On freeing data, unwillingly
Comments(4) So 10 days ago now, the UK Government announced that they would be freeing up some Ordnance Survey data as part of a “drive to improve transparency”. You can read reactions to this all over the geospatial blogs (see Mapperz excellent round-up), some ambivalent, others mostly positive.
Me, I’m with the ambivalent crowd. On one level, assuming it’s actually useful datasets made available, and assuming we get to use them how we’d like to (using it in our internal GIS, for example), it will be great, and my colleagues will find their jobs considerably easier. However, like a lot of things the UK Government has done recently, this was handled badly (the lack of Ordnance Survey input into the original announcement, speaks volumes). If it’s not going to be done willingly, will we see the same quality? The same commitment to survey the rural areas? Presumably, the cost of the remaining datasets will go up to cover the projected loss of revenue (whatever that is), so a lot of small businesses could end up in a worse situation than they are now- with expensive data that they can’t afford, or free data that’s not fit for purpose because it’s not complete or recently updated. Like Peter Batty, I don’t want our data brought in line with that of the US, as it’s currently much better!
I’m glad there has been a change, as I broadly support the Free Our Data campaign, and have done since it started, but I wanted it to be well thought through and planned for, and not a snap decision to try and win some votes. How naive of me…
I’m also concerned, like Peter Batty, about the affect on OpenStreetMap. I’ve been an enthusiastic convert to the benefits of OSM, since hearing of Muki Haklay’s completeness studies, and with seeing some of the beautiful maps and really cool innovations that it has spawned recently. At conferences in the UK this year OSM felt like the juggernaut it was impossible to ignore, and why would you want to? OSM is now a valuable part of the UK geospatial ecosystem and I would hate to see that change.
Ah – that old rallying cry “Give me free data, but not in a way that implies any change”.
The key point that Tim Berners-Lee made at No.10 is that the derived data problem is going to gotten rid of. It’s been driving government departments mad.
Lack of OS input into the announcement? Mm, yes. Imagine their delight at hearing that the government is acceding to the case made by people who think it should not charge for data licences. OS wasn’t represented at the No.10 meeting, though its regulator – National Archives (which owns OPSI) – was. National Archives didn’t seem upset; quite the reverse.
OS will still do Mastermap (and still be selling it for money – the same money, I think) so its commitment to rural areas will have to remain the same. Basically, the Treasury is underwriting the cost of doing this free data move. I’ve heard estimates but want to get them firmer before publishing them.
Peter makes good points (I wish I’d discovered his blog much earlier). The answer to all your questions and wishes: pitch in when the consultation starts. Make your desires clear. Make your fears clear. State how you’d like your desires met and fears calmed.
OSM could benefit hugely though if it is allowed to take in the OS dataset for 1:10,000 upward without copyright provision. That I’d like to see.
Derived data is the big one for UK local government. If this freeing of data early next year does not do away with the derived data issue we’re still ‘stuffed’ in local government as we will be forever beholden to Ordnance Survey. Most local authorities have spent years deriving data from OS mapping because up until the last few years we’ve never had a choice really when it came to map suppliers, particularly for large scale mapping.
Now that we have a choice (eg UK Map, and OSM for smaller scale raster mapping) the thing that prevents us taking the ‘new’ road is the derived data issue. We need to keep using our derived data because it’s the product of years of work and a lot of money. But if we still have to keep on paying for full Mastermap licensing just so we can retain the use of our derived data, we’ll never be able to break free and begin using other mapping products, let alone ‘freeing our data’.
Many local authorities have forward thinking staff who would like to publish data (particularly) mapping data but because of the draconian licensing regime pushed out of Southampton we’re not able to.
A good example of freeing data up is what we’re doing with street naming and numbering data. Because we ensured that we had derived data rights on our new air survey, all new street naming and numbering properties get posted up to OpenStreetmap even before they’ve been built. This includes the address details (obviously not the postcode at the moment!) and the building footptint.
The release of generalised, small-scale and mid-scale data under an open license is unlikely to have that significant an impact on OpenStreetmap; it’s the MasterMap-level large-scale vector data that is the “competing” product.
I share your concern that the opening up of generalised products may cause National Mapping Agencies to seek price rises on large-scale more accurate data. NMAs are all suffering the effects of the public sector debt crisis – income is receding as sales to other government departments fall as well, and the sustainability model for quality data isn’t really being addressed yet…
@Charles- I don’t fear change, but I do fear winning the ideological battle but losing the war, for the sake of what looks like election promises. I remain quite uncomfortable about the way this has been and continues to be handled. However, you are correct, that I should make sure my concerns are known in the public consultation.
@Jo- I’m not sure. I use OSM data as backgrounds for all sorts of online maps and such like, because I don’t have the corresponding 1:50K, 1:250K etc mapping from the OS, so for me the competing products are the smaller scale ones. I think people that need Mastermap data will stick with it because they know it is updated regularly etc. We’ll see…