This post is already quite old. As such its contents may be outdated, obsolete or outright wrong. Please handle its content with care and question any recommendations it might give you.

Also, if you find any dead links please let me know, I'll try to replace them with links to an archived version .

Open Government Data Wien – First Thoughts

Open Data Initiatives have been blooming all over the world recently starting from the US and Great Britain. Last Week the City of Vienna joined this trend and has published a first round of datasets including census, budget and location information.

Almost immediately people started digging into the data and before one week has passed we can already admire some interesting and useful projects.
Looks like people are mainly interested in Toilets and Budget Data

Of course I also had a look a the data and came up with some first thoughts.

Two kinds of Data

By now there are two kinds of data: statistical data (census, budget) and location data. My first guess was that people will be trying do use statistical data for visualizations and create tools for every day use out of location data. By now it looks like my guess was right. Similar to my recent Push Snowboarding Visualization I tried to find a less occupied niche as I believe it would be a shame if the use of that data only focuses on Budget Visualizations and Toilet-Finders. Don’t get me wrong – those are great applications and really interesting and useful. It just doesn’t make sense to have 20 variants of each.

So what I started thinking about was making sense of the location data beyond just searching and displaying it on a map as it is. I looked for ways to extract further information from what we have. One idea immediately came to my mind:

Infrastructure coverage

Imagine you are looking for a new house, a flat or just a hotel to spend some nights. Normally you would just check on the internet which areas are deemed nice and start searching on some real estate website mainly focusing on price and the house/flat/room itself. Especially if you don’t know the city well.

But aren’t there some other things that are quite important as well? What if you’ve got kids and want to have a kindergarten or a school nearby. Or the other way around – you’re looking for a calm place apart from screaming kids. You’re a semi professional swimmer and want a public swimming bath just around the corner. You’re in the city just for a weekend of cultural exploration and want to be as close to museums as possible.

The list goes on and on and actually for these situations we already have most of the data we need. So why not give it a go?

The perfect spot

So I fired up Processing and started exploring what we have. As I really just did a quick and dirty sketch of my idea to get it running for tomorrows OGD Platform Meeting I won’t go into (technical) details for now. I plan to clean up the code and publish it as soon as I can find time to do so.

Anyway I just want to go through my sketch quickly and outline some of the findings I had.

Of course the first thing I did was to load all the location data and simply display it. For testing I did choose the hospital data as some hospitals are really close and some are far away from any other, which seemed perfect for testing.

Next I calculated the minimum distance to the next hospital. The darker the map, the closer the hospital. I also experimented with some more colours to get more detail than what was available just with shades of grey.

I also played around with things like segmenting:

Then I finally returned to what was the initial idea: Finding areas that conform to certain criteria. Therefore I needed to combine various datasets.

So I visualized the scenarios outlined earlier on:

Someone looking for silence:

  • at least 1km away from any school or kindergarten
  • and at least 3km away from hospitals (ambulances!)

A swimming-enthusiast:

  • at most 1.5km to the next public swimming bath
  • or at most 1.5km to the next bank or beach

More examples can be found on flickr.

Final thoughts

I believe that this application and Open Government Data in general do have a lot of potential. I am aware of the fact that e.g. for the amount of silence there are plenty of other factors apart from kids and hospitals. But this can easily be expanded if more data like traffic, factory noises,… become available. I will definitely try to explore this further and publish the code as soon as I do find time to clean it up a bit.
In the meantime I’m eagerly watching what other people are coming up with. Feel free to comment on this as well – I would appreciate any kind of Feedback!