Spatial Galaxy

Well, the QGIS workshop at FOSS4G2007 is history. We had a capacity crowd and covered a lot of ground in a short 3 hours.

Rumor is there are some pictures and heaven forbid, audio from the workshop floating around. Maybe they’ll surface at some point this week.

I have a few LiveCDs left over and some of the coveted QGIS carabiners. If you run into me at the conference and want either, just ask.

The conference goes into full swing tomorrow, with the plenary session and the first presentations beginning in the afternoon.

Day 0 - Things are hopping in Victoria. Yesterday I helped a big group of volunteers set up 160+ PCs for the Workshops and the Labs. People filtered in all day and the process of putting faces to names was interesting.

Workshops start at 0900 today (Day 1) and run till 1600, then the OSGeo Annual General Meeting begins at 1630. I imagine by the end of the day, nearly the full contingent of conference attendees will be stalking the streets of Victoria.

This is an experience report–your mileage may vary

I decided to give JUMP another try today. So I downloaded the latest release (1.2) and unzipped it into a directory. Looking at the JUMP Installation Guide reveals the document is written totally for Windows users. No problem, but I’m using a Mac.

Looking in the bin directory there is a shell script named JUMPWorkbench-mac.sh. OK, make that executable and give it a go:

$ ./JUMPWorkbench-mac.sh
: bad interpreter: No such file or directory

I’ve seen that before–the shell script has DOS line endings and that makes for an unhappy script on OS X. That can be easily fixed in a number of ways; I chose to load it up in VIM and set the file type to unix:

:set ff=unix
:w

Now the script will execute and JUMP starts up. Of course the first thing I try to do is load a shapefile. No go–clicking on the OK button after selecting the shapefile results in nothing happening. The dialog doesn’t close or change, and nothing is added. No exceptions and nothing in the log either. Checking the mailing list reveals that at least one other person had this problem, however there is no solution in any of the posts.

OK, so let’s try a WMS layer. I enter in the old faithful NASA JPL WMS server (http://wms.jpl.nasa.gov/wms.cgi) and click the Next button. This time an error pops up telling me “No description was provided (Null Pointer Exception)”. Clicking the Details button gives me a stack trace but nothing to indicate the root cause of the problem.

I tried a PostGIS layer, but of course I didn’t have the PostgreSQL JDBC drivers on my system so that didn’t work.

I was able to load an image and create a new layer.

I guess maybe JUMP isn’t ready for the Mac–or maybe my system is just cantankerous. Maybe I’ll try Windows or Linux next…

Marco Hugentobler at the Institute of Cartography, ETH Zurich has announced the QGIS MapServer project.

From the website:
“QGIS mapserver is a server module for geographic maps. The content of vector and raster datasources (e.g. shapefiles, gml, postgis, wfs, geotiff ) is visualized according to the request parameters. The generated map image is sent back to the client over the internet.”

This project is very much in the early stages, as it requires a specific development version of QGIS. There is a Windows binary available for download as well as packages for Ubuntu (make sure to read the caveat).

You can also view a sample map request that displays the countries of Europe.

Is desktop GIS software a rusty old car with no wheels? Bouncing around the blogosphere sometimes leaves you with that impression. All the excitement these days seems to center around mashups, hacks, and mapping in your web browser. It’s definitely cool stuff. A number of folks think this is the future of GIS, even when it comes to doing analysis.

Part of this trend stems from a desire to deliver mapping to the masses. If you’re a consultant or service provider, it’s a good thing. Being able to provide your “customers” with a toolset over the web makes sense.

Read the rest of this entry »

Here is the process I used to quickly build (OK, but it was faster than usual) QGIS HEAD on Feisty Fawn. What’s QGIS HEAD? It’s the current development version that includes the tasty Python bindings that allow you to write both QGIS plugins and stand-alone mapping applications.
Read the rest of this entry »

And so it begins. Chad has made a plea to Microsoft to help sort out issues with the latest World Wind release and Vista. The new security features are causing problems and I suspect that World Wind won’t be the last project to have to deal with it. So far the QGIS project has yet to get an experience report from anybody using Vista. Who knows what that will bring….

A great divide separates the typical Open Source developer and user.
Each has differing expectations, assumptions, and priorities. The interaction between developer and user can be helpful, cordial, confrontational, or antagonistic.

Of course this all stems from being on opposite sides of the fence. The key to a successful relationship is communication and understanding (not exactly a new revelation). Unfortunately its not possible for one developer to communicate directly with thousands of users. So developers resort to blogs, websites, forums, and mailing lists in an attempt to foster some sort of dialogue. This works, although it can become so consuming that development time vanishes away in an attempt to keep up the lines of communication.

Read the rest of this entry »

Why would you want to run the Windows version of QGIS on Linux? Because its there. Actually, it may be a useful way to test the Windows version without firing up the dusty old Win32 box.

I did this more out of curiosity than anything else. I installed Crossover Linux (http://www.codeweavers.com/products/cxoffice) on an Ubuntu Dapper box. During the install process you are given the option to install Windows software. Of course QGIS isn’t in the list of supported software, but there is an option to install Unsupported Software.

The QGIS installer ran fine, however there was a problem with the fonts — they were invisible. I clicked my way through the install, unable to read the options/buttons. Obviously this is an issue if one didn’t know where and when to click.

Once installed, QGIS ran fine, although there were some font issues, specifically with the Plugin manager. I was able to load a PostGIS layer, zoom, pan, and identify features.

While Crossover Linux uses Wine under the hood, you can also run QGIS under Wine itself.

I remember growing up and reading predictions for the new year developed by some prognosticator, supposedly in the know. Of course most of the time it was all wrong, but often made for interesting reading. With that in mind, here are my top 10 predictions for Open Source GIS (OS GIS) for 2007.

Top Ten OS GIS Predictions (in no particular order)

1. OSGeo will be a synergetic force, fostering new cooperation and collaboration between projects.
2. OSGeo will struggle with being “all things to all people” but will settle into its niche by year-end.
3. Quantum GIS will release version 1.0, five years after its conception.
4. The FOSS4G 2007 conference in Victoria will be the biggest ever, bringing together all facets of the OS GIS community. Commercial GIS vendors will be lurking about the conference floor in significant numbers.
5. The Big Commercial GIS vendors will continue to take notice of OS GIS although by year-end its likely they “still won’t get it”.
6. A “killer” web mapping interface will be developed and cause a mini-revolution across the geospatial web space.
7. An odd alliance will develop between an OS GIS project and a big commercial GIS vendor.
8. Paul Ramsey will finally throw up his hands and implement raster support in PostGIS.
9. GRASS GUI debate/discussion/development will heat up again, the result being a new improved interface.
10. I will be exposed as a charlatan and unable to predict anything in the future.

There you have it. Maybe your crystal ball is better than mine. If so, here’s your chance to go on record.