Policy and technology: a "longue durée" view

Random thoughts on policy for technology and on technology for policy


May 2011

Digital Assembly tweets follow a power law distribution

I’ve been monitoring the hashtags for the Digital Agenda Assembly.

You can see them here.

If you plot the data, you obtain a power law distribution. Interesting.

How can you design a policy-making approach that takes into account this uneven distribution of attention and makes the most out of it?



Gov20 for joined up government: collecting examples

I am trying to catalogue examples of Gov20 approaches for joined up government (managing programmes across organisational silos). Clearly sharepoint is not making the trick.

Can you help me listing examples?

– intellipedia from 14 US intelligence agencies to one single collaboration space (maybe one step too far, very specialistic)

– enterprise 2.0 solutions like blogs, social bookmarking (not very impactful in my view)

I am particularly interested in tag-based solution that allow many-to-many unpredictable collaboration/networks to emerge. Also would be nice to have application of social network analysis.

As usual, this is more a working note to myself, but ideas are welcome. I am using the delicious tag joinedup20, in case you want to help.

Two public datasets I really want to be made open

Many of the Open Data initiatives appearing all over Europe ask citizens about what datasets they would like to be made public.

In my work as a consultant, I would like to have 2 datasets which would highly benefit the effectiveness of my work.

First, each survey done for government should release anonymised data for free. The EC, as many governments, carries out plenty of policy related studies. Why cant we have, as a default, the raw data from these survey available? I am not referring to official statistics, but to ad hoc surveys. For example, can we have the datasets from this study on innovative SMEs in Europe?

Secondly, I would benefit from web statistics about government tenders/funding. It would be very simple to publish, beside every call for tender, data on how many the tender/funding information has been downloaded. This would give some information about the potential number of competitors, and thereby orient the decision on which tender to invest.

These are useful datasets for me, but also for any consultancy to have. They would improve the overall effectiveness of the system.

Also, they demonstrate that there can be no rigid definition of Public Sector Information, as datasets are continuously generated from unexpected sources (such as web access data).

Blog at

Up ↑