Free Our Data: the blog

A Guardian Technology campaign for free public access to data about the UK and its citizens


OPSI finds against OS in row over AddressPoint licensing

[Update: if you've come here via the Guardian article on spam, then first, apologies; secondly, the location you actually want to go to is http://noc.net.umd.edu/cgi-bin/netmgr/whoami - and that will tell you what IP address you're on. Again, sorry for the error, which was introduced during editing.]

We’ve seen the OPSI report into Intelligent Addressing’s complaint (148K PDF) about the lack of transparency in the AddressPoint licensing terms.

Essentially, Intelligent Addressing (hereafter IA) says that OS seems to have a different licencing regime internally than externally: that a company outside OS can’t get access to AddressPoint on anything more than one-year contracts, whereas OS can use AddressPoint indefinitely. That puts external organisations like IA, which wants to license AddressPoint in order to compile the National Land and Property Gazeteer (NLPG), at a disadvantage.

And OPSI agreed.

Para 35:

In addition, the Complainant alleges that the lack of transparency extends to the website in that it is not clear how the Framework
Agreement interacts with the specific use contracts. Furthermore, the
Complainant states that there are few published details about the
partner licensing options.

Para 37:

The PSIH states that the website contains significant openly accessible
information, but does not publish the licensing options as the majority of
partners have actively managed accounts, and the PSIH works closely
with them to discuss their tailored licensing solutions.

The PSI Regulations also state that “Where conditions are imposed they
shall not unnecessarily restrict (a) the way in which a document can be re
-used or (b) competition� (Regulation (12) (1)). It appears to OPSI that the
terms of the licence unnecessarily restrict the way in which AddressPoint
can be re-used and unnecessarily restrict competition since such terms
are unnecessarily prescriptive. Some conditions imposed by the PSIH
restrict the way in which AddressPoint can be used and this restricts
competition between the Complainant and the PSIH since it immediately
puts the Complainant’s products which use AddressPoint at a
disadvantage to those of the PSIH. One such condition relates to time, as
the PSIH can grant licences with a longer term to end-users than others
such as the Complainant, who is limited to twelve month end-user
licences as licensees want more security in the length of time they are
able to use the information. This means that the Complainant is at an unfair disadvantage. Another example is cost as the Complainant has
to pay a higher licence fee for the use of AddressPoint. OPSI does not
consider that such restrictions are necessary.

You can read the full complaint and response. But here’s OS’s response, hot off the press:

“On the OPSI report, as Carol Tullo [head of OPSI] was saying last night [at the RSA/Free Our Data debate], it was only published last Thursday and so we still need to discuss it with OPSI in the context of the points we are already exploring from the reverification process you reported on in March.

That said, we do have a number of concerns around matters of fact and law in the report and are considering the options open to us including our right of appeal to the Advisory Panel on Public Sector Information (APPSI). The report purports to make wide-ranging recommendations regarding our licensing policy on the basis of a narrow investigation of a single, and atypical, set of circumstances.

Please be aware also that although they have a right to make a complaint to OPSI, Intelligent Addressing are not a licensee of Ordnance Survey information; they are a contractor to Local Government Information House, who are the licensee. We believe that OPSI’s interpretation of some of the concepts, terminology and clauses within Public Sector Information regulations could make it difficult for us to properly differentiate between the terms and conditions used for products and services licensed directly to end users for their internal business operations, from those applying to partners who generate commercial revenue by exploiting the data in their own products and services (often by adding value to our data); for example, a direct utility customer using the data to manage multi-million pound assets of pipes and cables would have to be licensed in a similar way to a small publisher using the data to produce consumer products retailing at a few pounds.

OPSI also appear to ignore the fact that third party data plays an increasingly important role in the content of our products and services, bringing added value and benefit to users. Royal Mail has a significant IPR interest in the products involved in the complaint. OPSI’s reading of relevant PSI regulation clauses could result in us having to implement pricing and licensing terms and conditions that would undermine the commercial interests of these third parties. This could in turn discourage them from offering their data into collaborative products and inhibit our ability to produce marketable data or to generate adequate revenue for reinvestment.

Our adoption of specific use contracts followed wide consultation with licensed partners from whom it gained broadly-based support. It enables pricing of data to be varied according to end-user perceptions of value for clearly identifiable areas of use. With this mechanism in place it guarantees that competing partners licensing products into the same market have paid similar licensing costs to us. The model assists in creating competitive markets among partners providing similar products and services and also enables users with specific needs to acquire their information at appropriate prices. Introducing a “one size fits all� approach may also be anti-competitive where others have already entered high cost/high value markets, and it may unreasonably inhibit more from doing so.

We are committed to honouring our full obligations as a Trading Fund with delegated authority to protect and exploit Crown Copyright and in so doing we are required to operate on a commercial agenda. We are proud to be accredited under the Information Fair Trader Scheme and are working with OPSI to ensure our licence terms and conditions remain open, transparent and fair.”

26 Responses to “OPSI finds against OS in row over AddressPoint licensing”

  1. Leslie Ramage Says:

    I find OS’s response very sad indeed albeit very predictable. Wouldn’t it useful if instead of an immediate rebuttal or a “consideration of options” , that there was some admission of culpability. A couple of things need to be clarified for the readers of this blog. OS’s partners are simply customers ‘buying’ into ‘partnership’ with OS based on a contract. It is a business scheme. It is not a partnership in the sense of buying into the ethos or culture of the organisation and pledging support. Last evening the Director General noted with some pride that OS now has 500 ‘partners’. These partners are customers, some of whom purchase data only because OS is the only show in town. Also on the issue of specific use contracts, this was a regime implemented to retain the aura of equal licensing terms for all customers…oh excuse me, partners. To say that the partners ’support’ this regime is extraordinarily hypocritical. When OS is the only purveyor of the base data needed to ensure a customer’s business, of course they are going to say yes. Doesn’t this border on monopolistic and anti-competitive practices?

    Which brings me to one of the most incredible acts in restraint of competition of all which most cogently bears out the complaint of Intelligent Addressing. To be honest, I don’t know why this wasn’t reflected in the report which was published last Thursday. http://www.ordnancesurvey.co.uk/oswebsite/business/copyright/docs/D03800.pdf
    In this document, the Director General lists the reasons under which a license for data may be refused. These exceptions are part of her personal commitment to trading fairly which is required to be verified under the OPSI Information Fair Trading Scheme. The exception states: “We may refuse to grant certain applications if you want to market a product whose intended use is the same as, or comparable to, that of any product marketed by Ordnance Survey itself or [most disturbingly] ANY PRODUCT WHICH ORDNANCE SURVEY INTENDS TO MARKET…!!!!!

    If this isn’t blatantly anti-competitive, I don’t know what is.

  2. Dan Macdonald Says:

    OS response to Opsi complaint

    I agree with Leslie. The initial response to any criticism of OS is always for them to dismiss the points made as being untrue, and to claim that any support for the viewpoint is entirely mis-guided as the true facts are not available or understood by those putting forward their views. Well, this time this approach may not wash! OPSI are the body responsible for overseeing the introduction of the Re-Use of Public Sector Information Regulations (PSI Regs) in the UK. They are responsible for crown copyright even where, as in OS’s case, it is delegated to other government agencies to operate. Under their remit they have introduced the Information Fair Trader Scheme (IFTS) designed to ensure that licensing of public sector information is fair and transparent. OPSI ARE the authority on these matters. On the complaint put forward to them OPSI have found against OS under both PSI Regs & IFTS.

    In their critique of the OPSI report, OS state “We are proud to be accredited under the Information Fair Trader Scheme…� You would not think so if you read their immediate and vehement rebuttal of the report OPSI published when they checked OS’s credentials against the requirement of the IFTS rules earlier this year. In essence this review made exactly the same criticisms against OS that have now been found to be true in practice in investigating IA’s complaint. OS’s response was one of the most arrogant dismissals of the powers of another government body – especially a regulatory body – that I have ever seen. It was the verbal equivalent of two fingers.

    OPSI now have a real test of their position on their hands. This is the first formal complaint their relatively young organisation has faced. In addition OPSI is now being subsumed into another government department – the National Archives. At a time of real importance in the debate about public sector information it is crucial that they hold their nerve.

    One final point. At the RSA debate, referred to elsewhere in this blog, one questioner raised the very central point about what is core data and what is value-added data. This is crucial to the debate about the use of public sector information, and presumably the eagerly awaited CUPI review from OFT will clarify this position. The point Vanessa Lawrence seeks to defend is that having Trading Fund status forces OS to adopt the business model and licensing regime they have chosen to adopt. Not so! Other trading funds take a very different view of core data & re-use (a point neatly made by the National Archives representative at the RSA debate). On a scale of Attila the Hun to Greenpeace, OS have chosen to take the most draconian stance available to them. They should not be surprised if one of the users of GI information has chosen to challenge them. IA’s championing of the cause has the tacit support of many of the 500 partners of OS, that Leslie Ramage points out have nowhere else to go.

  3. Stephen Palmer Says:

    I was interested in the OS’s response that ‘The Royal Mail have significant IPR in the AddressPoint’ but they do not recognise the IPR of Local Councils in Britain who actually create every address in the UK

  4. Martin Malliet Says:

    I remember writing in 1998 in a ‘possible response’ to the Green Paper on PSI:

    “Only if an agreement can be reached on clear cut ‘rules’ that can be applied and monitored in a straightforward manner would any EU regulation make sense. There is no need for regulation in the form of policy ‘guidelines’ that need too much policy to be correctly implemented. This probably means that a US-style Paperwork Reduction Act – government sticks to its trade and leaves commercial exploitation of public sector information to private initiative – is the only viable option.”

    When I look at this dispute between Intelligent Addressing and OSGB, I wonder if there will ever be a chance that disputes are actually solved under this PSI directive? Or is it that where we only had disputes before, we now have disputes and bureaucrats ruling on them without any significant differences in the outcomes?

    This PSI directive has established PSI as an infinite resource, most certainly, for all sorts of unintended purposes, it seems to me.

    I mean, it was a VERY bold move on behalf of these EC officials, to try and put all public sector information resources, from local government all the way up to the EC itself, into a sort of commons by means of a simple directive imposing marginal cost dissemination. (In the US I suppose such a move would require a constitutional revolution.) So the directive was watered down to a compromise, a compromise that isn’t very different from what we had before. Or am I wrong?

    Still, I think it would have been a very clever move, had it succeeded. A move that would have created the opportunity for user interests (including public sector user interests), assisted by private sector intermediaries, to gain the upper hand in their struggle with bureaucracy. For one, I don’t think we would have needed INSPIRE, if we had had PSI: the new balance of power would have brought about gradual change towards a desirable outcome without much need for top down coordination.

    Now there is a review data for this directive on PSI re-use: 1 July 2008. It appears that politicians all over Europe are getting nervous because of the Lisbon agenda and all those knowledge based economy mirages not materialising as intended. Maybe one could remind them of PSI and let them have another look?

    Why is re-use by the public sector itself, when it is part of the public task, whether commercial or not, excluded, and what does this exclusion then mean? In the report on IGN France the inspectors used the argument that the directive does NOT apply to the by French law mandated re-use of IGNF data by other government departments to say that IGNF commercial prices also did not apply, defending themselves against the interpretation by IGNF officials that the directive did apply and that therefore IGNF commercial prices could not be contested.

    http://www2.equipement.gouv.fr/rapports/themes_rapports/administration/2005-0034-01.pdf

    In this dispute between Ordnance Survey and Intelligent Addressing, OS argues that because commercialisation of the data is the initial purpose within its public task for which the data are collected, the directive does not apply as there can never be any re-use, because re-use is defined as use for another purpose than the initial purpose within the public task; to which OPSI replies that the buyer is inevitably using the data for another purpose than the initial purpose, so that use by the buyer constitutes re-use. What a strange thing!

    But I must say that I had started thinking myself that when data are collected for the initial purpose of commercialisation, with commercialisation being within the public task of the data collector, then the directive would not apply to these commercial transactions; and that a public sector organisation, in order to escape from the marginal cost dissemination rule that seems to be written into the Belgian Bill, would only have to argue that the data were never collected for any other purpose than commercialisation; and that we could then corner them, especially when they’re heavily subsidised, by saying that they cannot at the same time collect data for the initial purpose of commercialisation, use taxpayer money to finance the data collection, and then abuse their dominant position by refusing access to private competitors at marginal cost (because these awful things indeed happen all the time).

    But I suppose OPSI is right: use by another party than the data collector himself is re-use.

    Except when the other party is a public sector body using it for its own public task (exchange). So when a public sector body exchanges its data with another public sector body that has received the public task to commercialise it, the directive does not apply to this indirect commercial activity, because it would not be re-use! This looks like a neat way out of the directive for commercial public sector bodies, they just need a sister organisation to ’smuggle’ the stuff out!

    Then there is this strange article 10 on non-discrimination: when the data are collected for an initial purpose other than commercialisation, and the data collector then uses the data as an input for commercial activities which fall outside the scope of his public tasks, he may not discriminate against other users; but when these commercial activities fall within his public task, he may discriminate. How strange! Because first, how can a public sector body have commercial activities that fall outside the scope of its public task? And how come a public sector body that does have commercial activities within his public task then obtains the right to discriminate against competitors?

    The more I look at it, the more I find it confusing. What a mess!

    And on top of this mess, you have all the messy transpositions into national legislation, can you believe it? This is bureaucracy at its best. I always say: in a bureaucratic environment, you can always employ 20 bureaucrats to tie complicated nots, and employ another 20 bureaucrats to untie these complicated nots again; both offices will produce convincing progress reports on their activities, showing how modern and technologically advanced they’re going about their business; and if those who bear political responsibility for it only look at it from a distance, they can of course go away with the impression that things are not that bad; but the productivity of the whole enterprise is NIL, if not NEGATIVE. (The idea comes from J.M.Keynes’ ludicrous proposal on how to solve unemployment by letting half of the unemployed dig trenches, and the other half fill them up again. I must go and look for the exact quote on the internet, because this is from memory.)

    At the moment I’m trying to spell out ALL the possible configurations of government commercialisation, i.e. charging beyond marginal cost. So that I can demonstrate clearly two things I already know: (1) that all charges beyond marginal cost can never be made transparent, and that allowing for charges beyond marginal cost while at the same time asking for transparence is a contradiction in the directive; (2) that charging beyond marginal cost while not granting access to the bulk of the primary data at marginal cost is always an abuse of a dominant/monopoly position.

    [I posted a copy of this message as a comment on the item 'Caro Tullo on free data: what's the point'.]

  5. Robert Barr Says:

    Stephen Palmer repeats the often stated view that “Local Councils in Britain (who) actually create every address in the UK”

    I don’t believe this is the case. Local councils APPROVE addresses, they don’t CREATE them. And anyway, naming something gives you no IPR, creating a database does.

    This is what puts Royal Mail PLC in a very strong position, but only because Parliament chose, when creating the new plc, to give it ownership of PAF (Postcode Address File) FOR THE TIME BEING, unwisely in my view. This is interpreted by Postcomm as meaning that Parliament retains the right to pass new legislation which would remove ownership of PAF (and therefore most addresses that matter) from Royal Mail.

    I believe that such a law should be passed and should force all addresses to be placed in the public domain. This should prevent any public or private organisation from having any IPR in raw address data, or any database or compilation of addresses.

    It is the bizarre concept that Royal Mail, Local Councils or Ordnance Survey can OWN addresses that has got us into this mess. The only reason any of them want to assert ownership is in order to have a right to charge for addresses. If Local Councils were funded to collect and approve addresses, by an address registration and publication fee, levied when addresses are first approved or are changed none of this stupid wrangling would need to happen.

    However, as neither the Treasury, nor OPSI, nor anyone else is prepared to see addresses as a special case, but only as a tradable commodity the nonsense will go on – sadly!

    Bob

  6. steved Says:

    In response to the comments raised by Robert Barr,

    “Local councils APPROVE addresses, they don’t CREATE them. And anyway, naming something gives you no IPR, creating a database does.”

    The street name originates from local authorities, and in the majority of cases this name is either thought up/created by/allocated (Call it what you wish) by the street naming officer. It is often put out to consultation, often including Royal Mail, which I can assure you takes a great deal of taxpayers money – particularly in those districts facing 20,000 new homes over the coming few years. This is far more than just approving addresses.

    As for IPR, I am aware you have been critical of the National Land & Property Gazetteer for a number of years, but surely you must acknowledge, it is a database, created by local authorities!? This information is passed to Royal Mail and Ordnance Survey, and indeed during the recent NSAI fiasco, was supplied to OS free of charge. Local Authorities have no interest in joining in the petty IPR debate, other than to stop the ridiculous situation where they are forced to buy back Addresspoint from OS, despite having no interest in it as a product and despite supplying much of the data used in its creation free of charge. Incidentally – The Land & Property Gazetteer at the Authority in which I work contains 86,000 addresses. Addresspoint for the same area is considerably less than 70,000 address.

    I cannot agree more with the concept:

    “I believe that such a law should be passed and should force all addresses to be placed in the public domain. This should prevent any public or private organisation from having any IPR in raw address data, or any database or compilation of addresses.”

    Local authorities have no interest in ‘owning’ an address or charging for others to use it. They have a statutory duty to create/allocate addresses in the first instance, and receive taxpayer’s money to provide this service. They are just fighting to ensure that the taxpayer is not paying for this service, and then paying again for local authorities to use addresses to provide that same service! Which is exactly what is happening now.

    Local government address ‘creator’

  7. Michael Cross Says:

    An interesting development on the addressing front: OPSI reports that talks have been held on reviving the National Spatial Address Infrastructure.
    See
    http://www.opsi.gov.uk/advice/psi-regulations/national-spatial-addressdatabase.htm

  8. Mavis Pidgeon Says:

    How peculiar!! Having tried to go to tinyurl.com/y5zd74 as advertised in todays technology section all that happens is I get to here which appears to defeat the purpose. What gives?

  9. Eddie Sheffield Says:

    Don’t you just love the internet. I’m not an idiot but seems I need an idiot guide ‘cus I don’t know mu IP address! and this page isn’t helping me find it!

  10. Ted Truscoe Says:

    I too thought I was getting to somewhere I coul;d find my IP address. Is this website operating as spam?

  11. John McCartney Says:

    Yup, I’m trying to find my IP address too. Is this the digital version of all those wonderful Grauniad misprints?

  12. john archer Says:

    checkout the shields up! tool at gibson research http://www.grc.com – that’ll give you your current IP – plus other goodies

  13. rik dowding Says:

    “why spam is out of control” indeed!
    Because the Guardian is part of the problem it would seem…..

  14. Alex Says:

    Awful, awful Grauniad.

    http://www.whatsmyip.org/

    ^^ go there, your IP is in big letters at the top of the screen.

  15. Jimmy, Dublin Says:

    you’ll get your ip address at the following link: http://www.lawrencegoetz.com/programs/ipinfo/

  16. Richard Jones Says:

    Unfortunately http://www.whatsmyip.org/ and http://www.lawrencegoetz.com/programs/ipinfo/
    give entirely different IP addresses. What gives?

  17. David Wright Says:

    I tried both, and I see what is happening.
    http://www.whatsmyip.org/ gives your IP address.

    However http://www.lawrencegoetz.com/programs/ipinfo/ gives the IP address of the computer that connects to it. For most people, their computer does not connect to the net directly, but via a Proxy Server which is shared by all their ISP’s customers, and which provides optimisation and control. E.g. for NTL customers like me, it will be 62.253.128.11 or similar. These are not the IP digits you are looking for.

    Short version: http://www.whatsmyip.org/ is the one to use if you want to look up your PC or home network’s address.

  18. jules Says:

    Thank you guys, have found out my IP address. Now what?

  19. Charles Arthur Says:

    Very sorry to everyone for this; an error in the editing process.

    If you want to find out your IP address, the link should have been http://noc.net.umd.edu/cgi-bin/netmgr/whoami

    We’ve changed this. Again, apologies.

  20. tok_bok Says:

    send me my ip/ thanks

  21. tok_bok Says:

    please send my ip

  22. Kris Says:

    Send my IP

  23. David A. Blanc Says:

    in order to find out if my computer is being hijacked

  24. Andrew Ballantine Says:

    What is the legal position for a group that sets out to collect addresses and postcodes from NON PAF sources?
    Would it infringe the Royal Mail’s IPR?

  25. Charles Arthur Says:

    @Andrew – no, I don’t think it would. Certainly that’s the line being taken by the open postcodes projects – see this post on this very blog for links to two of them.

  26. Lucy rodrigues Says:

    Please can you send me my Ip address to find out if my computer sends spam.

Leave a Reply