Our summer tour has veered off over the Atlantic once again so Pete Forde and I (Communications Director Momoko Price) could attend and speak at the 2011 Open Knowledge Conference this week. We’re honoured to have been included in the Open Knowledge Foundation’s impressive speaker list and put among such stimulating company.
Earlier this month I got a chance to interview Seb Bacon, one of the developers for the OKF’s Comprehensive Knowledge Archive Network (CKAN) software, which has been adopted by institutions like the U.K. government for the backends of their data hubs. If you’ve used data.gov.uk, you’ve used CKAN.
MP: How did you get involved with CKAN?
SB: I’ve been doing various sorts of civic-society projects for a while; I do a lot of stuff with mySociety, which are a bit like the (U.S.-based) Sunlight Foundation. Then I joined the wider network of which the OKF is a part. Rufus (Pollock), who started the Open Knowledge Foundation, sits on the U.K. government’s Transparency Board, as does Tom Steinberg, who is the director of mySociety. These various groups are all about open software, open knowledge; open data. I drift between these groups myself.
MP: Wasn’t CKAN behind the design of data.gov.uk?
SB: So we have CKAN.net, which is the data hub, and then we have CKAN, which is the software that drives CKAN.net, which you can download and use to make your own data hub. We have a bit of a branding problem, in my opinion!
MP: You encourage people to open up their own data hubs?
SB: Right. OKF is a non-profit organization, but it makes some income from doing consultancy on the CKAN software. The OKF has released the CKAN software as open-source, and continues to develop open-source and uses income from customizing it for groups like data.gov.uk to fund further development.
MP: How is using CKAN more advantageous than just publishing data to your own website?
SB: If you look at the OKF, who are behind CKAN, its principles are all about openness. That means open source, but it also means open decision-making and so on. What that means is for people who choose to use CKAN they can be directly involved in influencing the direction of the software. The users are as much a part of the community as the developers.
In terms of the actual software itself and what it does, the idea of openness is embodied in the model as it comes out of the box: it’s a wiki-type model — the quality of the catalogue improves collaboratively, and you don’t actually have to have an account to start editing the metadata.
There’s also a huge amount of effort put into the API, which is very complete, and that means it’s very interoperable with other systems.
MP: What is your ultimate hope for CKAN in terms of its growth?
SB: Well, we don’t have some kind of world domination master plan; I think we’re all into collaboration — a functioning ecosystem needs lots of different players in it. On the other hand, we want CKAN to be as interoperable as possible, so it may be natural for it to federate other catalogues.
We’re also looking at the actual story of what we call “data wranglers” or data hackers, really, who want to do things with the data. We do want to do a bit more on the “doing things with datasets” and that’s obviously something that overlaps with BuzzData. We’ll be doing some of the same things, some differently, but we want to move somewhat into that area, and be interoperable with other systems that do that.
MP: At BuzzData our vision is a kind of Github for data that can put a dent in the insanely redundant work that happens with the one-way, top-down data distribution system that predominates today.
SB: Yeah, it’s sort of slightly similar to what we’re hoping to do, but we’re going to be looking at more institutional datasets, at the same time recognizing that those need to be altered to be useful. I think some of the concepts are going to overlap.
MP: How far has the reach of CKAN gotten so far?
SB: I think there’s been about 30 different installs, mostly around Europe. Lots of major capital cities across Europe are thinking about using it, it’s a big growth time for that at the moment.
MP: Where do you think the open-data momentum in the U.K. is coming from?
SB: It’s been years of slogging. All the people involved have been at this for 14 years or something, flogging the same horse again and again, about openness and opening up data, and feeling like they were really getting nowhere, and saying “For God’s sake, we’ll do it ourselves” and setting up the websites. Slowly some of these websites got noticed by the media. It’s just been hard work and time, to be honest.
MP: What keeps you going when progress is slow?
SB: I don’t know, a perverse certainty that you’re right? (laughs) A doggedness, I guess. All the people involved are from an open-source, technical programming background, so there’s this idea that you can solve any problem with the correct application of technology. This isn’t always true, but it’s an interesting prism to view things through. I think the idea of seeing the internet as a place where you can make political change happen, that’s what’s common with all the people involved in the movement. They’re geeks who believe that you can use the Internet for social good.
It’s as simple as that in a way.