dwine

Actually not true. If the website gets a million unique visitors, there is about a 60% chance of someone winning.

And why is this even called Infer? It's not doing inference...

BTW, Infer.NET is written in C#, but there is an F# wrapper.

What was the social network with public relations between people that failed because the "follow" was symmetric?

I'm a computer scientist, and I call the (very old) study of graphs "graph theory". How is "network theory" different?

And is it just me, or is it painful to read this? It has the feel of a student taking an intro course on optimization, then exclaiming, "everything is an optimization problem!" I mean, sure, this is true, but it's tautological and feels forced to me.

And sorry, but Twitter's main innovation was discarding symmetry? People have been subscribing (an asymmetric relation) to things on the internet and otherwise for ages.

The data there has the scores from ~5000 games played over the course of each season, and the model he links to also seems quite reasonable to me: http://blog.smellthedata.com/2009/03/data-driven-march-madne...

Don't think of it as two data points. Think of it as two data sets.

And from a fellow researcher: http://hunch.net/?p=1172