[OSM-talk] SVG from OpenStreetMap
tom at tom-carden.co.uk
Fri Mar 17 14:56:10 GMT 2006
(cc'd to talk at openstreetmap.org for archives etc)
> The experimental SVG output looks very impressive. I have resisted
> commenting previously on the aesthetics of the output from a
> cartographic point of view because the OSM project has taken a time to
> reach a critical mass of data collectors and thus data to output.
Sure. I think that's been the correct approach, and I've kept quiet about
aestetics myself too. I'm hoping that with the excitement around the Isle
of Wight workshop we'll reach a tipping point soon where people with a
traditional cartographic background can really start to contribute.
> now I would hope that interested individuals will continue the process
> of making this data into a potentially more functional output(s). There
> are obviously still problems to cross, and I imagine these will be
> compounded in heavily (data)-populated areas. Traditional cartographic
> problems of generalisation, over-crowding and type positioning (for
> instance) will need to be addressed.
Absolutely. We have some interesting ideas up our sleeve for
generalisation - some truly novel stuff that won't require a complete
database to be useful. Watch this space.
> Suggest you put this experiment in the "Neat stuff done with OSM data"
> on the wiki.
Good idea. Done.
> Steve Chilton, Learning Support Fellow
> Learning and Technical Support Unit Manager
> School of Health and Social Sciences
> Middlesex University
> phone/fax: 020 8411 5355
> email: steve8 at mdx.ac.uk
> Chair of the Society of Cartographers
> 2005 SoC Summer School, Cambridge
> -----Original Message-----
> From: talk-bounces at openstreetmap.org
> [mailto:talk-bounces at openstreetmap.org] On Behalf Of Tom Carden
> Sent: Friday, March 17, 2006 1:02 PM
> To: talk at openstreetmap.org
> Subject: [OSM-talk] SVG from OpenStreetMap
> For those who don't follow the OpenStreetMap wiki too closely, I just
> wrote a post to the new OpenGeoData blog highlighting Etienne's
> interesting experiments with XSL transforms for generating SVG from the
> OSM API's XML. (Excuse my TLAs).
> I think this is really interesting for a number of reasons:
> * SVG is a well supported format in editors such as Illustrator and
> * SVG is built in to Firefox
> * The XSL transform happens on the client side, so it doesn't take extra
> server resources (above the extra call to the API)
> * SVG has built in support for rendering text along paths, making
> drawing road names etc that bit easier
> There are obviously some optimisations to be made in the SVG (Etienne is
> already working on that I think), and the projection needs to be
> generalised for any given API output (maybe needs help from another XSL
> head?), but I think it's an excellent start and want to encourage people
> to keep looking at this kind of thing.
> There's no reason I can think of not to put an XSL style sheet directly
> into the OSM API output, and link to it from the map viewer for the
> current bounding box. The only thing I'm not sure about is how easy it
> is to save the resulting SVG (I had to use the 'copy XML' option from
> the DOM inspection tool) and if it's ever going to work in other
> To help with getting the projection right, I've suggested to Steve that
> the osm element in the XML output should have the bounding box details,
> to avoid clients having to calculate them first.
> Any other suggestions? Technical answers on a postcard to
> dev at openstreetmap.org please :)
> talk mailing list
> talk at openstreetmap.org
More information about the talk