2007 (old posts, page 5)

Geospatial Venus

James, are you looking into Sam Ruby's new Planet branch, Venus? You can see its features, including a meme tracker, at Planet Intertwingly. Pretty sweet, eh? Mail me for Python help.

Meanwhile, my increasingly awesome feed (brand-new Atom categories, complete with schemes, and more liberal license starting today) is being massacred at the old Planet Geospatial.


Re: Geospatial Venus

Author: James Fee

Oh I've been looking at it. I love the whole J/K key control.

REST and Geospatial

David Smith, who has generally been a proponent of SOA and Web Services on his blog, considers REST:

Certainly the geospatial community will need to sit up and take note of where this leads... Having written my own WFS and WMS services and clients from scratch, though I am still no Web Services guru, I still had to wonder about the wisdom of the approaches being used in the whole paradigm.

See my web category for more about REST and geospatial.

Catalog Free

I finally read Adena's interview with Michael Jones, who in fact does invite us to start spatially marking up our data and services in the way I proposed. I'd better start reading Directions more often.

I'm excited about going forward catalog-free, but I'd be even more gung-ho if we were using Web standards (and maybe GML) instead of a repurposed proprietary XML language designed by a geo-visualization company.


Re: Catalog Free

Author: Tom Kralidis

Thanks for the info, Sean. What's so wrong with a "catalog" anyway? :) And isn't what Google is doing a catalog? Instead of marking up FGDC / OGC Capabilities docs, it's KML. Instead of an OGC Catalog approach, they're using their own API. Of course, I'm blindly assuming you're talking about OGC-ish catalog approaches.

Re: Catalog Free

Author: Sean

Tom, my favorite explanation of the weakness of web catalogs is Clay Shirkey's Ontology is Overrated. I can't recommend that article enough.

New Pleiades KML

I'm off on vacation for a few days after finishing the latest Pleiades site migration and release. Our master KML file is now at http://pleiades.stoa.org/places/nwlink. That's 304 places in ancient Lycia and Pisidia (southwest coast of modern Turkey) from map 65 of the Barrington Atlas. There are 102 map pages to go, and we'll eventually have about 50,000 ancient places (with extensive bibliography) online next spring. In case you have a Barrington Atlas handy, the map 65 grid is available at http://pleiades.stoa.org/places/BA065grid.kml.

It will take a while for Google's index to catch up with our changes, but thanks to mod rewrite and a little scripting, anything you find through a Google Earth search is properly redirected to our new URLs.

Interesting Web of Services Position Paper

Substitution of 'W*S' for 'WS-*' in Nick Gall's position paper yields a bunch of good lessons for the GIS standards bodies. For example:

Unfortunately, Web Services, at least the W*S style, are "Web" in name only. While W*S enables tunneling over HTTP (used merely as an XML message transport), in almost every important aspect, W*S violates (or at best ignores) the architectural principles of the Web as described in the W3C's Architecture of the World Wide Web, Volume One and in Tim Berners-Lee's personal design notes.

Via Pete Lacey.

Imagine a Geo-Web Without Catalogs

Google Earth with KML search may not be an SDI, but it sure as hell looks like a data or service catalog killer. If every W*S published a KML document like:

<?xml version="1.0" encoding="UTF-8"?>
<kml xmlns="http://earth.google.com/kml/2.1">
        <h3>Example Web Map Service</h3>
          This is a KML feature that points to the web map service
          at example.com.
            W*S Service Online Resource URL

we would be able to discover services using Google Earth, or by searching the index of a yet-to-be-developed, open KML crawler.


Re: Imagine a Geo-Web Without Catalogs

Author: Allan

And then a great wailing and gnashing of teeth arose from the semweb crowd...

Re: Imagine a Geo-Web Without Catalogs

Author: Sean

Wailing about the "alternate" link type? Then let's define a new one. It may look like I'm attaching new semantics to <ul> and <li>, but those are just for humans readers. The link (<a>) is the important part.

Re: Imagine a Geo-Web Without Catalogs

Author: Allan

I don't know.. I like the human-readable stuff. I was thinking maybe some people would want to see more things like capabilities XML or 19119 metadata. But they can go ahead and do that. Overall, I think it's great that people can play around with using KML like this.

Open Source GIS and Vista

Issues with Vista will be popping up in our projects soon.

I recently upgraded to Windows Vista / IIS7, and found very little help on the internet regarding installing PostgreSQL / PostGIS, most 'help' consists of useless trolls and flame wars. Seems the open source community would rather fight each other than try to play nice with the new Microsoft OS platform.

Useless help? You mean like this -- why in the world would you want to run your RDBMS, Plone instance, or any other server application on Microsoft's defective-by-design multimedia appliance?


Re: Open Source GIS and Vista

Author: James Fee

Latest and greatest Sean. You know how today's youth is. Heck most of them can't even remember a world without XP. For them Vista is their Woodstock.

Ancient Places and Google Earth Search


I'm finding a fair amount of other placemarks in the neighborhood of some interesting ancient places from Pleiades. Check out the ancient settlement of Caunus [direct network link], and the host of interesting photos in the Panoramio layer. If you search for "roman" in this spatial context you get a number of Pleiades placemarks, but there are a few from other sources, notably the "theatre of Caunos", and the co-located Panoramio mark. There are also nice shots of remarkable rock tombs near Karpasyanda.

Selected places from map 65 of the Barrington Atlas, covering Lycia and Pisidia, are at http://pleiades.stoa.org/places/nwlink.