[OSM-dev] [Fwd: Project idea for openstreetmap in Google SOC 2008]

[Martijn Pannevis]

Frederik Ramm wrote:
> Hi,
>
>     I haven't fully understood the proposal but the author writes:
>
>   
>> Moreover static map in a mobile device will be far costly.
>>
>>     
>
> With OpenStreetMap, the only thing that is costly about a static map  
> in the mobile device is the storage space. So it sounds to me that  
> this idea is trying to "work around" the non-existence of free maps;  
> since our aim is providing free maps, the idea sounds a bit  
> incompatible to OSM, or at least unnecessary?
>
> Bye
> Frederik
>   
I don't totally agree on that. Even if all map data is available, you 
still have to either store it on a mobile device, or transfer it to it, 
both of which can be costly (Even on my N95, not the whole planet.osm 
fits, and it certainly can't be processed). This can be an issue.
Altho I think the project proposal isn't that clear, what I think it means:
A map is heavy to download or store on a mobile device. To work around 
this, you could only show yourself relative to certain POI's in your 
neighbourhood. This will be less data to transfer (or store) than all of 
the map, and can still give you a good impression on where you are.

This is something that isn't solved entirely by the availability of OSM 
data: If complete, it still is a 'lot' to transfer, while a subset of 
that data could also indicate your position.

We (as in: my company) are currently working on an application that uses 
OSM data, but also has a similar mode: Don't show the map, only the 
POI's, for low bandwidth mode.
So, I do understand the problem, however, I wonder if it is a really 
relevant problem for OSM itself to solve.
Kind Regards,
Martijn Pannevis.