Rules vs. Lists

Rob Freeman lists at
Fri Jul 4 10:30:10 UTC 2008

That's actually a pretty good reference Dan. Thanks.

For my taste there's a little too much emphasis on the practical
efficacy of the approach, and not enough on why it might be so, but
the general idea is along the same lines.

Do you know of any discussions why this might be the case: why for
some systems we might need to eschew theory and work directly with

My own take is that for some systems there may be more rules implicit
in the examples than there are examples themselves. So, not so much
"The End of Theory" as the birth of the theory that there can be lots
more theories buried in a set of data than we've ever imagined we
needed to look for before.

But people really don't like this kind of meta-theory, so I'm trying
to keep it as concrete as possible. That's why I'm focusing on the
practical problem of counting the number of rules you can abstract
from a given set of examples. If it turns out there are more rules
than examples, then that is something concrete we can deal with.


On Fri, Jul 4, 2008 at 7:53 AM,  <dlevere at> wrote:
> I just 'published' a short bit on this as part of a debate on EDGE,
> responding to work by Chris Anderson.
> The discussion as a whole, not so much my reply, might interest FUNKNET
> readers.
> -- Dan
> --------------------------------------------------------------
> This message was sent using Illinois State University Webmail.

More information about the Funknet mailing list