Accountability Hack 2015 #AccHack15

Last years Accountability Hack was so successful that we decided to do it again over the weekend of 21-22 November at the NAO.

The event is a joint venture between the Office of National Statistics (ONS), the National Audit Office (NAO) and Parliament  – it is also part of Parliament Week.

If you are interested as attending as a developer over the whole weekend; or just coming to the show and tell on the Sunday afternoon you can sign up on Eventbrite.

On the afternoon of 10 November there will be a warmup event in Portcullis House (opposite Parliament) in the Grimond Room.

There will be more updates as more details are finalised.

[schema type=”event” evtype=”UserInteraction” url=”” name=”Accountability Hack” description=”A software development event – part of Parliament Week – organised by the Office of National Statistics; the National Audit Office and the UK Parliament” sdate=”2015-11-21″ stime=”09:30 am” edate=”2015-11-22″ street=”The National Audit Office, I57-197 Buckingham Palace Road, Victoria, ” city=”London” postalcode=”E15 3DZ” country=”GB” ]

Is ‘digital transformation’ similar to the industrial revolution?

I apologise if you are expecting some great insights as my point is rather modest.

When I studied the Industrial revolution for A level History one thing stuck in my mind. Earlier books on the subject had titles that confidently stated ‘The Industrial revolution 1785-1815’ or similar. The period was clearly defined.  As were the people and technology – Arkwright and the spinning Jenny; Watt and the steam engine; Stevenson and the Rocket. Legitimate legends one and all. The changes were – automation, mechanisation, lots of digging and drilling; mills and factories – the decline of the hand-loom weavers.

However the more I read the more I noticed people saying, ‘well maybe the time period is a bit restricted?’ ‘I can see some earlier industrialisation in the early 1700s’. By the time I finished reading the ‘revolution’ was being pushed back much further back to when the Cornish tin mines existed. Equally the end date started getting pushed forward and the depth and width of the changes to society were questioned. In fact was there really a revolution at all?

So is there a comparison with ‘digital transformation’. I am using the inverted commas deliberately. What is digital and what is transformation?

Is/was Mike Bracken James Watt? Are developers being industrialised for the public sector? Are they previously independent ‘craftsmen’ in a cottage industry made cunningly to feel at home producing industrial code in centres of ‘digital excellence’? Is flexible/remote working and shiny devices the modern digital equivalent of factory schools – a way for ‘benevolent employers’ to tie workers to an industrial routine?

I have a distinct feeling that in decades someone will look back and say – ‘how many consultants made fortunes peddling the snake oil of digital transformation‘? On the side of the bottle it said it would cure – high expenditure; demotivated staff; fix legacy systems; guaranteed to give you satisfied users. Equally how many authors will have made an easy buck  with ‘Digital transformation for dummies’; ‘Lean digital transformation’; ‘The digital waterfall transformed’.  Or conference organisers coin cash from events called ‘How digital transformed my X organisation’; ‘What have we learnt from digital transformation?’; ‘Digital transformation from the users’ perspective’. (These are all made up by the way).

You get the point.

As I said a while ago the similarity is probably that the Industrial Revolution turned out to be difficult to define, describe or date – I expect something similar to be found with ‘digital transformation’.


ODUG is dead, long live ODUG (Open Data User Group)

I heard about the Open Data User Group a couple of years ago and last year was fortunate enough to be chosen (in a transparent manner) to be a member.

The thing that impressed me at the first meeting was the range of expertese about open data in the room both from the public and private sector. It was the imposter syndrome writ large. They also turned out to be a great group of people and the chair Heather Savory was an inspirational leader. All the members did any work on their own time; and Heather worked full time in a role that was in fact part time. So there was a lot of personal committment all round.

There were ups and downs plus a huge amount of frustration about the obstruction and obscurantism across the public sector to making data publicly available. There was a lot of ‘really?’ ‘they said that?’ ‘how long?’ ‘in what format?’ ‘but we raised this x time ago surely they have done more than that?’.

However like a mayfly ODUG’s short three year life span has expired. As a result there is no significant group representing #opendata users in the UK. This cannot be right at a time when more and more conversations come back to data, openess, quality, economic growth etc.

You can help though, yes you can. We have written two notes: one highlights some of the outstanding issues for data in the public sector; the other lists our lessons learnt.

These documents are worth both reading and sharing to keep up the pressure for a group to succeed ODUG. We are not precious, any group is better than none, though preferably with some sharp and pointy teeth.


Who can we trust with #opendata in the UK?

The opendata landscape is complicated as Simon Briscoe has pointed out in his excellent map.

However recently some of the players have disappeared as Giuseppe Solazzo explains and these are the groups like the Open Data User Group that in some way represented ‘data users/consumers’.

Some might argue that all will be fine now that Mike Bracken is the Chief Data Officer. However Mike is a busy man in effect with two jobs and presumably he is more likely to concentrate on projects where there is overlap between his two roles? Is it really likely that he will benevolently take a wider view across the whole of the open data landscape and drive wider change?

So who can we trust? Who is going to represent data creators, consumers and curators?

There are plenty of interested parties around such as GDS, ONS, ODI and the Cabinet Office but they clearly have their own agendas.

So I ask again who is going to represent the wider open data community?



What happened at the first Public Sector API meetup?

On 30 April we held our first Public Sector API meetup at the National Audit Office.  We had about twenty people turn up from a wide variety of backgrounds.

A few people have asked what happened at the meetup? So I will highlight a few points here though I did miss a fair amount as I was letting people into the building.

We did discuss in passing how to keep in touch and share ideas so I have set up a Google group. There is already a discussion taking place about tools and techniques.

In addition we also thought it was worthwhile trying to crowdsource what public sector APIs exist. So we have started a list in a Github wiki. Feel free to chip in with any more public sector API that you are aware of.

We are hoping to link across to the wiki from the page on Government Digital Service Design Manual pages about API.

These are pretty positive outcomes so far but perhaps the biggest one it that there is a group of people with a common interest in APIs who are finding it a bit easier to find and talk to each other.

Oh and we did go to the pub…

Is there a problem with the sponsorship of public sector events?

We ran some user testing of the GovCamp website last week. One of the topics that came up was the size and positioning of the sponsors logos.

Actually the comments did more reflect on the difficulties we had last year adding the logos to the site rather than any bigger issues.

However it did spark a discussion about sponsorship in general some of which have been raised in the past.

So here are some thoughts/questions which feedback is welcome. They are definitely not directed against any particular sponsor.

As far as I know at least the first two GovCamps did not have any sponsorship. Jeremy Gould kindly held them in the basement of MoJ – I have no idea who paid for the food at the time – maybe even Jeremy himself?

As the event grew a bigger space was needed which leads onto issues such as needing to pay for a venue; then security, then food; then sponsors..who need visibility so then t-shirts are needed to display logos and money is needed to pay for the t-shirts. You get my drift.

Is there a thing such as a ‘good’ sponsor who the public sector should accept money from and who it should refuse. So it is it a smallish startup ran by someone well known in government digital is that alright? What if the company offering is a big SI who might not have such a great public profile? Does it make a difference where the money comes from?

Are the people planning to attend influenced by which sponsor logos they see on a website? Should the organisers of public sector events draw some kind of moral line and say I cannot work with x kind of sponsor? What if their boss tells them not to?

Should there be a percentage of sponsorship which one organisation should not be able to exceed? Should sponsors serve ‘fixed terms’ so that there is not an over reliance on particular companies?

What kind of information should be provided to sponors about attendees? Their name, job title, interests, contact details? If someone is attending on their own time at the weekend are they even to be linked to their weekday job title?

During the event is there a way that sponsors should behave. Sit in a corner and ignore the event and try to pitch to likley business or attend all the workshops? How much visibility should be given to the sponsors logos – splash them all over the event with many stands or pop-ups; or is something more discrete sensible?

So there are a few questions. What do you think?

Why do we need a public sector API group?

Somewhat rashly last week I decided to create a Public Sector API Meetup Group.

Why, I hear you ask?

The idea has been strongly influenced by Paul Hallett who set up the #LondonAPI meetup group where I have attended, I think, every meeting. They have been great meetings and much credit is due to Paul for lining up some great speakers who have been very willing to share so many excellent tips.

Why do we need a public sector group then?

Partly because, as far as I can see, most of the people who go to Paul’s meetup are from the private sector. Is that a bad thing? No. In fact for me this has been one of the strengths – seeing the commercial drive behind APIs and the rigour around making businesses work profitably around APIs. There is lot the public sector can learn from this approach.

I am hoping that because I know mainly people in the public sector we can start to replicate what Paul has done who knows maybe we only need one group in the long run?

In the meantime these are the issues I think we can look at as a group.


Are we working to the same standards in the public sector?

Should we be?

It is clear that there are divergent thoughts on this topic amongst developers but presumably in the public sector we have some responsibility to think about interoperability.

Do standards generate certification much like the Open Data Institutes #opendata certificates?


I keep expecting to find a directory of public sector APIs but have not come across one yet. Maybe I have been looking in the wrong place? If so please point me in the right direction. Programmable web have a great database of APIs and related content and there are some listed from the UK.

Do we need our own directory? Should we add our content to theirs? Do we need a standalone directory for the public sector. What kind of links should there be with


Perhaps the biggest thing for me, inspired by Jeremy Gould, is to bring together people with a common interest around a topic. This is always the most powerful of pushing things along and engaging with people.


A great point raised by Peter Wells @peterkwells

How does anyone know that our APIs exist? If there related to public services surely we have a responsibility to make sure the public know they exist? How do we make this easy and intuitive?

The (public sector) API economy

Of course what we should be aiming at is to contribute to the API economy and build services around freely available open data. This great article Wholesale Government: Open Data and APIs which is equally applicable to the UK has plenty of great suggestions.


Does any of that make sense? If so come along to the first meetup on 30 April at the NAO – just because I could easily find a room.


How much does the public sector spend on data?

I gave a short internal presentation this week about open data. As I was writing a few thoughts came to mind based on chats with data users.

One of the intriguing questions is – how does the public sector spend on data? This could be data being bought from other public sector bodies or the private sector.

For example the BBC buy weather data from the Met Office. How much do bodies pay in licence fees to Ordnance Survey every year? Someone suggested that it might be several million pounds a year. Of course the Trading Funds have a financing model to follow and are expected to generate income.

What about private sector data? Apparently it is quite common to pay for Dun numbers (Dun and Bradstreet) in government. How much does this cost each year? Do we know how much is going to be paid to the Royal Mail for postcodes address information?

Has anyone ever investigated this topic and done a map looking at the flow of data/money in and out of the public sector?

Just curious and whether there is something here about the wider benefits of open data to the public sector.

Thoughts and comments more than welcome.

Why I support National Hack the Government

Slightly by accident I went to the first National Hack the Government in the Guardian building seven years ago. As I did not know what to expect it was with some trepidation that I went up the Guardian’s escalator.

Of course I had a brilliant time both because the developers like to have fun and the things they built in a weekend were an eye opener as to the art of the possible.

Since then I have always tried to attend unless I have been out of the country.

Again slightly by accident I ended up hosting the London part of the event last year. I saw this as a bit of a milestone for the NAO as it was the first time, as far as I know, that the building was fully open for a weekend event. Not just any event but something completely outside the normal experience of a supreme audit institution.  Many colleagues across the Office helped make the event a success.

But why support the event in such a concrete way? Well for one it meant that we pushed our technical capacity further and tested, stretched and improved our wifi for future events.

We got to know software developers and continue to stay in touch with them which helped us to run an internal data hack. This then led on to our joint Accountability Hack with Parliament and the Office of National Statistics – and of course hosting the London part of the event this year.

National Hack the Government is about encouraging innovation, doing things differently, being willing to learn from others and share knowledge. That pretty well sums up the NAO itself.





What should the UK’s Chief Data Officer do in their first 100 days?

Ok I am making an assumption that there is going to be a CDO but bear with me on that.

Will a CDO be the saviour of data in the UK?

Here is a starter for ten list:

  1. Work out who is responsible for data publishing and transparency – GDS (The Government Digital Service) or the Transformation Team in the Cabinet Office?
  2. Do something with – a great idea but a bit like a marathon runner it seems to have hit the wall.
  3. Please sort out public sector data publishing – are central and local government working to the same standards?
  4. Give some oomph to the National Information Infrastructure, even a new name. Data is the life blood of the economy it should be sexy not boring.
  5. Mandate data registers for every public sector body as will now be the case in the US (and why not throw in the private sector as well why you are at it.)
  6. Data quality is appalling and a perennial issue – it should be everybody’s responsibility not just the person who pushes the publish button.
  7. Open, open, open – yes with the relevant caveats around privacy and security everything should be open.
  8. Education – so we are teaching coding to kids where is data in all this? Most organisations cannot function without data and data management but this is not entirely clear when you talk to public sector bodies. So education, education, education – about data.
  9. Make sure the UK stays at the top of the table for open data – but by using better metrics than the current lightweight assessments.
  10. Create a network of Chief Data Officers.

So that’s ten points written in ten minutes – what else is missing?


Create, curate, consume