- From: Erik Wilde <dret@berkeley.edu>
- Date: Thu, 22 Jan 2009 09:48:48 -0800
- To: Alexander Mayrhofer <alexander.mayrhofer@nic.at>
- CC: "Thomson, Martin" <Martin.Thomson@andrew.com>, public-geolocation@w3.org
hello alex. Alexander Mayrhofer wrote: > Plus (warning - slightly OT) - barometric altitude is completely useless > in some situations, most prominently in the cabin of a commercial > airplane. > Some GPS receivers insist on using the barometic altitude (and ignore > GPS altitude), which makes your GPS track stick at 2000 Meters, > essentially rendering the elevation measurement completely unusable, and > off by several kilometers. but isn't that exactly the kind of scenario that you could easily deal with if the application had control over the location provider, whereas there is nothing you can do if the API itself uses some fixed method? it's hard to know for a device to know when it's on board of a plane, but it's pretty easy for a user to enable "airplane mode" in which case the API's barometric measurements will be ignored. simplicity always comes at a price, and i think in this case it makes more sense to pay that price in an application rather than in the API. the API could even go as far as proposing sample algorithms (such as the one presented by martin) for people who are not interested in more nuanced geolocation information. applications with a need for better control over how geolocation information is used would then still have the possibility to implement their own simplification, or to not simplify at all and present all information to the user. cheers, dret.
Received on Thursday, 22 January 2009 17:49:50 UTC