Data Policy Ordnance Survey

The OS free data licence

I have had a couple of questions about how the free OS data is licensed, here is the license which as you can see is basically a creative commons attribution license.

This confirms there are no derived data issues.

In fact this license makes OS Opendata more “open” than Openstreetmap.

Written and submitted from the Where 2.0 Conference (37.331N, 121.888W)


11 replies on “The OS free data licence”


I will do a longer post in the next couple of days, but I’d say they got it pretty much right!

Most people were not looking for free access to MasterMap and actually the iconic raster versions of Landranger and Explorer maps are not particularly suitable for use on the web.

I’m pleased with the openness of the license and the download/request a DVD service is just what is needed and no more.

When you say ‘more “open” than Openstreetmap’… it depends on your point of view! Ordnance Survey OpenData is more open for commercial publishers looking to use mapping in their own copyrighted products, much less so for end-users of maps.

Hi Andy,

Yes of course, I was just looking from the point of view of the license, however it’s not hard to imagine a community effort growing around a version of this data with the appropriate editing tools. I think this is what Steve may have in mind for OpenOS ?

Maybe so, though it looks like the OpenData licence *is* compatible with OSM’s CC-by-SA licence, so in theory OS OpenData information *could* be imported into OSM. The trouble is that any OSM-derived work that included Ordnance Survey data would probably have to carry an explicit Ordnance Survey attribution as well as the usual OSM one. Since it would be hard for the publisher to determine whether a particular map included any Ordnance Survey data or not, that would mean adding such an attribution to any OSM map of a UK location, “just in case”. It’s easy(ish) for Google Maps to handle this sort of thing automatically because at Google you know exactly where your data is coming from. In OSM it would be a nightmare.

I was really thinking about the effect of OSM’s CC-by-SA licence, which effectively favours the end-user over the intermediate publisher.

Let’s say I want to publish – and sell – a paper map of a cycle route. If I base it on OSM, I am obliged to pass on the CC-by-SA licence, which means that anyone can legitimately make copies of my route map and they won’t owe me a penny. They can even sell their own copies of it. That’s great for the end-user, but as a publisher I might see it as damaging my sales.

On the other hand, if I base my route map on OS OpenData mapping, I can retain full copyright control of the original work in the resulting product. Bad news for cyclists in search of freebies, but (maybe) better for my sales.

Incidentally, that’s a classic example of where the 1:50 000 Landranger would have come in very handy. It remains to be seen whether the new VectorMap District will be adequate for maps of cycle routes or country walks.


Yes you have hit the nail on the head with the “problem” with CC-by-SA from the perspective of the publisher or shall we say value adding contributor, but from the point of view of the OSM community it protects the project from widespread commercial use which is something the community desires.

It will be interesting to see how the two “open” street maps now develop moving forward.

I too am looking forward to seeing what Vectormap District turns out like, while I was at the OS we saw a vector version of the Landranger as crucial to many future developments, it’s amazing to think this will now be made freely available !

Hi Ed,

OS have a sample of VectorMap District at and some other info.

I’ve had a look and, for me, it’s a real dissapointment. If the sample is anything to go by it’s a very poor relation to Landranger and Explorer. If you have any interest in anything outside roads or settlements you’ll feel a bit let down – no footpaths, field boundaries, landcover. The coutryside, it appears, is empty.

I’m interested in exploring the derived data a bit further …
This is clearly relevant for all sorts of uses – not least overlaying on google maps

As I understand it – any data derived from the new Open mapping won’t have derived data issues – but what about data historically that was:

1) Derived from e.g. 1:50K map in the past – is this now free of derived data issues as the product its based on is now “free”

2) What about data that has been captured from OS mastermap – this is likely to be the real issue. Many, many datasets have been digitised – in part at least – against master map data. Many of these add very significant new data (e.g. a habitat map) – but of course may share portions of mastermap field boundaries in places – whats the position with this ? Any views?


James, this is only a guess, but I think the Ordnance Survey would see it this way. When you created your data, what map were you looking at? Was it – at that particular moment – a copyrighted Ordnance Survey map? If so, then (as things stand presently) it is still data derived from a copyrighted source, and subject to OS’s “derived data” rules. At the time you created your data set, you accepted the restrictions imposed by the copyright statement on the base map. (You could only get around this by throwing your current data set away and re-plotting from scratch using one of the OpenData datasets as a base.

OpenData does not include the 1:50K maps, so the answer to your first question is even more clear-cut.

One outcome of the govenrment’s recent consultation is a promise that the derived data policy will be reviewed and revised. However, the Ordnance Survey itself gets to propose the new policy, which will then be agreed with OPSI and the Cabinet Office.

Add a comment?

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