We have been busy designing our workshops for our OD4D project on USER CENTERED METHODS FOR MEASURING THE VALUE OF OPEN DATA.
The end aim of the project is to come up with a methodology for developing metrics to determine if open datasets can provide real value to users in a specific context. The workshops are the first step as they will help us find what really matters about open data in a specific context. The aim is to identify what is needed from open data to make a real difference to core concerns to “ordinary” organisations. That is organisations that are not especially “into” open data or indeed technology but have other pressing needs. Our first example being organisations supporting the homeless in the UK city of Winchester.
Many open data meetings and events (e.g. Hackathons) are aimed at promoting open data and developing a community. These are vital and worthwhile aims. But they have two characteristics which make them unsuitable for our purposes:
* The attendees have some kind of special interest in open data. They may be enthusiasts or simply been given responsibility for open data; but in either case they come looking for applications for the data.
* Typically attendees are presented with some examples to encourage them, told about some data and encouraged to find good uses for it, frequently in the form of an app. They inevitably end up working from the data, to the information it can provide, to possible problems that information might address.
Data => Information => Problems
Sometimes this is very successful. But the danger is that it produces sexy solutions to relatively minor problems – problems selected because they are amenable to such solutions – not because they are vital concerns. And it leaves open the question – can open data help with core problems? Almost every problem can be alleviated with better information – so what would need to change for open data to provide that information in a useful way to the people with the problems?
We want our workshops to reverse the order.
Problems => Information => Data
We will identify a small number of problems that really matter to our attendees, then find out what information would make a significant difference to addressing those problems, and only in the second workshop look at how open data might supply that information (and what is needed for it to do that effectively). The output from the first workshop will be a short list of major problems and key items of information that would help alleviate them. Between the workshops we will sit down with people who know a lot about the available data and see how close we can get to providing the information. We will do this at two levels. One is simply examples of the datasets as they exist at the moment. The other is examples with our best effort at providing support and visualisation. Then at the second workshop we will ask the attendees to work with the examples and see if they can use them. There is a real danger they will not be able to get anywhere with the data as it is at the moment – hence the need for a supported version.
It is vital to have a open mind as to what really matters about the data. It could be the content, it could be how up-to-date it is, how secure it is, or something we had not anticipated. So we want the attendees to think how they could use the data in their real working environment – what time, what physical environment, who else is there, what else will be going on. One thing we will not be addressing is how easy is to find the data in the first place. It is a vital concern but we can’t cover everything.
Finally, in addition to furthering our research we want the attendees to come out with something useful. Luckily the workshops will include people who have responsibility for providing open data to the Hampshire voluntary sector who can listen to their needs and with luck help to meet them.