Have a listen to hear about some of mySociety’s projects, and to learn Tom’s views on the current rush toward greater access to data.

As usual, links to the resources we discuss are available in show notes on the Talis site.

Related articles by Zemanta

• Parliamentary website TheyWorkForYou launches redesign (onlinejournalismblog.com)
• Mapumental: visualise any neighbourhood in the UK by transit times, house prices and “scenicness” (boingboing.net)

Have a listen, and see what you think.

As a rule, they tend to either offer a pretty interface that facilitates end-user exploration of a constrained set of data, or they proudly offer a massive list of files for download, or they publish an API and encourage you to interact via that. Each of these modes of access has a place, but none is going to be appropriate for most uses of most data, most of the time.

As such, it was refreshing to talk last night with Kevin Merritt. Kevin is CEO of a Seattle-based startup called Socrata, and he is convinced of the need to support multiple modes of access and interaction.

Socrata has utility in a number of markets, but the focus of our conversation was in the context of Government data. Have a listen, to see what Kevin thinks can be done to make initiatives such as the US Government’s data.gov site more useful.

Production of this podcast was supported by Talis, and links to items discussed are included in show notes available on the company’s Nodalities blog.

Related articles by Zemanta

• Linked Government Data (w3.org)

Tim Berners-Lee has been amongst those rallying to the cause, and working with Governments here and overseas to realise the opportunities in — first — simply getting data out and — second — ensuring the structure and linkages required if Government data is to form a useful foundation upon which others really can build.

Tim O’Reilly, too, has been pushing his notion of ‘Government as a Platform’ for some time, driving toward next week’s gov2.0 summit in Washington, DC. His arguments reached a broader audience with yesterday’s guest spot on TechCrunch.

O’Reilly points to some of the uses of Government data in his TechCrunch post, and I’m continuing my own efforts to secure podcast interviews with some of the more interesting examples that I come across.

One of the things, I think, that is most interesting about this ‘platform’ that O’Reilly describes is the myriad ways in which it potentially benefits so many different constituencies.

Government itself should certainly become more efficient, making better use of its own information ‘simply’ because it’s so much easier to see what’s there. John Sheridan of the UK Government’s Office of Public Sector Information (OPSI) touched on some of these issues in our recent conversation, and the potential for discovering synergies across the Departmental and Agency divide is surely just beginning to be realised.

‘Activists’ of various kinds will have easier access to information in support of their various causes. Some of this information will, undoubtedly, be used to embarrass the authorities, and much of it will be skewed to present the truth in rather odd ways. Rather than simply leading to more informed activism by a vocal minority, however, David Eaves in Vancouver and Sunlight Labs’ David James both argue that better availability of data will make it easier for everybody to hold Government to account.

Researchers, such as Jim Hendler’s team at Rensselaer, are turning to resources like Data.gov in search of interesting technological problems and large pools of data upon which to test new techniques and ideas. The resulting data exhaust (in Hendler’s case, RDF versions of Data.gov resources) is then available for others to use in further innovation.

And (perhaps) most interestingly of all, the well understood trinity of open source software, commodity hardware and near-ubiquitous connectivity is coming together with increasingly available data to make the public sector information space — for far too long the expensive preserve of the big Consultancy firms and their ilk — interesting to start-ups and innovators.

Sunlight Labs‘ Apps for America2 contest is into its final stage (don’t forget to vote!), and the range of applications received is a clear indication that innovation in and around Government information is both possible and long overdue. Amongst the three finalists, This We Know appeals to my Semantic Web interests because of the particular technological approach they’ve adopted, but there’s plenty to admire in all three.

The test will come down the line, when working with Government data is no longer incentivised by competition glory and prize money, when it’s no longer the hot new source of Big Data for academic exploration, and when the activist arms race levels out at a new plateau of comparable informedness. When that day arrives, will enthusiastic entrepreneurs still be competing to extract additional capability from the Data.gov and Guardian apis, or will all of the hits to these sites emanate from the cubicles of Accenture, IBM and EDS?

Whilst I certainly hope that these mainstream exploiters of public sector information embrace the possibilities, it would be a sad day if independent grass-roots innovation atop the Platform of Government were so short-lived.

As such, I’ll be watching next week’s proceedings in Washington with interest.

Image of the Scottish Parliament by Stéphane Goldstein, shared under Creative Commons license on Flickr.

Continuing my ongoing series of conversations about the increasing availability of Government data, I recently spoke with David James of Sunlight Labs.

Sunlight Labs is part of the non-partisan Sunlight Foundation in Washington, DC, and undertakes a range of projects to shine a light on the dealings of the US Federal Government.

We discuss some of the Labs’ current projects, including their new National Data Catalog, and consider ways in which this and the Federal Government’s Data.gov portal increase the availability and utility of Government data.

Like other conversations in the series, production of this podcast was supported by Talis. Show notes for the conversation are available on Talis’ Nodalities blog.

Related articles by Zemanta

• Thinktank Aims To Crowdsource Government Earmark Analysis (news.slashdot.org)
• How Sunlight Foundation Achieves Government Transparency – Web 2.0 Expo San Francisco, CA – 04.02.09 – Ellen S. Miller (nsrtk.blogspot.com)
• Sunlight Labs Offers $25,000 For Data.gov Apps (developers.slashdot.org)
• Sunlight Foundation pushes ahead with data.gov (cnewmark.com)
• Data.gov Launches to Mixed Reviews (wired.com)
• Data.gov to Face a Challenger From Sunlight Labs (readwriteweb.com)
• Report Federal Pork, Win A Kindle (wired.com)
• Data.gov Needs to be a Community Effort (socrata.com)
• data.gov, Vivek Kundra, Sunlight Foundation get props at NYTimes (cnewmark.com)

As we discuss in this podcast, the programme is ambitious but eminently achievable, and builds upon a tradition that has actually been a lot more open than it may sometimes appear.

Production of this podcast was supported by Talis, and show notes are available on their Nodalities blog.

John describes work at OPSI and elsewhere in the UK Government, and also draws upon his role as co-Chair of the World Wide Web Consortium’s e-Government Interest Group to consider some of the broader issues.

He concludes by issuing an invitation to anyone interested in working with Government to ensure that data is made available to a timetable and in forms that really meet the needs of potential beneficiaries here and overseas.

Related articles by Zemanta

• Councils urged to free e-data (guardian.co.uk)
• Open government data and Digital Britain (opencontentlawyer.com)
• The Socioeconomic Effects of Public Sector Information (zzzoot.blogspot.com)
• Talking with Mark Birbeck about RDFa and its use in Government (cloudofdata.com)

There has been a recent burst of enthusiasm for making raw data produced by and for Government more ‘open,’ and this must surely be welcomed. Long-running grass-roots efforts such as Tom Steinberg’s mySociety and The Guardian‘s Free Our Data campaign continue to innovate, but in an environment that is suddenly more receptive to their ideas. Edge-case adoptions of RDFa and other ‘semantic’ specifications, perhaps, are at last moving from being merely the preserve of a few isolated enthusiasts.

Sir Tim Berners-Lee now walks the corridors of power in London and Washington and elected officials (and even the Opposition parties) at least claim to be listening to his call for ‘Raw Data, Now!’ and his talk of Linked Data, URIs, and the rest. How far we have come, but we have much further still to go.

‘Open’ and ‘Transparent’ Government is nothing new. It’s been talked about for a very long time, and there has been some progress. Part of the issue, I think, comes down to interpretations of ‘open.’ Just because it’s possible to download some Government data doesn’t necessarily mean it’s practical for most interested parties to do so.

If a national library puts all of its catalogue online for free, but requires you to query it via an obscure industry protocol, is that ‘open’ ? If they then throttle access so that it would take an inordinately long period of time to ‘copy’ their catalogue, is that ‘open’ ?

If a National Statistics agency makes all of their research freely available, and provides access to thousands of opaquely named csv files by listing them on a web page, is that ‘open’ ?

If a Government department makes all its research reports available online as Microsoft Word files, is that ‘open’ ?

A purist might strenuously assert that none of these are ‘open.’ Most, certainly, are far from ideal… but they still serve a real purpose in making the innards of Government more accountable. How good should be good enough in 2009?

Going the other way, does a Health Authority have to make my medical records visible to the world before it can be called ‘open’ ? It seems almost unthinkable, but extremes of viewpoint do have an annoying habit of quickly becoming that absurd.

The current enthusiasm for ‘Open’ is closely associated to Tim Berners-Lee’s talk of Linked Data and the newly pragmatic Semantic Web, and Berners-Lee provided a short note last week on his current views. Contrast Tim’s discussion of the ways in which Government data should be linkable with the Sunlight Foundation’s attack on the US Federal Government‘s transparency flagship, Recovery.gov, for not making any real data available in the first place.

If we can’t even get the existing raw data out of Government as often as we’d like, there’s a long way to go before Berners-Lee’s grander vision can be achieved. He recognises this, of course, writing;

“Government data is being put online to increase accountability, contribute valuable information about the world, and to enable government, the country, and the world to function more efficiently. All of these purposes are served by putting the information on the Web as Linked Data. Start with the ‘low-hanging fruit’. Whatever else, the raw data should be made available as soon as possible. Preferably, it should be put up as Linked Data. As a third priority, it should be linked to other sources. As a lower priority, nice user interfaces should be made to it — if interested communities outside government have not already done it.”

(my emphasis)

To get much further, and to make that progress sustainable, there’s a requirement for a very real shift in attitudes at the heart of Government. Openness (of data or anything else) shouldn’t be a tactic to distract from worse news elsewhere, or a short lived knee-jerk response to the latest embarrassment. Rather, it should be a deep-seated presumption to underpin policy, systems design and more.

Data from Government should, quite simply, be freely and easily available. As a matter of course, and without prevarication. Unless there is a compelling reason to do otherwise.

For all the talk of ‘open,’ that is very far from being true today. The presumption is ‘closed.’ The mindset is (largely) ‘closed.’ ‘Open’ has to be fought for, and ‘Open’ has to be justified. ‘Open’ has to be championed, endlessly, tirelessly, thanklessly.

The exact opposite should be true. Then (and maybe only then?) Berners-Lee and his colleagues can build something wonderful.

Related articles by Zemanta

• Open government and Linked Data; now it’s time to draft… (johnbreslin.com)
• Sir Tim Berners-Lee talks to the BBC’s Rory Cellan-Jones about Linked Data (cloudofdata.com)
• A World Where Everybody Has Put their Data on the Web (futurechat.in)
• Happy 20th Birthday, World Wide Web (readwriteweb.com)
• Linked Data in Health Care (ehealth.johnwsharp.com)
• Sir Tim, the web and silos (bbc.co.uk)
• SemTech2009 impressions (ivan-herman.name)
• Linked Data is Blooming: Why You Should Care (readwriteweb.com)
• Reflections on SemTech 2009 (w3.org)
• UK Government Moves to Put Data on the Web (w3.org)
• Tim Berners-Lee: “We need data on the Web to work better together” (semantic-web.at)
• Talis Connected Commons: Linked open data repository opens up shop (mndoci.com)
• Happy Birthday, Web of Documents. Now bring on the Web of Data! (cloudofdata.com)