Five short links

Photo by Andreas

Mapumental – A really useful visualization that promises to help you find the ideal place to live, letting you interact and question the data in an intuitive way. This is exactly why I built OpenHeatMap, an interactive data map opens up so many possibilities. Now I’m eagerly awaiting the live product.

Guide to finding a technical co-founder – Vinicius’s guide rings very true to me, and most of his points also apply to finding a business co-founder if you’re technical. The short story is, it’s hard. Really hard. I spent a lot of time and effort on the hunt, and still failed. It will help a lot if you’re flexible about what you want to work on, then you might get lucky at co-founder dating events or via other random connections. My biggest regret is that I didn’t set myself a deadline, I kept at it long after I should have given up on the search and either dropped my idea or focused on getting traction instead (which is what I eventually ended up doing).

7,000 Neighborhood shapefiles – I only just came across this resource from Zillow, it’s incredibly handy who wants to talk a geographic language that users understand. We all talk about neighborhoods in our daily life, but this is the only open resource I’ve found that gives you definitions you can use.

Who is the greatest diva of the last 25 years? – I love this mostly because of the brutally honest description of their methodology (“We used computers!”). Truth be told, most of the visualizations you’ll see on the web don’t have much more behind them either, including some of mine on a bad day. Always try to look beyond the veneer of authority that charts and infographics give to any argument.

Twitter contour maps – Talking of substance before style, here’s a spartan visualization with some great data and ideas behind it. I’m looking forward to the next version of Twitterplaces from Ben, he’s digging deep in some interesting territory.

OpenHeatMap now visualizes Canada and Mexico


After the UK, the most-requested countries for OpenHeatMap have been Canada and Mexico, so I'm pleased to launch state/province level coverage for both of them. All you need to do is create either a Canadian province or Mexican state column in your spreadsheet and I'll take care of the rest!

Unfortunately I haven't been able to find more detailed information for either country, I'd love to find electoral boundaries for both, but I'll keep looking. I also have leads on Turkey, Australia, New Zealand, Germany, Austria, Switzerland and Ireland, so I hope I can cover even more of the world.

If you don't want a thematic map, and are happy with blobby heatmaps instead of filled polygons, you can already use OpenHeatMap with international street addresses, just put them in an address column.


OpenHeatMap now visualizes the UK

I've had a lot of requests from friends in the old country wanting to use OpenHeatMap to display data about areas in the United Kingdom. After quite some time wrestling with the Ordnance Survey's recent release of open data I'm pleased to announce that you can now visualize everything from counties to constituencies to council districts! To celebrate, I've created a visualization of the recent election, showing the percentage of the vote that each party received in every constituency:


To go along with it, I've put together a quick 'making of' video tutorial, showing how it only takes a couple of minutes to transform the Guardian's spreadsheet of election data into an interactive map:

I just want to say a big thanks to everyone who helped me gather this data, especially Matthew Somerville of MySociety who went above and beyond to help me understand counties, Simon Rogers for making the election data freely available, along with Cameron Neylon, Colin McCrory, Chris McCray, Leigh Dodds, Bill Roberts, John Goodwin, Richard Stirling and Laura Oliver.

Do you wish you could visualize data about regions of your country? Drop me an email and I'll see what I can do (and maybe rope you in to investigate if your government makes the data openly available!)

Build a map, win an iPad!

Photo by Foxy Coxy

Did I mention how much I like the folks over at This Week in Relevance? They've just launched a competition where the best OpenHeatMap wins an iPad! I love that idea so much, I'm going to give a free t-shirt to everyone who enters too.

If you're looking for inspiration, just today I've helped people build maps of UEFA cup rankings, amphibian habitats, oil well starts, college alumni, and even someone's friends on Facebook, and that was without the lure of a lovely Apple gadget at the end.

As always, I'll be online as much as I can and available to chat via the awesome SnapABug (I had four simultaneous conversations going at one point this afternoon), or you can email me via or IM me at petewarden on Skype.

Where are LA’s most effective schools?

After seeing the LA Times analysis of how effective local teachers are at improving their students performance, I was left wondering how the different neighborhoods performed. School rankings based on pure academic achievement aren't that informative since they correlate very closely with the wealth or poverty of their students family, but the 'value-added' approach seemed like it might produce more informative patterns. Here's the map I built, using the Times' ranking of schools from one to five, where dark blue marks the most effective elementary schools:

As always, here's the code I used to gather and format the data to load into OpenHeatMap:

There's an obvious cluster of under-achieving schools between Culver City and downtown LA, but I was pleased to see some pretty poor areas like Compton showing some effective schools. Northridge seems to have a lot of great schools too, along with Culver City and Santa Monica, though there are a few surprises with 'one' rated schools tucked in amongst the rest.