UK Web Focus (Brian Kelly)

Innovation and best practices for the Web

Thoughts on Wolfram|Alpha Personal Analytics for Facebook

Posted by Brian Kelly on 5 Sep 2012

Recent News: Wolfram|Alpha Personal Analytics for Facebook

Tony Hirst alerted me to the recent post on Wolfram|Alpha Personal Analytics for Facebook. Facebook is, of course, one of those services which generates strong opinions, rather as Microsoft used to do. In the case of Microsoft the criticisms have been centred around its proprietary file formats and its misuse of its dominance in the desktop computer environment. For Facebook, the criticisms have focussed on Facebook being a “walled garden” and its misuse of personal data.

Facebook Was a Walled Garden

It was back in 1993 when Novell claimed that Microsoft was blocking its competitors out of the market through anti-competitive practices. However as described in Wikipedia the European Union Microsoft competition case resulted in the EU ordering Microsoft to divulge certain information about its server products and release a version of Microsoft Windows without Windows Media Player, in addition to paying a fine of £381 million. Microsoft also eventually migrated its proprietary file format to XML and the Open Office XML format which became an ISO standard in December 2006.

Might we see similar changes happening with Facebook? Back in December 2008 I asked Just What Is A “Walled Garden”? – a post which generated interesting discussion on the pros and cons of walled gardens, with Ben Toth commenting:

I don’t like the phrase at all. Firstly it’s one of those phrases which gives the impression of being meaningful but in practice doesn’t bear too much analysis. Secondly, walled gardens were a pretty clever Victorian technology for creating micro-climates in order to boost food production (, so it seems a shame to use the term in a negative way. Finally, all gardens have walls of one sort or another – an un-walled garden wouldn’t be a garden. So the phrase is a tautology.

Max Norton concluded the discussion by observing that:

to leap to judgement just because something can be described as a walled garden is hasty. While my instinct is towards openness I try to be pragmatic about these things and where I feel there are gains to be had in using “walled garden” solutions I’ll use them.

A willingness to accept the benefits that can be provided by walled gardens can clearly be seen by fans of Apple products, with, as described by the Wikipedia entry for Walled Garden (technology) Apple’s iOS devices are “restricted to running pre-approved applications from a digital distribution service“.

In October 2010 I pointed out that Planet Facebook Becomes Less of a Walled Garden following the announcement that “Facebook lets users download data, create groups“; news that was welcomed as “A step in the right direction, by the vice-chair of the DataPortability Project“.

Back in September 2011 ZDNet published an article which provided an update on Facebook’s export options and argued that Facebook finally makes your exported data useful. Since there are also tools such as SafeGuard which enable you to export data from Facebook and other social networking services it seems that we can say that not only can a walled garden provide a safe managed environment, but that it would be wrong to describe Facebook as a walled environment.

Accessing Facebook Activity Data

There are now a number of ways of migrating one’s personal data from Facebook. Facebook provide advice on how to do this, and this approach has been described in an article published in C|net. Meanwhile applications such as Social Safe provide alternative ways of accessing one’s Facebook data – and I learnt that I updated my Facebook profile picture on 13 December 2007.

But it was Tony Hirst’s tweet which interested me that most, since the Wolfram|Alpha service goes beyond the simple exporting of one’s content (status updates and images and videos which have been uploaded) and provides information and visualisations of one’s activity data.

Figure 1: Facebook activities, by time and day of week

Once you have given permission to the Wolfram|Alpha app to access your Facebook data visualisations of how you use Facebook are provided, such as the day of the week and time of posting status updates, posting links or uploading images. As shown in Figure 1 it seems that I tend to use Facebook mostly between 6pm and 9pm, which is not unexpected as I use it primarily for social purposes.

Figure 2: Facebook apps used

Figure 2 shows the Facebook apps which I use. It seems that the one I use most is the app which provides an automated status update when I publish a new post on this blog.

This information simply gives me a better understanding of my use of Facebook. This personal understanding of one’s Twitter use was the angle taken in a post on the Mashable tech blog which described how This App Knows More About Your Facebook Account Than You Do.

Figure 3: Visualisation of my Facebook community

However of greater interest to me is the way in which the Wolfram|Alpha app provides a visualisation of my Facebook community and the connections between the members of the community.

In Figure 3 you can see the various communities, which includes my sword dancing and folk communities and my profession contacts. I can also see the various outliers, of people who have few connections with others, which includes the landlady of a pub I often visit.

Such visualisation of one’s connections will be familiar to anyone who keeps an eye on Tony Hirst’s work in this area. In the past Tony has made use of Twitter APIs in order to visualise the growth and development of Twitter connections, including connections based around an event hashtag.

Facebook and Twitter Social Graphs

Assuming that you are willing to trust Wolfram|Alpha, their Facebook app may be of interest to anyone who would like to gain a better understanding of their own use of Facebook – as well as understanding what Facebook may know about you. Apart from the automated updates when I publish a new blog post, I update my Facebook status in the evening, often when I’m listening to live music in a local pub. Being able to process such information in an automated and global way will be of interest to the service providers who are looking to optimise targetted advertising.

Beyond the individual’s interest in such tools, clearly of greater interest will be developments around the global social graphs provided by Facebook, Twitter and, to a lesser extent, Google+

Tony Hirst has addressed this issue recently when he asked Is Twitter Starting to Make a Grab for the Interest Graph? As Tony pointed out:

If targeted advertising is Twitter’s money play, then it’s obviously in their interest to keep hold of the data juice that lets them define audiences by interest. Which is to say, they need to keep the structure of the graph as closed as possible.

Will Twitter’s increased control over their APIs mean that there will be less opportunity for developers such as Tony Hirst (and Martin Hawksey with his developments based on processing the Twitter data stream) to continue their work which helps to provide a better understanding of how social networks are being used to enhance teaching and learning and research activities? And will, ironically, we find that Facebook provides a more open environment for such work?

NOTE: Following publication of this post Tony Hirst informed me of his posts on Getting Started With The Gephi Network Visualisation App – My Facebook Network, Part I and Social Interest Positioning – Visualising Facebook Friends’ Likes With Data Grabbed Using Google Refine which described his experiments in analysing and visualising Facebook data.

Twitter conversation from Topsy: [View]


2 Responses to “Thoughts on Wolfram|Alpha Personal Analytics for Facebook”

  1. Tony Hirst said

    Hi Brian – thanks for all the plugs..;-)

    For folk wanting to play with raw Facebook data, here are a couple of hacks:
    – visualise/analyse how your friends connect on Facebook
    – visualise/analyse what your friends Like on facebook

    It’s worth pointing out that for the social network analysis, Facebook will only give you data relating to how your friends connect to each other, whereas Twitter lets you (but not a grwoing list of other social networks – ) grab a list of IDs of the all the friends/followers of any public Twitter account. (Actually, Facebook also lets you test if two specified individuals follow each other, but API rate limits limit how many connections you can test… LinkedIn InMaps show a map of how your friends on LinkedIn connect, but I don’t think you can get hold of the actual friendship connection data?

  2. Thanks Brian (and Tony) – didn’t know about this.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: