Data & Stuff // Neil Houston

For those who aren’t aware what cablegate is, then in brief it’s a set of diplomatic transmissions.

The cables, which date from 1966 up until the end of February this year, contain confidential communications between 274 embassies in countries throughout the world and the State Department in Washington DC. 15,652 of the cables are classified Secret.

The Wikileaks team had performed some analysis of the data. Around where the cables had been sent from, and what classification given to them. As well as looking at what the most popular ‘tags’ for the cables were. Unfortunately their analysis that they placed on Tableau Public was removed, see here for the Tableau Statement.

This seems a bit odd, considering the analysis was at a summary and statistical level. For instance, there were X number cables sent by Y. Though not revealing the actual content of them, which is obviously in some cases secret, or classified. This is in contrast to the analysis I performed of the Wikileaks Afghanistan War Diaries. Which shows both a summary level, and further detail about the confidential files. This analysis is also hosted on Tableau Public.

So take a peak at the analysis I performed and placed on Tableau Public (the links click through to the Tableau Public versions).

Cables by Origin

Tags Used

I’m a user of Tableau Desktop, a product by Tableau Software – it’s a great versatile tool for visual analytics, from getting a quick feeling of data trends right through to versatile summaries, and complex graphs. It’s always had a catch though, the pricepoint. [$999 or $1800].

It’s been used to produce some pieces of analysis on this blog, and I use it regularly for work. What’s great is that alongside their latest commercial release, is that they unveiled Tableau Public. This is a free version, that allows you to download the software, create your analysis and deploy it to their web servers.

You can then grab the embed code, and display visualisations on your blog; which people can then interact with. This platform takes it one step further than just a static image. It’s on par with IBM ManyEyes, but doesn’t use java and, to me at least, is much easier and intuitive to use.

The product has the same feature set as the commercial version, with the only limitations being a 100,000 row limit and restricted data sources.

So for those who have chastised me for using ‘expensive’ software before, don’t worry – you can have a go with one of the latest challengers on the block, as according to Gartner.

P.S. There is an obvious lack of a Tableau Public embedded chart in this post, that is because it does not support the Apple platform. Damn. Instead, here is one they produced: