The Map Is Not The Territory

A blog by Christian Willmes.

FOSS4G 2014 Portland recap

| categories: conference, geospatial, open source | View Comments

I was lucky to get a talk in at this years FOSS4G academic track, which allowed me to travel to Portland, Oregon. The talk was about a project in which we applyed Köppen-Geiger climate classifications to paleo climate model simulations, using GRASS GIS. The talks were recorded and shared on vimeo, so enjoy! :).

Köppen-Geiger classifications of paleoclimate model simulations — Christian Willmes, University of Cologne from FOSS4G on Vimeo.

The slides of my talk are accessible from the foss4g 2014 slides collection on github and from the CRC 806 database.

The FOSS4G program itself was huge. Eight (8) parallel tracks plus an extra track for Invited talks and Keynotes. So it was not at all possible to attend all the talks that you had an interest in. You can find the recordings of mostly all talks on vimeo and they are also linked directly from the schedule. This great organization and technical skill which deliverd almost all the recordings in that short time to the crowd is by itself a great acheivment of this years FOSS4G hosts, the OSGeo PDX team.

I was wondering about the organization of the academic track in the program schedule, because there was none, at least to my knowledge. So I could not see which presentations in the programm were from the academic track, I could only guess based on the titles and abstracts of the talks. This was sad, because I would have liked to attend most acedemic talks to maybe network with other academics using and developing FOSS4G.

I also have a subtle feeling, that the academic track has a not so well standing for many of the core FOSS4G participants. I think this is a pity, because those pesky academics ;-) are teaching the FOSS4G tools to their students and use them for - and promote them in - their academic works. This creates new users and maybe even developers for the projects and the community can only win from this. Thus, I would strongly recommend to better promote the academic track on future FOSS4G conferences, to increase visibility and thus also the quality of the academic track. To have the oportunity of publishing the submissions in the Transactions in GIS Journal is a big advantage and makes the submissions for academics to FOSS4G attractive. And those who are bored by academic talks do not have to attend them. There is no downside of having a prominent academic track on FOSS4G conferences, as far as I can see.

On Saturday after the conference I attended the code sprint, and had the opportunity to work a bit on the planned OSGeo Journal' OJS system update and on the update of the Wiki usermap, that we/I try to migrate to a Semantic Mediawiki based implementation (which seems to be a bigger story, on which I might have an additional post in the near future). Anyway, I enjoyed to sat down there, breathe the air of an OSGeo code sprint and gave the OJS update a try, but I did not prepared anything beforehand and I ran into some MySQL level problems durig a test run of the update on my local environment, which I could not solve on the code sprint.

In recap, I had a very good time in Portland and at FOSS4G, during the conference as well as the two days before and after the conference. Before the conference, on my first day in Portland I went to the Portland Timbers vs. San Jose Earthquake match at providence park joining the FOSS4G Timbers field trip. It was a quite entertaining match, with a final score of 3:3 (6 goals!). On this way thanks to the organizers of this field trip again! After the conference I had an awesome roadtrip from Portland, along Eugene, Crater Lake, the Oregon and north California coast, the Redwood trees, Napa Valley to the Bay Area and San Fransisco, from where my flight back home was heading.

Have fun!


 

comments powered by Disqus

Read and Post Comments

SLD production for use with GeoNode/GeoServer

| categories: webdev, geospatial, open source | View Comments

The Problem

Normally I use QGIS as a Desktop GIS for producing geodata and visualization of it for the deployment of web services. For MapServer WMS there is a great toolchain by using the MapServer export plugin. But since a while I am working with the even greater GeoNode application for publishing geodata, which uses GeoServer for the deployment of OGC Services (including WMS).

From a first look, there is no problem, because QGIS supports SLD export out of the box, and GeoNode / GeoServer accepts SLDs for the styling of WMS services. But sadly the SLD produced by QGIS is version 1.1.0 and the GeoNode / GeoServer only accepts SLD 1.0.0 at the moment.

<StyledLayerDescriptor xmlns="http://www.opengis.net/sld" xmlns:ogc="http://www.opengis.net/ogc" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" version="1.1.0" xmlns:xlink="http://www.w3.org/1999/xlink" xsi:schemaLocation="http://www.opengis.net/sld http://schemas.opengis.net/sld/1.1.0/StyledLayerDescriptor.xsd" xmlns:se="http://www.opengis.net/se">
QGIS generated SLD 1.1.0 header.

A solution

So I was looking around for a tool which can generate and edit SLD 1.0.0 conformant styles. I looked into other GIS desktop applications I have at hand. I did not even dare to think that Arc* does support interoperable styles... indeed it does not. ;-)

But AtlasStyler to the rescue! AtlasStyler is a nice small Java application which offers an intuitive GUI for editing SLD Styles, and best is, it produces SLD 1.0.0.

Screenshot of AtlasStyler GUI.

Additionaly, If you have some basic CSS knowledge, the created SLD file can be easily adjusted in a text editor of your choice. This is maybe needed for some more complex SLDs. I had to adjust for example the stroke width of the geometry outlines, because the AtlasStyler GUI only allows natural number (integer) values for the stroke-width parameter.

<sld:StyledLayerDescriptor xmlns="http://www.opengis.net/sld" xmlns:sld="http://www.opengis.net/sld" xmlns:ogc="http://www.opengis.net/ogc" xmlns:gml="http://www.opengis.net/gml" version="1.0.0">
AtlasStyler generated SLD 1.0.0 header.

Hope this little work around for QGIS -> GeoNode data publishing toolchain is of use for the one or the other around...

Have fun!



 

comments powered by Disqus

Read and Post Comments

« Previous Page