mikel has been experimenting with getting openstreetmap data on a Garmin handeld GPS. This is a massive step forward and points to all sorts of uses.
Also, the map of Weybridge made with XSLT and SVG described previously has made it on to the Weybridge wikipedia article! So, what a milestone! OSM maps have made it back to the mothership and been judged worthy to go on a wikipedia page.
One of the cool things about wordpress 2.0 is that you can drop an mp3 in to it and it magically becomes a podcast, including the RSS feed bits (so yes, you can subscribe to this feed in iTunes). I thought it would be useful to talk to people within openstreetmap and slightly further afield in the geo community as openstreetmap grows beyond everyone knowing each other. If you have a story to tell please get in touch, it’s likely I might ask to talk to you anyway You can discuss this cast, and maybe help transcribe it if that’s useful here. The first cast is me speaking to Ed Parsons who is CTO of the OS and says some pretty intersting things, but has detractors. You can find the 21Mb mp3 here. Enjoy.
An idea discussed by a few bright people like tom, mikel and ben is that of getting people to map as fallout from some other process like sharing their GPS traces. Or manipulating those traces, getting statistics or commenting/organising them. As tom calls it ‘the flickr of maps’, rather than ‘the wikipedia of maps’ which would be, from one angle, mapping for mappings sake. To this end I’ve been hacking away at nice sharing features for your GPX / GPS traces. You can have a look. RSS feeds, tags, descriptions and little pictures of your trace are now available. There’s some work left yet to be sure. It’s not super polished, but it is getting better and better.
OpenStreetMap contributor Etienne Cherdlu (aka 80n) has been working on converting the OpenStreetMap API output into SVG using XSL tranforms. So far so good!
The projection on this example is only really suitable for Etienne’s area of interest, but this gives a good example of where we’ll be headed for the Isle of Wight workshop, and shows what is possible right now using OpenStreetMap tools and a little bit of hard work. Thanks Etienne!
‘Wouldn’t it be better to create an open-source database of geographic and other data?
Much though we admire the stoicism of the people at XXX, when you compare it to the Post Office’s thousands of postcodes – which it has to verify – and the Ordnance Survey’s billion-odd bits of data, which would cost perhaps £200m in taxes to keep updated to their present quality – that is, about £4 per taxpayer per year – you have to say that it makes more sense to free the existing data than to reinvent the wheel. It’s a very large wheel.’
This misses the point of what’s useful. We don’t have to have millions of postcodes to be useful. We don’t need to know where trees are to the milimeter to create a map that’s 99% useful. Even Ed was saying this, I think, when he was saying that what OSM is doing is not the same as what the OS is. To the best of my knowledge the OS leaves streetmap work to mastermap derivation by third parties. They do mastermap.
It also misses that OSM and FTP are collaborative efforts with social and innovative technological angles.
If the campaign is about money then as a friend said the other day, let an economist decide IP law. There is no ‘right way’ to do government IP policy. The US has public domain, we have crown copyright, others have others. public domain is not magically better, it’s worse in lots of ways than, say, viral licenses like CC. I’ll leave the BSD vs. GPL argument.
But again, some economists should decide the IP law. Not the rightsholders (the OS) or us as consumers because we’re equally biased. What’s the overall benefit to the UK economy? That’s the question.
Personally, having been involved in campaigns, I think it’s going to take *ages* to have any effect. This will stir up a hornets nest. The OS may be big and evil, but as people said on Ed’s blog (I think): the treasury isn’t going to pay for this. They’ll privatise the OS if anything. Then what? Then we’re all suddenly competing with someone who has all the data, most of the expertise, and no need to play nice. Because I think at the moment they _do_ play nice with some people, compared to their options should they be privatised.
This is not an argument that the debate is pointless or that the ‘campaign’ should not be supported, but that we should be careful what we wish for. Oh, and that we shouldn’t dismiss the people at XXX.
So personally, I’d avoid polarising this issue about public domain vs. crown copyright. My personal judgment is that I will have more effect on the availability of free mapping data by working on and promoting OSM than lobbying europarl or within Westminster.
OpenStreetMap is way bigger than me though. I could spend time lobbying. What does everyone think? Are these seperate parallel inititives or more closely intertwined?
Part two of taggable gps trace support is now live in openstreetmap. There’s still quite some work to do: RSS is missing, you can’t search by bounding box yet but it’s functional.
Richard posts an interesting set of thoughts on geodata licensing, specifically CC-SA which is what openstreetmap uses. Recent threads on the OSM mailing list mirror some of the age-old BSD vs. GPL debates and unsurprisingly fail to reach a consensus. I’m hoping that debate before or during the workshop will settle down to something we can work with.
Lots of updates went through today like the API now supports things like streets, tags (keys and values) are a more realistically handled. This brings the API to version 0.3
I spent the rest of the day fixing up the gps trace pane so you can add descriptions to your traces, tag them, and make them available for public download. Still some work to do, pending input from the mailing list.
Ed notes (and dislikes) this piece in the Guardian about how taxpayer funded map data should be Free. The article isn’t limited to geodata and fails to mention openstreetmap like previous articles so perhaps it has an industry source. They make good work of the juicy target – the OS – and mention the Peter Weiss article.
More interestingly they bring up other examples like the Hydrographic Office’s somewhat mad scheme of copyrighting tide tables.