Sean Gillies Blog

I am helping Howard Butler run a Python GIS hacking workshop (scroll down a bit) this coming June 16 at the 2005 Open Source Geospatial Conference in Minneapolis. This is the 3rd year for an event that began as the MapServer Users Meeting. In 2003 I presented the ZMapServer project and in 2004 lead a workshop on MapServer's scripting interface. Hobu is also a perennial presenter and the organizer of this years' session of lightning talks.

We have 4 hours to hack away at some interesting problems:

• Geometry and raster processing with Python, GEOS, and GDAL.
• Geocoding, RSS parsing, and map rendering with MapScript.
• Wrangling OGC Capabilities, GML, and WFS with Elementtree.
• Free time near the end to explore specific interests of attendees.

Howard tells me that we're well on our way to filling the workshop seats. If you are interested, don't delay.

My proposal is submitted, but no word yet on whether I'll get to present on the PCL and ZCO projects.

Donnie Marino is hosting the first meeting of the Front Range MapServer Users Group tomorrow, March 9, at his home in Longmont. It's an open-ended social gathering from 7pm onwards. Sounds like it should be a good mix of users: academics, engineers, web programmers, open source zealots, and those finding the right use for open source software in otherwise COTS shops. I'll finally get to meet 3 people that I only know from the #mapserver channel on freenode. If you missed Donnie's post to the mapserver-users list, send him an email to get directions.

Fort Collins' own Linux and Python advocate Sean Reifschneider blogs PyCon 2005. Interesting photos and fun commentary. My suggestions for code sprints at the MapServer Users Meetings has become a running joke among MapServer developers, but these photos have inspired me to run the old flag up the pole one more time. Past two years, the MapServer project has scheduled releases to coincide with the meeting; this leaves most of the developers in no mood for programming. Any sprinting would probably have to be on some project other than MapServer itself.

update: Hobu just informed me that he proposed sprinting at the OSG05 organizing meeting and it was shot down immediately.

Good news: heard from Steve Lime this morning that my proposal to present a talk about PCL and ZCO at OSG05 has been accepted.

Three books related to MapServer and/or Open Source GIS are due to be published this summer. It's fascinating how a community with no paper books can suddenly produce a trio. Almost like how Hollywood can burp up a pair of asteroid/volcano/virus disaster movies simultaneously -- except that these books won't suck.

Tyler Mitchell's Web Mapping Illustrated (oreilly) was the first that I noticed. My understanding is that Tyler is covering everything from data processing and desktop GIS with OpenEV to web applications using MapServer. I proof-read his chapter on mapscript and found it very good. He's not re-publishing the tired old community scripts; these are fresh and clearly coded Python, Perl, PHP, and Ruby examples. I'm definitely getting this one. Should be available in time for OSG05.

Next, i was clued-in to Mapping Hacks (oreilly) by Schuyler Erle, Rich Gibson, and Jo Walsh. This book addresses not the mainstream analyst/developer, but hackers, inventors, artists, and hobbyists. Their interest in scripting MapServer seemed to grow with time and I'm very impressed with the sample code that I found on their companion site: http://mappinghacks.com. Has also been fun watching Schuyler latch onto Python. Or is it Python latching onto Schuyler? Being a fairly conventional GIS type, I think I have more to learn from this book than from Web Mapping Illustrated and am eager to pick it up. Should also be available in time for OSG05.

Lastly, there is Bill Kropla's MapServer: Open Source GIS Development (apress). I know very little about this book or its author, but do know the technical reviewer who assures me that it will be a comprehensive guide to MapServer installation, configuration, and templating. No publication date announced, but probably this summer post-OSG05.

The workshop that Hobu and I are running is full, so now it's time for the first homework assignment. If you are new to the Python language and platform, please spend a bit of time with Mark Pilgrim's Dive Into Python. Chapters 1-6 are essential background. Chapters 7-11 are not essential reading; we'll cover the important aspects as needed. Chapters 12 and higher are outside the scope of the workshop. We plan to spend about 20-30 minutes summarizing from Dive Into Python before we get into the GIS hacking.

Steve Lime has cut a draft of the OSG05 program, and done a great job as far as I'm concerned. No conflicts for me. I'm scheduled to present in session 11 during the first slot of day 3 along with Schuyler, and some guys who offer "Talking to MapServer with PERL". I hope that's just a typo of Steve's. Are there still people who go around shouting "PERL"?

Here's a more interesting use of the cheapo WFS from last week: mining the next generation MapServer website's portal membership for the latitude and longitude properties (just added by Hobu) and publishing them as a WFS.

Here's the layer definition for MapServer users who'd like to try it out:

LAYER
   NAME "members"
   TYPE POINT
   STATUS DEFAULT
   CONNECTIONTYPE WFS
   CONNECTION "http://zcologia.com:9001/mapserver/members/features.rpy?"
   METADATA
     "wfs_service" "WFS"
     "wfs_typename" "member"
     "wfs_version" "1.0.0"
   END
   LABELITEM "zco:fullname"
   PROJECTION "+init=epsg:4326" END
END

The service root is

http://zcologia.com:9001/mapserver/members/

and the online resource URL for capabilities is

http://zcologia.com:9001/mapserver/members/description.rpy

So far it's just Hobu, Yves, and me:

As other members provide a location, the service will update (every 5 minutes). The picture above will not. I've tried no other clients save uDig, which likes my capabilities, but gags on my feature response. Your mileage may vary.

ZCO 0.5.1 and PCL 0.8.2 releases are available from http://sourceforge.net/project/showfiles.php?group_id=71074. The major new feature (see issue12 in the new issue tracker) will help new users properly configure their data stores.

The source for these projects has been moved to a Subversion repository hosted by Fort Collins' very own tummy.com, who do a lot of other great things for the Linux and Python communities. I've also moved from Sourceforge's issue tracker to a new tracker at http://zcologia.org:8900/primagis/ that is shared with PrimaGIS. It's filling up much faster than the old one.

The UNC-CH Ancient World Mapping Center brings a GIS perspective to ancient studies, and its maps for students are far better than the maps in my Penguin Classics copy of The Histories. I've downloaded the map of the Roman Empire's Northern Provinces cira 100 A.D., georeferenced it, and loaded it up as a layer available in the ZCO demo. Select the global_mosaic and rve11 layers to produce the following effect (here zoomed into the Aegean Sea):

Combine these maps with the better map interface of PrimaGIS and its support for custom annotations and one could have a rather nice annotated online atlas of the Ancient World.

Denizens of the all-ESRI blogosphere are writing about their tentative forays into open source GIS software. Predictably, the square peg doesn't fit in the round hole, and the result is disappointment. Hopefully, Fee will keep plugging, get beyond the lack of drag and drop, and see the obvious upside of PostGIS.

PostGIS is a great tool, and continues to get better and better as its user base grows in size and sophistication. The developers have clearly chosen to expend more resources on stability and performance than on useability, and it's been a good choice. Fee's problem is that he expects it to be a drop-in replacement for ArcSDE. This is not the objective of PostGIS and I think it would be a pointless waste of resources to try.

I've been having good luck with MapServer User Meetings and Rock. In 2003, I stayed an extra day to see Yo la Tengo at First Avenue, Minneapolis' historic music venue (other side of downtown from the conference venue). This year I'm fortunate to have a chance to see Sleater-Kinney play the night before the Open Source Geospatial Conference kicks off. If you're also an early-arriving rock fan, drop me a note and I can try to get you a ticket.

The lightning talk session went well. Hobu kept us on time and schedule, and after a succession of modest 9 volt talks Schuyler finally brought some real thunder to close it out. The audience appreciated his proposal for a distributed WMS cache, but it was his old time revival preacher delivery that made the entire session.

After this, Markus Neteler made an excellent case for users to try the new and improved GRASS 6. He asserted that they had cut the codebase by more than half, throwing out cruft while adding a new vector engine and yet more raster processing and analysis tools. The MapServer project has been unwilling to take on rewrites of such an extent, but might want to reconsider after seeing what might be gained.

After last year's conference there was some discussion about whether or not the MapServer/Open Source Geospatial community needed a Foundation to look out for the software, its users, and its developers. Some saw the Apache Software Foundation as a model, and were able to bring its current President, Dirk-Willem van Gulik, to tell us all about the ASF and compare the community of Apache users to our own fledging open source community.

We learned that open source GIS projects have asked to join the ASF. They were not enumerated by van Gulik, who simply stated that they had all been quickly rejected. The ASF's great success at promoting and defending the Apache brand is the result of extremely rigorous code requirements which our community's projects cannot, apparently, meet. This shouldn't be seen as invalidation, because the ASF model is not for every community. It arose from the very specific needs of the Apache users who had to find a way to work through the chaos of the internet boom and bust. Competition between their employers was far more cutthroat than we experience in the open soure GIS community, and extreme measures were needed to protect the Apache code.

To me, van Gulik's talk was right on. Instead of bringing answers, he reminds us of questions that we should be asking not of him, but of each other. Which brand is it that we want to promote and defend? Do some of our projects need to be more open? Is our community one of individuals, or of corporate bodies? There are lessons from the ASF, but we are really going to need something more organic.

Update: For those of you coming from mappinghacks.com, I have a follow-up.

Sol Katz, a pioneer in free and open source geospatial software, died in 1999. His family has agreed to allow us to honor him and give credit to other contributors to the community with an annual award in his name.

The first recipient of the Sol Katz Award for Geospatial Free and Open Source Software is Frank Warmerdam. It is unanimously agreed that his geospatial data translators have been the catalysts for the rapid growth of our community. I've benefitted from Frank's software, and also from his generous advice and mentorship. Congratulations, Frank, and thanks for everything!