<<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>>
          <<  Lancair Builders' Mail List  >>
          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>>
>>
Terry Pickering asked about the searchable index for the LML postings.  As
I mentioned a few weeks ago, I'm migrating the web archives to a new server
that uses a more robust version of the email-to-HTML converter to help with
threading, and which also uses a subject oriented search engine for looking
up posts.  Unfortunately, in order to get all of this stuff working
properly it's necessary for me to wade through the entire archive (which is
presently topping out around 1350 messages) and manually correct all of the
subject headers to keep the relational aspects of the database intact.  As
you might imagine, this is no small task, and as of this writing I've
completed about 20% of the work.  Once that's done I'll be able to update
the archive based on monthly input by cleaning things up one month at a time.

This is the primary reason that I make so many references to the "Subject"
headers in the List Posting Guidelines.  Every time someone responds to a
particular post and changes the subject, the thread is broken and a new one
started.  I know there's always the temptation to come up with a new
'cutesy' header for a response, but if we could all resist that temptation
it would make everything work considerably smoother and make searching for
particular subject matter much easier.  As I also mentioned in that same
post, the entire Lancair Mail List homepage has been updated to include a
mission statement and vastly expanded List Posting Guidelines, all of which
are aimed at making the LML a highly functional resource.  If you haven't
seen that page more recently than about 3 weeks ago, I'd highly recommend
you take a look (http://www.olsusa.com/Users/Mkaye/maillist.html).

Terry also asked about an annual index of message headers, and that is
certainly something that I have in my long term plans for the LML.  I'm
hoping to be able to make the compilation of each year's postings available
on CDROM complete with just such a searchable index.  If everything goes as
planned, that database should be able to be searched by keyword based on
content in addition to a "Subject"-based search which would undoubtedly
yield many more hits and wind up retrieving information that is buried in
the message bodies but not necessarily identified in the header info.  That
same database will also be available on the LML homepage, but that's a
whole nuther project <g>.