"Tim Rhodes" <proftim@speakeasy.org> wrote:
> I, once again, suspect--that one could find evidence of memetic
> replication of phrases or turns of speech within the archives of
> this very list if one possessed the proper tools. Anyone know
> what those tools might be?
Not off hand. I have a couple of text-based editors that can do searches
pretty well, but counting? I could probably write a program to do so --
but I would need local access to the archive (e.g. I'd have to download
it), and I would also need more direction than that. I might also need
significant amounts of processor time... 13,000 messages!
The trickiest part would probably be deciding which "turns of speech" to do
searches on... and how we should quantify the results.
Suggested data to gather:
1) Turn around time (how long before it gets repeated?)
2) Wear out time (how long until all reference to it is gone again?)
3) How to measure the "distance" between two expressions of the meme
i) by bytes?
ii) by messages?
iii) by time?
iv) all of the above!
4) Perhaps a plot of frequency over time? For instance, I think a search
for "level 3" over the entire length of the archive would turn up a neat
boom/bust cycle...
However, for now, I'd like to finish *reading* the archives before I go on
to *processing* them.
I haven't done anything on that project for about a month, but my current
results are now available on my web page
http://qlink.queensu.ca/~6ceb3/Virus.htm
Any other good ideas?
ERiC