[OSM-talk] Some question about OSM

[Nicola Ranaldo]

> You ask some good questions, but I must ask -- why not help develop one
> of the OSM off-line editor projects already out there?  Though Imi
> currently keeps tight control over repository access, JOSM is an
> increasingly mature project that could use more eyes.  Or maybe add your
> routing and editing features to the online editing applet?

Due to my experience the only project i can contribute is osmrender, howewer 
it's qt3based.

> Given the lovely activist and volunteer disposition of the OSM
> community, I highly doubt you'll get help on damn near anything unless
> you GPL or PD *first* and ask questions later.

Ok let me do a "minimal" code cleaning/well-commenting :)
Consider the project was born on my specific feature-request, and them could 
not be so intuitive for all users.

> > *) wath's the best approach to do street and place names rendering?
>
> The "best approach" is probably to join forces with Etienne and improve
> the SVG rendering, and then work to incorporate this same rendering
> system into JOSM.

I'm new in the community, i do not know your work around SVG rendering, 
howewer, in a real-time fast osm client, creating an svg and rendering it 
could be slow.

> > disposition of text will be evaluated by software in real-time?
>
> Yeah, probably.
>
> > will be some > key/value helping rendering software ? (for example, a
> > tag could say render this town names at x,y, or <important> do not
> > render it when scale > z and so on)
>
> Again *theoretically*, we want to limit the amount of data that is
> specific to a particular type of rendering system.  Consider the ideal
> of HTML in the early days ("content specification") vs. now-a-days (CSS
> as page layout).  I would hope we lean toward the former philosophy as
> long as possible -- with renderers interpreting *geographic* OSM data
> and metadata intelligently, rather than expecting the data to have
> explicit rendering instructions.

I agree, but i think we should take care to exceptions too.

> > *) my route planning code actually works on every raw segment/node, is
> > this right?
>
> That depends.  Is it fast enough looking at everything?  If so, that's

Ok, suppose it works on every raw segment (filtering them if unpaved, 
restricted etc. etc.), and suppose on a segment extreme there is a node 
belonging to a roundabout and we want a nice vocal message, i have to tag 
this node to instruct well the software, and so, i have to tag all the nodes 
of the "circle", or link the "circle" nodes to an inner unlinked node 
rappresentating the roundabout.
In all case i need a "formal" rule. OSM is free in tagging, but if we does not 
consider a "core-routing" tags we'll not gain from collaborative editing.


	Niko