[lug] Help make an email summarizer!

[Chris Riddoch]

Hi, everyone.

I'm working on my masters' thesis, and the project is to create a tool
that can automatically summarize very busy email lists.  Starting with
a large collection of messages, I'm organizing them, and sifting
through them for the most important pieces to quote in a shorter,
plain-English summary of the discussion taking place, a la Kernel
Traffic summaries.

Please visit http://www.syntacticsugar.org/abridger/ to help make a
good, free software, summarizer.

This is a call for volunteers to help me train a system to recognize
what <i>kind</i> of email any given message is.  I've made an archive
of the Linux Kernel mailing list available on my website, alongside
forms that you can use to help me identify what categories any given
email falls in.  There's more information available to help you learn
how to use the website.

There are currently no summaries available; that comes after being
able to identify types of messages, and I still need to write code to
analyze the information you're helping to provide, and to produce the
summaries.  Once that's done, you'll be able to help evaluate the
summaries the system is producing, to help qualitatively identify
improvements or problems in the system.

So as not to be off-topic, the most useful code for my project will be
made available under a suitably free licence -- most likely the GPL,
assuming I don't uncover license incompatibilies in the libraries I'm
making use of, or something similarly silly.  I'll figure this out
soon.

If you've come to hacking society in the past few months, you're
probably sick of hearing me talk about this already. Sorry. ;)

-- 
epistemological humility
   - Chris Riddoch -