OS Consultation – winners, losers, irony and a bit of schadenfreude 12


The dust hasn’t settled yet from the government’s response to the OS Consultation but already I can hear the mumblings and grumblings. Inevitably 441 respondents (including 218 individuals) are not going agree on much if anything, the response document says “different stakeholders have different, and sometimes conflicting, objectives” So that might suggest that there are winners and losers. What follows is a quick trawl through the response highlighting some of the bits that may be the subject of debate, delight or disappointment.

We don’t want free!

Right at the beginning:

There was strong support in the consultation responses for the opening up of at least some data, with 300 (68%) respondents indicating that they support the release of some or all Ordnance Survey data for free use and re-use by all.

I am baffled at the 141 responses who were opposed to opening up any OS data! If we assume that most of the individual responses were in favour of opening up data that suggests that a significant proportion of the organisational responses (not just companies) were opposed.

The ins and outs of OpenData

The package of data that comprises OS OpenData (TM) – a few people have already commented on the irony of trademarking the word OpenData – included a couple of surprises.

  • The withdrawal of the 1:50k and 1:25k rasters and their replacement with a mid scale vector product will have delighted digital folk who never really wanted the ugly raster products but has disappointed the rambling folk (their pastime not their communications) who wanted the rasters to be released so that there could be alternative (and presumably lower cost) sources of printed maps. I think they missed the point that the release of the OpenData was intended to fuel the new digital age of innovation, democracy and accountability as envisioned by TBL and Prof Shadbolt not just to reduce the cost of hill walking maps.
  • Royal Mail have been dragged screaming to the party, the CodePoint Open data set has made Ernest Marples very happy whilst not many people will be overly bothered about the other fields removed relative to the full dataset.

Several respondents thought that OS should be made to fund OpenData through cuts elsewhere in the organisation (I will keep my speculations on the motivations for that line of response for another post). They will have been disappointed by this decision as will any who hoped that the outcome of the consultation would be a deconstruction of OS:

A long-term funding arrangement is a pre-requisite for any changes to be viable. Central government has negotiated a commercial agreement with Ordnance Survey for the licence for OS OpenData and for its ongoing maintenance.

The announcement of the free to use TOID look up service should please the linked data community and could help the vision of DNF start to gain traction.

You can download all of this wonderful geogoodness from the OS site or they will cut it to DVD and mail to you for a small fee. If you find the OS site overloaded you can also download from the mysociety mirror. If you only want a tiny bit of the data then the enterprising guys at emapsite (an OS partner) have a “cut a little bit out” download service available for a small charge. See how the options are flourishing within a day of release.

Public Sector Mapping Agreement

It looks as if we are going to get a One England and Wales Mapping Agreement covering the whole of the public sector and procured/funded centrally

Government’s preference is to obtain all its underlying core, national geography data and mapping services in this arrangement, whilst leaving open the ability for organisations in the private sector to separately add further value to Ordnance Survey’s core product sets.

Not good news for the commercial bodies who wanted to compete with OS (possibly using raw data collected by OS) in the provision of mapping to the public sector. It is difficult to see how commercial producers are going to be able to provide competitive alternative mapping solutions to the public sector unless they focus on substantial added value or niche requirements. Maybe I have missed something but I think OS will see this as a big win. it also sounds like common sense as far as the public sector is concerned removing the stupid issues of data sharing across the sector and cutting the tiresome costs and bureaucracy of procurement.

The elephant in the room – Derived Data

Although Derived Data was not a specific topic in the consultation I guess that it was a common feature in most of the responses. It is a bit disappointing that OS have been working on this topic since the announcement of their last “New Business Strategy” in April 2009 and yet there is no detail announced just:

Ordnance Survey will also be proposing changes to the derived data policy for the commercial sector, including ‘Free To Use’ data, as part of its work on revised pricing and licensing. Further details of when these changes will be implemented will be communicated later in the year. In addition, Ordnance Survey will work with Cabinet Office and the Office of Public Sector Information (OPSI) to agree guidelines and processes which would permit certain other datasets created by public bodies and containing licensed paid-for Ordnance Survey data to be made available for unrestricted, including commercial, re-use. In agreeing these guidelines, the parties will need to balance a number of issues, including the significance of the Ordnance Survey data taken, the impact of release of any dataset on Ordnance Survey’s commercial business and any legal or regulatory consequences for Ordnance Survey.

Not really what we wanted to hear but perhaps the final outcome will be a sensible and pragmatic approach to the “elephant”. Having spent some time talking to people in OS I know that the devil is in the detail. On the brighter side there cannot really be any derived data issues associated with OpenData so if you want to capture postboxes, dog pooh, observations or points of interest just get started on StreetView or VectorMap and you will be free to share (OS if you are reading this does a point derived from OS OpenData need to be attributed?)

Tariff rebalancing

There were two main options for tariff rebalancing – increase the cost of MasterMap to the public sector (a pretty unlikely scenario in current financial environments) or reduce the costs to the private sector to those paid by the public sector (despite the much larger values purchased by the public sector). Outcome – no change:

Due to a combination of comments received from consultation respondents and further economic analysis conducted internally and by LEK Consulting, the Government has decided not to pursue immediate upfront tariff rebalancing, as the broader welfare benefits are not likely to be as significant as previously assumed by the Cambridge study

That looks like one in the eye for Prof Pollock!

INSPIRE

OS will take on the technical delivery role for the UK’s INSPIRE implementation.

.. using Open Source software and the G-cloud where appropriate. Ordnance Survey will design, specify, create and acceptance test repeatable INSPIRE compliant hosted data services, together with an overarching portal to support the View and Metadata Services. These platforms will then be offered back to the community as Open Source software

That will have the open source community cheering, although some may wonder whether OS are the best equipped to undertake this platform. With up to a thousand local authorities and other data providers being required to comply with INSPIRE it makes sense for someone to deign and test a standard INSPIRE stack that all of the public sector can deploy (and adapt as they see fit). I imagine some software vendors will be resetting their budgets for 2010 and 2011, They will be even more concerned to read

Ordnance Survey will use the Open Source platform to establish an operational service environment, set up and hosted on the G-cloud, and will enable other public sector information providers to configure a local manifestation of the service environment for their own obligations and requirements. Ordnance Survey will also support other potential users through paid-for consultation, offered alongside other commercial service support providers.

OS 2 – Software consultancies 0

Winners

The Free Our Data lobby have to be pretty pleased with this outcome and respect should go to Charles Arthur and Michael Cross who demonstrated how a persistent, incisive and occasionally humorous campaign (not that I think they got all of their arguments right) could stimulate change at the centre of government.

The OpenData community have got most of what they wanted. Democracy and accountability are definitely winners. Whether this unlocks the flood of innovative new businesses that some have predicted remains to be seen.

Open Street Mappers in the UK should be quite pleased, they ought to be amongst the first to make use of the new free data. OSM should have complete  GB coverage (albeit at varying levels of detail and accuracy) within a few months.

Public sector and tax payers will be winners from simpler procurement, easier data sharing and reduced costs of implementing INSPIRE

The Open Sourcers must be pinching themselves, this could be the tipping point for Open Source in the UK (funny that Ordnance Survey and Open Source both have the same initials)

Losers

Anyone who was hoping that OS would be deconstructed into a DataCo and a ProductCo will be gnashing their teeth and will have to wait a while (quite a long while in my opinion, no one is going to have the motivation to tinker with the OS business model for a good few years)

Anyone who thought they might make a profit from the privatisation of OS. That looks a long way off now, and a good thing too.

OS partners who built a significant part of their business on simple reselling of datasets that are now free

Software and services companies who thought that implementing INSPIRE would open up opportunities at local and national level. The knock on effect, if open source gets widespread adoption in public sector via INSPIRE, will require a transformation in some businesses’ models. We could be talking about this long after the release of OS OpenData has dropped off the agenda.

Utilities and telcos who were hoping for a reduction in the cost of MasterMap

The Treasury – if the tax returns from the new innovative businesses don’t exceed the costs of OpenData

An ironic moment or two

Just as the news about OS OpenData was spreading in the UK and everyone was anxiously queuing to download the data (in case it was an April Fool’s joke and vanished on the 2nd) Peter ter Haar, the OS Director of Products, was at where 2.0 talking to Steve Coast, the founder of Open Street Map, in this interview.

There is an irony (probably not intentional) when Peter describes the release of OpenData as “cool”! Can you imagine an OS Director describing this U turn as “cool”? Still good on Peter for embracing the changes. I loved the little dig at the OSM community’s struggles with their new licensing model when Peter tells Steve that the new OpenData license is probably more open than Open Street Map’s.

Massive Winner

Much to the surprise/disappointment of many who wished them ill, the OS seems to have come out of this as a big winner. It doesn’t look like they will be privatised, nor will they be broken up. They have won the funding for government for the free data sets and most of their core activities will continue. They are now going to build out their technical and delivery capabilities around INSPIRE and open source and that should provide them with some new and interesting opportunities.

And that is my schadenfreude moment – all of those who wanted to see OS change but continue to maintain a high quality National Map should be pleased at the outcome. On the other hand those who backed the Free Our Data campaign, not because they cared about open data but because they wanted to make profits either in competition with or on the back of OS may well be ruing the day they joined the campaign. I seem to remember saying early on “be careful what you wish for”


12 thoughts on “OS Consultation – winners, losers, irony and a bit of schadenfreude

  • Andrew Zolnai

    “Anyone can make something sound complicated, it takes a genius to make something complicated sound simple”. Thanks Steve for an erudite yet clear summary…. this is indeed the beginning not the end of a process! A cottage industry evolved in North America around freely available (not free) data, and I doubt either country banked on substantial returns to federal coffers. What I’m trying to say is that the public service of opening the data playing field is worth all the wrangling past and future. PS: love your blog’s new look!

  • Robin Waters

    Very good summary, Steven. The INSPIRE stuff is, as you say, going to hit some business models. However at least OS has come ‘out of the closet’. Despite having one of the biggest teams of (OS funded) INSPIRE experts helping the EC with INSPIRE you would have searched almost in vain for anything about INSPIRE on the OS web site until this week. Now this will put OS in the driving seat. Exactly where it will take us and how it will relate to an independent Location Council – which still has(?) responsibility for the UK Location Stategy (remember that?) and, theoretically, INSPIRE, remains to be seen.
    I would suggest that OS needs to take very careful account of other public sector data providers’ views if it is not to get in another ‘derived data’ mess.

  • Charles Arthur

    Splendid analysis/summary – thank you Steven.

    “On the other hand those who backed the Free Our Data campaign, not because they cared about open data but because they wanted to make profits either in competition with or on the back of OS may well be ruing the day they joined the campaign. I seem to remember saying early on “be careful what you wish for””

    Well, they perhaps didn’t see where the campaign was heading, or understand its aims.

    I think we could say we’re satisfied with the outcome: OS remains as the national mapping agency (good) doing quality work (good) and people have free data to work with (good). So yes. Oddly, though, the feeling isn’t one of popping champagne, but of “lots of work to do, got to prove this works”.

  • steven

    Thierry

    I agree that we may wait a long time for these new geoenabled businesses that will fill the tax coffers through their activity. If they do appear I wonder if they will be based in low/no tax jurisdictions?

  • Thierry_G

    Great summary, Steven.

    Given the different opinions and many responses to the consultation I think the UK government and OS have done a remarkable job of coming up with a sensible compromise solution. There are many winners and very few significant losers (if any).

    New opportunities will no doubt materialise as a result of this move, but I think Martin Daly (@mpdaly) hit it on the head with his tweet yesterday:

    “Where do I point my browser to see the running total of tax revenue accumulating after OS OpenData release?”

    Will that elusive long tail now finally open up some new GI-based business opportunities? Or will the making of location-based money remain the preserve of the big players like Google or Facebook? (who, by the way, employ accountants who know exactly how to minimise their tax payments).

    Very interesting & exciting times indeed. Given the market size you’d think there should be enough space for everyone but it’s probably good to have a plan B up your sleeve just in case. Like killing puppies with Martin. http://twitter.com/vicchi/status/4338391912

    Thierry

  • Ian Painter

    Steven,

    Commenting on your reply “Agree that open source is not free but heavy license and maintenance models are certainly going to have to defend their value against a real alternative.”

    Its easy to defend value. Take our most recent customer, 6 months trying to implement Open Source WFS, swapped to our stuff up and running in 3 days. There’s not much in either model : go open source spend consultancy services and/or resource time to get it running, or spend licence fee. As for maintenance, open source maintenance contracts / pricing are no different. Take a look OpenGeo http://opengeo.org/products/suite/buy/ that’s comparable if not more expensive than most closed source products I know. We’re certainly cheaper. I can feel the beginning of this years Geocomm rant !

  • Jo Walsh

    This looks like a solid analysis, thankyou for writing it.

    As you suggest, software and service companies do not have to be disadvantaged by an open source move by OS towards INSPIRE – an open-direction “business transformation” is working out elsewhere – in Germany, companies like WhereGroup and Metaspatial are employing many to do training, systems design and support for open source spatial data infrastructure (SDI) projects in different German states. In Spain there’s also extensive use of free software in SDIs at a regional level – the Valencia administration even created its own open source GIS project, gvSIG.

    Lifting the restrictions on derived or produced works, if only just among the public sector, will create the conditions for a proper SDI in the UK, a federation of local authorities, with Ordnance Survey providing quality assurance for data collected at many different sites.

    As for the estimates in “Models of Public Sector Information Provision via Trading Funds”; there have been studies in Australia and New Zealand on the value of public sector information and the open approach, the former claiming that the opportunity cost due to spatial information not being fully exploited could be as high as 7% of GDP. The latter estimates a four-fold return in tax revenue, on investment in supporting the mapping agency to publish open data. Guessing these conclusions are comparable. I don’t really follow the point about welfare benefit and tariff rebalancing (e.g. putting up prices to the public sector), I thought the Cambridge study concluded that a marginal cost (e.g. tending to zero over the internet) model would be most efficient for map data.

  • Stu Mitchell

    Great article – it’s worth remembering that there are inevitably some losers here, and not those necessarily the obvious ones either.

    There’s also the set of companies that provide data which are currently competitive with OS data. For instance, those suppliers who supply 1:250k digital raster maps; still popular and used in certain products. Now there’s less of a reason to consume these.

    • steven

      Good point Stu

      Omelettes (or even soufflés) and broken eggs come to mind though. Is there really much of a business in 1:250k rasters?

  • steven

    Now we are cooking with OpenData in GML! Love it Ian.

    Agree that open source is not free but heavy license and maintenance models are certainly going to have to defend their value against a real alternative.

    Exciting times. Makes me want to get back in the fray!

    steven

  • Ian Painter

    Great summary Steven.

    OS as an open source software house for INSPIRE will be interesting to say the least. Not sure what this has to do with maintaining the national map but hey ho times are a changing.

    As I see it the winner will be whichever open source platform the OS chooses, the others open source platforms will take the same hit as the closed source community. The OS taking the lead on INSPIRE makes total sense, but to mandate open source looks anti-competitive to me. There is already an open competitive market of open and closed sourced software products for INSPIRE. There are pros and cons on both sides, why not let us compete on best value? Some will choose open source some closed. Open source is not necessarily free, nor is it necessarily the cheapest option. Its just another business model.

    Anyway, I’m loving OS OpenData, I’ve always bitched at OS through its INSIGHT programme on its product design, now I can redesign them myself, watch out for OS Open Data in GML!

Comments are closed.