> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On
> Behalf Of J. van Baardwijk
[snip]
> Since you seem to know a lot (definitely more than I do) about link
> analysis and analysing the dynamics of on-line communities, can you
> recommend some good reading material on this topic? Something
> that does not
> require a college education to comprehend?
Look at books on data mining. They're usually oriented toward marketing
applications, but the principles apply to many other areas. I've never seen
anything in a book about using these techniques against on-line communities.
I know of only one serious researcher -- Marc Smith. He works for
Microsoft's research group. See http://research.microsoft.com/~masmith/ and
look at Netscan.
> A one-message thread would have a subject header like "This Is The
> Subject", but there would not be any messages with the subject
> header "Re:
> This Is The Subject". So, a filter would (in pseudo code) look
> something like:
>
> EXCLUDE IF
> EXIST "Subject"
> AND NOT EXIST "Re: Subject"
>
> Just theorising here...
Not very useful pseudo-code unless there's a magic "exist" function... ;-)
The problem isn't excluding them, it's excluding them *easily*. If I add
threading data to the database, that'll make it easy. Until then, it
requires a lookup for each one.
>
> >That probably isn't clear unless you speak SQL...
>
> Sorry, me no speaky the SQL very good. (I know a bit about it as it comes
> with my attempt to get my CompSci degree.)
But isn't that what you say about English, too? Just kidding.
Nick
