Gi forum Raper Lessons of Open data from London

Realising the benefits of open geodata: lessons from London's experience

Jonathan RaperPlacr Ltd.

giCentre, City University LondonMayor’s Digital Advisory Board

Revolution in geodata

User paysPhysical distribution

Web distribution

Marginal cost/ free

2000

2010

Open geodata technologies

Open geodata apps

Open geodata content

Open geodata = open data

• The problem of “closed data”• How [we opened | are opening] the data• What “open” means for data• Why open data depends on politics• How open data can be sustained• Examples of open transport data from London• The implications of open data

Problems of closed-ness

The road to #opendata

• ‘The reason that we are now seeing major shifts in policy is that government has been hit by a perfect storm, with waves of change coming from different directions, which together are compelling change’

• Digital Geographer January 2010

1st wave: hyperconnected mobile society

• ‘the huge success of smartphones and app stores means people now have the platform in their hand for the delivery of real-time information services’

• ‘There’s an app for that’: expectations have been raised that government delivers its services digitally

• Many new mashups for mobile that put the government to shame e.g. Cyclestreets

2nd wave: the digital economy

• ‘Digital goods have different economic properties, notably, they have low marginal costs of distribution after the first copy’

• Companies (try to) protect digital goods and spread the cost over all purchasers

• When governments collect data to fulfill a public duty eg postcodes to deliver mail, any further releases can be at marginal distribution cost…

• Guardian ‘Free our data’ campaign

3rd wave: open source philosophies

• ‘the open source movement (has) beliefs (that) derive ultimately from principles of freedom, self help and an opposition to monopoly’

• OpenStreetMap is the classic open source response to government using crowd sourcing to create open alternative to traded map data

• Has ultimately led to Ordnance Survey free map data releases

4th wave: growth of the open society

• ‘a new generation has developed a paradigm of open social life that has confounded the “broad-casting” model of the 20th century’

• The rise of Facebook, Twitter and Foursquare create collaborative models of publishing and communication never previously envisaged

• Social media are in the process of re-shaping open societies eg social transport models

1700 Copyright

1800 Moral rights

1900 Patents Free trade Public libraries Privacy

1950

1960 Free markets Civil liberties

1970 Intellectual property Efficient markets FOI

1980 Standards GATT Online communities

1990 Data protection WIPO Public sector info

2000 Public domain Monopoly Open government Open society

2002 Digital Rights Man’t Walled gardens New Public Man’t Broadcasting

2004 P2P sharing Open source Blogging

2006 Creative Commons eCommerce Gov 2.0 Web 2.0

2008 Wikipedia Micropayments Gov as a platform

2010 Open licences “Free” models Communicative gov

Social media

2012

Pre-history of open data

How open data was released in UK

• PSI re-use requests• Freedom of info requests• Civil disobedience• Press campaigns• Political persuasion• Economic self-interest

Open data- what do we mean?• Released with an open licence• Updated at intervals, not single data dumps• Free of explicit and implicit costs• Based on publicly available specifications• No proprietary dependencies• Scalable distribution, not 1st come 1st served• Release is in the public interest

What data should be opened?• Any dataset with– a necessary public sector first purchaser– low marginal costs of distribution– no personal data content– no security implication

• Presumption of release... typical objections– What if we are blamed?– What if they use the data wrongly– What happens if our decisions are challenged

• Open data means “openness”• Open data means ending monopolies on usage

London DataStore

Open government reality

Web 2.0 Gov 1.1

Engagement with TfL, Met, LDA• Requests for data were met with incredulity• Followed by organisational denial at the top• Then middle management fight-back eg MyTfL• http://placr.co.uk/blog/2010/10/

journeyplanner_api/• The press got involved • Re-assertion of political authority• Pragmatic negotiations begin• First wave of data releases• Now TfL are an open data reference• Initiatives in PTEs+ Scotland/Wales

London DataStore

DataGM

Plan for open data releases…

• ‘Let not the best be the enemy of the good’• Start with actionable information: allow access

that can make something useful to most people• Follow on with datasets that drive transparency:

build public support for the programme• Then open the public sector “too difficult” box:

let someone else try to develop data• Publish the best: London DS inspirational uses

Mashups

Interactive guides

Public analytics

Feedback

Placr data

Open transport data in UK• Transport data is available to users– Live train & metro departure boards– National Traveline bus timetables– Traffic and accident data– Journey planners e.g. TfL– Reference data eg NaPTAN

• Some of this data not reusable– Developers stillscraping datasets– Publicly funded data mixed with private IPR e.g. NR

• New opportunities for transport

New services built on open data

• Independent performance statistics• Fare splitting to reduce prices• New visualisations of service opportunities• Commentary on ‘official’ versions of events– Re-ordering of journey planner options– Advice when service breaks down

• New service compositions– Driving, tolls, parking– Bus, train, taxi

Open data business models• Input costs low (open

data is free!)• Processing costs high

(need skilled labour)• Delivery costs low (cloud

computing, app stores)• Sales revenues are low

without v. high volumes• Closing the gap

– Freemium app offers– Ad-funded communities

around services– Data Store sales

Performance indicators

Implications

• Open data unleashes huge innovation• Challenges state monopolies to be efficient• Creates new opportunities for businessesBUT• Huge amount of work to clean up the data• Needs platforms for neutral data distribution• Needs engaging new business models• Needs ongoing political support

Contact

• Placr Ltd (http://www.placr.co.uk/• @MadProf on Twitter• Journal of Location Based Services

http://www.informaworld.com/jlbs