Archives for posts with tag: ClimateSOS

I write a work-logbook. It often helps me to resolve issues I know I had before, and allows me to look back and see what I did when, and how. It also helps me a lot to try to write down every day what I actually accomplished, and if not, why.

Why do I write about that now? Because when I opened it today, the last entry was for the 15th of  July. That’s A LONG TIME AGO. I was actually busy, partly working (OGC discussion papers for SOR and SIR – contact me if you want to know more!) and partly travelling (a little bit, but not too much…), so it’s not that bad, but still. Anyway: I want to get the project finished in September and will be working almost full time for sos4R in the next weeks.

Here are some milestones (just from the top of my head, so subject to change) I’d like to take: Read the rest of this entry »

After the last post about data processing, I will not work much further on inserting data into the ClimateSOS – even so, there are plenty of features of interest and observations, really enough data for testing.

However, I’d like to save and share the current state with some example documents:

The current Capabilities are quite large already, due to the relatively many features. The division into offerings seemed to have worked well:
Read the rest of this entry »

Since the last post I mostly worked on processing the ds570.0 dataset. I implement a parser in Java to feed the ClimateSOS (http://giv-sos.uni-muenster.de:8080/ClimateSOS/) via the transactional profile – meaning I user the SOS operations RegisterSensor and InsertObservation via http POST to insert data. I base the code on the transactional feeder from the OX-Framework, and will also make the programme available for download here (see below).

The data insertion into the SOS takes me longer than planned, but for good reasons. First, I took some time to think about how to model the data that I have in O&M. I have to be able to parse the observation markup in R later and cannot prepare for all eventualities that are possible in O&M. Therefore I outlined an idea for an O&M profile for simple measurements (I will write about that soon). Second, I also spent some time to think about how to model the data in R once I requested and parsed it – luckily I am not the only one needing such a feature – see the sptX project.

So, a lot of things happening “on the side”, that do contribute though not directly.

But back to the parser: Read the rest of this entry »