What is Science?

**Tamur** · 05-20-2008, 21:00

The thread on the end of the world made me wonder what we're considering to be science. Is this the 18th century broad definition of philosophy -- any systematic knowledge-seeking, leaving the forum open for social sciences? Is it the narrower 20th century+ definition of empirical-only inquiries?

Personally I'm interested in keeping it open to the social sciences, because then I can discuss my favourite topic: social network analysis!

**Papewaio** · 05-20-2008, 22:56

I think as a gaming forum we should keep it open to pop science.

If it becomes political/religious debate then that is for the backroom. So a dose of empirical evidence to keep it in the Frontroom.

**cmacq** · 05-20-2008, 23:59

Originally Posted by Tamur

Personally I'm interested in keeping it open to the social sciences, because then I can discuss my favourite topic: social network analysis!

Please, what is social network analysis?

**Papewaio** · 05-21-2008, 00:11

Now that you've asked the question you've got me interested too... according to Wiki

Social network:

Originally Posted by Wiki

A social network is a social structure made of nodes (which are generally individuals or organizations) that are tied by one or more specific types of interdependency, such as values, visions, idea, financial exchange, friends, kinship, dislike, conflict, trade, web links, sexual relations, disease transmission (epidemiology), or airline routes. The resulting structures are often very complex.

Interesting, I was just reading about

Artificial Neural Networks... I assume the two can be neatly tied together.

**Tamur** · 05-21-2008, 03:50

The resulting structures are often very complex.

You can't imagine until you've dealt with one. They can be hairy monsters.

I'd definitely recommend the Wikipedia (english) article on social network analysis. As Papewaio quoted, a social network is fairly simple in concept: a group of nodes tied together by bonds of any kind, with strength ratings attached to each bond. Bonds can be one-way or two-way.

What I like about it as an analytical tool is that the nodes can be anything, and the bonds can be anything, so long as the bond is some kind of relationship that every node can have with every other nodes. Sometimes a bond is obvious: distance between cities, for example, or number of emails sent from one person to another. Sometimes it has to be massaged a bit: percentage similarity between genomes, for example.

As a scientific tool it's incredibly flexible, and a great way to take another look at data which may look stale in tabular view.

Highly recommended

Oh, about neural networks --- they're a bit different in form because most neural nets are linear in nature (i.e. based on dependencies) where SNAs are multiradial in nature (i.e. based on lots of interconnected centre points)