[OSM-dev] [OSM] Informations about osmarender for Google SoC

MilesTogoe miles.togoe at gmail.com
Sun Mar 23 18:28:43 GMT 2008

Mario wrote:
> Wow, the discussion is getting more and more interesting :)
> I'll reply to you all at once, cause the replies would be tangled.
> The little brainstorming I've done is not intended to be for this 
> summer. I've written those guidelines to clarify my ideas on the 
> application: for example, I could develop the core of the project (ie an 
> API to handle the rules file) and write down some GUI ideas. But I think 
> it's useful to know "where do we want to go tomorrow" (eheh ;)) to 
> better design the core and/or the GUI skeleton and to know how much it 
> would get. :)
> I think, just to start, that it's reasonable to achieve these goals:
> 1) The core which handles rules file
> 2) Some work to define the general usability of the final interface
> 3) Get to work some GUI skeleton, which could have, for example, only a 
> textarea to edit any specific CSS. (I have in mind that all the GUI 
> should be i18n aware, or maybe not)
> 4) Eventually, a demo preview (or a part of it) like Frederik and 80n 
> have proposed.
> Unfortunately, I can't help much on the discussion about how it should 
> be implemented (XSLT/Java/Inkscape Plugin -> Perhaps I can help in 
> finding a way to edit CSS styles easily in inkscape, like proposed by 
> 80n in the last mail), but, obviously, this is very important to have a 
> rough idea of the time the work will get.
Brainstorming is good :)  And it helps us help you.    As info, if you 
decide to include some Inkscape integration on one of the steps, you can 
expect to get a lot of help from the Inkscape team (Inkscape has had 3-5 
GSoC successful students every year and the team gets quite involved).  
80n will be an awesome guide for you on this. 

I'm not sure exactly what you mean by step 1 but I think your steps 1 & 
2 could possibly be combined to form a "big picture" and task plan 
(which you're already getting a good start on) - maybe come up with a 
few detailed task plan alternatives (ie desktop vs web vs plugin, ...)  
and then submit to the OSM community to get feeback.  You may not expect 
to get total consensus but enough ideas to plunge forward in a focused 

Step 3 could allow for submitting GUI mockups again to the community for 
feedback and selecting a GUI design to go with.

Step 4: I also think a working demo is something you should plan on this 
summer.  And you might want to include submitting your code into a 
development branch - that way it starts to get used under "real world" 
conditions.  This creates a good "deliverable" for the project. 

btw Inkscape has a wiki or blog page for students to blog about their 
progress - I think this is a great idea for OSM and you might mention 
about blogging your progress in your application.  

More information about the dev mailing list