W3C home > Mailing lists > Public > public-geolocation@w3.org > June 2008

Re: skeleton Geolocation API

From: timeless <timeless@gmail.com>
Date: Fri, 27 Jun 2008 07:16:21 +0300
Message-ID: <26b395e60806262116m2a8f5ceft5aacea40b61c86de@mail.gmail.com>
To: public-geolocation@w3c.org

On Fri, Jun 27, 2008 at 3:56 AM, Aaron Boodman <aa@google.com> wrote:
> I am most interested in providing a way for web applications to get
> the current location of the user.

this certainly isn't my goal. while i'm not speaking specifically for
our device implementation.

Here are some design considerations:
1. i do not have enough internal storage space on my device (n810
[2gb], or n800 [0gb]) to store a complete high resolution map of the
universe (or globe). Even omitting water and ether, it's still not
practical :). (Although I had one not too bad collection of pictures
in 1gb, thanks.)
2. even assuming that i did have enough storage to store one set of
maps (satelite or street maps), it turns out that people in many areas
can't find themselves on a map.
3. while our n810 has gps, the n800 does not, and i'd like both to be
able to participate equally.
4. my goal is for web apps to enable users to ask them questions.
5. my goal is not for web apps to be able to ask devices questions.

I'm currently trying to plan (potentially hypothetical) trips to:
* Vancouver
* Boston
* Los Angeles
* Redmond
* Cannes
* Turino

and a number of other places. It would help me a lot if i could use a
geolocation enabled browser as i sit here in Helsinki.

note: while I have been to two of the places on this list, I've
*never* been to the other 4, which means that even if I were "able to
locate myself or places I've been to on a map", I could not really use
this information to try to tell the device where I need to be.

Some of the places I've listed are fairly major, but as I'm about to
plan trips, I may only know the names and may have absolutely no idea
about what things are near them.

> I get the impression that this is
> what the other UA vendors here are interested in as well, but they
> should chime in one way or the other.

I don't know that people know how they expect things to be used. I do
understand people saying "oh yeah, it'd be nice if we could let google
maps work for mobile devices".

For another example, I've been lost at times in various places. I
remember being lost "somewhere in the Philadelphia suburbs with only a
cell phone" (and also "somewhere between Santa Cruz and San Jose".
While it sure would have been nice to be able to use a GPS enabled
device, I didn't have one. I did have at best a name of the road I was
on, and the name of the next road I saw. But as the roads were very
small and their names were common (and in Philadelphia roads like to
get up and move, and hit eachother repeatedly) and the same names
could appear in dozens of communities. Imagine you're lost and your
only point of reference is "main street, some small town usa". It sure
would be nice if your device supported gps, but for whatever reason,
you don't have gps, you find one open wireless network which doesn't
directly provide positioning and seems to have tunneled ip presence on
the other side of the globe. For convenience, you only have an n800.
Your cell phone ran out of battery, and it's night time. (This means
that a future solar charging enabled device will not help you).

I've traveled through a number of places in scenarios like this on
both coasts. I travel as a cyclist, which means I "avoid highways",
"use backroads" and can easily get stuck between two nowheres at

Ideally, I should be able to somehow "figure out where I am" with
whatever accuracy I need to use, and then enable my browser to share
this information with the next service sites I'm going to visit.

In some of the cases, I'm likely to actually want to give out high
accuracy information (so that the taxi, bike repair, or friend can
find me). In some cases, I just want to know of places to stay and
don't really want to give my exact coordinates to anyone. Perhaps all
I know is that I'm between certain major points and would like to know
what POIs I might be able to use to find my way.

> Having a URI scheme for location is an interesting concept, but it
> does not seem like the simplest way to expose this information.  I
> also see no privacy benefit to exposing information this way.

I'm pretty sure the uri scheme wasn't intended as the only way of
expressing location. I think the goal was to express position relative
location with accuracy limited by the relative named location's size.
If you specify a dorm, you've pinned yourself to a much smaller
location than if you specify a campus.
Received on Friday, 27 June 2008 04:17:01 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:33:49 UTC