Guest Blogs

Determining the economic indicators that could be used to measure the economic activity from the re-use of public sector information is but one part of the process of preparing to undertake regular economic measurements. To undertake a measurement one needs to identify the value chain that is the subject of the intended measurement. The economic studies undertaken over the past decade have adopted the cascade approach as a way of establishing the value chain that may exist. An alternative approach is that which the European Commission intends to adopt when the Commission initiates regular economic measurements at three monthly intervals commencing in the September to October 2010 timeframe.

The approach the European Commission has adopted is to use a sampling technique whereby a subset of any particular value-chain is pre-established prior to commissioning an organisation (via an open competitive competition) to undertake regular periodic economic measurements. The European Commission led PSI Group has formed a number of economic indicator subgroups. One of the tasks these subgroups have been set is to determine the sample value-chain that relates specifically to an information sector. That is to identify the various bodies on both the supply (public sector) and the demand (the re-users) sides of the value-chain. Once the bodies in the value-chain have been identified the next task is to seek approval from each body in the chain that they are willing to participate in the periodic measurements. That is the organisation will provide information when the economic measurement is undertaken. The information to be provided will be for a small number of indicators chosen by the Commission but based on recommendations from the PSI Group.

Determining the entire length of the value chain requires an appreciation of the type of information involved and the information flows between organisations within the value chain. For example Business Registers (Company Registration information). It is for this reason that the Commission has established five subgroups each of which is handling one thematic category of public sector information. As a result the membership of each subgroup comprises of stakeholders that have knowledge pertaining to the particular information sector that the subgroup is responsible for.

One’s first thoughts when constructing the value-chain maybe that the main difficulty is determining the demand side of the value chain. In particular the organisations adding value after the first organisation down stream of the public sector information holder that added value. This is based on the premise that the value chain starts from one public sector body. This is the simplest approach to developing the value chain but in reality the value chain even on the supply side maybe far more complex. On the demand side the value chain also includes service providers that provide services to the bodies that are adding value to the information as the information passes along the value-chain. For example legal services, language translation services.

The number of public sector information holders is numerically large and these public sector bodies are distributed across different levels of government from the local level right though to the Member State level. In some European Member States the public sector information may flow across four or more levels of government. One consequence of this is that depending on the category of information there maybe within the public sector a value-chain that exists as each public sector body undertakes its assigned public task. In the process each public body in the value-chain adds value by adding further data to that passed to it by an upstream public sector body. From the public sector information re-use perspective the data can be obtained from each of the public bodies within the value-chain on the public sector (supply) side of the overall value chain. A further complexity is that each European Member State has a different organisational structure. As a consequence this will also likely to impact the value-chain. This then raises the question as to how many value-chains to measure in order to take into account these Member State differences.

One of the value-chains to be constructed by the PSI Group subgroups is that related to Address Information. In this context the work of the European Union eContentplus funded project European Address Infrastructure (EURADIN) is of interest. The EURADIN report titled: Business Model: Social and Economic Benefits, considers 5 value-chains related to address information:

1. The Navigation and Location Based Services value-chain (Section 7.4.1, page 34)
2. The E-Commerce value-chain (Section 7.5.1, page 46)
3. The Geomarketing value-chain (Section 7.6.1, page 49)
4. The Delivery Logistic value-chain (Section 7.7.1, page 51)
5. The Telecommunications value-chain (Section 7.8.1, page 53)

The report indicates that even within an information sector various value-chains are likely to exist.

In considering the value-chain related to the re-use of public sector information it becomes apparent that this is a complex area to understand and measure. It is not surprising that the techniques for measuring economic activity with respect to the re-use of public sector information are still evolving.

Views on the above and other questions and concerns related to measurement of PSI re-use economic activity would be very welcome.

If your organisation would like to participate in the regular measurement then please contact the European Commission's facilitator for this activity by email. The resources expended by participating once every three months are considered to be minimal.

Over the past decade a number of economic studies related to public sector information have been undertaken. Some studies have adopted a top down approach, others a bottom up approach and yet others a thematic approach. A number of the studies have also included comparisons with Countries outside of Europe. The European Union (EU) by its nature embraces 27 Member States and such comparisons are inherent when assessing the economic activity at the pan European level from the re-use of public sector information.

The information published by Eurostat demonstrates that the economies of the Member States within the Union vary. Other variations exist across Member States such as for example:

• The structure of the public sector;
• The public task of public sector information holders;
• Whether the information is held centrally by one public sector body or is held across a range of public sector bodies;
• The degree to which the public sector information is held electronically.

The question then arises as to whether these variations have an impact on the value added chain related to the re-use of public sector information? If the answer to this question is YES then is there a need to take this into consideration when establishing economic measurement indicators?

An interesting example of comparing value chains has been published in the International Journal of Spatial Data Infrastructure Research under the heading Value of Geographic Information in November 2009. Assessing Geographic Information Enhancement, by Bastiaan van Loenen and Jaap Zevenbergen of Delft University of Technology in the Netherlands. The paper develops a methodology for considering value adding within a value chain. The objective is to improve the understanding of the conclusions reached from economic studies undertaken on the re-use of public sector information especially when the studies have made comparisons across different jurisdictions. The methodology developed is then applied through a number of case studies that compare the value chain in three EU Member States with that in the USA.

One of the cases studies considered in the paper was that of the value chain related to Cadastral parcels the results of which are shown diagrammatically in Figures 7, 8 and 9 in the paper. The case studies show the difference between European Member States value chains with those of the value chains in the USA. In Europe the public sector Cadastral Institutes have undertaken many of the value added processes before the data becomes available for re-use where as in the USA these value added processes are provided by the private sector. In other words the data becomes available for re-use far earlier than it does in Europe. As a result one would then expect to see higher value adding processors in the value chain in Europe as the lower value adding processes have been undertaken by the public sector itself. The paper also refers to European Union activities such as the European Union Land Information Service (EULIS), which potentially enables a value chain comparison to be made between the Federal level of the USA with the European Union level.

The results of the Delft University research paper would indicate that there is a difference in the public task of each of the public sector bodies included in the case study. As a consequence this has an impact on the analysis of the results of economic measurements especially when making comparisons across different jurisdictions. The study would suggest that there maybe a need when determining the economic measurement indicators to take such variations into account. The variations may have different impacts for each thematic category of public sector information.

Views on the above and other questions and concerns related to measurement of PSI re-use economic activity would be very welcome.

If you would like to participate in round table public discussions on the PSI re-use value chains at the public meeting that will be held on the 8^th June 2010 in Madrid then please contact the meeting co-ordinators (Ministerio de Industria, Turismo y Comercio, Spain (MITC) or ePSIplatform).

The PIRA report published almost ten years ago (October 2000) explained a number of the issues faced when undertaking the economic study that involved measuring the economic activity from the re-use of public sector information. The report then went onto explain that in the absence of hard financial information economic studies fall back to estimation techniques and that this was not unusual and cited the national statistical agencies as an example.

One of the measures that the PIRA report used was that of the transaction cost related to obtaining the public sector information. The elements of the transaction cost may include amongst others:

• The time spent finding the information;
• The time spent on the Internet;
• The time spent determining whether the information met the re-users needs;
• If the information was that sought then the time establishing the conditions under which the information maybe re-used; and
• If the information was not free of financial charge then what was the price of the information?

The European Directive that came into effect in all Member States of the Union on the 1^st July 2005 was structured to address many of these issues such that the time would be minimised. The MEPSIR economic study tested in part the compliance with the PSI re-use framework using a similar sample of public sector bodies in each Member State at the time the Directive came into force in Member States back in 2005.

If the PSI re-use framework has been implemented as intended and this is coupled with the increasing amount of public sector information held electronically together with the advances made in technology and its deployment then one would expect to observe an increase in economic activity three years later! Regrettably the MICUS economic study undertaken in 2008 did not show that there had been as yet any significant change.

An interesting example of measuring the transaction costs in 2009 has been published in the International Journal of Spatial Data Infrastructure Research under the heading Value of Geographic Information in December 2009. Reference research paper titled: Methodology for Measuring the Demand Geoinformation transaction Costs: based on experiment in Berlin, Vienna and Zurich by Alenka Krek of the University of Hamburg. The paper details how the transaction cost is measured and undertakes three similar transactions in Berlin, Vienna and Zurich. The paper has published the three transaction costs together with a breakdown of the components of the cost. The total transactions costs have been reported as:

• Berlin costs €1,195.52
• Vienna costs €657.97
• Zurich costs €29,831.31

Although the research paper makes no reference to the EU PSI re-use framework as set down by Directive 2003/98/EC one would have expected that compliance with the Directive would have in part reduced these costs, which remain high.

Perhaps the reason why the studies undertaken to date have not shown any material change is in part due to the transaction costs as cited above. As a result only large companies maybe have the financial strength to re-use large amounts of public sector information.

From the studies undertaken to date and the research paper it is apparent that measuring the transaction cost is time consuming and as such may not be suitable when undertaking regular measurements of economic activity arising from the re-use of public sector information as now envisaged by the European Commission.

If one wishes to regularly measure the economic activity related to the re-use of public sector information there is a need to identify a set of economic indicators that can be utilised when undertaking economic measurements. So what indicators could be used?

One approach could be to consider the economic indicators that have been used in economic studies that have been already undertaken and published at the pan European and Member State levels. The task would involve reviewing each study in turn to record the economic indicators used and then to collate the individual set of indicators identified to establish a combined list of the economic indicators which in turn would provide a frequency of use. The process could then be repeated by reviewing further economic studies that cover specific jurisdictions as well as economic studies within particular public sector information domains. The process could be repeated until it is clear that there is a set of economic indicators that are regularly used irrespective of the jurisdiction or the thematic information sector measured. As a result of the process it becomes self evident after several iterations that to continue the process of reviewing economic studies is not materially changing the set of indicators that have already been identified.

If one considers that the re-use of public sector information is a new developing economic activity then it is possible that the indicators used in past economic studies may not be applicable today. As such the set of indicators derived from reviewing past studies may not be useful for measuring the economic activity today or over the short term horizon of 18 to 24 months. Why would this be the case? There may be a range of reasons for this that will include amongst others for example the:

• Changes brought about from the implementation and compliance with the European Union PSI re-use framework;
• Impact of other European Union Directives and frameworks that are applicable to public sector information. For example the INSPIRE Directive;
• Impact of technological advances; and
• Quantity of public sector information that is held electronically.
• With this in mind the potential list of economic indicators would need to be assessed and tested in order to establish whether each economic indicator is suitable in today’s environment and for regular measurement of economic activity arising from the re-use of public sector information.

The process described above is that which the European Commission PSI Group’s Economic indicator work group has adopted. The first two iterations of reviewing past published economic studies and research papers in one particular thematic sector have been completed. The first iteration reviewed six economic studies – three at the pan European level and three at Member State level. The second iteration reviewed nine research papers from the geographic information sector. Further iterations are in process. The process so far has identified over 25 indicators of which only six have been used in three or more of the studies or research papers reviewed. The six indicators identified ranked in the order of use include:

1. Number of staff employed (6 occurrences)
2. The Organisations financial turnover (5 occurrences)
3. The data supply income (4 occurrences)
4. The number of licences (4 occurrences)
5. The total income (3 occurrences)
6. The PSI procurement cost (3 occurrences)

The list already raises a number of questions for which answers are now being sought:

• Are there indictors missing that one would expect to be regularly used in measuring economic activity related to the re-use of PSI?
• How robust are the indicators listed to change?
• How easy will it be to use each indicator in regular measurements?
• Are the indicators applicable to the supply side or the demand side or both in the value-chain?

Views on the above and other questions and concerns related to measurement of PSI re-use economic activity would be very welcome.

If you would like to participate in round table public discussions on the PSI re-use indicators at the public meeting that will be held on the 8^th June 2010 in Madrid then please contact the meeting co-ordinators (Ministerio de Industria, Turismo y Comercio, Spain (MITC) or ePSIplatform).

The European Commission has over the past decade commissioned at least three economic studies that have measured the economic activity from the re-use of public sector information (PSI) within the Europe namely: the PIRA, MEPSIR, and the MICUS studies. Each of these studies have employed to varying extents the cascade methodology supported by a questionnaire. The process slowly builds up the actual value added chain (separate bodies in the value chain that add value to the PSI) for each PSI holder incorporated within the study.

The cascade methodology first asks the PSI holder about their down stream PSI re-use customers. The identified customers are then asked about their down stream customers hence the term cascade. The PSI re-users immediately down stream of the PSI holder are also asked about how cost effective they have found the upstream PSI holder from whom they have obtained the PSI that they re-use. The methodology also involves either face-to-face interviews or telephone interviews to gather information.

Overall the cascade process is expensive and time consuming especially if the study embraces all PSI thematic sectors and 31 Member States of the European Union (EU) and the European Free Trade Area (EFTA). This then raises the question as to whether an alternative method could be deployed that could be run swiftly on a periodic basis – for example at six monthly intervals?

The European Commission put this proposal to the PSI Group working group on PSI indicators during November 2009. To meet the requirement the working group agreed to test the economic measurement of five thematic PSI re-use value chains. Each of the five value added chains would involve a small number of public and private sector bodies across a small number of Member States in order to make the measurements manageable. The results of these tests will be presented at the public meeting that will be held in Madrid on the 8^th June 2010.

The five value chains chosen were:

1. Cadastral Information
2. Case Law information
3. Company information
4. Postcode Information
5. Weather Observation Information

The working groups are now seeking EU Member State based organisations that would be willing to share their information and to participate in the test runs of the economic indicators that could be used in regular measurements of these value chains at six monthly intervals commencing from the summer of 2010.

If you would like to participate then please contact the European Commission's facilitator for this activity by email.

The 1^st July 2010 will be the 5^th anniversary of the implementation of the European Union Directive 2003/98/EC in Member States. (The Directive of the 17^th November 2003 on the re-use of public sector information (PSI)) The recitals in the Directive state that the objective is to stimulate economic activity within the Union and that this is likely to be further stimulated by technological developments over time. (Reference: recitals 3 and 5). The recitals mention growth in jobs (people employed) in the digital content sector and a growth of Small and Medium sized enterprises - two potential economic indicators!

The European Union Committee of the Regions during the February 2010 Plenary Session adopted a global response (The Opinion) to the European Commission’s Communication published in May 2009 that reviewed the implementation of the Directive. The Opinion confirms the objective of the Directive and points out that it is essential to determine a way of measuring the economic value of the information in an objective manner. (Reference: Clause 14) The Opinion has in effect backed the Commission's action plan announced in the May 2009 Communication to undertake further economic studies and regular measurements of economic activity over the period up to the 1^st July 2012.

However the Opinion also pointed out that in order to develop the market in PSI re-use it was important to implement policies that apply licensing and charging models to facilitate and maximise the re-use of PSI (notably through the application of marginal costs) (Reference: Clause 12). The Opinion at the same time welcomes the Directive's objective to minimise the administrative overhead placed on public bodies by the availability of the information. The Opinion thus hints at the way that the PSI re-use framework should be implemented such that these two requirements can both be met. Translating this into practice requires Member State PSI policies and public sector bodies to adopt:

• Non transactional licences recognised globally; (the use of Creative Commons type licensing so no burden in generating, describing and entering into licences with other parties)
• No financial charges (so no burden of collection of the financial charges);
• The information to be available online and electronic; (no prior communication with the public sector body so no burden – a move towards data.gov type services)

The above-mentioned implementation methods at the same time remove the economic indicators that have been used traditionally in economic studies and measurements. There is little point in implementing the minimised overhead framework only to then start to implement economic activity meters to support economic measurement. As a result the determination of the economic activity has become a challenge in itself!

Over the past five years a number of Member States have made good progress towards implementing the minimum overhead framework and this is likely to accelerate over the period running up to the 1^st July 2012.

The Opinion also stresses regarding the principle of a competitive PSI market that it is essential to ensure that private service providers face the same conditions as public institutions, to enable access to public data by private users. (Reference: Clause 13). To comply with this requirement means that Member States must adopt PSI policies that do not permit their public sector information holders to act commercially or if they do that they ensure that the public bodies that do act commercially are physically separated from the public task activity. From the economic measurement perspective a clear separation supports the measurement of economic activity from PSI re-use as the public sector commercial activities can be grouped with other economic actors that are engaged in PSI re-use or benefit indirectly from PSI re-use services. If there is no separation then this gives rise to a further challenge when undertaking measurements of economic activity. Economic studies over the past five years have been faced with a reluctance of both the public and private actors to release financial information related to PSI re-use.

The PSI re-use framework as set down by the Directive supports both commercial and non-commercial re-use. The latter sets yet a further challenge to measuring and estimating the economic value from re-using PSI.

Facing such a challenge then leads to the question: Why bother to undertake such economic measurements? Especially if it is likely to lead to economic estimates that are immediately challenged!

Putting aside human curiosity one reason why economic evidence is required is that those that set and implement policy would like to know whether the policy is meeting the high and lower level objectives or not. Hence the reason that the European Commission supported by the European Parliament and the Committee of the Regions wishes to undertake further economic studies and measurements.

The European Commission during the latter half of 2009 has commenced the preliminary steps that are preparing the way for undertaking these measurements and commissioning economic studies. During November 2009 the Commission through the auspices of the PSI Group has commenced work on these preparations.

To support the Commission preparatory actions to collect further economic evidence a public meeting will be held in Madrid on the 8^th June 2010. The objective of the meeting is to bring together those that have an interest in measuring economic activity to consider economic indicators. The desired outcome from the meeting is to agree a small number of indicators that can be used at regular intervals to determine the trend in economic activity related to PSI re-use. Further more whether the indicators need to vary depending on the thematic domain of the information (or sector) or there can be one set of indictors that can be used across all thematic domains?

If you wish to actively take part in proposing and justifying such indicators at the meeting then please contact the European Public Sector Information Platform meeting co-ordinator as soon as possible as the program for the June meeting is now under formation.

1. What indicators would you use if you were set the task of measuring the economic activity from PSI-re-use?
2. Would the indicators that you have chosen be sustainable overtime? Or pose the question another way could the indicators that you have selected be used at six monthly intervals over the five-year period from 1^st July 2010 to 1^st July 2015 without change?
3. Are the indicators that you have chosen applicable to only one thematic information domain or all information domains?
4. In your opinion has the PSI re-use framework in its first five years stimulated economic activity? If so what is the evidence that supports your opinion?

GeoVation is an Ordnance Survey (OS) initiative that’s funded and managed by OS with input from “external champions”. According to the GeoVation website (www.geovation.org), while GeoVation is founded and currently financially supported by OS, its mission is to: “’encourage and support innovation for social, economic and environmental benefit through the use of all geography’; not just the products and services of Ordnance Survey.” Apparently this decision was the subject of quite a lot of discussion in the OS boardroom. My view is that GeoVation is a more credible initiative if it is not exclusive to OS, so they made the right decision.

GeoVation’s aims are to:

1. promote the use of geography in innovation
2. make innovating with geography easier
3. provide support to new innovations that use geography by connecting people with geographic expertise to those who require it in realizing their objectives.

Membership is open to all. There is a small management committee organizing GeoVation’s activities right now.

In early October GeoVation announced “the GeoVation Awards Programme (GAP)”: a competition to develop new ideas that use geography. Entries opened on the 6th of October 2009 and closed on 4 January 2010. The judges were looking for ideas that demonstrated their innovative potential, sustainability and how they use geography.

The Awards were supported by OS, Ideas in Transit, ESRI, and the Technology Strategy Board, and attracted 150 ideas for 60 ventures. The judges created a short list of 9 finalists, who were all asked to present to and answer questions from the judges and attendees at Tuesday’s event, for the opportunity to win one of four cash prizes, with the money to be used to develop the winning ideas.

Full descriptions of the short-listed projects can be found on the GeoVation blog, here, http://www.geovation.org.uk/blog/. They were:

1. Mapland England & Wales: a massive (13m x 11m) walk-on map of England and Wales with full topographical detail.
2. Online Route cards for hill walkers: A site to provide walkers with tools to help plan appropriate routes, and the ability to have automatic messages sent to a designated emergency contact if the walker doesn’t arrive on time.
3. Mission: Explore London: a project to engage young people with geography through playful and thought-provoking missions.
4. Mapping Britain on Horseback: a British Horse Society initiative that aims to collate all equestrian routes in the UK.
5. Goodfindr iPhone App: using geography to source soft commodities in season.
6. London Blue Plaque: a search facility showcasing the hundreds of Blue Plaques erected over the years to commemorate famous people and places that have helped shape London and the world.
7. MaxiMap: a large-format educational floor map.
8. The Open Climate Data Repository: a service to provide a set of web-based tools to allow the public to view, analyze and feed back on climate data.
9. Open.HistoryMap: A project to map historic features, events and maps against a background of today’s maps or aerial imagery.

The entries ranged from existing, profitable businesses to ideas and vague plans for the creation of sustainable services.

The original plan had been to award four cash prizes: one for £10,000, two for £5,000, and the community choice award, voted by the audience, of £1,000. In the end, the judges decided to split the money differently, awarding £3,000 to London Blue Plague; £7,000 to Mission:Explore; and £10,000 to MaxiMap. MaxiMap also won the community prize.

I was pleased to see MaxiMap take home first prize: it’s run by two women, one of whom was a schoolteacher, who have created a simple and easy-to-use tool to help overcome some of the challenges of teaching children geography. Their presentation was clear and concise, their financials were sound, and they’ve already tested their idea and achieved some success. Well done, MaxiMap! And of course all of the other finalists.

The GeoVation champions also deserve congratulations, not only for creating an interesting event and encouraging innovation: for every person who attended the awards showcase (which was free), GeoVation donated £10 to MapAction. The day raised £750 for MapAction (www.mapaction.org), and chair Steven Feldman encouraged all of us to go home and match that contribution, to help the charity do more on its upcoming trip to Haiti. If you’re not familiar with MapAction, I’d encourage you to take a look at their site, and also consider a donation.

GeoVation plans to put on these awards next year, with the expectation that they will be more independent from OS. It will also be interesting to see where all of the finalists are with their businesses and projects in a year. If you’re interested in GeoVation, or the awards, check out the GeoVation site for information on the finalists’ progress, and upcoming events.

We assembled an excellent panel to discuss Labour and Conservative PSI policies and Locus’s perspective on issues surrounding PSI licensing, pricing and regulation. We also left plenty of time for attendees to ask questions, and raise their own issues for discussion.

Our panelists were:

Francis Maude MP, Conservative Shadow Cabinet Office Minister and Shadow Chancellor of the Duchy of Lancaster. The Minister leads the Party’s Implementation Unit, which is tasked with determining how policies will be implemented should the Conservatives win the General Election;

Professor Nigel Shadbolt, Professor of Artificial Intelligence (AI) and Deputy head (Research) of the School of Electronics and Computer Sciences at the University of Southampton, and the Prime Minister’s Information Advisor; and

Sir Bryan Carsberg, Locus President, and former Director General of the Oftel and the Office of Fair Trading.

I chaired the event.

The presentations were thought-provoking and there was plenty of stimulating discussion. Here are some of the highlights:

Nigel Shadbolt spoke first, going through some of the key elements of the manifesto for government data that he and Tim Berners-Lee have published, (and which I referenced in my last blog; you can find it here: http://www.guardian.co.uk/news/datablog/2010/jan/21/timbernerslee-government-data), and talking about the new data.gov.uk website. Some interesting points:

• “We’re now moving from a web of documents to a web of data”.
• Most of the government PSI that has gone onto the website is in spreadsheet format.
• On release day, the data.gov.uk website had 700,000 hits; the web servers had to be replaced three times to keep up with demand. Within the first 40 minutes of the site going online, users had created three new web applications using government data. The site is currently in Beta: it’s not “complete” and it’s hoped it never will be.
• The site is all open source, done quickly and at low cost. It uses the same platform as Wikipedia.
• In creating the data.gov.uk site Professor Shadbolt and Tim Berners-Lee had to grapple with the issue of what data should be free. Their view is that if creation of the data has been paid for by the taxpayer, then the public should have it. Further, if information is available through a Freedom of Information (FoI) request, why should it not be published as an RSS feed?
• When it comes to determining what data should be made public, the question should be not why, but why not.

Next up was Francis Maude, who opened by stating that “there is a decent chance of continuity” — that the Conservatives would very likely continue the great work that’s been done to release government data already. Other points made:

• The Conservatives recognize that having open data is uncomfortable for government ministers. “Transparency is not always agreeable.”
• We need public money to be spent better. Government spends £4 for every £3 it generates in revenue. That has to change. “We see transparency as a friend of that process.”
• We need to enlist the public to crowd-source consultancy, to be “armchair auditors” to scrutinize government spending.
• It’s understandable how we got to the current situation. OS and the Met Office are required to cover their costs; they then felt empowered to create added value and have squeezed out the private sector. This is not for the benefit of the economy as a whole. There’s in-built conflict in this arrangement.
• If we have the data, we should just put it out there. The Conservatives would release information on the salaries of the 35,000 most senior government employees and all contracts over £25,000 in value; provide an XML feed of council decisions; and publish the COINS database (this is the Combined Online Information System. It contains the Treasury's detailed analysis of departmental spending under thousands of category headings). Some of this will be uncomfortable.
• We need to balance getting the data out there with making everything just right; we would be inclined to trade off some quality in the interests of making data available more quickly.
• We propose to carry forward the work Nigel and Tim have done.

Our final speaker was Sir Bryan Carsberg, Locus’s President, and formerly the Director General of Oftel (Ofcom’s predecessor) and the Office of Fair Trading.

Sir Bryan opened with “Carsberg’s principles of regulation”: The best form of regulation is the promotion of competition. Monopolists often have an interest in postponing uses of new technology. Liberalization has been a tremendous boon in the telecom sector; there would be enormous benefits if we can open up use of information. Other points included:

• Pricing: Locus has always been cautious on price. It’s a no brainer that data financed by tax-payers should be released free; In other cases, we believe data should be provided at marginal cost.
• Locus is mainly seeking fair competition, whether or not the information is free.
• Information should be made available in an easy way: in telecom, we had the CLASS license. That provided ready access to the base infrastructure and the constraint that providers compete fairly. All players had access to the raw materials on the same terms, at the same price, with no-cross-subsidization by the incumbent.
• You could argue that many of the Trading Funds shouldn’t provide value-added services at all.
• No matter what the outcome of the OS consultation, “This will not work without regulation.”
• We can understand the reluctance to set up a regulator: it could be costly. Right now, though, we have OPSI, but it doesn’t have the power. So we’re incurring the cost and not getting the benefit.
• It needn’t be enormously expensive to regulate this sector. If there are big penalties for breaking the rules, that can be very effective.

The panel discussion was followed by a very lively question and answer session.

I asked to panel to comment on Sir Bryan’s assertion that we must have regulation if current PSI holders are to continue to act commercially. Nigel Shadbolt’s response was that quite a lot of time had been spent on the management of data to be released, the technology, and getting it all out under a Creative Commons license; how to regulate is still to be determined.

Francis Maude said he was reluctant to set up a range of new bodies: He feels that the OFT is good at this. But, “we’re open-minded.”

One participant said it would be “terrible if the whole environment changed on the first of April.” We’ve had years of consultation, but we do need to be able to prepare for whatever changes are brought in. We also need a set of principles that we can be sure will survive changes of government .

Other topics of discussion included:

The consultation on the future of Ordnance Survey: A key point: How much Ordnance Survey data should be made available for free, and the potential negative impact on existing commercial providers if, for example, OS maps, rather than raw data, are made freely available. Nigel Shadbolt responded that he and Tim Berners-Lee couldn’t avoid the issue of location and place when looking to open up government data. He felt the OS consultation didn’t need to be “as complex as it is.” He also finds it “odd to pay for information that you need to live your life.” There was general agreement in the room that raw data, such as boundaries, should be released. There is an argument that detailed mapping should not be released for free, but that all players, including OS, should have access to the raw data to create those maps, and should compete on the same terms to provide them.

Nigel said, :”Tim and I were told not to deal with OS. But we knew we had to. We will have to grasp the nettle…. We must decide what are the ‘totemic datasets’”.

How to find the right cost model to make data available for free: Francis Maude expressed relief that those of us who spend time of this issue also find it complex. Sir Bryan referenced the Cambridge Study, and noted that the cost of making basic data sets available is actually small. Some participants, though, disagree with the findings of the Cambridge Study, which assumes price elasticity of demand: it was noted that while there has been a 30% real price fall in MasterMap over the last few years, there has been no increase in use.

This led to a discussion of the whether or not Ramsey pricing (pricing according to the elasticity of demand of different customer groups through price discrimination) is appropriate where data is charged for. Sir Bryan noted that the OFT study concluded that the market could be greatly expanded if data was made available at marginal cost. He could see how Ramsey pricing could distort the market. Others considered it a mixed blessing for the mapping sector. Nigel Shadbolt suggested more work was required from economists.

Postcodes: There was a fair amount of discussion surrounding access to the postcode database. One participant said, “There will be trouble if we don’t release postcodes soon: OS and the Royal Mail have crucial pieces of data. Have we got a plan for dealing with postcodes and addresses?” The Ordnance Survey consultation document is vague on this issue. Nigel Shadbolt suggested that we should determine what the developer community would find most useful. However, if government doesn’t find a way to release postcodes, it’s likely the community will crowd-source them. Other participants made the point that there is a cost of maintaining the postcode address file, and that must be covered somehow.

Our Locus event highlighted the considerable progress that is being made in PSI policy in this country, but also emphasized that there is a long way to go in the next decade to unlock the full potential of the UK PSI marketplace. In particular, government must “grasp the nettle” of freeing up access to the postcode database and Ordnance Survey raw data so all can benefit fully from the PSI that has already been released.

According to the press release, “All of the data is non-personal and in a format that can be reused by any individual or business to create innovative new software tools, such as applications about house prices, local amenities and services, or access to local hospitals.”

Read the full press release, including quotes from Stephen Timms MP, Sir Tim Berners Lee and Professor Nigel Shadbolt, who will be speaking at the Locus event this evening.

And of course the website is here: http://www.data.gov.uk/

Data sets include information on crime, health, education, property prices, travel, the environment, agriculture, etc. It’s time-consuming, but fascinating, to browse through the list to see types of data that are being collected, and are now available for use. The plan is for the site to continue to add data from government departments each month.

In conjunction with the launch of data.gov.uk, the UK’s Office of Public Sector Information (OPSI), part of the National Archives, has drafted a new set of terms and conditions for the site that are “interoperable with any Creative Commons Attribution 3.0 Licence.” The new model will replace the existing Click-Use Licence; OPSI expects to launch the new terms in May 2010.

The Guardian has an interesting article on how data.gov.uk came to be; Professor Shadbolt and Sir Tim Berners-Lee’s manifesto for government data; and an interview with both on the announcement and thinking behind the release of the data. These can be found here:

http://www.guardian.co.uk/technology/2010/jan/21/how-official-data-freed

http://www.guardian.co.uk/news/datablog/2010/jan/21/timbernerslee-government-data

http://www.guardian.co.uk/politics/video/2010/jan/21/uk-national-data-website-launched

And if you’ve never seen Tim Berners-Lee’s TED talk on the subject of public data provision, where he asks the audience to join him in calling, loudly, for “Raw Data Now!”, it can be found here:

http://www.ted.com/talks/tim_berners_lee_on_the_next_web.html

Locus is very pleased by the announcement of the data.gov.uk website, and the release of so many government datasets. Of course, we, like many other parties, still believe that for this data to fulfill its potential, government must also find a way to make Ordnance Survey map data (raw data, in particular) and the postcode database available, without barriers. The data.gov.uk site is, however, a very good first step.

The initiative is part of a campaign to boost the transparency and accountability of City Hall, and will be formally launched at the end of January 2010. Plans are for the Datastore to contain over 200 different sets of information, such as crime rates, abandoned vehicles, schools data, hospital waiting lists and recycling rates. Here’s a link to the Datastore website: http://data.london.gov.uk/; more information can also e obtained by emailing datastore@london.gov.uk.

In a statement, Johnson said, “"I firmly believe that access to information should not just be the preserve of institutions and a limited elite. Data belongs to the people - particularly that held by the public sector - and getting hold of it should not involve a complex routine of jumping through a series of ever decreasing hoops.”

According to the Datastore site, the policy will be “to only withhold data where to release it would infringe privacy legislation or a contractual obligation eg commercial confidentiality.” The license conditions under which data can be used will vary for each data package; however, the site says that “in general the data held in this site can be used for most purposes, provided the Terms and Conditions of this site are not infringed.” The data comes in several different formats; each data package has its own licensing terms.

Locus is very pleased to see this initative coming to fruition, and we’re looking forward to seeing the rest of the datasets that are to be released at the end of the month. We believe that enhancing access to Public Sector Information (PSI) for re-use brings economic benefits as well as improving services to citizens. This move by the Mayor of London sets an excellent example which we hope to see repeated across Government in the coming weeks.

Locus, as some of you will know, was established to encourage the public sector (primarily in the UK) to maintain a trading environment that is fair and equitable, in particular in relation to the licensing and re-use of public sector information. Our members are private sector companies that are committed to working with PSI holders to maintain and develop a vibrant, information-driven UK economy that ultimately works to the benefit of the public sector, private sector and end consumer.

Locus acts as a forum for exchange of information, keeps its members up to date with latest policy developments, and provides advice and guidance. Each year we hold several events that are open to non-members, to raise the level of discussion on a particular topic. In January, we’re hosting what we expect to be a landmark PSI event examining the future of the UK PSI marketplace in light of the upcoming General Election. Right now the Labour Government is pushing forward a number of plans to give away data, and to change the structure of Ordnance Survey. However, if we have a change of Government in May (or before), that would very likely all change, making it difficult for those of us who work with PSI to know what the future will hold.

We’ve therefore put together a top-notch panel to discuss Government and Conservative PSI policy. Our speakers are:

• Francis Maude MP, Conservative Shadow Cabinet Office Minister;
• Professor Nigel Shadbolt, the Prime Minister’s Information Advisor, and
• Sir Bryan Carsberg, Locus President, and former Director General of the OFT and Oftel.

I’ll be chairing the event.

We believe this timely and exciting event will be an excellent opportunity to hear about Government and Conservative PSI policy ahead of the next General Election, and to provide policy-makers with input at a key time for PSI in the UK.

The event will be held on Monday 25^th January, 5.30-7.00pm, at Olswang LLP, 90 High Holborn, London, WC1V 6XX. For further information, or to register for the event, please contact Harriet Crosthwaite, of Luther Pendragon on

Tel : +44 20 7618 9136

or email: harrietcrosthwaite@luther.co.uk

This promises to be a very busy month for PSI in the UK; over the next few weeks I’ll provide blog posts with Locus’s views on the Communities and Local Government Consultation on Ordnance Survey; the London Datastore (announced today); derived data, and other topics as they come along.

Locus thanks the ePSIplatform for the opportunity to contribute our views to the PSI debate.