Do you want to know how the average Open Government Data Application looks like? Now it is quite easy thanks to the DERI guys and their presentation at the Using Open Data W3C workshop about a recent survey they conducted:
A web based map visualization that relies on a single static dataset and was developed by an individual who made it available for free.
Probably if you have thought about it before, you would bet for all of the above. Indeed, that is a very good description of roughly 90% of the Open Data Applications we have currently available, and that is the problem. If we examine the definition close up and analyse it step by step, several inherit problems with current Open Data policies may arise:
- Web based: In fact this one is great, well done, no changes here please. Of course mobile apps are of the utmost importance nowadays, but just applying Best Practices and the Open Web Platform you will always be able to reach mobile web users with low efforts and a superb user experience.
- Map: yes, every time a new app arises you immediately think Wow! Another map. Of course, there are some memorable exceptions, but you know what I mean. The question is: Is there any live after maps and statistical graphs? and no, I’m not thinking about all those beautiful (and mostly useless) animations, a trend that fortunately we are leaving behind us. No thanks, then I prefer to keep with maps. For sure we need to improve on this.
- Visualization: You still will always be able to interact with that map, but usually that’s all. Data nowadays moves exclusively in one way, and that way is from Governments to the world, with notable exceptions again. You can always look at the data and visualize it, nothing else. Have a look, but don’t touch please! When will we start to see two-ways apps? Remember that this should also be about collaboration and conversation.
- Single Dataset: and when we say single we have both meanings in mind, just a single class of data and just a single data source. It looks like data have not yet reached the mashup era and we are loosing the benefits of cross-data apps. You can probably blame the lack of diverse data sources and common standards for this. We need more data integration to unveil the real value of Open Data.
- Static Data: people say that PDFs are the place where data goes to die, and the same can apply more or less to current Open Data apps. They feed from datasets that never are going to be updated or use a static copy of them ignoring any further update. Please, live data is what we need, care about it while you’re planning your Open Data infrastructures and processes.
- Individual: a solitary developer, freelance or researcher is what you will probably find at the app credits. If the business community is not yet convinced about the economic potential of OGD we indeed have a problem.
- Free: it looks like a good thing, no? But, what’s then the business model then for Open Data apps? Advertising? How are we going to convince anybody about the economic potential and benefits of OGD if there is nothing but total-free? And this time we are focusing just on economic benefits.
Of course, all of these are general comments and you can easily find exceptions for each of them. Nevertheless it remains a good representation of how an Open Data App looks like nowadays and, the most important thing, these characteristics show us some challenges in the Open Data universe. Several of them are quite complex and can’t be solved in a few words, but once identified we can discuss how to address them and change the Open Data outlook.