Nestoria Blog

A few months ago we featured London Profiler, a website created by researchers at University College London. For this month’s Nestoria interview we have the pleasure of talking with Dr Alex Singleton, a research fellow at UCL’s Department of Geography and Centre for Advanced Spatial Analysis about London Profiler and the various other projects the department does.

Alex, thanks for taking the time to speak with us.

1. What is the Centre for Advanced Spatial Analysis and what are the types of projects you work on?
The Centre for Advanced Spatial Analysis (CASA) is an initiative within University College London to develop emerging computer technologies in several disciplines which deal with geography, space, location, and the built environment. Researchers at CASA are involved in a broad range of research activities including developing models and simulations of urban areas, pedestrian movement, implementing 3D Globes, the geography and ethnicity of names and multiple projects around geographic Web 2.0 type technologies.

My A-Level geography teacher once described the discipline of Geography sitting at the “hub of the wheel of knowledge”, and although an undoubtedly biased view, this conceptualisation has flowed through into my personal research activities which tend to cross an interdisciplinary range of areas. However, broadly as a human geographer I am interested in how new data sources and technology can be used to help us visualise how the world looks, and from this information, how we can build better models of how the world works.

2. What’s new with London Profiler? What was the goal of the project and what sorts of feedback have you had from users?
The growth of Google Map type technology, and the ability to re-use these interfaces and data in third party systems has transformed Geographic Information Science; opening up the visualisation of how the world looks to a plethora of new users. However, despite this improved access to general map and satellite imagery, there are still lots of really interesting local information about the neighbourhoods in which we live which are not available through these new interfaces. This is really the crux of the research problem which I have been working on in the London Profiler project.

The London Profiler website is a repository of data about London Neighbourhoods, covering a huge variety of different domains such as health, crime, ethnicity, deprivation and education. The majority of the data we display are in the public domain, however, we have simply improved their presentation. Using technology developed in CASA we are able to transform any spatially referenced data into the tile based system used by google maps for their background (map or satellite) information. Our data is presented as another layer on top of the Google background maps, which we can fade in or out as required.

The website also provides the ability to add your own data on top of our maps from any publicly hosted KML file. For example, the KML feeds which Nestoria provide have enabled us to create a basic property search function on our site. This perfectly demonstrates how synergies between different data providers using web 2.0 methods of sharing and reusing information can create innovative new services. For example, if I were looking to buy a house; by adding the deprivation data on London Profiler and then searching for properties for sale, I could examine how the price of properties for sale might change depending on how deprived an area was. For people moving into London, who are not familiar with the local geography, this provides a very powerful tool.

We still consider the London Profiler site to be very much in beta, and in the future we have a range of plans including the addition of more data, testing alternate mapping technology and expanding the coverage’s outside of London - A national profiler would be very exciting!

3. What can sites like Nestoria do to better

• cooperate with academic researchers?
  Supplying an open API is a great way to start building interest, however this is only useful to academics if they actually know about it. Therefore I think co-operation could be improved, firstly through promotional efforts to develop your brand awareness within academia about the types of services, but more specifically data, you can offer to academics for use in their research. This may involve coming along to some academic conferences, or perhaps delivering joint papers. Secondly, an important way in which we have conducted much of our previous knowledge transfer is through the joint funding of PhD students working on collaborative projects. These activities can take a variety of forms and offer a good foot in the door of many leading academic institutions.
• make our data available for academic projects?
  My co-researchers and I have been impressed by the data we have accessed through the Nestoria API. This model of access is very different from how those nationally distributed data academics normally use are released, and in reality this places Nestoria well ahead of the game. It would be great if we could see census or some of our other prevalently used data disseminated using similar technology.

  I would not want to imply that there was anything wrong with the current Nestoria offering, it is very good. However, as a geographer I might look at how aggregations of the data you collect could be gathered across a number of spatial units (e.g. Wards, Output Areas), perhaps with the ability to query these information temporally. This could be really useful.

4. What are the challenges you think a vertical search engine for property like Nestoria faces?
It will be interesting to see how business such as this scale, as unlike those recently publicised web 2.0 successes (e.g. Facebook) where the value is linked to the size and interaction of membership (which could grow indefinitely), a vertical search engine’s value is based on the volume of relevant data you can collect, and the frequency of clicks through this information you can garner.

The first challenge with this model is that information collected is theoretically capped at a maximum as there could (in the case of Nestoria) be a situation where you capture all property for sale within a country, thus limiting any further growth through data capture alone.

A second challenge lies in how you get the information collected out to your potential users, so they can click through and generate your advertising income. This is slightly more challenging and requires a good understanding of your potential customer base. The provision of API and those other dissemination routes such as Google Widgets / Facebook integration are excellent examples of how this can be achieved.

Finally in terms of risk, it appears to me that the vertical search engine business models are perhaps too dependent on advertising revenue derived through click through “adverts”. I suspect that to mitigate some of this risk in the future it will involve the development of new income generation streams outside of offering consumers a free service. The data collected by Nestoria undoubtedly has great geographic value and there are multiple ways in which a range of corporate services could be created without undermining the ethos of the current Nestoria model.

Great answers Alex, thanks for offering us your insights. For those that haven’t yet seen it I strongly recommend checking out London Profiler, it is an amazing tool.

Alex is correct that there is a limit to the number of houses on the market, and in the UK and Spain in most regions we are well past that limit. A bigger technical challenge we now face is correctly identifying duplicates as we now very often have the same listings from multiple sources. We hope our recent release of embeddable historic house price charts and support for the hListing microformat are just two examples of us trying to release our data in a more usable format. This is an ongoing project and one we hope to have more to report on in the future.

We’ll keep refining the data quality and user experience. We look forward to seeing what new innovations academic groups like CASA present.

past Nestoria interviews: Tom Steinberg, Lelia Ferro, Lloyd Shepherd.