SLA Event (Part 1): Open Data: Powering the Information age

This is the first of a two-part post about a Society for Location Analysis (SLA) event last night (21st Feb). It was held in Hampshire County Council’s headquarters in Winchester, but the SLA – and its sponsors CBRE – kindly picked up the tab.

There were just under thirty present – enough to make it a ‘proper’ event, but sufficiently few to keep it nice and informal. Amongst our number were location specialists, information managers, researchers, techies and others just interested in a conversation blending geography, information, location and open data. Even the organisations were an unusual mix, including several councils, the Army, Police, Fire & Rescue, private consultancies, academia, and a couple of large retailers.

There were several excellent presentations which I couldn’t hope to do justice to, so I’m just going to share a few of what I felt were the highlights. Besides which, the slides should be online shortly, so you can see for yourself.

Introduction

Peter Sleight

The SLA chairman Peter Sleight introduced the event, explaining that 2012 is the tenth anniversary for the SLA, and that there’ll be ‘a bit of a do’ in April to celebrate. He also mentioned that this was the first time the SLA had a meeting with more attendees from the public sector than the private sector.

Peter went on to share his perspective on open data. I’m not  familiar enough with the background to explain the detail, but I think it’s probably fair to say that he’s pretty keen on opening up address data for businesses and feels that there’s still a lot more that can be done to remove barriers, particularly for smaller companies who can’t afford fees.

Peter then went on to introduce the next speaker, Professor Nigel Shadbolt from the University of Southampton

Open Data: Powering the Information Age

Professor Nigel Shadbolt

Nigel started by asking for a quick show of hands to find out:

• how many present were technologists (very few),
• who had heard Nigel speak before (very few)
• how many present use spatial analysis or spatial statistics (lots)

He then took a quick look at the background to data and suggested that “the guys at Google” are comparable to the great Victorian engineers. He rattled through examples and analogies far quicker than I could hope to keep up (I do so hope he shares his slides!)

Nigel loves data and loves its power – for example Search and click analysis. He noted that Google Trends was great, but that sometimes the best bits of products, having first established themselves, then seem to disappear (at least from open products)

He referred to Flu mapping in real time using powerful analytics as great examples, but emphasised the need to work with large corporates on their data as well – it’s not just the public sector which holds valuable public data.

Nigel then went on to talk about Flickr uploads shown on an openstreetmap, and showed an example of them plotted on a detailed map of London. He played the brief video clip of the OpenStreetMap – Project Haiti, which demonstrates just how much can be achieved in a very short time through crowd sourcing. In that example, people used open source software with GPS devices to crowd source a high definition map of an earthquake zone. That brilliant achievement was only possible through open standards, open licenses and open data

In 2009 Nigel and colleagues produced the Postcode Paper which they took to Cabinet – 80 percent of the data in the paper was technically illegally produced, even though it was  data with great public value. That helped to demonstrate why barriers and obstacles had to be removed to help open up new opportunities.

Detailed studies haven’t yet been published on the economic value that can be derived from open data. Two studies have been commissioned by Ordnance Survey that report on the economic impact of ordnance survey open data, but the conclusions haven’t yet been made publicly available.

Nigel made an interesting observation about the economics of open data – they needn’t be traditional, and that it’s more about elasticity, and how you generate value from it.

A few other quick points he made:

• Cities are very interesting as they are natural generators of coordinated data – there are fewer obstacles get in the way. (N.B. That reminded me that CityCamp Brighton is coming up 2nd-4th March)
• Civil servants and their local authority counterparts are trying very hard to get the data published – it takes lots of effort!
• The unrestrictive license is the most important thing about open data – the open government license is the “crowning glory” – Nigel suggested always aiming to make data available under the Open Government Licence (OGL) wherever possible
• Linked data – good example in Hampshire already with ordnance survey creating 5 star data that can be linked to. (N.B. John Goodwin has done lots of work on that, much of it in his own time).
• Nigel wants a Companies House URI
• Data will have a long tail – data that will be pointed to that will help make sense of all the other data
• Even if you decide not to publish data as open data, allow the reference points in your data to be the same as the reference points within public data. Then, should you change your mind in the future, it will be much easier to publish as open data.
• Significant new data releases, e.g. Weather, transport and health The freeing up of weather data can free up a secondary market  (it’s a big secondary market in the US)
• The location of the post office tower was once an official secret, even though it was plainly visible!
• Find your post box – crowd sourced accurate data through people visiting postboxes with their smartphone and recording box number and GPS coordinates
• Public services will increasingly crowd source the data we can’t produce ourselves.
• Citizen science is happening everywhere People taking old nautical reports and entering them in to databases.
• Data probably has a long tail – just because relatively few people use a particular data set doesn’t mean it doesn’t have lots of value
• Midata – the next battle round will be consumer data – making people’s own data open to themselves
• New economic circumstances can be brought about by opening up individual’s data (to themselves)
• There is a range of data from Big Data (a term which Nigel hates) through to very small data. Most of it will be somewhere in the middle. A spreadsheet can tell you a lot if it’s the right data.
• Not all data fits along one dimension
• Open data can be a cheaper way to deliver services – you can achieve the end in a different way
• The “sainted Berners-Lee” said: “The value of government data is in its widespread use, not in its sale”
• People find the stuff that’s interesting, and they link to it
• CEOs are starting to see the benefit of open data for themselves Spending data has become a source of inspiration for CEOs
• There is a lot of innovation going on with open data, it’s just not that visible yet

That’s all for this week (I’ve run out of steam for tonight)

In the second half of this post I will cover the other two excellent presentations given by Ian Holt and Chris Parker (aka @Geovation).