James Fee is using Planet to create a new geospatial news aggregator: Planet Geospatial. I'm a regular watcher of Planet Python, and am game to try Fee's new page. Russ Nelson has been doing something similar since the end of 2004 (if I remember correctly) at http://planetgis.russnelson.com, but I've fallen away from it because it's dominated by boring press releases from gisuser.com. Either Russ has configured his Planet incorrectly or he's being played by gisuser.com. Their future dates keep the releases at the head of his aggregator. The inability of some people to provide summaries or content excerpts in their RSS (Atom users, I'm pointing at you) is also rather annoying. Nothing like two pages of Java ArcObjects code to make one hit the back button ;)
I'm as concerned about a pandemic as anybody, but am appalled at the bird-flu fearmongering in Directions Magazine aimed at scaring up the money for yet another expensive global web mapping system. I'm willing to bet that, in the wake of a flu pandemic, we're going to be wishing not that we had a fancier web mapping system, but that we had a more robust physical public health infrastructure. I'm talking lab equipment, stockpiles of reagents, well-trained and motivated lab technicians, and well-funded community health centers.
The author of the article has a long track record of ignoring natural pandemics in favor of sexier and more outlandish threats to The State (previous articles at the bottom of that page), but is trying his best to make up for lost time. To top it off, he reveals to us that natural outbreaks may not be as natural as they seem:
The threat is real. The crisis is just around the corner. Whether naturally occurring or brought on as a terrorist act, this must be addressed. What? You don't think Al Qaeda would use the bird flu as a weapon against the US? Think again. Remember the West Nile Virus (WNV)? Reports that the US strain of the WNV is more deadly than anywhere else in the world have not been fully investigated. Could the WNV outbreak in the US have been engineered? I am not sure anyone has the answer. But it is something to consider.
Al Qaeda attacked us with West Nile Virus? That's just nuts. Who was behind the 1918 flu? The Illuminati?
In WGS84 and the Web, Allan Doyle notes that people are waking up to the fact that the current world geodetic system isn't going to be around forever.
Literal positions (whether in the http://www.w3.org/2003/01/geo/ or proposed http://georss.org style) are going to lose their meaning over time, whether from a change in geodetic systems, or continental drift. Rather than express location in XML like:
I think we need:
Users can be relieved of the responsibility of migrating their float values from one coordinate system to another on a per-document basis. The good folks at the fictional h2g2nis.net can take over the duty of preserving the accuracy and relevance of the Fort Collins location.
I'm sure that this is not a totally original idea, and would appreciate being clued in to prior art.
I've enjoyed corresponding with Allan Doyle, and will be a regular reader of his new Think blog. His first post has drawn the Ordnance Survey's Ed Parsons into a discussion about enterprise use of open source. Parsons would consider using MapServer, etc, if only there were accountable support for it in the UK like we have in North America. Allan follows up with a list of geospatial consultants, and the comment
I'm sure there are others, so people can help out in the comments. Maybe as the saying goes, when the student is ready, the master will appear.
Anymore, when I read "master", I think "Pei Mei". Sometimes mastering MapServer does feel a lot like punching a board from dawn till dusk.
It's no accident that the US and Canada have a large and healthy ecosystem of open source geospatial consultants. The Canadian government, particularly, has invested in open standards and in support for open source geospatial companies like DM Solutions. The free geodata that we enjoy in the US and Canada also play no small role. It's up to the Ordnance Survey, Britain's National Mapping Agency, to make similar investments in a open source ecosystem for the UK.
Jean-Denis Giguere, Kai, and I recently made some labeling improvements that make it possible to make decent street maps using ZCO or PrimaGIS. Currently available only from our SVN repo, but should be released soon as PCL 0.9.1.
I updated this site to the latest Zope 2 this morning and also took the opportunity to upgrade ZCO to the svn trunk. This breaks all the user sandboxes, and the previous demo, but will let me move ahead on the new and improved demo. Let me know if I've overlooked anything else.
This is nothing compared to Hobu's site upgrade.
This past weekend I connected the PCL and ZCO/PrimaGIS repositories to CIA, the open source notification system used by the MapServer project. Stats and Feeds about revisions to our repos are now online: PCL, ZCO and PrimaGIS. I also hacked Micah Dowty's ciabot_svn.py into a reactor for our Roundup tracker so that messages about new or changed issues are also echoed to CIA: Issue Tracker.