Ne:new feature philosophy [was: Templates]

[David Relson]

On Fri, 2 Jul 2004 10:57:27 -0400
Gyepi SAM wrote:

> On Fri, Jul 02, 2004 at 08:11:01AM -0400, Tom Allison wrote:
> > Personally, I'm a little leary of all the new features that are
> > being pushed into bogofilter.
> > I'm tending to lean on the "ain't broke, don't fix it" mentality
> > here. Maybe that's a bad thing for Open Source projects, but
> > bogofilter is already wildly successful in it's ability to capture
> > spam.
> 
> I suspect that this issue has come up quite a few times with
> other projects as well and one could make good arguments for both
> positions. Adding a feature is sometimes the only way to show its
> utility. Even the failure of a feature sometimes helps to move the
> project forward.  Of course, not all features are useful for everyone.
> One hopes though that they are useful to enough people! I have not
> always agreed with every new feature that has gone into bogofilter nor
> even some of the design changes, but since none of them reduced its
> functionality, there's no cause to object.
> 
> >  The only 
> > deficiency that I see with respect to other spam filters is 1) 
> > daemon-capability in order to run more easily "site wide"
> 
> One more feature to add... ;) I am personally interested in designing
> and implementing this feature. ESR introduced the first try in an
> attempt to get around the slowness of using a text based data-store
> and the Judy hash library. When we switched to BDB and wordhash, the
> need went away. Nonetheless, it is a useful feature, but not high
> priority for me at the moment.
> 
> -Gyepi

Tom and Gyepi,

I'm inclined to say that very few new features have been added to
bogofilter in the past 6 months.  Much of the work has been cleanup --
removal of backwards compatibility code and rewriting of code chunks
that have evolved and, now that we know what they should be doing, can
be designed and written better.  

Once the new database transaction code is in the main line, Matthias and
I are planning a count down to releae 1.0.  After that release,
experimentation can begin with word pairs (or chains), daemon mode, etc.

Certainly having additional contributors will cause more rapid change.

Regards,

David