Sizing an online community (or how big is a web site)
Have you noticed that some on line communities are successful and some not.
Some start off successful, and then get eclipsed, others burst out onto the scene but then disappear.
The fact of the matter is that for every successful on line community there are a thousand that are failures.
Taking a scientific approach before you can work start to answer the questions about the growth of a population you must first have a accurate tool to put a size on it.
Beware – This question is more complex than it sounds – it is not as simple as stating the number of members, nor is it as simple as saying the number of pages. The relationship between content, information and its value is what is important.
Lets first consider sizing a really simple web site
The above example website is a simple as it comes. It has 14 bytes of data … but data is not information, . If there was just random data on the website it would have no meaning and therefore would not contain any information as it simply does not inform you of any thing.
To you as the reader it has only one piece of information, which in this case is the association of the object “you” with the verb “understood”. This association between two things is the smallest atom of information. I’m going to give this a value of 1 internet gram (or 1 ig).
Now consider world where there are only three websites.
- youunderstood.com
- yousmell.com
- and yousmile.com
each of these web sites have the same amount of information and associate the object “you” with an attribute. The weight of each site is 1ig and the total weight of the internet is 3ig
Now consider putting a search engine into this world
The search page itself has a weight it associates the word “search” with the input box, giving the page a weight of 1ig. Imagine that our search site was very simple, it has no results page. Instead it just goes straight to the first link it finds in its index. So … what is the weight of the search engine I hear you say.
its this
(the weight of the engine) +
(weight of each word to url association in it index)
which in our simplified word where there is only 4 words (associated 3 times on each of the web sites) is
1ig + 3*(2 * 1ig) = 7ig
From this it is clear to see how a search engine can add so much value with a very small amount of effort.
Where it gets really interesting is when you look how the weight of the search engine changes in relation to the size of the internet. Imagine that instead of 3 web sites that there are 4 web sites. This would give a value to the search engine of 9ig or to put it mathematically the rate of change of weight of the search engine is directly proportional to twice the number of sites on the internet.
Now imagine that the search engine was indexing information (i.e. associations) rather than words, also imagine that the pages had an almost limitless amount of associations, as they do in the real world. You are left with the conclusion that the rate of change of the search engine is proportional to the amount of associations multiplied by the number of urls.
Now imagine an on line community, but again in a simplified word where there are only 3 people.
Tom, Dick and Harry.
So the most simplified social networking site is a site that simply lists your friends. There is one association per friends that you have.
This means that for Toms social networking page he has a weight of 2ig. However if Tom is in the social network and he is friends with both Dick and Harry then they must also be in the network (infact their absence from the network is also information). So the total weight of the social network is 6ig, if again we add in another person then you will see that the weight of the social network is 12ig.
Infact the weight of the social network is proportional to the number of people in the social network multiplied by the number of people in the social network -1.
This is a formula for exponential growth !
