[Marble-devel] Questions About Marble Installation

[Paul Carlisle]

We will very likely be using Marble for an upcoming project that requires the 
display of a user's position on a map. Resolution will never have to be very 
high - at the finest, we probably won't have to show an area any smaller than 
roughly 50 miles on a side. The application will be written in C++ using Qt 
and the MarbleWidget class, running under Unix.

Unfortunately, this application will almost never have Internet access, so we 
will be forced to provide a large image, or our own set of image tiles.

After playing with a simple demo program, I note that the Blue Marble dataset 
only seems to contain a single JPEG image that is 2700x1350 pixels. This seems 
awfully small for the level of detail seen in the zooms. Is it drawing the 
data from somewhere other than the <install location>/data directory? Or is 
Marble just using a really good scaling algorithm?

I note in your documentation that the largest image that the widget will 
automatically tile is 10800x5400. If we choose to download, say, the full-
resolution NASA Blue Marble images and provide our own tiles, is it sufficient 
to simply place the tile directly tree in the data area, or are there 
additional steps needed to make the application locate the tiles?

Also, what is the recommended way of telling Marble where the data is stored? 
Is there an environment variable, or can the path be set at runtime? I'm 
referring to the widget, not the Marble executable. We will likely not be 
given root access when we install our application and data, so we need to be 
able to configure Marble to look where we will be forced to install.

If this project succeeds, we may explore the possibility of building our own 
Open Street Map server in order to provide the follow-on level of detail some 
users will want. For now, however, high-resolution imagery will probably be 
sufficient.

Thank you in advance for your help.