Draft response to Ordnance Survey consultation

Introduction

I am Simon Brooke, of Carlinscraig, Auchencairn, NX797514. I am a software engineer with a professional interest in location-aware systems, but I'm responding to this consultation as a private citizen, not on behalf of my employer.

List of consultation questions

Question 1: What are your views or comments on the policy drivers for this consultation?

The key policy driver must be the one that dare not speak its name: users are already deserting the Ordnance Survey for 'free as in beer' mapping providers - such as Google and Microsoft - in droves. Within a decade, if nothing is done, the Ordnance Survey will have become a rump and an irrelevance: the only users who would continue to use its data would be those who were compelled to under law, and it would be increasingly difficult to protect even this limited market from the cold wind of competition legislation.

The Emperor already has no clothes, and the public can see this. The fact that Google are only mentioned, in the entire consultation document, once - in table 2.23 – and that openstreetmap.org is not mentioned at all shows the depth of the Ordnance Survey's denial. It's true that new mapping data providers don't yet provide some of the richness of data that the Ordnance Survey provides, but it's equally true that the Ordnance Survey does not provide anything like the richness of data that the new providers have.

The Ordnance Survey may, for example, be able to tell you whether a church has a steeple or a tower; but Google's 'Street View' will show you a photograph of the actual church, which renders the point a little moot.

Most popular walking paths in your neighbourhood? Ordnance Survey doesn't know, Google does. Best route to drive from here to Bangor under the traffic conditions prevailing this afternoon? Ordnance Survey doesn't know, Google does. What bus will stop next at my nearest bus stop, and where will it go? Ordnance Survey doesn't know, Google does.

Question 2: What are your views on how the market for geographic information has evolved recently and is likely to develop over the next 5-10 years?

Ubiquitous location-aware devices will become universal. Within ten years every phone and every car will know where it is, and be able to display that information to the user in a multiplicity of forms, including maps.

It will be able to overlay on a scene information about points of interest, as Wikitude (http://www.wikitude.org/) is able to do now. It will be able to locate nearby points of interest of a wide variety of kinds on request, as several geographical search engines – including Google's and Microsoft's – already can. It will be able to locate the user's friends and business contacts, as Google's Latitude product already does.

And it will – inevitably – do all this for free. Because that is what Google, Microsoft and Open Street Map will charge, and because being free and open, they will attract an ecosystem of third party providers of information, contributing to a richness and depth of information with which the Ordnance Survey will not be able to compete.

Furthermore, as Galileo and GLONAS come on stream and Differential GPS becomes increasingly available on consumer devices, ubiquitous phones will increasingly have the geographical precision of current day survey devices. As a digression, while I'm aware that the current plan is that Galileo's full resolution will only be available as a premium service, I suspect that the same social and commercial drivers which will force the Ordnance Survey into opening up its data will also force the European Space Agency into making its full resolution free.

Against this technological background it's likely that the market for 'hard copy' paper maps will shrink rapidly. It won't go completely – fell walkers will not want to rely on an electronic device which might fail in adverse conditions, and there will still be a demand for wall mounted maps both for display and for planning. But this will become a niche business.

Question 3: What are your views on the appropriate pricing model for Ordnance Survey products and services?

There is a market niche for more expensive and better. There is no market niche for more expensive and worse. And unless the Ordnance Survey information is used and contributed to by third parties, it will inevitably fall behind. Commercially, you can't compete with free.

Question 4: What are your views and comments on public sector information regulation and policy, and the concepts of public task and good governance as they apply to Ordnance Survey?

The existing Ordnance Survey data is based on a dataset developed over centuries largely at the taxpayers' expense. Clearly the maintenance of this dataset is an expensive operation, but in equity it is unreasonable that the funders of the bulk of the dataset should be charged for it as though it were a commercial product.

More than this, however, it's uncompetitive for UK PLC. Many of our overseas competitors have been able to develop products based on ubiquitous free reasonable quality geographic information, and have gained a substantial head start on us. Setting the OS data free now will not change this, as for most commercial purposes good-enough geographical data is now already free in the UK. There is data the OS has which other providers don't currently have, but the difference is eroding month by month.

Question 5: What are your views and comments on the products under consideration for release for free re-use and the rationale for their inclusion?

There is absolutely no point in releasing any raster data. It is of extremely limited utility, and can in any case be reproduced from the data set quickly and easily with a suitable rendering engine. This suggestion simply emphasises that the Ordnance Survey simply are not facing the real problem.

The raw data from which the rasters are prepared is what is needed. The Meridian2 data is clearly a useful subset of this and its release is to be welcomed, as is the 1:50,000 Gazetteer; but I would hope that the Ordnance Survey would release a data set containing all of the features shown on a typical 1:25,000 scale Pathfinder map. Failure to do this would render the whole exercise meaningless.

Some form of rendering engine ought to be released with the data, but it could be a low-feature 'lite' version; this suggestion is expanded in my response to Questions 7.

Question 6: How much do you think government should commit to funding the free product set? How might this be achieved?

This is the tricky one. There's two elements to it; one is the existing data set at the time of liberation, the other is the continued maintenance of that data set into the future.

There is a real question as to whether it is necessary to continue to have a survey function. Google are already able to survey simply by aggregating data about the movements of hundreds of thousands of hand held devices. TomTom is able to do the same, mainly with in-car devices. Users already upload track logs to Google, and to third party users of Google's data such as Map My Run (http://www.mapmyrun.com/). All this contributes to Google's data set. Simultaneously, users already contributing data to Open Street Map (http://www.openstreetmap.org/), which is freely available to any user.

If a free dataset is available, and there is a widely recognised means of providing overlays and mixins to that dataset, then academics such as ecologists and archaeologists, commercial entities such as developers and businesses with physical infrastructure, land managers and the public at large will find it to their mutual benefit to make the geographical data that they create in the normal course of their activities available as mixins or overlays for that dataset, and software to enable this to be done (and rendered) will rapidly be developed. No-one is more keen to have the position of underground cables mapped precisely than the electricity company or telecoms businesses who would be called out if someone put a digger through it.

So the issue – except perhaps in rarely travelled areas – is not so much surveying as validating; of adding data from disparate sources to the core dataset where it is felt to be of public utility to do so. Under this model the Ordnance Survey ceases to be a creator and publisher of data and becomes instead a quality assurer.

Question 7: What are your views on how free data from Ordnance Survey should be delivered?

In my view the whole of the existing raw data set at the date of liberation should be made available under a license similar to the Creative Commons Attribution or Attribution/Share alike licences on suitable distribution media at a cost reasonably reflecting the cost of reproduction of the media. There is no reason for the taxpayer to fund hosting this data online since other organisations such as Google and Microsoft will undoubtedly be prepared to do so at no cost, provided the license chosen allows this (as it should).

Whether the rendering engine which renders the data into the map formats which users are used to should be made available at the same time and on the same terms is a slightly different question. In my view it would be to everyone's advantage if it were, but if it were not, with the data available third party renderers would be developed quickly.

However there would undoubtedly be a market for better, or custom, renderers and the commercial successor map-publishing business to the Ordnance Survey ('ProductCo') would probably want the intellectual property in the full-quality renderer, so the publicly released renderer could be a 'lite', feature-poor, version.

What is key is that in the process of releasing the data the OS should endorse a data format in which overlays or mixins to the published data can be made available. This will allow third party bodies, both public and private, to participate in the communal process of maintenance.

Question 8: What are your views on the impact Ordnance Survey Free will have on the market?

Small.

Realistically the Ordnance Survey is rapidly becoming a bit player except in legislation-protected or specialist niches. Commercial road atlases will improve with the new data – a bit. But they are dying under the onslaught from GPS providers like Garmin and TomTom, who are in turn withering under the assault from the Internet search companies.

Those companies – Garmin, TomTom, and especially Google and Microsoft – will of course incorporate data from the released data set, and the quality of their data will be greatly enhanced by it. But as it's their data that the overwhelming bulk of users would use anyway, that has little impact on the market. It's also worth pointing out that the late arrival of this data means there are and will be no UK based players in this game. By hugging our cards too close to our chest for too long, UK PLC has lost out on a major world market.

Question 9: What are your comments on the proposal for a single National Address Register and suggestions for mechanisms to deliver it?

Surely the Royal Mail are the right body to maintain this data? They must do so in any case. Why duplicate? Whether the Royal Mail should be renationalised and, as a government agency, release its information under similar terms is a matter, I'm sure, for separate consultation, but it seems to me sauce for this goose applies to not merely the Royal Mail but also (at least) the Meteorological Office and the Hydrographic Office.

Question 10: What are your views on the options outlined in this consultation?

In the real world, any attempt to maintain the status quo or even to reduce the pace of change will simply be overtaken by the market. The data can either be released now while it still has a small but significant marginal utility, or in the future once it has become worthless. The time this should have been done is twenty years past – then it could have made a real difference.

Question 11: For local authorities: What will be the balance of impact of these proposals on your costs and revenues?

[Not applicable.]

Question 12: Will these proposals have any impact on race, gender or disability equalities?

None that is apparent to me.