[OSGeo-Discuss] Mobile GPS data collection...

[Mateusz Loskot]

Bob Basques wrote:
> You need to make these types of collections systems easy (preferably 
> easier) to use than whatever system is being currently used by the 
> customer.   I've been pondering building a collection system having
> to due with automatically geo-referencing photographs with direction
> and bearing all being captured.  Any device listed here that also has
> a camera would be a candidate for such a setup.  The biggest obstacle
> to implementation is something that is deemed to be too much work to
> use won't get very much play by the field workers.  I would like to
> build something that is push-button simple.

Bob,

Generally, that's my idea too.

> Always on device, continuous tracking, with compass heading of
> device, all being captured when the shutter is clicked.  Combining
> this with a method to apply the meta data of the user (device login)
> and time index in an automated fashion. Point and click, and the data
> is ready for re-distribution after storage, via a map based lookup.
> 
> The more automation put into them, the more likely they will be used
> in the field.

I'd add to the data collecting functionality list features like:

- voice control - user should be able to control the application and
device functions using voice commands

- speech recognition - writing on PDA is not very convenient, so it
could be good to add speach recognition, so user would be able to save
notes in the database/file

- another functionality of speech recognition I see is controling data
input, for example user gets the drop-down list of options, she could
use voice commands in similar manner as letters input on desktop (type K
to move the list focus to the first option starting with K letter).
So, saying Ka would work similarly, move focus to the first option
starting with K letter. Saying a whole word would try to find recognized
element on the list, etc.

- navigation with GPS could be controlled with
voice too - "route to Chicago" - start routing from current
position to Chicago, etc.

Recently, many components and native APIs for various mobile platforms
has been developed for speach recognition, so it should not be a
difficult challenge anymore to implement such features.

> I believe that if the software and hardware systems are applied 
> correctly, and if the collection process can be automated to a 
> significant degree, the need to manually enter things via a keyboard
> is reduced and can even be eliminated in some cases.  combining the
> capture process with a multiple choice CGI form would suffice for
> most data capture I think.

Yes, I agree with this opinion too.

I'm really excited about this discussion and idea to start some Open
Source GIS project for mobile platforms.

Best regards
-- 
Mateusz Loskot
http://mateusz.loskot.net