(attachment forwarded from a geo/mapping hackers list.)
I'm interested to learn of any RDF vocabularies that folk
here have knowledge of (especially backed by running code,
data, community etc...). There is, as some of you will know,
a small lat/long-etc vocab at http://www.w3.org/2003/01/geo/
that was produced by Interest Group collaborations (mostly in
#rdfig IRC chat sessions a couple years ago). Lately I have been
investigating the uptake of that namespace in the RSS scene,
and looking at Google's KML (formerly Keyhole, now used by
the rather fun http://earth.google.com/ application).
If you have an RDF vocabulary or dataset that has a mapping,
geographical, lat/long etc component (including place name
databases), do let me know, ideally in this thread of by updating
the public Wiki page at http://esw.w3.org/topic/GeoInfo
I'm particularly interested in hearing from any W3C Members
on this topic, and in collecting perspectives on how "lightweight"
(RSS-friendly) extensions might relate to more sophisticated standards
such as the OGC's GML work (from which KML seems to draw). Mixing
geographic with other non-geographic data is one of the advantages we would
expect from using RDF; I'm interested to put that to the test by building
some demos on top of a SPARQL database. Suggestions welcomed!
cheers,
Dan
Forwarded message 1
Work on standardizing geoRSS, and bridging with OGC standards is very
commendable, if for nothing else than strengthening the support and use
of the format. However I question whether any increase in complexity
from the current "practices" of geoRSS will be adopted by the 90%
without demonstration of significant new benefits.
And any new formalizations need to acknowledge and incorporate present
work, mostly based on the w3c Geo vocabularly
[http://www.w3.org/2003/01/geo/] and the large amount of existing data
in this format. geoRSS as is has traction -- can we make
standardization backwards compatible?
Among the publishers are the USGS
[http://earthquake.usgs.gov/recenteqsww/rss.html] and European
Commission [http://tsunami.jrc.it/model/index.asp], and loads of
individuals and small orgs. Yahoo's new Mapping API
[http://developer.yahoo.net/maps/documentation.html] is based on RSS
2.0 plus the geo namespace. Hackers are plotting RSS feeds on Google
Maps too [http://blog.bulknews.net/georss-gmaps.cgi].
Even lighter weight and dead simple is the geotagging method seen on
flickr and del.icio.us, where coordinates are entered as tags in the
form "geo:lat=*" and "geo:long=*". In a complete abuse, these show up
within the content of <dc:subject> in this rss feed
[http://del.icio.us/rss/tag/geotagged]. But it works, for the users,
and for aggregators and mapping! On flickr there are more than 30k
photos geotagged
[http://geobloggers.blogspot.com/2005/07/rise-of-geotagged-tag.html]
For worldKit, I've informally extended the geo namespace for lines and
polygons [http://brainoff.com/worldkit/doc/polygon.php].
Beyond this, the georss.org spec would give the ability to define a
non-default coordinate reference system, semantics and relationships.
Perhaps polygons with voids. All worthwhile, but crucial to that 90%?
Don't get me wrong -- I'm happy to see an RSS/GML hybrid and I'd
definitely add parsing support to worldKit for it. Other efforts to
bridge these domains have perked my interest as well
[http://en.wikipedia.org/wiki/GML_FAQ_for_RSS_Geeks_and_others]. But
just as important is putting w3c geoRSS and existing flavors on firm
footing.
Mikel
--- "Josh@oklieb" <josh@oklieb.net> wrote:
> Bryce,
>
> A standard which is simple, but extensible in the way you describe is
>
> being developed at http://www.georss.org hosted by EOGEO). We hope to
>
> flesh out the representation and its applications in the next couple
>
> of weeks.
>
> Josh Lieberman
_______________________________________________
Geowanking mailing list
Geowanking@lists.burri.to
http://lists.burri.to/mailman/listinfo/geowanking