriffic's reliability. This offers
very interesting possibilities, but I'm not sure how well it would work.
Original comment by jonr...@gmail.com
on 16 Jun 2010 at 3:22
Thanks for the response. I realize the GPS data from the cell towers is
approximate and not necessarily reliable. I was just thinking it would be cool
to say "turn On WIFI" when i got near home rather then pick a time where i may
or may not be at home yet. Just a thought that's all
Original comment by cmgr...@gmail.com
on 16 Jun 2010 at 3:27
That's a feature I'd like to see in Timeriffic eventually.
However the issue imho is not as much reliability as much as battery drain.
Even in coarse-position mode (aka cell/wifi based) you need to have the app
periodically poll the location.
Reliability: agreed that it's an issue too, but imho lesser than battery. Take
Lattitude in Maps and look at friends in city-level privacy mode. Positions can
be quite approximate, to the point of being useless.
Another approach is to use the current Wifi SSID network name, see issue #33.
Original comment by rdrr.l...@gmail.com
on 16 Jun 2010 at 6:23
Don't get me wrong; this would be a wonderful feature if it could be made to
work well with all the quirks of GPS, cell towers, and WiFi -- but more
importantly, without affecting the mostly rock-solid reliability of today's
Timeriffic.
Problem is that there are enough diverse use cases that could make this a
complicated enough proposition that it wouldn't be dependable.
I don't know how Latitude's city-level privacy is implemented, but I suspect it
uses an algorithm to "blur" the location rather than falling back from a
precise, fine-grain GPS fix to a course-grain cell tower fix. I may be wrong.
My experience (Motorola Droid) is that the GPS takes some time to detect
sufficient GPS satellites to get a location fix. Problem is, the more accurate
you want to be, the more frequent polling you have to do, and the more battery
used in the process.
If you're using GPS coordinate data (whether from the cell tower or from the
GPS receiver) in Timeriffic, it would be interesting if you could leverage the
"Starred places" that a user can define in Google's Maps interface that comes
with the device, rather than implementing your own application-specific list of
places. (I'm all about reusing data and/or logic!)
The connected WiFi network name might in some cases be a reasonable criterion
to trigger settings changes, but in some cases it would not. I work on a large
university campus that is highly networked, but which comprises some 7.7 square
miles. Alas, the network name is not going to be suitable to allow a person to
distinguish between sitting in his/her dorm room, in a lecture hall, in his
advisor's office, or at the bowling alley in the student union. And most of
these buildings are fairly impervious to GPS reception.
Aside from the challenge of dealing the availability of accurate position data,
the biggest challenge I see is in making a user interface that's not
inordinately complex. For example, if I have time of day settings, settings
based on a GPS or cellular position (or the mere availability thereof), and
settings based on WiFi network name, which take precedence?
This sounds like a slippery slope, though it would be wonderful if you could
pull it off in a way that is easy to use and robust. Good luck!
Original comment by jonr...@gmail.com
on 16 Jun 2010 at 4:10
Issue 66 has been merged into this issue.
Original comment by ralfoide
on 2 Oct 2010 at 6:16
Original issue reported on code.google.com by
cmgr...@gmail.com
on 14 Jun 2010 at 8:05