[Geodata] public geodata piece for geo:connexions, feedback welcome

jo at frot.org jo at frot.org
Wed Apr 23 19:11:29 EDT 2008


dear geodata list, + mpg, 

I finally finished a draft of the public geodata column i promised 
mpg i would have finished a week ago, for the series of OSGeo-themed
columns that GeoConnexions magazine is running.
http://wiki.osgeo.org/wiki/GeoConnexion_Column

I see i already ran overlength trying to get too much in. But if
anyone has some quick lastminute commentary i would like to hear it.
I guess i'll write something similar for the Open Knowledge blog.

------

Next Generation Mapping Agencies

Government needs geodata, whether it's collected for census-taking or
taxation purposes, the monitoring of environmental change or the
allocation of land-use subsidies. Public sector agencies produce and
consume a great quantity and variety of geographic information.
Attitudes towards re-use and re-dissemination of this data notoriously
vary. In recent years, we've seen a slow trend towards an "open"
approach gather pace among policymakers and public agencies. What is
the likely impact of this, and how is it sustainable? 

In Spain, new legislation published on 8 March 2008 guarantees all
citizens the right to access and re-use information collected by IGN,
the Instituto Geografico Nacional. This applies to many different
kinds of geographic data publishde by the National Mapping Agency -
terrain models, aerial imagery, geodesy, transport networks, land use
information.

Loosely translated from the ministerial 'boletin': "these are all
kinds of 'public sector information' and their use, in whatever case,
will have an open and free character, given an acknowledgement of the
origins and ownership of the data." Those wishing to incorporate such
data in commercial projects must pay license fees.

IGN is in turn reliant on the cooperation and support of a whole
network of agencies feeding data into regional and national Spatial
Data Infrastructure projects. Networks of local government agencies
are discovering significant cost savings in the "network effect" of
increased sharing of geodata. 

The UK's situation has been described by experts as "a lot of data,
but little infrastructure". For ten years the UK national mapping
agency, the Ordnance Survey, has been a kind of semi-private company
called a "Trading Fund". Trading Funds are wholly government-owned,
but have responsibilty to cover their operating costs, and generally
behave in the marketplace as any private company would.

The UK Treasury recently commissioned a study of funding models for
these agencies as part of a cost/benefit analysis of private provision
of public data. The Treasury is fundamentally responsible for
decisions on funding and support for public sector information
provision. In contradiction to prevailing policy wisdom, the report
strongly recommends the distribution of "unrefined" or "raw" data - at
marginal cost, for which read the cost of dissemination. For
distribution of digital data over the internet, marginal cost is
effectively zero.

In the past we've been presented with stark alternatives for the
support of these agencies, necessary to the effective functioning of
so much government business, and with potential to underpin a lot of
business activity. On the one hand, free data, available in the public
domain and without restriction - but placing production and
maintenance costs entirely in the hands of government and taxpayers.
On the other hand there is the option of commercial licensing, and a
"cost recovery" policy which aims to extract operating costs directly
from users - even when a high proportion of usage is for other
government funded purposes.

The Treasury study identifies a middle way, inspired by that already
used to support the production of meteorological data in the UK, where
government re-users of the information contribute to a kind of
collective subsidy fund. "If some sort of analogous PGS (Public
Geodata Service) subsidy were being considered it might also be
possible to have contributions spread over a wide variety of
organisations (and departments within government)"

This Public Geodata Service model could help support agencies across
the US - where a strong tradition of public domain availability of
public sector information may be threatened by new pressures upon
government budgets. 

More "new hope" lies in the increased maturity of collaborative
mapping projects carried out by large, loose groups of collaborators
over the Internet. The OpenStreetmap project is the prime example of
this, and in selective areas has vector mapping coverage with accuracy
and currency exceeding what can be gained commercially from any
sources. As discussed in the April edition of geo:connexions, there is
a trend of realistic enterprise interest in Web 2.0 "neogeography".

Public sector agencies as well as large corporations can reach out to
these techniques and communities in the interest of cheaper and more
efficient maintenance of data holdings. A public sector information
holder may be the only authoritative provider of a data set, and data
about such abstractions as administrative boundaries and land coverage
may never be left reliably in entirely private hands - the public
administration always has a role.

Thus, though the news in Spain on open availability of national
mapping data is a very positive start; yet the complete "welfare
improvement" implicit in more open access to public geographic data is
only achieved by making it freely available for every kind of re-use,
including commercial re-use. A new class of data licenses, similar to
open source software licenses, are being published to facilitate this.
The Science Commons' CC0 and the Open Data Commons PDDL license both
take the concept of a Public Domain Dedication and apply it to a
formal agreement that can be supported in any national legal
environment.

In stating the case for the zero-cost availability of public geodata
at the point of provision, the UK study talks of "A chain of
cumulative innovations in which an innovator at each stage can only
extract some fraction of the total surplus ... another possibility is
that the innovation effort is distributed across many different firms
or individuals - "componentised" innovation - as an explicit example
one could think of an open source project working to produce GIS
software".

Ten years ago, open source was perceived as a matter of enthusiast and
academic interest. Now billion-dollar companies run on open source
platforms, and the Gartner Group predicts the quiet market domination
of open source in the years to come. "Open data" is exhibiting a
similar trend, with hosted services built on an open source and open
data basis underpinning viable business models. As ever, the
government market is a stable and fruitful place for such startups to
be. 


 
-- 


More information about the Geodata mailing list