[OSM-talk] Mapnik style in slippy map

[spaetz]

On Thu, Aug 02, 2007 at 02:44:47PM +0100, Steve Chilton wrote:
> Cheers Frederik, that is really useful.
> What would be even more useful would be a count of paired occurences (k=v).
...
> I thought using "find" on the text file would work but it only gives 1 pub, and 3 stations (!), so don't know what it is counting.
> Could spaetz (or anyone else) have a look and see if it is possible to generate that paired occurence data please?

Hi all, nice to see my little script is of interest. It lists only 3 stations because this is a list of tags, not tag values, ie. the k in k=v :-). So only three people used station=whatever rather than railway=station.

Since then, I have improved my script to differentiate between the tagging on nodes/segments/and ways, so I can see how often a tag is used on ways vs nodes etc. This lead to a couple of nice graphs which I would happily share if people are interested.

Including the frequencies of tag values, ie. higghway=primary vs highway=secondary would increase the complexity of the analysis by magnitudes as I would need to keep of keypairs in memory, rather than only the 1900 different tags or so. Also the display of such frequency tables would require more fancy filtering, tables etc. I agree that it would be useful, I just never got around doing it.

For such an analysis it would be better to import the tag data into a data base and do the anaysis there, rather than have a dumb script go through the planet file and count word frequencies.

spaetz
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20070805/aa79e3f8/attachment.pgp>