Disruptive Proactivity

When Will Perrin first heard about the plans for an Open Data Institute, this was his comment:

I agree with him somewhat, but have been saying that for a long time; but it’s not really that clear. It’s also an area to which I’ve been giving a lot of thought, for reasons we’ll come back to.

This week’s announcements are welcome, and a prerequisite for what should happen next, but they, aren’t quite enough. That is not to say that they should have been more, but simply that others need to build on this exceptionally strong foundation.

If 5 impossible datasets are released before breakfast, and no one says anything, did it really happen?

Back in 2007, Clay Shirky spoke about how Perl will continue to exist tomorrow as along as there is someone around who loves it, so wants it to exist tomorrow. It’s not new to believe that things can be done for love alone. Segment video or full audio.

For the last 900 years, the Shinto Shrine in Ise, Japan has been rebuilt every 20 years, and will continue to be for as long as there is someone to rebuild it. There are people who love OpenStreetMap enough that they want it to exist today, and will want it to exist tomorrow. Open data will remain open, as long as it is used, but generally doesn’t have enough love around it yet.

Like Perl, there are people who love OpenStreetMap, or WhereDoesMyMoneyGo, or trains so much that the things built on it will continue to exit. And there are communities there, tended and nutured by OSM, OKF or similar. However, for many of the datasets just being put online, there isn’t. In the same way that some books in libraries are just books on shelves. But like a book on a shelf for a decade between borrows, it can be useful to those who do want it.

Geek Love

There aren’t enough open data geeks to love everything at the level needed, it needs issue geeks to start to love open data, or simply those who are paid to work on an issue, to really need it. To love it because of the questions themselves, not the answers within.

If there are people who value a dataset, people who actively love it, then it will continue to be available absent significant reason (or screams when someone tries). It is most effective when it’s not just those who those love open data. They love data the way librarians love books, but what is really needed is people to love a dataset the way that a researcher loves a book.

That passion, when present and matched with external contextual knowledge, will push through problems with data quality, metadata, and all the other issues that can slow it down. But, fundamentally, someone has to really care about the data. Projects like LinkedGov and OSM, characters like @linkedduck, work by lowing the strength of feeling required for engagement; making it enough to hold “brief affection” rather than “undying love”.

COINS will continue to be loved for as long as Lisa cares enough to talk about it. I was chatting to someone from Treasury who was disappointed that a COINS update hadn’t got that much coverage or immediate usage. It doesn’t take much, but if you’re going to release an update the week Lisa’s on holiday, she’ll not be covering it. One person is enough, but it’s also somewhat fragile.

An influx of people could come from the quant-nerd core – people there care about some idiosyncratically specific things – but generally, OKF cares much more about openness in general, than a specific dataset. It’s good that OKF do, and everything that’s been achieved couldn’t have been done without them, but it’s not enough alone. It’s a community of support. The best hackday projects come from people who care a little about an issue, and put something together, but it all too commonly ends there.

To demonstrate real use, to demonstrate real value in those datasets, they need to be loved by people who care about something bigger than the data, and for longer than a weekend. People and organisations for whom it is a means to an end; campaigners who have dedicated their lives to fighting for equality, or animal rights, or the environment, or whatever.

Love the right thing

The train nerds (Jonathan, Peter) don’t primarily love train timetables, they love all things to do with trains (possibly excluding delays). Their incredibly successful (and time consuming) work on opening train timetables (and fares etc) are due to being important prerequisite for the stuff they really care about. If people see value in some data existing, if they will include it into their lives.

Currently, that’s not happening.

The realtime bus/borisbike/tube data for London is really useful, but how many people is it actually helping today who aren’t friends with transport nerds? Getting it to a critical mass doesn’t need more geekery, it needs more outreach and community.

If the staff of No10 (and 11, and the opposition front bench), had an app that told them when they had to leave their desk to be on time for a bus/tube to their next meeting (using real time delay info etc), I suspect they’d see rather a lot of direct value in that existing. It is one way, and possibly the only way, to safeguard it into the future. (Of course, if you have a chauffeur and PA who know what you’re doing, you don’t need an app).

‘To be at yoga for 18:30, you must be on the train from Kings Cross at 17:14, leaving your desk at 16:46, get the number 8 bus down the street, and get off outside the tube stop – at the speed you walk, you’ll make both the bus and train.” And changes the instructions when the bus is so late you can catch the 7 before it.

There may be an app that looks at my calendar to figure out where I should be, and does the above; but I’ve not seen it.

Geek for a day

“Geek for a day” from nominet foundation is a good start, but it depends on how issue-led the areas are, and one day is probably not enough for the charity to get a huge amount of substance out of it unless they’ve got the groundwork in place. “Geek for a week” is possibly still not enough. There needs to be sustainability and ongoing infrastructure. Most hackday ideas are prototypes, with no hope or process for going forwards, even where that is facilitated as much as it could be.

Peek at a Geek, might work better.

An ex-day-job event recently introduced me to this project (STAR/METRICS) by the NSF/NIH – see the fourth presentation here. What that does, fundamentally, is makes the people who make decisions about where science funding goes, get direct value out of using systemic data that is created anyway, it’s just not shared. What they see, is information that is directly relevant to their (political) needs, generated as a side-effect of the science. And, fundamentally, if the Ministers understand why it’s important, it’s probably safe into the future. That might be one lesson from the last year or two of data, as it’s survived the general election, and sorted itself out afterwards.

---

I’m not sure which bits of this post are useful, but it seems to be two sides to the same thing.

Yoga update

In reference to the sleep and yoga post which appeared yesterday, the question, as currently formulated, is “how can I help people who care about an issue of justice use data to get it?”. I don’t think that’s the final formulation yet, but it feels like an adequate answer for now, and one which will bear the load of some additional layers of thought. Coffee?