I created a free text index to a mhonarc archive of unrev-ii and
reported that and the changes (listed in the previous version of
this warp entry) to the CGI interface to the ba-unrev-talk list,
fishing for comments.

unrevdb
unrevprojectplan

---------- Forwarded message ----------
From: cdent@burningchrome.com
To: ba-unrev-talk@
Date: Sun, 28 Jul 2002 02:48:53 -0500 (EST)
Subject: Re: [ba-unrev-talk] further research and thinking related to
    unrev-ii archive


Kathryn and I have done a bit more work with the unrev-ii list
archive. Nothing major, but since this is something I requested
from the Bootstrap Alliance on the existing forums, I figured I
ought to report that I've done it on the old list.

Our work has involved processing messages from the unrev-ii
archive that we have parsed into a database. We have a few
interfaces to that data, including a CGI which will eventually
allow community members to code the messages for value and
aboutness. We'll use that information to help create a faceted
access structure for the archive.

That CGI has been updated to fix a bug. In the past when a set of
message were retrieved individual messages from that set would be
displayed with a "prev" and "next" link. Those links led to the
previous and next messages in the entire dataset, not the
selected set. That has been fixed. It wasn't so much a bug as me
being lazy in my coding. Now, if you feel so inclined, you can
retrieve the entire collection of messages from Eric Armstrong
(he is the single most frequent contributor by number of
messages) and easily read them in chronological order, jumping
out into the threads for context, if desired.

A few months ago I asked about getting a free text search
interface to the forums hosted on bootstrap.org. Since I had it
from the database parsing, I parsed the unrev-ii archive into
mhonarc and indexed it with ffw. FFW pays attention to name tags
in HTML to the index is down to the purple number. This may or
may not be effective. If people try it, I would appreciate
hearing comments. The CGI interface for the search system is very
crude compared to modern stuff, I'm just using the sample code
provide with the (old) system.

If people are interested I can create similar search interfaces
to the ba-unrev-talk and ba-ohs-talk lists.

These various things are linked to the following page:

  http://www.burningchrome.com/~cdent/uviz/

Please let me know if you have any comments.

On Mon, 15 Jul 2002 cdent@burningchrome.com wrote:

> Back in April Kathryn La Barre and I announced, on this list,
> some work we were doing with computational methods for generating
> faceted access structures to the archive of the unrev-ii (the
> list that is the precursor to this one) email list.
>
> Since then a paper describing the work and its relevance to the
> PORT project was written for the Pragmatic Web Workshop at the
> 10th International Conference on Conceptual Structures. Much of
> the workshop was held over email. For Kathryn and I the email
> collaboration led to an extensive revision of the paper. We had
> something of a revelation about the nature of knowledge
> representation and discourse that may be of interest.