So, are people still using ubh?

[Doug McLaren]

I've been using ubh2 for many years now, and have hacked it up to work
pretty much exactly like I want it.

However, RR recently outsourced their news server to newshosting.com.
Now, this is a good thing, but it created some problems -- most
notably, going from 1.5 days retention to 30 days retention has caused
the headers to increase greatly in size. Now ubh2 averages spending
around two hours merely filtering the headers from some of the larger
groups (the -I/-X options.)

So I'm investigating ubh3. So far, I'm pretty impressed with how it
works, but I see a lot of room for improvement. ubhdb started out
really fast, but it's really slowed down later in the run, probably
because my headers and xref_headers tables have over 8 million rows
now. (But that's what happens when you insert into a large table with
indexes ...)

Is ubh3 still being maintained? Looking at the CVS, it looks like
nothing has been changed in two years. And a few years ago (Feb 23,
2004) Gerald said he's not adding any new features, and not interested
in patches. Has any of this changed since?

I see a few minor changes I'll want to make to ubh3 if I'm going to
switch to it, and wonder if there's any point in submitting patches.
If Gerald is not interested, are others?

(For starters, I'd probably add a quickstart guide, move database
coordinates to a config file, add disk space checking, make -I/-X
searches case insensitive by default, perhaps make it work for
databases other than mysql (though mysql is probably the ideal match
for ubh.)

Have people replaced ubh with anything else? (Am I beating a dead
horse here?)