- From: Erik Wilde <dret@berkeley.edu>
- Date: Wed, 21 Jan 2009 09:22:40 -0800
- To: "public-geolocation@w3.org" <public-geolocation@w3.org>
hello.
i am still not quite sure where to look up open issues, could somebody
please point me to the right place?
a while ago i raised two questions about altitude, one was about the
model of altitude (WGS vs. EGM), and the other was about GPS vs.
barometric altitude. then we had the more general remark about the fact
that there can be various location providers, and in fact both of the
above issues could be summarized under that more general topic (even
though WGS vs. EGM also could be regarded as a simple unit issue; but
alas one that requires decidedly non-trivial conversions).
the comments on that list mostly seemed to argue that having multiple
location providers (such as multiple altitude values, or even multiple
locations) would complicate the API. and i guess there is no denying
that this would be the case. however, i am still curious to hear from
the proponents of the "there can only be one" (location) approach how
they assume implementers of the APIs would deal with the problem of
having to decide on for example "the altitude", when the device has two
conflicting readings from GPS and barometric sensors.
it seems to me that having only one location is a simplification that
either can take place in the API's implementation, or in the code using
the API. there are advantages and disadvantages to both approaches. API
simplicity is definitely an advantage of doing the simplification in the
API's implementation, and i am still curious to hear how the API's
implementation is supposed to do this.
cheers,
erik wilde tel:+1-510-6432253 - fax:+1-510-6425814
dret@berkeley.edu - http://dret.net/netdret
UC Berkeley - School of Information (ISchool)
Received on Wednesday, 21 January 2009 17:23:28 UTC