Have you ever wondered why the topics reaching the frontpage of some social networks are very often about the same subject?

When you read the frontpage of reddit, you will notice mostly political and world news articles. Propeller too concentrates on political topics. Digg in its turn, is full of technology news. This is something, every blogger knows and the analysis at SocialMediaTrader proves. But why is that?

I mean there is nothing in the Terms of Use at these social networks, saying that only submissions of a specific genre are required. Everybody is free to submit anything and the wisdom of the crowd, will bring a post up or bury it. But still in a magical way, the posts reaching the front page are not diverse.

In this post I will analyze and visualize the network of the digg users, to show why technology is the favorite topic in Digg.

How relationships define the popularity of a topic

There are 3 requirements for a submission to reach the frontpage of any social network:

1. The quality of content.
2. Who submits the posts.
3. Who are the friends of the submitter.

These are points we are all aware of. When a top user submits a post, he gives it - in a sense - an additional popularity boost. His authoritative figure in the social network, says to the other users, that this particular post is worth their time.

But a point we are sometimes missing out, is that the friends of a top user does not only define the popularity of a post, but sometimes also the popularity of a genre.

Lets take a look at the top users of Digg.

The Digg Example

I decided to study Digg, since it is one of the least diverse social networks. In order to find the reason of this, I took a deep look at the network of the top digg users (according to Chris Finke’s data on digg).

• Step 1. I concentrated on the top 10 Digg users and saw if they were friends with each other.

• Step 2. I selected a random user from the list with a high popular ratio. The user I chose was sepultura, 28th top user with a submission/popularity ratio of 61.1%.

• Step 3. I visualized the data into a network, to see how highly these users were connected.

• Step 4. I collected the most dugg topics of each user the past 30 days.

Click to enlarge

I wanted to see if the top users are mutual friends and therefore view and promote each others submissions.

I also chose a random user with a high ratio, in order to see if he also was befriended with those top users. If so the popularity of his submissions could also rely on his friends.

If the top digg users were all befriended and interested in the same topics, it would actually be the reason for this monotony in subjects.

Visualizing Digg’s top users

The result of the visualization shows that the top 10 digg users are actually mutual friends and are in fact creating a very condence network.

Click to enlarge

• Mklopez (rank:10) is mutually befriended with all the others apart from p9s50W5k4GUD2c6 (!!) (rank 7).

• supernova17 (rank:4) is mutually befriended with 7 others, while msaleem (rank:2), mrbabyman (rank:1) and digitalgopher (rank:5) with 6.

• In fact, when these top users are befriended it is always mutual friendship, with the only exception of CLIFFosakaJAPAN (rank:6), who is a fan of Zaibatsu (rank:3).

• Only aaaz - the 8th top user - is mutually befriended with only 2 other top diggers.

• Sepultura, (rank:28) the random digger with high popularity ratio, was mutually befriended with 5 others: mrbabyman, msaleem, supernova17, zaibatsu and mklopez.

So it seems, that Digg’s top 10 users are mutually befriended and therefore are aware of each others submissions. Sepultura, might not be that popular but his popularity ratio is extremely high (61.1%). Is it a coincidence, that he is mutually befriended with the most connected top diggers?
The next step was to see, what are actually the interests of these users. Not surprisingly, apart from P9 and aaaz, all the other users (including sepultura) have as one of their main interests Technology News.

P9 and aaaz seem to have mostly political interests (second most popular topic of digg) and are actually the least connected of the group. P9 is mutually befriended with 4 top diggers (aaz among them), while aaaz is befriended with only p9 and mklopez (the most connected).

The Conclusion

The information collected show how the top diggers are actually in position of controlling the whole thematology of Digg. They have the same interests and are therefore friends with each other. It is only natural that they also like the same content. There is nothing wrong with that.

The power of the top diggers doesn’t rely on the fact, that they are popular but mostly on the fact that they have formed their own network.

But since their influence is so obvious and Digg isn’t specifically a technology network, I am not surprised that Digg enabled couple of months ago a more strict algorithm for a submission to reach the front page.

Image from SocialMediaTrader

I tried to do the same research with StumbleUpon, to see if the top stumblers are also that befriended as in Digg. It is certainly not the case. I suppose that’s why StumbleUpon has a much more diverse range of topics.

I just read a very interesting post about data portability on Mashable.

In short the author argues, that data portability is boring because:

• the average internet user probably isn’t active in many web 2.0 sites.
• the average internet user probably doesn’t want to take his friends along every web 2.0 service he signs up for.
• what rights do users have to control where shared data goes?

So he says, that there is actually no demand for data portability.

I can see his point and I actually agree completely. Data portability is probably referring to extremely active users.

Moral Data Portability

The way I have understood data portability, it is about data. You can take your facebook friends and put them on kaioo. You then use kaioo as your platform, while not losing contact with your facebook friends.

So I see a moral side of data portability. If I want to move my data/friends from one site to another I want to be able to. It is not only about who owns the data, it is also about not monopolizing your network.
Data portability gives you the option to decide which platform suits you best. This doesn’t necessarily mean being active in 10 different social networks. It just means choosing how you communicate with your friends in the community you are in.

I believe that has a value, which every average user could understand.

Reaching the average user is another subject. Data portability must extend to data usability too. The process of moving your contacts from one network to another must become as easy as signing up.
That’s where data portability might lose the game.

I just finished listening to an interview of Tim Berners Lee on the Semantic Web (63min). Very insightful information on the future and development of the Web. The relation of web 2.0 and web 3.0, the technological development of the semantic web and privacy are some of his points, that I found mostly interesting.

He underlines the ability of the semantic web to connect data with metadata. A function, which will make our work easier, faster and therefore more productive.

It’s the connection from the data to the provenance of the data, and not just for the name of the document that it came from, but the actual properties of that - the licensing, what it’s supposed to be used for, what it’s appropriate to use it for, whether I got it because I’ve gone through an authentication process, and actually whether it’s private data, which I should not actually publish at all.

Tim Berners Lee already sees some of the most popular social networks taking advantage of semantics.

They haven’t just allowed you to tag something with somebody’s name, they’ve allowed you to capture the difference between somebody who took the photo and somebody who’s in the photo, so that the power of the reuse of the data has been much greater.

He also talks about the development level of the semantic web, saying that the technologies necessary are already there.

I think, really we’ve got all the pieces to be able to go ahead and do pretty much everything. I suppose, really you should be able to implement a huge amount of the dream, we should be able to get huge benefits from interoperability using what we’ve got. So, people are realizing it’s time to just go do it.

The only piece missing from the semantic puzzle is the actual implementation of these technologies to current systems. An effort, which - according to Tim Berners Lee - is easy and financially worthwhile.

But, the thing that’s holding us up is that, there’s data which the companies have got on this, sitting and going round and round on its disks. Or it’s in their SQL systems and needs to be exported in a way that we can get at it in linked RDF as a SPARQL. And then, that could be reused.

One more remark, that drew my attention was about privacy. Tim Berners Lee advocates for openness of data - at least in a company level. He supports exposure and integration; companies should give the ability to the people to do queries on their data. This can give them a great advantage against their competition.

If a company has got this feeling where people don’t want other people in the company to know what is going on, then, it has already got a problem, this just exposes the problem.

On a more individual level he argues, that users should get hold of their data. That’s where the semantic web really diverts from the web 2.0 model. The web 2.0 model incorporates sites, which have their data and they don’t share it. The Web 3.0 gives the power back to the user

Web 2.0 is a stovepipe system. It’s a set of stovepipes where each site has got its data and it’s not sharing it. What people are sometimes calling a Web 3.0 vision where you’ve got lots of different data out there on the Web and you’ve got lots of different applications, but they’re independent.

There are much more information to be collected in this interview. This were just the stuff I found most interesting. Marshall Kirkpatrick concentrates on other subjects.

So don’t be satisfied with any short summary, listen to the actual interview or read the transcript.

What defines the value of a social network? Is it its users or is it the network itself? Who should have the last word in a self-organizing community? The users or the administrators?
The digg debate, although resolved raises some very important issues on the future of self-organizing platforms.

The users

The top diggers yesterday revolted against Digg, because it enabled a new, more ’strict’ algorithm. The debatable part of this algorithm is, that when a post is dugg by you and your 100 friends, it doesn’t necessarily mean that it will reach the frontpage.

So what the algorithm brings in, is diversity. Top diggers disregarded this variable, since it treats their posts (and consequently their network) unfair. The logic behind this revolt is, that

Top users are top users because they submit high qaulity material. They should not be required to get more diggs simply because of the great job they have been doing to reach that top user status.

I don’t see this as a plausible argument. These communities give control to the end user; every end user and not only a ruling few. Regardless of the quality of content these top-users provide, they are considered authorities in their platforms (may it be digg, stumbleupon, slashdot, etc) and their opinion is highly praised. This authority alone, is a reason for their content to be promoted. Sometimes regular users are digging top-user content, just because it is top-user content.

The network

Digg (and all sites of the kind) started out with a main principle of self-organization. They provided the tools for the creation of an active community, which discovers and promotes content - and let it self-organize.
Indeed, the presence of administrators is very subtle in most cases.

But the functionality of such communities doesn’t rely only on self-organization.
Indirect control is sometimes required. When users post spam, when they have multiple accounts, when they behave improperly, an administrator is needed to set things right.

Similarly when these networks are actually managed by few top users, then another main principle of the system is at stake. Diversity.
Without diversity, the content is usually about the same subjects and very often from the same sources.
But when Digg decided to control this behavior and give the opportunity to other users to bring content forward, the ruling diggers revolted.

It is of course logical. They have worked hard to create a steady friends network, which in a way guarantees lots of diggs for every post they make.

The resolution

After this mini-revolution, digg founders came forward and explained their position. The way I see it, they diplomatically remained on their position on the subject, saying that content will not be directly undermined when posted by a top digger, but it will have trouble reaching the front page if it is promoted only by their friends. I find it fair.
In the drilldown, where the discussion was held the question posed at their last post is:

If Digg is a game then we are ready to play for keeps. What happens if the most powerful users in the community decide to leave? Will others join? Is Digg anything without us? Let’s prove it.

Would the community fall apart because the top users aren’t a part of it anymore? Would the quality of the content degrade?
I don’t think so. Digg is a very popular social network. The move, that digg did opens the road

• for more users to submit content
• for more users to join the network (hence more diversity)
• for more users to participate actively in the community

The issue

I believe that the value of a network is mostly defined by the users participating in it. But the way the network indirectly regulates the community is a fundamental aspect for the success of the system.
What do you think? Should digg stay out of the way and let the community evolve the way it was evolving, or was it a good decision to endorse a more strict perspective of popularity?

IT IS UNOFFICIALLY OVER.

I’m done writing my Master Thesis. The subject is Attention and Participation in the social Web.

A subject, which interested me in several posts of mine in this blog, as some of you might have noticed. A question, that is in every blogger’s mind:

When there are so many blogs out there, how do I get attention to my blog?

And I don’t mean, these “33 Ways to increase your blog traffic”. I find these posts very useful, don’t misunderstand me. They are speaking from experience. But, my main point in this thesis was to approach these questions more…well…scientifically.

My Problems

The issue, that occured to me while writing was, that many disciplines are involved in understanding and decoding the blogosphere and social networks in general: Sociology, social psychology, mass psychology, network theory, emergence, media studies.

Due to limited time (and pages), I had (and have) the feeling that I approached each discipline only at the surface. That before getting deeper into a subject, I got out and continued with another. I didn’t get to the core of each field. I saw each study, only from the perspective of attentiveness. Logical you might assume, since any other approach would abstract me from my main subject. But, still it is a worry I have.

The Wikilutions

That’s why (before presenting the thesis to my professors), I’m giving it to the public. So that everyone can read it, see how this whole network works and give something back.

It would be absolutely selfish to just give a pdf document of the thesis and simply asking for feedback. It is of course an option (you can download the *pdf here), but primarily it is about interaction. So I have uploaded the whole thesis as a wiki.

The purpose of the wiki is twofold.

1. to create a database, explaining in ’scientifical terms’ the functions and structure of social networks.
2. to invite people from different disciplines to add to the project.

Read and Participate

• So if you’re the reading/printing type of guy: download the thesis as pdf here.
• If you’re more the participate/write/critic type of guy: join the wiki community here.

Your feedback and contribution will be highly appreciated