Guardian article just plain wrong !!

Once again the Guardian Newspaper in the UK (which I read incidentally) has got caught up in the free geospatial data debate.

I quote..”Our taxes fund the collection of public data – yet we pay again to access it. Make the data freely available to stimulate innovation, argue Charles Arthur and Michael Cross ”

The OS is no more funded from taxpayers than many large software companies who work largely with government customers and have had the development of their core software funded over a number of years of development contracts, this is particularly the case with the large GIS software vendors!

The article is so full of factual errors I don’t now where to begin, but what comes across really clearly is the lack of the customer viewpoint!

Why not ask the customers of OS data what they would prefer – the status quo where they and they alone pay to license the high quality data they need, or the much admired situation in the United States where the provision of spatial data is funded by political mandate, which has over the last couple of administrations, resulted in data which is decades old and not maintained to any level of consistency.

The much admired data in Google Maps, MapPoint etc. comes from commercial vendors, which Google etc. have had to license, the base government supplied data does not meet their needs.. Remember the famous “Where is Apple” discussion last year, a result of government funded data used by Microsoft being so out of date it did not show the location of Apples’ offices in Cupertino !

There is no such thing as “free data”, in the end somebody has to pay for the expensive business of collecting and maintaining national geospatial databases, ask a politician what they would prefer to spend a limited tax funded budget on.. Hospitals and Schools or funding the collection of geospatial databases you know what they will answer !

Written and submitted from The Marriott Hotel, Huntsville, using the hotel in-room internet connection.

34 Replies to “Guardian article just plain wrong !!”

  1. How about a clear Funding of the OS page on the OS website to make this kind of journalistic error inexcusable?

    If true, it is interesting that 47.5% of OS revenue comes from its local government customers. Clearly then, the “government funding” of the OS is in a different league here from that other well-known trading fund, the Post Office. These same bodies also provide much resellable data to the OS for free, which, interpreted by some accounting rules, would surely push that percentage well over 50%.

  2. Ed,

    I’ve heard you bang on about this time and time again; I’m still very much to be convinced.

    Of course somebody has to pay for data, but who should pay, how much they should pay and the length of time they should pay for the data are all valid questions. The fact remains OS MasterMap is too expensive and is restricting the good GI could do for society.

    I refer to the example prices on the OS website; to put OS MasterMap data for Edinburgh on two computers for the period of just 1 year it would cost £10k! Sure the Royal Bank of Scotland making £500 a second could afford this, but the RBS aren’t the innovating people we need to get using GI. I’m sure I’ve read posts to your blog suggesting that GIS vendors restrict the uptake of GI with expensive software; I think you need to look closer to home.

    I do feel Arthur and Cross make a valid point when they remark…

    “OS has a monopoly on centuries’ worth of data thanks to taxpayer funding. Without that head start, it would be in the same position as any other startup today: facing the choice of creating a brand-new dataset, or finding someone who had collected it and licensing it from them.”

    However, I actually think Arthur and Cross miss the point somewhat banging on about tax payers money, the crux of the matter is GI must be cheaper. In this case, I find their charge about the use of taxpayers money somewhat hypocritical. If we looked at the percentage of Guardian readers employed in the public sector, I’m sure you’d find ultimately, taxpayers money pays for many copies of their paper everyday, yours excluded Ed 😉

  3. Rob,

    You make some good points, I agree our data is relatively expensive but it is expensive to collect large scale data nationally and maintain a currency where 99% of change is in the database within 6 months.

    You are right in that we need to find a way to get our data into the hands of innovators for much less or even free. I’m working hard on making this happen perhaps the way to do this is to extend our developers programme.

    SteveC trackbacks should be working now ?

  4. Rob – never mind the public-sector readers; I think it was pointed out a few years back that the Guardian’s annual profit (ok, surplus) equated almost exactly to the public-sector spend on recruitment advertising in their G3 Society/Media/IT etc. sections.

    On a general point, I’m always slightly surprised how black and white this debate is. Public geodata advocates say “we want it all free, and we want it now”; OS says “you can’t have it free, not now, not ever”. Isn’t there room for freeing _some_ of the OS’s data – say, a small scale map of the UK, plus postcode data to ‘AB12 3’ resolution? That would enable a vast number of mashups while protecting the most lucrative portions of OS’s business.

  5. I can remember back in 1991/2 talking with Census Office people who confessed that the Census’ cost recovery effort only looked so good because of other parts of government and the public sector buying the data. I seem to recall they recovered 15 million pounds but that 12 or so million came from the public sector (taxpayers).

    You’ll note that in 2001 the majority of census data down to ward level was made freely available to the public – leading to sites like and many other innovative uses of census data. The only downside was that it probably made junk mailling slightly cheaper 🙂


  6. As one of the authors of the piece, I’ve got a response.

    The article is so full of factual errors I don’t now where to begin, but what comes across really clearly is the lack of the customer viewpoint!

    I’m a customer – I use, or would like to us, OS data. I can think of zillions of uses for it. Except it’s hedged around with all sorts of absurd copyrights. I think it’s crazy that the OS got a monopoly of data and then gets to charge us to make use of it – and sues us (who gave it the leg up to collect it before the “trading agency” concept was introduced) if we use the data in a way it doesn’t approve of.

    So it’s expensive to collect? Yes, but as is obvious, there’s a benefit to the economy in collecting it and making it available to the public. People will find great uses for it. The expense is miniscule compared to, say, defence or social security costs. And good data might reduce those costs.

    For example – Hansard (the written record of Parliament) used to be under hefty copyright. Then it was made available online, for free. Which has led to the brilliant which lets you see what your MP has done, how they have voted, in ways that Hansard would never have thought of.

    How expensive is it to run Hansard? Who cares? It’s important to know what your MPs do.

    I did look at the OS page. I do know about trading agency status.
    As to customers – there are plenty out there who aren’t happy about the OS’s status. Perhaps you need to meet a few.

    As to the Microsoft/Apple/Google Maps issue – as I said in the article, “if you view [the UK] as belonging to its taxpayers, and meriting rigorous mapping for their benefit, there are no ‘uneconomic areas’ – only places that people haven’t started to use yet.”

    The OS should set up an API that would give people access to data that, after all, you oblige local authorities to collect – as is also pointed out in the article. Or shall we wait for the OFT investigation?

  7. Charles,

    Really pleased you responded.. and that you read the blog.

    So as a user what part of your license to use OS data causes you problems ?,

    The OS does not have a monopoly in the creation of spatial data, Yes it would be hard for a new organisation to start collecting large scale data, but not impossible and with technological developments this is becoming easier every day. Outside large scale data, we face competition in all our other markets and have in the past lost supply contracts to both UK and overseas based competitors.

    I agree that services like theyworkforyou are great, however the collection of geospatial data is not the same as the production of Hansard, once you capture the actions of a MP for the record.. the job is done, it will not change.

    Geospatial information is always changing and therefore the expense of collecting it is a completely different matter, like painting the Forth railway bridge it is a never-ending activity revisiting the same places at least every few years.

    Richard makes a good point above, about the need for lower cost less detailed mapping for citizen based applications, the OS business model today is built around the need to collect the most expensive detailed data. As I have said before I think there is a role for open source geospatial databases like openstreetmap to fill the gap.

    OK I know there are a few OS customers who do not agree with the status of the OS, but that is not the same as unhappy with the data OS produces or accepting a reduction in data quality which would be a direct result of funding from taxation.

    And finally the OS does oblige local authorities to collect information, if they chose to provide us with information about new development we are grateful to use it to direct our surveyors to go any survey it – there is no obligation however.

    Happy to continue the debate, I’d be happy to sit down with you and put the other side of the story ?

  8. Ed,

    I am not familiar with the position of the OS inside/outside UK government, nor do I understand in detail your mandate. This isn’t directed specifically at you.

    From my perspective (I work for a local government, but am not speaking for them) if the data is being collected to support government operations and is a requirement for the government’s business, then the data should be made available to the public free of charge. For many small organisations, the cost of actually servicing the requests means that there is little or no cost recovery.

    I’m sure that most people have read this:
    which originates in my part of the world.


  9. I agree that the article in The Guardian is wrong on many levels. The core of the hypocricy in it is the claim that because of the huge economic beneifts, if the data will be free the Treasury will pay for its collection.

    Saying this in a country where London Underground investment came only through PPP, where NHS hospitals are pushed toward PFI and where education infrastructure shows the signs of 20 years of neglect is either melicious or stupid. We know that the Treasury will not pay for geospatial data collection and as Ed noted, this is what going on in the US and other parts of the world.

  10. Muki, I really don’t understand what you’re saying. Why is it hypocritical to say that the wider economic benefits that accrue (like more companies generating more jobs paying more taxes) are worth the smaller sacrifice of not making taxpayers re-pay for their data?

    The PFI is not regarded, beyond the companies which rejoice in PFI contracts, as a desirable thing. And it’s not related. We’re talking here about a national asset – our geospatial data.

    Ed, Hansard has to be updated all the time. Those MPs keep talking.
    And I don’t mind continually paying to get newer versions – through my taxes. Hey, I pay to get better defence equipment all the time. What’s different about the GI of the country where I live and pay tax?

    The OS may not have a monopoly in creation (actually, capture) of spatial data. But it has a bloody good 200-year head start plus the following wind of the Treasury. Try working from one of those “competitors” for a few months and see how you feel then.

  11. The situation with OS data is actually very difficult for a lot of reasons. I’ll give one example from earlier this year when I wanted to use sections of a map but couldn’t.

    I am a committee member of the Cambridge Cycling Campaign. This is a non profit organisation which has a 700 subcribers who pay the lowest fee we can arrange to cover the costs of four magazines a year which cover local campaigning issues.

    There is a new development in Cambridge near where I live which involves more than doubling the widths of some existing roads, and which will have a huge effect on the amount of traffic in this area.

    After much effort (over many months), we managed to get paper copies of the plans for the new development. This arrived in the form of an A2 sheet covering an area in real life around 2 miles wide.

    So, I wrote an article for the magazine intending to include small closeups of road junctions from the sheet to show how these will affect cyclists.

    There was no copyright notice at all on the map, but because virtually all maps in the UK derive from OS data, we got concerned that OS data might be involved. The last thing we need is to be sued.

    I spent over a week contacting different people within the OS to find out what the situation was. How much licensing the data might cost, what conditions this would put on us using the data etc. etc.

    I got three different answers on three different phone calls, but then a few more calls lead me to believe that reproducing sections of this map would require one of several licenses. There are a lot of different licenses to choose from depending on who you are and how you work.

    For instance, with some licenses it depends on how many sites your organisation has. I explained that we have no sites as such as we do what we do in our spare time from our homes. So, that’s apparently one site per member of the committee.

    Then there are odd restrictions of other licenses which say you can’t reproduce the content in publications which have advertising. We have ads in our magazine from local bike shops. This helps to bring in enough to cover the issues we print.

    We also have a cover price. That’s another “no” for some licenses. A very small number of our issues are sold. Often these are past issues sold on a stall in town. Again, it’s a way of our organisation making another small amount of money. This wasn’t possible with some OS licenses.

    Some licenses cover the number of copies you make.

    Some let you put stuff in a web site, others on paper. Our magazine is published on both.

    “Fair dealing” for news reporting looked like a good option, but I was assured that this allowed us to make just one copy for our own use. Not a whole lot of use for sending out to a few hundred concerned people, then.

    Honestly, it’s bewildering. All this for data which in large part is not OS data because the OS data definitely doesn’t cover the roads which have not yet been built. However, at some point on the maps we have they link with the existing bits which are (almost certainly) from the OS, so we’re stuffed again.

    Finding a license which we could afford and which would allow us to actually do anything was not possible in the few weeks I had to do it. Maybe there is one, maybe there isn’t. It’s rather difficult to tell.

    In the end, I went to the site, looked at what was going on, looked at the bits of the paper plan which are new and therefore definitely not OS while trying not to look at the bits which are (probably) OS and also relied on my memory of what it’s like (I’ve lived a few hundred yards away for 10 years), and hastily made drawings which approximate the actual plans. I didn’t make quite as many of these as I would have put in excerpts of the actual plans because it took too much time to do so. I can only hope that the OS don’t still try to claim that what I have done is to copy their map. I certainly didn’t try to.

    Anyway, the overall effect of the OS copyright in this case has been to make debate of the pros and cons of this new development much more difficult. It is vital that such things can be discussed in our democracy, and in this sort of case, the OS copyright is stifling our democratic process.

  12. David,

    I’m really sorry about your experience, to be honest it seems your issues are not so much with the fact that there is copyright on OS data, more as an organisation the OS screwed up giving you poor advice.

    I am the first to admit that the the licensing of OS data is complex and needs simplification – and this is a personal view rather than a OS position – I think the OS needs to look at better ways of making its information available for non-commercial use, but that I afraid would not help you in this case.


  13. In the open source worldview there is a concept of a ShareAlike license. Data can be made fully and openly available for all uses including commercial uses, with the constraint that if improvements and enhancements are made to the data, those changes are made available in the public domain, covered by the same license. The Public Geodata License ( provides a good model for this.

    This isn’t the only model for making geographic data available in a more open way, but it’s a good one for state-collected information and public sector information in general. A ShareAlike license applied to distribution of Ordnance Survey data would help fulfil the purpose for which it is suppose to exist, to collect and redistribute the most current and accurate description of the UK that is available.

    The OS have access to incredible data that would enable open source geospatial developers to build applications that would knock the socks off Google Maps. Charging the public for data they have paid to collect is a short-term approach which is blocking the creation of a lot of economic value, as well as tremendous innovation in the tools that state agencies use to communicate with each other and share information with the public.

    I thought the Guardian article was spot on in its assessment of the situation, and apart from trivial things like Nims instead of NIMSA, describing the agreement by which OS receives public subsidy, i didn’t seem to contain any false statements. I actually thought the amount of income through sales that OS got from other government funded agencies was over 50%, not 47%, but it’s close enough.

  14. HI Jo,

    Thought you would join the debate 🙂

    NIMSA is not a subsidy, it is a contract carried out at cost and audited as such.

    At the end of the day this debate always is about who funds, regardless of the licensing model, somebody needs to pay for the collection of high quality geospatial data, the user or the general taxpayer and I just don’t think we are going to agree on it.


  15. Wow! Where do I start? As a long time fan of the Guardian I was surprised that they had taken the lobbyist’s PR and printed it without any apparent critical research, but maybe I am underestimating Michael and Charles. Glad to see the range of opinions being expressed on this subject, here’s one more.
    Many people enjoy knocking the OS within the geo-community in the same way that they also knock the dominant software vendors. I guess if you are big in a market sector you have to accept that. Unfortunately so much of the criticism is misinformed and poorly thought through particularly with regard to pricing, licensing and funding of the OS. The collection and maintenance of the large scale base map of the UK is a challenging task particularly the capture of real world change within 6 months, and the achievements of OS are greatly underestimated by many of their critics.
    A valid question to ask might be whether the current organisation of OS provides the most efficient method of maintaining the national map at the current levels of accuracy and currency. I am not aware of any comparative statistics for other national mapping agencies, but whenever I meet GI practitioners from other countries including the US they all speak of OS with the utmost respect. Perhaps Ed knows of some efficiency study?
    If the OS changed from trading fund status to a wholly funded agency that provided data at no cost to users I believe there would be an inevitable trend towards lower data quality standards. The first time the treasury were looking for budget savings they would lop 10 or 20 million off the OS funding and survey frequency, accuracy and other activities would have to be reduced. I doubt that would benefit many of the users of the current large scale products.
    Many countries including the US which is quoted in the article as the reference model for the free data approach do not maintain data of anything like the accuracy or currency of OS data. The freely available US data referred to is of such poor quality that numerous private companies then earn a living enhancing it to a standard that is usable, I am not clear how that provides a better economic model. Incidentally none of the data underpinning sites as Google is the freely available Tiger data, could it because it is not accurate enough to be suitable for those applications?
    In the US the absence of a definitive base map has resulted in each, town, county, state and local utility having to develop their own base maps none of which integrate successfully with each other, resulting in massive duplication of effort, failures in joined up government and enormous commercial opportunities for a dominant software supplier and aerial imagery companies. If you ask the users of this “free” data in the US they would switch to the OS model without hesitation.
    Does the current OS pricing and licensing make data unavailable to some innovative services or unaffordable to some users? Possibly, but if these services and their commercial sponsors cannot justify a portion of the costs of collecting the map data perhaps the problem lies with the services and the value that their prospective customers put on them, not the pricing of the data. Do all of these innovative services that are waiting to launch really only work if the OS data is free?
    Who will benefit if public sector data is free? Will it be the end users or the commercial entities that exploit the opportunities to repackage, reformat or add value to the data? Making OS data available free could have some unforeseen consequences – perhaps a disruptive force would enter the market offering some of the same services as the proponents of this change but with a radically different business or revenue model (list of examples withheld because I might want to try one if the opportunity arose). One might also ask whether the jobs created by unleashing the innovative talents of the proponents of this change will be in the UK or in some low cost offshore location.
    A last thought. Locus is not, in my opinion, a trade association – the trade association for the Geographic Information industry is the AGI. .Locus is a lobbying organisation for a small number of companies who either wish to compete with OS or add value to its products. The Locus web site directs any communications to Quintus a public affairs and political consultancy.
    I thought that we had moved on from the instinctive “private is better” mentality of the eighties, but then that was about “selling off the crown jewels” now it seems we are proposing to give them away! Hopefully the pending OFT report will balance the public interest in the broadest sense with that of the private sector.


  16. Steven, at the very heart of your argument you say:

    “Does the current OS pricing and licensing make data unavailable to some innovative services or unaffordable to some users? Possibly, but if these services and their commercial sponsors cannot justify a portion of the costs of collecting the map data perhaps the problem lies with the services and the value that their prospective customers put on them, not the pricing of the data”

    First, a small correction: the word ‘possibly’ should be ‘certainly’.

    Second, your argument is explicitly built on the assumption that only paid for services delivered by commercially sponsored entities can create value for the public: this is demostrably untrue.

    Finally, you don’t make any mention of opportunity costs or externalities. I suspect this is why you have fallen so easily into a bifurcation fallacy by which the only options you aknowledge are current-US-style or current-UK-style, rather than something better than both.


  17. Tom, a fair point.

    I focussed on commercial services because I was responding to an article placed by a group of commercial companies seeking to promote a change to OS funding and licensing status. These are not open source companies or “not for profit”.

    How will you measure “opportunity costs or externalities”? The opportunity cost of freeing up OS data could well be the gradual reduction of quality.

    There are several models of licensing and pricing that would sit between everybody pays and its all free that would support expanded use of OS data.

    I know this is a geo-blog but why has nobody mentioned the Royal Mail, Vehicle Licensing Authority, Land Registry, Environment Agency, etc. Try and deal with some of those organisations and you might change your view of the OS.

  18. It is standard commercial practice to value an enterprise’s assets and then amortise/depreciate them at some percentage over the lifetime of that asset, showing this on the annual balance sheet. Can I ask Ed if the OS balance sheets for the period since trading agency status actually do this with respect to the mapping database? If so, at what rate? What would the asset value, what amorisation period, and what appropriate depreciation rate? Or has the asset value increased? If OS did what their competitors do and account for what they claim to be ‘theirs’ in this way, what would be the impact on the supposed ‘profit’?

  19. Hi Dave,

    The OS, like many other organisations whose business is exploiting data, don’t capitalise the value of the database.

    It is more valuable to look at the asset value of the systems developed to maintain it over that period.

    For full details of the potential value of the database as an asset are detailed in the annual accounts, for last year at

    and indeed for all years as a trading fund on the website.


  20. As a long time user and critic of OS data, and usually a pain-in-the-**** of OS, I have to say that many of the comments about the price, the restrictive copyright and derived data conditions and the variable quality of OS data, pale into insignificance with other GI data sources. I sometimes wonder if anyone has ever tried to acquire and use geological, hydrographic and air photography from other GB GI data vendors. Sure, the OS data is expensive to acquire and restrictive in use, but, heck, they are not the only ones. And, if it is all so damn easy to acquire and publish cheaply, why hasn’t anyone else done it? Openstreetmap may be a start but I’m sure my grandson will be an old man before anythinguseable comes of it.

  21. I believe that the OS could strike a balance and offer Meridan and Strategi for free to the public. This then might be comparable to data made available in the US market, though in fairness, the physical landscape doesn’t change very much over time. The las time I checked, the Mississippi River still runs from the State of Minnesota to the Gulf of Mexico. Yes, local authorities do capture and sell their data to the commercial sector, and many in the commercial sector continue to use the blog maligned USGS 1:24K series of maps. Most of these large scale datasets that the local authorities generate there also hold far much more attribute information than MasterMap, though in fairness there might not be the same level of detail captured from one authority to the next, but there is a collective effort in this direction.

    Anyway, I think it would be good PR if the OS provided these aforementioned products and perhaps a generalised version of Landline at low cost or free. There should also be a serious review on the restrictions of sharing datasets derived from OS products.

    Also, I would invite readers to check this local authority’s website. Pricing schemes are provided for their datasets. Some datasets are listed as free, including Lidar datasets!

    More free data at this url which includes every street and road in the State of Minnesota

  22. Simon,

    I don’t agree with public geo data position, the current draft of INSPIRE will still increase the availability of information across Europe while maintaining the ability to continue to adequately fund its collection through licensing where necessary.


  23. I notice that there are at least two arguments above that go “The OS licensing system may be bad, but have you seen the other guys?”

    Now, forgive me for being presumptuous, but I could have sworn that such ‘least-bad-is-best’ thinking was supposed to have been confined to the history books. Punch in the arm instead of a kick in the teeth? Me sir, me sir!

  24. Maybe “least-bad-is-best” – or more to the point is “least-bad-is-better-than-none”. I did mention data “owned” by other government bodies in my first message. Has anyone compared access to that data with access to OS?

  25. Just a quick comment on

    “… and requires specialized software, making is inaccessible to schools”.

    Did you know that there is quite a bit of free software out there that will let you view / query / play / teach OS MasterMap. We have a free OS MasterMap viewer that has had over 3500 registered downloads. It also has numerous educational organisations using it for teaching purposes.

    You can download our viewer here:

    And the official sample data here:



  26. Just an observation – part of the problem seems to be that some people perceive the OS as having an unfair advantage. One solution which I havent heard discussed recently is that of freezing the large scale database at the time OS became quasi commercial. If the database was then valued (wow) then it (collected at public expense up to that point) could either be made available for free to anyone who wants it or OS could be invited to licence it on the same terms as anyone else who wanted to use it – that would seem to even things up?

    I dont think there is anyargument over the quality of the data and its comprehansive national coverage, the only issue seems to be who pays (or for some people who pays twice)

  27. An excellent and articulate blog that is spoilt by one of the GI software vendors exploiting it for advertising purposes. (Thanks again ES”Built on gobvernment funds”RI). For me the question isnt the history / ownership (??Elgin Marbles??) but the statement in the original article “Much is of great commercial interest”. So what we have here are the vendors wanting data to make their systems complete at point of sale. Will the overall cost go down? I don’t think so, it will just transfer the money, that makes OS a self funding agency, to a group of private software developers. Yes we will all continue to pay twice but now it will line the pockets of US Corporation individuals.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.