Linked Data at Ordnance Survey
In the early 1990s there began to emerge a new way of using the internet to link documents together. It was called the World Wide Web. What the Web did that was fundamentally new was that it enabled people to publish documents on the internet and link them such that you could navigate from one document to another.
Part of Sir Tim Berners-Lee’s original vision of the Web was that it should also be used to publish, share and link data. This aspect of Sir Tim’s original vision has gained a lot of momentum over the last few years and has seen the emergence of the Linked Data Web.
The Linked Data Web is not just about connecting datasets, but about linking information at the level of a single statement or fact. The idea behind the Linked Data Web is to use URIs (these are like the URLs you type into your browser when going to a particular website) to identify things such as people, places and organisations, and to then use web technology to provide some meaningful and useful information when those URIs are looked up.
Linked Data
Ordnance Survey’s Research Department, which is where I work, has been interested in Linked Data for quite some time, and the release of OS OpenData means that the fruits of these labours can now be realised, providing Ordnance Survey with an opportunity to make a major new contribution to the Linked Data Web.
As a first toe in the water we decided to produce a gazetteer of the administrative regions of Great Britain. Each region is given a unique identifier in the form of a URI and described in terms of its name and the spatial relationships it has to other regions.
So, for example, if you look up The City of Southampton (identified by this URI) you’ll find a list of all the wards contained in Southampton along with the counties and unitary authorities that Southampton is adjacent to.
Last week saw another milestone release when we published URIs for every postcode in the country and linked these to URIs for the administrative regions. The URI for our head office postcode (SO16 4GU) is:
http://data.ordnancesurvey.co.uk/id/postcodeunit/SO164GU
Looking this up you’ll see that Ordnance Survey is based in the ward of ‘Redbridge’ and also ‘The City of Southampton’.
This offers great potential to data publishers. By linking to identifiers for places and postcodes in your data you can enrich the information you hold.
Imagine you have a list of schools and their postcodes. By connecting to the URIs for those postcodes you have a whole new way to view and analyse your data. Through the link to the postcode you now know the ward, district and county those schools are in. This is a very simple example of how merging two sets of linked data can deliver benefits.
We are already seeing quality linked data being published elsewhere, notably from the likes of the BBC and data.gov.uk to name two.
As more data is published the opportunity to create interesting applications based on combinations of these datasets grows. We suspect location will create a key information hub for many of these and applications…imagine the possibilities!
If you’re interested in Linked Data and want to find out what you can do with it, there is more information on my blog.
Themocracy
[...] This post was mentioned on Twitter by John Goodwin, AGI LPSSIG. AGI LPSSIG said: Any good examples of local public services engaging with linked GI data? Linked Data at Ordnance Survey. #opendata #geo http://bit.ly/aPqhmC [...]
Hi John
Nice work… and a move beyond the pizza analogy. We (in the current ESDIN partnership) are interested in Linked data and how this fits with an emerging European Location Framework. Do you have plans to connect with European data sets? Through the association there are a number of our 55 Mapping agencies that would be interested. As would our growing customer base (particularly from the EC family).
Keep it up! and the free data is the greatest chance for your work to get the exposure it deserves…
David
How can the public make use of open and/or linked data?…
There are several ways in which the public can make use of Open Data and gain more benefits particularly if the data is linked. Here are few examples that I have found interesting.
Public Transportation as Open Data [1]: See this video to get more ins…
Hello John,
This is very interesting! It is nice to see the OS being so active on the Linked Data front. I wonder, has the OS (or anyone else) ever tried to publish data sets containing geometry more complex than points?
Frans
Hi Frans,
Thanks for the comment! Yes we have published polygon geometries in our linked data. Take a look at the URI for Southampton:
http://data.ordnancesurvey.co.uk/id/7000000000037256
This is linked to a geometry via the
http://data.ordnancesurvey.co.uk/ontology/geometry/extent
predicate. The geometry
http://data.ordnancesurvey.co.uk/id/geometry/96957
then stores the polygon as an XML Literal in GML. The geometry URI is linked to this literal via the
http://data.ordnancesurvey.co.uk/ontology/geometry/asGML
predicate.
You can look at the RDF/XML here
http://data.ordnancesurvey.co.uk/doc/geometry/96957.rdf
Hope this helps…
John
Thanks John, this is very helpful. It is interesting to see (I have only just noticed) that for simple WGS84 points you make use of a W3C RDF vocabulary. It seems such a thing does not exist (yet) for more complex geometries. Was the choice for GML for those cases an obvious one for the OS?
It seems to me that using GML to encode complex geometry is somewhat in conflict with a basic rule of Linked Data: information returned from dereferencing an URI should be easy to use, i.e. be coded in the standard format (which is RDF).
As it is now, I think it would take quite some trouble before the extent information can be put to use (e.g. drawn in a map). Especially for one unfamiliar with geospatial data.
Do you think GML suffice in the long term, assuming that the whole Linked Data thing becomes as commonplace as the WWW is now? Or is there a need for a standard (OWL/RDFS) vocabulary for geographical geometries (as suggested by the comment in AbstractGeometry class definition in goemetry.owl)?
[...] Although we created new geographic identifiers for the IMD Lower Super Output Areas, these are cross-linked to the local authority identifiers from http://statistics.data.gov.uk and Ordnance Survey’s Linked Data for postcodes. [...]
[...] Linked Data at Ordnance Survey [...]
[...] something which goes beyond the content of the EAD documents themselves – prompted mainly by the recent announcement by John Goodwin that the Ordnance Survey had extended their linked data dataset to include [...]
Hey there, I think your site might be having browser compatibility
issues. When I look at your website in Opera, it looks fine
but when opening in Internet Explorer, it has some overlapping.
I just wanted to give you a quick heads up! Other then that,
great blog!
Thanks for your feedback. We’ll look into it!
I wanted to thank you for this great read!
! I certainly enjoyed every little bit of it. I’ve got you saved as a favorite to check out new things you post…