Department for Transport - Our Open Data/Transparency Story
Following the Coalition Government’s formation in May 2010 there was a seeming blitz of policy initiatives coming from the Cabinet Office, Transparency, Open Data, Empowering Consumers, MiData………most of which seemed to come in my direction, it was time to draw breath and assess how best to tackle these various demands. On reflection, although there was a real sense of urgency and pace; in reality this was part of an on-going process. When I ran National Rail Enquiries, it was the most phoned number in Europe (people loved transport information), in setting up Transport Direct we needed to acquire data and the permission to re-use data from at least 100 data providers (sharing data) and prior to 2010 we had already released the National Public Transport Access Nodes dataset (NaPTAN) as Open Data and developers loved it!
But now the net was to be cast wider (all transport data was deemed in scope), the release process needed to be quicker (PM letters on summer 2010 and 2011 demanded a delivery schedule) and data quality and timeliness needed to be improved. So it was clearly going to be a challenge, data owners had spent the last ten years jealously guarding their data as a much-loved asset (advertising on websites and value of on-line data were predicted to be major money generators in the early 2000’s) and we would need to convince the transport industry of the sense and business value of joining the Open Data revolution, rather than just relying on the excellence of their own information services (Transport London, National Rail Enquiries etc).
Steve Gooding, my DG, was very clear, we will do it and we will do it very well! So, we set up a Transport Transparency Board to oversee the process, to get input from the Open Data community, the centre of Government and to give a forum for talking to the data owners about the path to data release. We released data owned by DfT, including our expenditure, contracts, organisational arrangements, accident data, traffic counts and from my own area national cycle routes and car park data. From our agencies we released data about MoT tests on cars, driving tests, vehicle registrations and driver metrics. But the Holy Grail for developers has always been the data about “real” transport, so at the end of last year Highways Agency released huge swathes of planned and real-time road and traffic data and National Rail published weekly rail timetables and in the current year Traveline has published a weekly national dataset of bus timetables from April and in June Network Rail made a massive set of data available including real-time train running across its entire network!
And the developers love it, for example on the first day of their rail real-time Open Data adventure Network Rail had well in excess of 100 developers register for access to the data feeds.
So we are well into our journey (pun intended) with much success but with much still to do; we need to decide how to deal with the definition of networks, decide where to draw the boundary with commercial data, spread further into areas such as aviation and be careful to maintain privacy. The enhanced provisions of the Civil Aviation Bill and the results of the rail Fares and Ticketing consultation will both need to be taken into account. I am also keen to develop a real dialogue with the Open Data Community so that we can understand what they like and don’t like about the data already released, what their next priorities would be and issues around formats etc, but most importantly to see what they have done with the data so that we can start to tackle some of those “Value for Money” and “Benefits” issues.
It is not quite yet business as usual, but the momentum feels pretty powerful and increasingly our stakeholders across the transport industry really are seeing the benefits of opening up.
Cheers
Nick
Nick Illsley | Chief Executive
Transport Direct Team
Department for Transport
3 comments
Comment by Owen Boswarva posted on
Nick, thanks for this post. I've been a fairly harsh critic of this Government's progress on opening up key reference datasets. However DfT has stood out as an exception. Well done.
One thought: it's important to recognise that developers aren't the only constituency for open data. Transport data does particularly engage the interest of developers of course, because as you have highlighted so much of it is useful in 'real time' and we need good APIs etc. for that. However, much of the economic benefit of open public data comes from direct analytic re-use, in support of business insights and better decision-making. I've found DfT's Accessibility Destination Datasets to be a useful addition to my toolkit for sociodemographic and geographic risk analysis, for example.
-- Owen Boswarva, 29/08/2012
Comment by exstat posted on
From what I can see, transport information is one of the great success stories of Open Data so far, and certainly more of interest to me than interminable files listing expenditure of over £25K, useful though these may be to somebody.
My only comment is that it appears that you have to register as a developer to get access to some of these data files (e g Traveline). Now my idea of open data is that (a) it is available to all and (b) what the user does with it subsequently is up to them, not the originator. So why do we have to register at all?
I can see there may be a resource issue if the potential application is accessing the data source continuously, so at that stage there may be a need to register, but if all someone wants to do is download the data once in a while, what's the problem?
As a data custodian in the not too distant past, I can see that it would have been sueful to me to know who the users were and what they were doing with the data. But statisticians had to move away from that the moment that placing resources on the internet came to pass, and the world did not come to an end. Why do the owners of these resources take what seems to be a somewhat paternalistic approach on this?
But I don't want to strike too negative a note on what seems a very positive overall story. I'm fairly amazed at the progress that has been made so quickly in this field, and at some of the truly inventive apps that have made use of the data.
Comment by Owen Boswarva posted on
I think part of the problem is public bodies tend to have a "typical" type of re-use in mind for their data, usually gleaned from early engagement with developers (and ideas for "apps" in particular).
So for real-time travel information for example the focus is on delivery through a feed or API. It's reasonable to expect developers to register for that service, because there are infrastructure demands and technical standards that the provider has to manage.
What gets missed is other types of re-use, which might only require a "snapshot" of the data every so often. That version of the data could simply be packaged for download and should not require registration.
It's important that open data providers don't prejudge the type of re-use too much, because we should be trying to encourage innovation with the data -- and the shape of innovation is by its very nature difficult to anticipate.
-- Owen Boswarva, 31/08/2012