Paper Worth Reading: Government Data and the Invisible Hand

[Shoaib Burq]

David Robinson, Harlan Yu, William Zeller and Edward Felten have a great paper (pre-publish preview) at Yale Journal of Law & Technology challenging the mainstream thinking regarding role of IT in Government Services. I gave a talk last year "Can government learn interoperability from Web2.0?" at GA that touched on these issues (best viewed with these notes). However, one point that I failed to make explicit that is highlighted in the paper is that "... federal websites themselves use the same open systems for accessing the underlying data as they make available to the public at large..." Yep, Government departments can solve quite a few of their internal IT issues by using Web2.0 technologies. I have been really impressed by Powerhouse Museum's great initiative in this regards as well. Dare I say, GovPIs are coming to town!

[Seb Chan]

Hi Shoaib

Plenty of interesting patterns in amongst the PHM stuff - you can get an Opensearch RSS of our collection but you might also be interested in our 'Synonymiser' which data mines queries made to our site and returns user generated synonyms. Check my papers (esp this and this) over at Archimuse as well.

Back on the Yale Law paper - the best response I've seen to it yet comes from Jason Ryan in NZ. I'll be blogging about it later today if i get the time.

Seb

[Shoaib Burq]

I agree but traditionally governments have been too conservative to try, as you say less obvious uses of data. They tend to follow a mandate handed down by parliament and do no more. However, if their IT staff organize their data along the lines of best practices suggested in the paper they solve their internal data management issues. As a side effect (if they make data free for use) they also spinoff non-traditional uses of data. For example I can query the power house museums flickr stream to see where the visitors to the photos are located by drilling into the visitor's profiles. I could even generate a heat-map. What sort of exciting patterns or interpretation have you found internally - can u talk about them?

[Seb Chan]

Hi Shoaib

Whilst the Robinson et al paper is useful my feeling is that its 'minimalist government'/'leave it to the market' line of thinking grossly undersells the potential of open government data.

If government IT is reduced to 'just' supplying data, and leaving combination and interpretation to the private sector, then many of the exciting combinatory value adds we are discovering with our own work won't be realised - simply because of the lack of a market drivers - especially outside of the 'obvious' areas.