[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GNATS these days

From: Chad Walstrom
Subject: Re: GNATS these days
Date: Thu, 01 Sep 2005 15:18:16 -0500

Jason wrote:
> I noticed that on Savannah, there's a later update to the code
> (4.1.x).  How is the project doing these days.  I see extremely
> little activity on the system.  Is that because people are opting
> for other project or task management tracking systems or is it a
> result of stability?
> I thought most GNU projects still relied heavily on GNATS for
> tracking.  Do you consider the project lively?  Stable?

GNATS is getting some development attention as of late.  The two
biggest lines of development right now are with Mel Hatzis' database
abstraction, triggers, and other enhancements and my relatively
unexciting yet necessary work on migrating to automake (with has been
stalled by real-life interference).  There is some code cleanup and
security cleanup going on, too.  A patch for PAM is awaiting entry
into the main code base, and there are some active and useful bugs
being entered into database at  Basically, GNATS is
moving ahead slowly but surely.

I read a relatively recent post about problem tracking design.  I
don't recall where, but the gist is that problem tracking has two main
components: facilitating discussion and tracking state.  GNATS is good
at tracking state, but mediocre with facilitating discussion.

Part of the reason for this is because GNATS hasn't really kept pace
with the changes in email formats over the years.  I would love to
address the email->gnats gateway issues, such as poor handling of MIME
and character-set/localization handling.  Time is an issue, and higher
priorities take precedent, obviously.

Still, GNATS is quite useful as is, and unique in a number of aspects.
The database format is RFC 822 formatted messages managed similar to
MH.  In fact, one hack for managing a local GNATS database (I wouldn't
recommend it) is to use MH tools, but exclude the sorting functions.
PR's are stored in files by their number in directories named after
their categories.  If you run 'sortm' or 'folder -pack' on a GNATS
folder, you'll break things horribly. 

GNATS has also supported remote databases through it's inetd server,
gnatsd, for quite some time now.  This allows remote management of the
database through shell scripts and web CGI front-ends, but also
removes the necessity to create local users (other than the gnats
user) to manage the database.

All in all, GNATS is stable, flexible, and useful.  If you're
interested in helping, by all means join in!

Chad Walstrom <address@hidden> 
           assert(expired(knowledge)); /* core dump */

reply via email to

[Prev in Thread] Current Thread [Next in Thread]