Many-Eyes: Data visualization for Non-Scientists
Last Thursday, I attended one of the ongoing Parc Forum speaker series events in Palo Alto. The presenting duo of Martin Wattenberg & Fernanda Viegas hail from IBM Visual Research and the technology they were showing off is Many Eyes. The concept behind Many Eyes is that advanced data analysis shouldn’t be restricted to the scientific community. Any and all people should be able to support their hypothesis and opinions with visual aids that enhance understanding. This philosophy fits in perfectly with the blogosphere and shift away from mainstream media as a news outlet. Now, Joe blogger, can post about the correlation of fossil fuel pollution and population density and back up his view with pollution data number combined with land mass, all overlayed onto a world map.
After the jump – Sample Analysis of the CNN Democratic debate & Macworld 2008 Keynote address
If “A picture is worth a thousand words”, an interactive picture’s worth in words must compare to the works of Tolstoy. The ability to easily group and explore data, whether it is tabular or text, does open up interesting new insights and make concepts much easier to comprehend. My favorite visualizations on the site, which was also highlighted in presentation, are the text analysis. There are two options, tag cloud or word tree. Tag clouds are ubiquitous on the web, so while this analysis isn’t revolutionary visually, it does a great job of summarizing long bits of text such as political speeches or testimonies. In fact, when Martin presented this, the examples he highlighted were the 2008 State of the Union speech by President Bush and the testimony of Alberto Gonzalez to the Senate. It is extremely interesting to see the number of occurrences and the word relationships throughout a long dialogue.
In addition to the technology, the most commendable feature of Many Eyes is their commiment to providing the technology for free to the public. The development is completely funded by IBM, but it’s creators have the mindset that this should be freely available to anyone and everyone. They’ve managed to convince their employer to keep it that way for now. Of course in return, IBM gets the benefit of a better product by letting the entire internet try their hand at using it and doing better testing/QA than any single company could do. As Martin and Fernanda hinted, this technology will also likely show up in future releases of IBM software. From personal experience, Lotus Notes needs all the help it can get.
Some future capabilities that were discussed were the ability to do visualizations of real-time data, and possibly opening up the API. As it stands, Many Eyes has the startings of an open-source project, but for now it’s still tightly controlled. Who knows, maybe IBM will take the final steps and release the source code for all to use and contribute. Either way, this is still a great technology and I encourage everyone to check it out and start making their points visually as well as verbally.
Screenshot Examples
To evaluate Many Eyes’ useage myself, I uploaded a few data sets and created some new visualizations. There are a lot of great examples on the site, but I wanted to get a feel for the process. I chose the last CNN Democratic debate, held on Jan 31st. and the Macworld 2008 Keynote Address. You can, in theory, embed the live visualizations into your site, but I wasn’t able to easily make it work here on WordPress.
No Comments Yet
Be the first to comment!


