How does Ordnance Survey justify its licensing costs when its accounts are disputed?
In this week’s Guardian Technology supplement we have Time to account for travel maps’ costs, which looks at the costs of Ordnance Survey licences. If the organisation is right about the importance of its work (it claims to underpin £100 billion of economic activity), then – as Tom Steinberg of mySociety points out – even a slight error in overpricing the licence could mean a huge fall in tax revenues.
That is illustrated by the cost of a licence for “time travel maps” that mySociety created. As the article points out,
How much would it cost to put those maps on a webserver that anybody could access? MySociety asked Ordnance Survey’s licensing department. It calculated that displaying 16 “map tiles” with the relevant data would cost between £837.81 (for a 1:25,000 scale) or £1,032.71 (for a 1:250,000 scale).
Does that sound good? Here’s the sting in the OS’s small print: “All prices for one user, one-year licence and exclude VAT. Terms of data use are internal business use, display and promotion as long as there is no financial gain.” So for a charity to put that on a webserver that might be used by hundreds of people (a typical server can handle 2,000) would cost millions of pounds annually.
And here’s the other thing: the National Audit Office does not accept the OS’s accounts, and has not done since it became a trading fund in 1999, because OS treats the National Geographic Database (which originated with taxpayers’ money) as an intangible asset – but puts no value on it.
That in turn means that the £9.2m surplus the OS shows was almost four times above its target return, of 5.5% on capital employed (tangible and intangible assets). Yet put in the NAO’s estimate of the NGD’s value, and you still get a rate of return that’s about double the 5.5%.
What does that imply? To us, that OS licences are expensive – they’re generating too high a rate of return. Read and see what you think.
- The following posts may be related...(the database guesses):
- What's the cost of sales and marketing in the government organisations? (26 March 2006; score: 32.86%)
- DCLG Select Committee snipes at Ordnance Survey; MoD says why not split it up? (7 February 2008; score: 20.22%)
- How much does it cost to display a map online? The correct(ed) answer (29 June 2006; score: 18.67%)
- Companies House: who's watching how it works the numbers? (18 May 2006; score: 17.66%)
- OPSI finds against OS in row over AddressPoint licensing (18 July 2006; score: 17.56%)
June 22nd, 2006 at 8:26 am
I think you have got the OS pricing wrong.
1:25k must be more expensive than 1:250k surely? I asked one of my staff who deals with OS licensing to estimate the cost of 16 tiles of 1:25k for web use in a not for profit site and got back a price of £293.60 per annum!
Hardly milllions of pounds. How did you check your facts? Your campaign increasingly appears to be a vendetta against the OS rather than serioous journalism or considered lobbying. These wild assertions are not helpful to any meaningful debate about the funding of OS and te recovery of costs through licensing.
How long will it take you to post a clear retraction of your claim that “or a charity to put that on a webserver that might be used by hundreds of people (a typical server can handle 2,000) would cost millions of pounds annually”?
June 22nd, 2006 at 8:50 am
The Guardian have got their figures mixed up, yes. The £837.81 is for 250k England cover with Codepoint (for postcode lookup); £1032.71 is 16 tiles of 25k with Codepoint. http://www.mysociety.org/pipermail/mysociety-maps/2006-June/000105.html has the email from Ordnance Survey itself, quoting the figures involved (”=A3″ is meant to be a pound sign in that message) – note that 16 tiles of 1:25k apparently costs £301.12 “for 1 user, 1 year licence and exclude VAT”, which presumably therefore cannot include web use. So I guess I should ask, where do you get your figure of £293.60 from? :)
June 22nd, 2006 at 9:52 am
The rates come from a spreadsheet that OS provide to resellers/partners for calculating license fees. The article did not mention CodePoint so the rate I quoted is for the 1:25k only. That rate is for up to 10 users. Can’t explain the slight differnces but not sure that it is material.
As I understand it if you are a licensed user, web use (display purposes only) counts as 1 extra user.
The leap from ca. £300 to “millions of pounds” is the point at issue. Someone has leaped to the assumption that because mysociety have been quoted for a single user that pricing scales up to “millions of pounds” for internet use. This is just not the case, it is so obviously incorrect that surely someone should have double checked with OS rather than guessing?
June 22nd, 2006 at 10:35 am
The Code Point costs put the price up to between £800 and £1,000 per year per user.
Since you’d have to hope for more than 1,000 users to be a useful service – yes? On a web server where each can handle 2,000 quite handily, you’d be looking at around £2m per year.
That’s the licensing cost assumption I was making. Perhaps it should have been more explicit.
June 22nd, 2006 at 10:36 am
Oh, and on..
How long will it take you to post a clear retraction of your claim that “or a charity to put that on a webserver that might be used by hundreds of people (a typical server can handle 2,000) would cost millions of pounds annually�?
Um, for as long as we don’t think it’s factually incorrect. Then we’d change it.
June 22nd, 2006 at 11:04 am
There are differing prices depending upon the number of users, yes. http://www.ordnancesurvey.co.uk/products/25kraster/pdf/d02086_25k.pdf appears to be the current pricing structure for 1:25k data. That file says that licensing 16 tiles for 101+ terminals for 1 year (I’m presuming that the internet counts as “101+ terminals”, it doesn’t explicitly say that anywhere, but it definitely doesn’t say anywhere that web use “counts as 1 extra user” – this should really be checked) is £37.63 a tile = £602.08 – so about double what the license is for 1 user. And that’s just for 16 tiles; if you wanted to run a service at that level for the whole country, it states the license is £104,550 for 101+ terminals for one year. Even if somehow the internet was only “1 user” (please quote source), that’s still £13,068.75 for one year.
CodePoint is only licensed for the whole of GB, and that is £26,342.50 for 101+ terminals for one year (with polygons), or £5,852.75 without polygons.
So perhaps not millions, but still an awful lot – I certainly couldn’t afford it! :)
June 22nd, 2006 at 11:59 am
So Mathew agrees with me that displaying 16 tiles of 1:25k would NOT cost “millions”.
I suggest that Charles or Mathew actually checks with OS as to what 16 tiles would cost to license for internet use rather than “presuming that the internet counts as “101+ terminalsâ€? (I have given you my understanding but I could be wrong).
Charles, now that Mathew has confirmed that in his opinion your statement is factually incorrect will you publish a correction?
June 23rd, 2006 at 4:53 am
All,
Let me clarify…
The use of data in the way described would be classed as internal business use, and is not for commercial gain – therefore there is NO additional licensing required, so the cost would be…
16 tiles of 25k data = £301.12
National cover of 250K data = £183.22
National cover of CodePoint data = £731.59
So to put this onto a website would cost either £837.81 or £1032.71 per year.
Charles, I look forward to reading your correction.
ed
June 23rd, 2006 at 9:33 am
Ed, just to clarify. Does “internal businses use” (as defined here?) really include putting arbitary numbers of maps on a freely, publically accessible website?
I don’t understand how that could count as “solely within the licensee’s specific organisation;”
June 23rd, 2006 at 1:39 pm
Hi Francis,
Yes it does.. the license allows “electronic transmission of a graphic image” as long as the original data is behind the firewall.
This is the mechanism used by Local Councils for example to produce web mapping applications of local services such as recycling sites etc. on their web-sites.
ed
June 23rd, 2006 at 3:14 pm
(Ed is the CTO of Ordnance Survey, in case anyone was wondering. And I work for mySociety, but was not involved in the travel time maps at all; I’m just asking out of my own curiosity.)
So if mySociety (or anyone), free of any OS licences, wanted to provide a service where, say, a user entered a UK postcode and was given an OS map superimposed with travel times to that postcode, from what you say above, an OS licence would cost £13,800.34 per year (1:25k for the whole country and CodePoint)? Please let me know if that’s wrong; I find the OS website and licences quite difficult to understand, I’m afraid. :)
June 28th, 2006 at 11:43 am
Interesting discussion especially considering that licensing of crown copyrighted information/data is supposed to be open, transparent and in plain English. Leaving out the particulars of the maths here, can we say that the relatively straightforward transactional scenario described above is easily translated to a license? Again the answer is no. And why must it be so? Obfuscation.
June 28th, 2006 at 1:55 pm
Rob,
I cannot disagree with you.. the OS must clarify its licensing to make such relatively straightforward requirements to license data simple.
We are working on it, but we have a lot of historical agreements to deal with and we need to make sure we comply with the relevant legislation so that all users and dealt with equally and fairly.
ed
July 1st, 2006 at 5:26 pm
Well, yes, Ed. A rather disingenuous response, I’m afraid. The specfic use licensing regime which is currently operant was created to do just that i.e. treat everyone equally and fairly which is one of the primary tenets of the OPSI Information Fair Trader Scheme. The current regime of OS contracts, I would posit, was created only to maintain a veneer of fairness and transparency because a bit of a scratch below the surface shows that this particular array of specific uses allows OS to call the shots on the terms and conditions. And this is where the devil is in the detail. Not to mention, that this licensing methodology hasn’t been in use all that long. OS must be still dealing with historical contracts prior to its institution. So that if this licensing regime is revamped yet again, there could possibly be historical agreements, now twice removed?? One must be very very skeptical.
July 1st, 2006 at 9:57 pm
Roy,
Ok I give up !!
July 3rd, 2006 at 9:54 am
You give up, Ed? Does this mean that OS is going to provide its data free of charge as this campaign has been encouraging you to do all along?
July 3rd, 2006 at 3:46 pm
Roy,
No I am giving up defending the OS from the likes of people like you… who despite the efforts of people at the OS to make a difference sometimes against massive organisational resistance, accuse us of being disingenuous!
ed
July 3rd, 2006 at 4:14 pm
Oh, dear. I’m afraid you seem to take it all so personally. Are you trying to make changes at OS “against massive organisational resistance”? Good for you, if you are. All I ask is that if I am incorrect that I be corrected. What changes are you trying to make anyway?
Roy
July 28th, 2006 at 8:44 am
Abuse of dominant position, equality and fairness
Ed Parsons wrote: “We have a lot of historical agreements to deal with and we need to make sure we comply with the relevant legislation so that all users are dealt with equally and fairly.”
Equality and fairness aren’t actually clear-cut concepts that can be used to determine what is equal treatment (WHAT has to be treated equally?) or a fair outcome (fair from WHOSE point of view?). In bargaining theory they’re more seen as additional arguments that help the parties involved to agree more quickly on a deal, a deal that is just one of an infinite number of solutions that are all equally efficient (within the core of solutions acceptable to all parties).
In the case of monopolistic commercialisation, relying on price discrimination to earn revenue, the only thing you can probably say is that equality and fairness amount to ‘whatever the monopolistic supplier can get away with’. Because all customers and ‘partners’ know this, they all have reasons to become distrustful, or more exactly: to behave strategically in order to get a better deal. As time plays an important role in data licensing, the repeated game of bargaining can become extraordinarily complex.
The problem with it is NOT that it is a shabby way of being equitable and fair, because it isn’t actually shabby (it only looks that way); the problem is that it is TOO COMPLEX and therefore very costly to manage (transaction costs).
Negotiating a cost sharing agreement for the production of a public good between a series of cooperating parties is equally shabby, only it doesn’t look that way anymore, because the negotiating parties can better understand what is on the table: a deal in which every party agrees to a contribution within its willingness-to-pay (cannot very well be pushed to pay more than its willingness-to-pay), while trying to keep its contribution as low as possible without jeopardising the negotiation itself; everybody knows that everybody will get a better deal as long as everybody stays on board; everybody knows that nobody will ever be able to know who got the best deal; nobody has a reason to feel cheated as long as the deal goes through. So the negotiating parties can more easily settle for an efficient solution to the cost sharing bargain while believing that they got an equal and fair deal (envy-freeness).
July 18th, 2007 at 4:35 pm
I got a quotation to reprint a section of a traditional OS folded map, sized 4″ x 6″, in a generic 10,000 print run publication … £900. But of course, they were nice enough to give us a “discount” of 50%. So £450 ($925 USD). Wow.