The so-called Dunbar’s number was prominently featured in the book The Tipping Point and is basically the average size of a social group that is still somehow manageable for a human. Or to put it simply: Your brain can just deal with about 150 meaningful relationships. While the number has caught on, it is more a statement about how much social complexity the brain can deal with.

Once he and his team came across this number and were looking for confirmation, they found a number of examples in which something around 150 seemed to be a recurring pattern for manageable group sizes of people. His examples included:

• the average village size in the Domesday Book
• the ideal size for church congregations
• the common size of the smallest self-sustainable militar unit, the company
• the maximum number of guild members in some online games
• the religious community of the Hutterites start a new colony when it grows above 150
• Apparently Gore Associates, the makers of Gore-Tex, start a new factory once one has more then 150 workers

Now, the crucial point is obviously what meaningful relationships are and here the definition becomes a bit more hazy. He offered several estimations:

• people you can go to in moments of great distress
• people you to trust and one which you could rely
• people of which you know immediately the position in your overall social network

Within these group size of about 150 they also found a hierarchical structure of friendship which basically means different layers of friendship, from very close friends to more weak ties with people you communicate with less often. Interestingly they observed a scaling factor of three, which means 5 intimate friends (support), 15 close friends (sympathy), around 50 in the next layer and then 150 which according to Dunbar denotes the inner circle of one’s relationships.

They are unsure about the reasons but possible explanations could be either the limits in brain capacity or simply the time requirements of managing relationships. He emphasized that their postulated relation between brain size and group size is only valid for primates because of the pair-bonding behaviour of those species. Apparently it takes a pretty large brain to solve the cognitively demanding task of finding a life-long partner.

Now while this was a really interesting and entertaining talk it left me wondering about what makes good research as well as what makes popular research. Not necessarily assuming that this is a zero-sum game, ie. research that is popular cannot be thorough and vice versa, I was at times struck by how vague the definitions were (e.g. of relationships), how much variation there is in the measured group size and how little justice the fixation on a group size of 150 really does Dunbar’s theory. And yet, it has caught on in the public mind.

Apart from the fact that this topic of relationships is stricking a chord in people I also think it is because Professor Dunbar knows how to make his research appealing to the broader public (see for example his article “Tall men have more reproductive success“). However, I do have my doubts about some of the conclusion drawn from the underlying research. For example, is it really insightful to study 43 individuals and whom they send Christmas cards to? Despite the huge variation of the results and the small sample it is used as yet another pillar that supports the overall theory.

Apart from pondering about what makes successful research – both with peers as well as the public – the obvious question for me is how technologies might change Dunbar’s number. Surely we could assume that for example social network sites help us deal more efficiently with the social complexities of our lifes. There has been a bit of work done for online communities.

Professor Dunbar expects that the general hypothesis is still true but that technology might slow down the decay of relationships. However, he is just in the process of collecting data on this topic with the DETESS project (Developing Theory for Evolving Socio-cognitive Systems) in which they for example equipped final year A-level students with free mobile phones to track their social networks (whom they call and text) in the transition period from school to university.

For now I believe that this theory merits a close re-examination, both for its general validity as well as for its applicability to the network society age. If we could come up with a useful experimental design this could be as ground-breaking a study as Duncan Watts re-examination of the Milgram’s small world experiments. Social network sites seem to be an obvious candidate for such a study but it hinges on the proper definition of relationships.

Let me know if you would be interested in brainstorming together – maybe we could offer some new insights!

In his view advantages of online dating are:

• extends range of potential partners
• cover every niche
• cheaper than inviting people for dinner
• avoid social awkwardness
• convenient 24/7 (match.com is used throughout the day and night but 12-2 seems popular, 7-9 subscription, Sunday most popular)
• intellectual connection over physical attraction
• on paid sites: policing

problems of online dating are:

• social stigma (but decreasing)
• technology is de-humanising
• security concerns (mainly personal data protection)
• overall market just £100m but 10m singles in UK (rising)

Jason argued that maybe online dating might be seen as a return to more traditional ways of dating instead of the current culture of depending on luck and alcohol for getting together. What is more, undoubtedly culture is changing towards more single lifes although biology (the instinct to reproduce) does not change that much. To summarize his argument, Jason basically believes that in the near future (by 2015) online dating will be the normal way of dating.

One interesting idea he had for future development would be a meta-search engine for dating. Also there is an interesting conflict for the future development of those sites. Their main value is to be able to choose from as many people as possible so this would favour a monopoly in a winner takes all situation.

He mentioned that they are also quite keen to sponsor research in this area (i.e. on their data) so if that rings a bell with you, get in touch with him.

I was quite satisfied with the response so far and it has also really been useful to connect to other people that do similar work. Social network sites as well as geography have been some important topics here. It has been especially enlightening to listen to Mike Thelwall’s analysis of 15,000 MySpace profiles that did help me to qualify some of my findings (such as that girls make better friends and older users have less friends). There are some other insights in his paper so you should check it out on Mike’s website.

There has been quite some interest in this kind of research and I just wanted to highlight a couple of really useful tools for analysis of MySpace and other network sites. First of all, for MySpace I am helping to develop a Perl module (WWW::Myspace) that actually acts as an API that interfaces with MySpace (via screen scraping) to provide access to the (public!) data on people’s profiles. It is now working rather well and can give you friends, comments, location information etc. If you want to give it a quick look, check my simple WWW::Myspace trial.

If you are not familiar to programming, you might want to have a look at dapper.net. It lets you select the sections of a web page that are relevant to you, extracts data and exports it into a format of your choice. I haven’t used it myself so far but it seems really useful.

If you have any comments or question, please leave a comment on this blog or send me an email!

“Last but not least, the subject under public scrutiny does matter as professors might well be made to face personal criticism in their role as public figures while teachers and nurses might have to be treated differently.”

Well, I might have been wrong. A couple of weeks ago the Regional Court of Cologne has decided that teachers too have to tolerate to be rated online by their pupils. In the case a German teacher had sued spickmich.de, a site similar to the British RateMyTeachers, that allows pupils to mark their teachers in nine different categories such as fairness of grades, quality of the lessons and difficulty of exams. So far more than 150.000 kids have used this service (according to the publishers).

The court made two important rulings: First, the ratings of the pupils are clear value judgements and are acceptable under the principles of freedom of expression as long as these are not defamatory. The court has given a lot of freedom and even allows polemic statements as the criticism of the pupils has the necessary relevance (after all, they are talking about people they meet and work with every day) and the evaluation of the school and its teachers is of general interest. Second, the judges did not follow the argument of the claimant that displaying her name next to the rating would constitute a violation of her privacy. They argue that this is information that could be easily obtained from the school’s website.

While this is an important ruling in favour of collaborative group judgements (something my colleague Malte Ziewitz is working on) there are a number of particularities about spickmich.de that merit highlighting to assess the ruling properly. First, there is a need to register on the site. The ratings are not publicly available and pupils have to register on the site in order to rate their teachers. Another difference to other rating sites is that the service does not allow for personal comments about teachers but only for grades between one and six (like the German grading system) in different categories. Finally, looking at the screenshots of the site it seems that what spickmich.de is really after is to establish a new social networking site for pupils that piggybacks on teacher ratings to reach a critical mass. It is telling that spickmich.de refers to itself as a pupil’s network.

Bottom line: I think this has been a good ruling in general as collaborative efforts with general interest goals should be encouraged and not stifled by the law as long as they do not harm anyone. Still I am slightly worried. The creators of the site do not get tired to emphasize that overall marks received by the teachers are rather good (average of 2.9) and that the site is intended to collect fair and objective judgements in order to improve education. But despite their arguably good intentions it remains the secret of the creators of the site how to reach a “fair and objective” judgement in categories such as sexy (or ugly), cool and funny (or embarrassing and boring) or easy (or difficult) exams.

While the association of German teachers made the predictable outcry about this effort one should make no mistakes: the role of teachers is difficult, public demand on them is high while appreciation of their work is low and although for years German teachers have been paid much better wages than their British counterparts, burn out syndroms are common, contributing to the German situation in which only 5% of teachers do actually work until there legal pension age. I yet have to see a fancy Web 2.0 application that would give teachers something in their hands to cope with their everyday work. Are there any social networking sites for teachers?

The opinion of the general public is in high demand as exemplified by a recent announcement of the British National Health Service (NHS) to introduce NHS Choices which among other things will allow patients to comment on their hospital treatment. The move has clearly been inspired by the success of PatientOpinion, a not-for-profit site which has allowed people for more than a year now to voice their experience with treatment received in hospital (and for which this new state-run competition raises some problems of its own).

Commendably recently even the Prime Minister’s Strategy Unit within the Cabinet Office has launched a review that is intended to highlight ways on how to leverage the Power of Information generated by state as well as the public.

However, not all is perfect for this new form of user empowerment. Especially one form of rating site has come under attack: sites which allow to comment on the performance of teachers, in particular a site called RateMyTeachers. These sites have come under fire as teachers complain about cyber bullying:

“Another teacher at the same school is criticised by a pupil because they “cannot speak English” and another as “the worst person ever” and “everyone hates her”.” (BBC, 3rd April 2007)

“Mr Brown said: “Comments are one thing but what about teachers who’ve had images of their heads super-imposed on to gratuitous images or who have had pictures taken and posted of their cleavages or underwear as they bend over, or who have had comments questioning their fidelity to their partner?”" (The Independent, 4th April 2007)

As every reader of newsgroups or forums has learned long ago many people have yet to master the art of making constructive criticisms and it is understandable that teachers take offence at anonymous comments like the ones quoted above. The concerns are serious enough for some to even call for a closure of such sites.

A new court decision in Germany could lead the way in assessing these issues in the future: The German website MeinProf.de which allows students to rate their professors and comment on their performance had been sued. After some nasty comments (including naming the professor a “psychopath”) that were quickly removed by the web site owners the professor in question went to court demanding the operators to pay 3,000 Euros (about £2,000) for any similar comment about him that might appear on the site in the future.

The court has decided that a general “cease and desist” for unacceptable comments is against the law. As a professor one has to face public criticism that cannot be prohibited ex ante. (see full coverage as well as press announcement – all in German I’m afraid).
(btw I wonder what Wendy’s take on the situation is…)

Several things have to be noted: In general this is a positive outcome for web sites that leverage the wisdom of the crowds as it offers some protection for the often not-for-profit operators of these sites. However, this does not justify defamatory comments on those sites and the court has emphasized the operators’ duty to remove those entries as soon as they are recognized. Last but not least, the subject under public scrutiny does matters as professors might well be made to face personal criticism in their role as public figures while teachers and nurses might have to be treated differently.

“To put the matter briefly, if secondary education is to be successful, it must successfully compete with cars and sports and social activities for the adolescents’ attention, in an open market. The adolescent is no longer a child, but will spend his energy in the ways he sees fit. It is up to the adult society to so structure secondary education that it captures this energy.”

Written long before the advent of the Internet with its many exciting distractions, this conclusions seems ever so relevant today. If we knew all that 45 years ago, what did we do in the meantime?

Why this sounds like a great idea, I just wonder where one would draw the line between help and actual surveillance. Not sure how I would react once some person from social services would post a comment on my blog reading: “Your posts clearly indicate that you are a highly troubled person and could need some professional help… get in touch with me …”

“When a teen is engaged in risky behaviour online, that is typically a sign that they’re engaged in risky behaviour offline. Troubled teens reveal their troubles online both explicitly and implicitly. It is not the online world that is making them troubled, but it is a fantastic opportunity for intervention. What would it mean to have digital street outreach where people started reaching out to troubled teens, not to punish them, but to help them? We already do street outreach in cities – why not treat the networked world as one large city? Imagine having college students troll the profiles of teens in their area in order to help troubled kids, just as they wander the physical streets. Too often we blame technology for what it reveals, but destroying or regulating the technology will not solve the underlying problems that are made visible through mediated publics like social network sites. “

Anyway, in order to make that a worthwhile exercise I will try to document what happens. So here we go, the first three things that are actually rather useful:

1. I can finally start to remember the names of people in my college.

2. I could check photos from Keble Ball.

3. I start to find out what people my friends know that I do know as well – a whole lot of new connections.

So, but in order not to make that a total defeat: The privacy issues with facebook are simply frightening. Okay, news feeds aside, do you really want to give facebook a blanket license to use your personal information for other applications? Anyway, three days from now I will probably even consider paying money for someone to use it …

“…features of social life – networks, norms, and trust – that enable participants to act together more effectively to pursue shared objectives… Social capital, in short, refers to social connections and the attendant norms and trust”.

However, on his homepage I found a list of 150 ways to build social capital which is much more instructive of the meaning of the term. On closer examination there are only 144 ways – 6 spots are still open. Suggestions?

It was set up by two (later three) German students in October 2005 and the company was recently sold to Holtzbrink, a German media company for something around €50 million. Despite its popularity studiVZ has been in the news repeatedly for a number of serious issues including security problems (user data was stolen from the server), sexism (a group was rating female members of the service with subsequent visits to their home addresses) and awkward behaviour of the founders (or how one would interpret a flyer for a birthday party imitating a racist newspaper of the Third Reich and Internet clips of unsuspecting women on toilets).

However, the thing that was striking me most was how a project like that can be successful, given that it is a blatant copy of facebook which has been around for much longer. However, during my discussions with some people at the re:publica conference (see next post) the main explanation seems to be that people really want a tool that considers their specific culture, that is mainly language and country. StudiVZ uses German language and its user base is German University students. Its success also points to shortcomings in the strategy of facebook that never managed to get a critical mass of users in the German market.

There are a number of other start ups that put their stakes on a similar bet:

• Mister Wong is a German version of del.ico.us that allows for social bookmarking. It’s strength according to their PR guy Christian Clawien is that it does focus on German websites that are more relevant to German users than the sites on del.ico.us.
• WebNews is a German digg clone that features user nominated news stories. As Thorsten Luettger of WebNews did point out to me, one of its advantages over the much bigger digg is that its in German and again features stories that are of interest to fellow countrymen. As an interesting side note, these guys got their first money from the Samwer brothers who earned (before selling ringtone subscriptions to teenagers) their money with setting up a German alternative to ebay (called alando.de) that was six month later acquired by ebay itself for $50m.

So the bottom line is: Geography still matters, the Anglo-American companies still get their marketing wrong and there is no global Internet culture that everyone could relate to regardless of provenance.