Oddly, Economist readers like data
The seventh in the series on the ideas we received
As I’ve written previously, I think data is fun. Understandably a lot of the submissions revolve around data, and the possibilities in this area are massive. With a ton of the stuff in Economist Intelligence Unit (EIU) vaults and a huge community of economists, what couldn’t we do? In the main, the ideas submitted revolve around three issues :
- make data free
- make it interactive
- show the links between data
“Information wants to be free” is a commonly held tenet by webbies. The expression was coined by Stewart Brand at the first Hackers Conference in 1984. It is worth reading the entire quotation :
On the one hand information wants to be expensive, because it’s so valuable. The right information in the right place just changes your life. On the other hand, information wants to be free, because the cost of getting it out is getting lower and lower all the time. So you have these two fighting against each other.
The argument goes that due to the elimination of distribution costs by the web we should now distribute this data for free. This would be a valuable service due to the EIU’s enormous info stash. This information is current and historical, giving us the opportunity to give away feeds of old data sets and up-to-the-minute info.
Giving away the data for free is the first thread of ideas, but the key is to make it interactive. An EIU data API would then put the tools in the hands of our users who would be able to create all manner of wonderful graphs, tables and visualisations. (What’s an API? An API is data that can be reused by developers. Companies give away their data in publicly usable APIs thinking that the ideas of the public at large will be more innovative than their own). By opening the data up to the collective intelligence of Economist Groupers, we’d definitely get some brilliant insights.
The value of this type of service can be seen in the sites Swivel and Many Eyes which share data sets and then give readers the chance to create interesting visualisations. Here are some of my favourites on Many Eyes:
- 100 top Bible Characters by Frequency and Dispersion
- CO2 Emissions in 2003
- Distribution of Guitar Shops by US State
The best headline we had from this set of ideas was Give The World Your Curves. Naturally, he was referring to the curves of the graphs in the Economic Indicators section of The Economist, and suggesting that they should be customisable on the web. This is clearly an idea that just makes sense. I remember a Canadian telling me how annoyed he was that he never saw Canada listed in any of our tables. The magazine has also dropped one page from this section, annoying numbers freaks. That wouldn’t happen at Curves Online because it costs nothing to have an extra page (that might not be the best name for the section).
The last group of ideas concentrated on the idea of showing the links between data. Super Size the Big Mac Index (another great name) wanted to use the fast food industry as a prism through which to go deeper into society. After extreme weather damaged the US tomato crop in 2005, tomatoes were removed from standard burgers. The idea leant itself to examinations of unionism, environmental issues and agro-industrialisation.
This deeper insight was in contrast to Paul Pedley’s (disclosure: Paul works for the EIU) desire for us to save him time, by quickly showing the inter-relationships between data. Information overload is a messy reality for any number of data professionals so a page that visualised these links would save time.
Naturally the issue here is that the EIU charges money for the data so why make it free? There are a few ways of looking at this. The easiest way would be to liberate some of the data. This could be toyed with by developers and manipulated in an interesting fashion and, by spreading this out, we’d in fact be advertising the EIU. More radical, though, would be turning a business with a few subscribers into one with many more monetised customers.
There are many ideas around data but here’s the nub: the tools that web developers now have in their hands mean that data can be freed from the two dimensional representations that we have today. Multimedia tools, from Flash to Google Maps, mean that data can be turned into compelling stories. Whether we do anything on this during Project Red Stripe remains to be seen, but I think The Economist Group should take advantage of this, and quickly.
(In addition to those mentioned in this post, we would like to thank all those who have contributed suggestions related to data.)
———————————–
Update : Show’s how much I know. It turns out an API is not the data from a site, but a way of accessing the data from a site. This shows not only my ignorance, but also that you don’t have to know all the technical stuff to love the web. The technology’s not the exciting stuff, the ideas that it enables are.
April 4th, 2007 at 10:26 pm
Excellent post. As a long-time fan of the Big Mac index, I am thrilled to see the Economist engage in the debate about open access and numbers - where they came from, what they are good for, who might want to use them and how.
I have always wanted to do more with data than see a chart or read an analysis - I want to see what new and exciting insight I might discover on my own. As the web evolves in capability and users become more savvy, the possibilities are endless. As you rightfully point out, technology’s not the exciting stuff, the ideas that enable it are.
Looking forward to see what you come up with! Especially if it includes an open API
Sara Wood,
http://www.swivel.com
April 6th, 2007 at 6:18 am
Troubling entry. However, funny if it was meant to be…
April 11th, 2007 at 12:58 pm
I like the technology as well.
April 12th, 2007 at 5:26 pm
Visualization is thinking small! Artificial Intelligence is the wave of the future.
The key is to develop a way to ensure the credibility of your data and provide a feed into systems which want to react real time to changing conditions. Automated business systems (with various checks, automated and human) could be altering their hiring, marketing, and production plans all based on latest data news releases.
Think about it!
April 12th, 2007 at 7:37 pm
On Free Data
One of the trade offs on data is that you don’t know if you need it until you have it.
Here are some ways that you can give partial access to data so that people can see if they need it enough to buy it.
1. All data is free part of the time:
1A. On odd days of the year, it’s available to IP’s that end in an odd digit. This can be generalized to any modulus.
1D. All data is free, but paid subscribers get their own server, with much faster access.
1E. The first N accesses in a given day are free, after that it costs. (N chosen to correspond to about an hour’s worth of data mining.) This allows student researchers and casual researchers to have good access, but means that the professions will pay to get continuous access.
2. Some data is free all of the time.
2A. Current data is free. Historical data costs. Or the reverse.
2B. Raw data is free. Processed data costs.
2C. Data for X is free, for the world costs. (X being my country, my province, my municipality.)
2D. Summary data is free. Detailed data costs.
3. All data is free all of the time.
3A. The sponsoring organization gains credibility/status.
3B. The sponsoring organization sells advertising.
3C. The sponsoring organization offers consulting services to help understand the data.
April 13th, 2007 at 10:28 am
Thanks for all this information guys. Very interesting models for making data available.
April 23rd, 2007 at 8:07 pm
Gapminder:
http://www.gapminder.org/downloads/applications/
has lots of interesting data and tools for visulising it.
Maybe you should do some mata-data layer - a mechanisim for websites to llist and describe the datasets they have so that different websites can visualise data from each other.