Twitter and Power Laws

The web seems to be one of many areas of human activity where power laws hold sway. The book by Barabasi gives a very comprehensible account, neatly sidestepping the mathematics.

The web is a network, in fact several different kinds of networks and the number of links that come from each network node can often be expressed by a power law.  For example we may be interested to know how the number of incoming links on a web page (the node) varies from page to page.  We might expect that most web pages had roughly the same number, not too far from some average, just as most people's height is around the average for the population.  But as Barabasi recounts that is not the case for links to web pages.  Most web pages have a few incoming links, whilst a few pages have a very large number.  If you plot the graph it might look like this: 

The mathematical expression of a power law is in fact rather simple as such things go.  It is something like

                         number of web pages = constant * number of links ^power

The value of power is typically in the range -2 to -3

This idea has been applied by Milovanovic to four quite different sets of data.  They all show, with varying degrees of conviction, a power law distribution.

The one which interests us here is a plot of the number of followers of 11.5 million Twitter users. This shows a very clear power law distribution with a power value of -2.25.  The power law in this case refers to the whole linked network of these millions of users.   That certainly helps us to understand the structure of the whole network. However we may be interested in the nature of the network as experienced by an individual user such as the distribution of their followers.  To do this we can plot the number of followers of each of the followers of this individual user - does this also give a power law?  Having done this for a few users the answer seems to be that typically as the number of followers becomes large, of the order of 10's of thousands, a power law distribution is approached.

But most users do not have 10's of thousands of followers.  What is the distribution of their followers?  They can be characterised in part by simple measures such as the proportions of their followers that fit into defined categories, such as the tail of the distribution, set perhaps arbitrarily at 100000 followers.

In networks such as that of Twitter users there would seem to be considerable possibilities to explore deviations from power law patterns.

Barabasi, A (2002), Linked, Penguin Books Ltd, London 

Milovanovic, G.S.(2010) Online Contributions to the Global Community: The Power Laws for New Media Available from:

http://www.milovanovicresearch.com/post/1217388296/online-contributions-to-the-global-community-the-power [Accessed 30/11/2010]

FriendorFollow - Excellent in parts

I'm looking at some Twitter apps to try to analyse what type of networks people in education develop using Twitter.

One I'm trying out is FriendorFollow. You put in any Twitter username (no passwords needed and a nice simple interface) and it provides 3 pieces of information for that username:

• A list of users you follow who don't follow you back

• A list of your fans - those who follow you that you don't follow back

• A list of friends - those who you follow and who follow you back

The results give thumbnails for each user; mouse  over on a thumbnail gives that user's profile. The number in each list is also given.

It's a useful and effective app.  The response is usually reasonably quick but can be much slower for people with large numbers of followers and friends.  Some people find it useful for managing their followers.

So far so good.  But there are some problems.  For each  list mentioned above a CSV file with a line for each user may be downloaded.  Potentially a very useful facility.  But for each user only two numbers are given:

• Followers_count - this is (reasonably) the sum of fans and friends, and is meant to be the same as the followers number in Twitter

• Friends_count - this is the sum of those you follow but don't follow you back plus friends, and is meant to be the number that Twitter calls "Following", i.e. simply the number you follow

So there is some confusion about terminology but the above explanation clarifies this (I hope).

More importantly against these numbers is a "status date" which I take to be the date when the numbers were obtained from Twitter.  This date is NOT the date when the numbers for your input username are obtained.  The numbers have been obtained some time in the past; they can be up to two years out of date.  So for most purposes they are quite useless. 


A couple of other small things about this app.  Putting in a non-existent user gives a system  error message; it doesn't say the username doesn't exist.  Also, surprisingly, the Twitter account for FriendorFollow seems dormant at present.


Overall a very useful app but be careful about some of the detail in the numbers

How bad is your online experience?

As new technologies come along every day, we sometimes forget just how bad is the experience we still get from many an established web site.

Take my frustrations last Saturday when I wanted to make an online donation to a charity. I started from the home page of the company I use to make charity contributions, wanting to specify a charity and then log in to my account.

So I searched for the charity - a well known one, the Disasters Emergency Committee. It wasn't found, nor were any abbreviations for it. Today in talking to the charity company I found there is indeed an error in their search subsystem. But in small blue type on the home page were the words "Pakistan Flood Appeal". So I should have found those words by reading the home page? Not on your life. It's well established that people don't read web pages so much as scan them - see the interesting eye tracking pictures in this article by Jakob Nielsen:

http://www.useit.com/alertbox/banner-blindness.html

So twelve years after the idea of banner blindness was first raised it seems that many web designers aren't aware of it.

Then  I tried to log in to my account. I got a "web server error" message. Obviously the server for the user accounts was down. It was back up again on Monday. I was told that the server is always available at the weekend. Yeh - right. Perhaps they hadn't thought about (or couldn't afford) hardware to ensure uninterrupted running.

My third Saturday morning problem was a curious one. I went directly to the charity and tried to make my donation by phone - an automated system. All went well until I was asked to give the 3 digit security code on the back of my card. I was making my donation from my charity account, for which they had issued a special card. I turned over the card. No security code! On Monday I checked with the charity company. Yes there had been a problem. When the charity had updated their automated donation system they had forgotten that the card for my charity company did not have a security code on the back. So talks were now underway to resolve the situation.

The lesson from these experiences?  There are still many organisations that don't have useable websites.
 

Blended Learning - interaction between classroom and online activities

There several kinds of e-learning. One is distance learning, where all interactions between tutor and students and between students themselves are electronic. So students never see one another face to face. By contrast are traditional classroom based courses which are augmented by access to a repository of information held on a VLE. In between are many blended learning courses which augment their face to face interactions by various forms of electronic communication.

Blended learning is often described as though there was no interaction between the face to face and electronic components of a course. Which is surprising because the culture within a classroom affects educational outcomes, as does the culture which exists within online facilities such as forums and wikis. We should not be surprised then if in a blended learning course the online and the offline cultures interact.

This is what a colleague observed recently. She asked the students in a class she met every week to add summaries of papers on defined topics to a wiki . The wiki allowed all students to see and edit each others contributions. There were no marks for completing this exercise. And yet all the students made their regular required contributions. This was even though, as some evaluation responses showed, some students thought that posting to a wiki was not a useful activity if they were not to be given credit for it.

The reason for the engagement of all the students with the online activity seemed to lie in what went on in the classroom, the offline activity. At each weekly meeting students were reminded that they should be contributing to the wiki. In a small class it was obvious who had and who hadn't contributed. Students were told that the final wiki would be of help to the whole class when it came to revision for the final exam. So those who were not contributing could be seen as letting down the other students. In short there was social pressure.

Social pressure also manifested itself in the use that was made of the wiki. Students posted, as they had been asked to do, summaries of papers. But they were also told that they could comment on each others contributions. There were no such comments. There could be several reasons for this. One might have been that since there were no marks for the postings, students simply put the minimum effort into the wiki. However another possibility was that students didn't want to criticise in a public forum fellow students that they met with on a daily basis. If this was the case it raises the possibility that the use students make of online facilities will be influenced by the nature and extent of their offline interactions. One might hypothesise that the contributions to a wiki or forum for a distance learning course, run completely online, would be less concerned with their affect on the feelings of other students. Of course, such an effect might fade over the duration of a distance learning course, as students got to know each other through electronic interaction.

It seems that the delivery of blended learning courses requires more than simply adding electronic facilities to an existing course. The interaction between the online and the offline can be quite subtle.

Montaigne on Liars

Book One Chapter 9 On Liars

Montaigne talks about having a poor memory. I sympathise, having a memory that blanks out both the important and the trivial. But I don't need to be too envious of those with wonderful memories for as Montaigne says “ excellent memories are often coupled with feeble judgements”. And then again the worse the memory the briefer the speech.

Then he comes to the main point – if you are going to lie then you had better have a good memory. I'm reminded of modern day politicians when he talks of liars being tripped up when those they have told different lies to come together and can compare what they have heard. In times past a politician in the far reaches of his constituency might say something to appease his constituents, knowing that it would hardly get to a wider audience. But these days it will rapidly appear on national television or be picked up by his/her opponents to be used as ammunition at the next election.

The very first blogger?

You might ask who was the first blogger. Quite by chance, on holiday and looking for something to read in a pile of books in a rented apartment, I came across a book I had always wanted to read – the essays of Montaigne. Sporadically I had come across references to it and they had often seemed intriguing. But I had never read it. Now was my chance.

I immediately realised that these essays (essais = tries) of Montaigne were in fact nothing but blogs; fairly lengthy blogs perhaps but nonetheless blogs. Each essay addresses a subject and tells us what Montaigne knows about it. Sometimes that isn't very much; at other times we get many pages. Always there are many quotations from the classical authors such as Senecca, Horace Virgil etc. But what makes the reading worthwhile are the opinions expressed: always positive and clear and often saying things that are as relevant today as when they were first written over 400 years ago. A single sentence will leap out of the page with a quintessential curtain-lifting thought. It is Montaigne's opinions that I am going to comment on in this and following blogs. I know that over the centuries many people have have written about Montaigne, so I don't expect to produce anything that is new. But Montaigne's own introduction to his essays starts “my sole purpose in writing this book has been a private and domestic one”. In that spirit I am simply recording my immediate reaction to some of his essays, to help me (and maybe a reader or two of these blogs) understand how they might have meaning today.

The edition of the essays that I found was the Penguin Classics one: Michel de Montaigne, Essays Translated with an Introduction by J M Cohen (1958). This is only a partial edition of about 1/3 rd of the essays, but certainly adequate for me to get started. The essays were published in three books, additions being added in each successive edition. You can download the complete works, in a 19th century translation, from Project Gutenberg. This is often heavier going than the Cohen version; 19th century English can be rather elaborate. But it does have some exquisite translations such as: “a man's accusations of himself are always believed; his praises never” which in Cohen is the more leaden “what [a man] says in his disfavour is always believed, but when he commends himself he arouses distrust”.

There are certainly sections of the essays that strike no chords with me: a comparison of Seneca and Plutarch or an analysis of the strengths of Tacitus. But leaving these aside there is much that seems of relevance today, even if you have to take account of social context. I'll try in a few brief blog entries to point out some of the Montaigne's thoughts that I find intriguing.

Manifesto for a Networked Nation - ageist

I posted the following to http://raceonline2012.org/ the website for Manifesto for a Networked Nation.

Having looked at the press release for Manifesto for a Networked Nation I see that it talks about "getting everybody of working age" online.

This is pure ageism.

If it had been to get every male online, (with the unstated assumption that females are less worthy of support) it would rightly have been condemned out of hand as sexist - and would probably have been illegal.

But here we have a policy that those over working age are of less importance. There all sorts of reasons why this is socially and economically disastrous besides being discriminatory. I would willingly explain here why this is but I'm doubtful if it is worth my time to do so if the person (people) who put together this site haven't naturally thought about this.

It is perhaps worth noting that the very concept of "working age" is now meaningless. The age at which the state pension is awarded is due to go up, but there many who routinely work beyond that age.

Judging by the number of posts there is very little traffic here, so I'll post this comment elsewhere.

Comment on Anatomy of a PLE

Added this comment to Steve Wheeler's blog on Anatomy of a PLE

I agree that the notion of a PLE is ill-defined, like many such notions in e-learning, so attempts at clarification are useful.

I also agree that the PLE vs VLE dichotomy is not a helpful way of looking at it. A VLE isn't only for content management. It can be, though often isn't, the centre of social interaction for students on a course and staff in a department. All behind a privacy barrier to the outside world. A person's contribution to a VLE should surely be part of their PLE. But those contributions are also needed by the institution, in the case of a student, as it certificates what the student has done.

So while the PLE will certainly include the use of web tools, some of which will be at the behest of the institution, it will also include the use of a VLE. The essential point is that the "personal environment" has to include the whole of the environment in which learning takes place.

On a more practical note I wonder how present PLEs will really support lifelong learning. I still have some books from my (long past) undergraduate days. How long will the present generation of web tools be around?

The Guardian front page - could do much better

I'm afraid I'm still on about the Guardian, with some hopefully what will be perceived as helpful criticism. This time it's the front page of the paper edition.

It is laid out in a way that I find very frustrating. Typically there may be three quite interesting stories, but none of them is complete on the front page. You have to turn to page 2 for the continuation. So to read one front page in its entirety you need to turn to page 2 three times.

It can get worse; you may have to turn to different pages for the continuation of each story.

The edition of Monday the 12th July reached a new low.

Story 1 - continued on page 4.

Story 2 - said "continued on page 6-7" Also a large icon indicated the whole story on 2-3. In fact it was continued on 2-3 and the main story was there.

Story 3 - said "continued on page 2" Also a large icon indicated the whole story on 6-7. In fact it was continued on 6-7 and the main story was there.

So the sub-editors got highly confused as must have been all the readers.

Does the Guardian get it

The Guardian newspaper sees its future in its online, free version. As someone who has bought the paper version for a very long time I thought I would try it out online.

I had something to explore. Specifically I had become very dissatisfied with one aspect of the daily paper - it has a disastrous coverage of the sport I follow which is athletics. A couple of examples: it wasn't able to afford the train fare to Birmingham to send a correspondent to the national championships; it didn't give the team selection for the European Championships. Although it did manage an article on the gender issues of one South African athlete. So I'm a long-term reader with a distinct feeling of being ignored. Should I change my newspaper?

Before I do that how do I get my dissatisfaction across to the people at the Guardian. Just as they didn't understand their readers about the Doonesbury cartoon strip I suspect they also are not in tune with how many readers have got passionate interest in sports other than football. The readers editor is for errors, not about content complaints. A letter for publication would likely to be too long to get published and the people in the sports department (whoever they are - who is the sports editor?) would not read it.

So I turned to the website. Surely I could interact with the sports editors through it? Under one link I found:

"The guardian.co.uk site provides a growing number of opportunities for readers who wish to discuss content we publish or debate more generally."

What this implies, and what comes over through the multifarious pages of guidelines, is that readers can discuss and thus create new content, but the Guardian is not interested in the readers view of how it approaches such issues as the sports it chooses to follow.

I don't expect that the Guardian will be particularly interested in the sense of alienation it has created in this reader.

The real point of this post is that the website reflects an approach in which the Guardian staff are not part of the community they claim to want to foster. It is in this sense that Guardian staff (web site managers? editors?) don't "get" social media. Until they actively invite comment on the whole content of what is being published, they will not fully involve readers.