package shepherding

[Lamar Owen]

On Friday 05 March 2004 03:59 am, Alexander Larsson wrote:
> That is very much not true. Do you think we should be doing nothing but
> bug work? Any software *always* have bugs, so that would mean we'd
> *never* do any feature work or any other kind of important work. We'd
> ship a piece of stagnant software (if we ever got our noses out of
> bugzilla to spend time actually shipping it). A bugfree version of a 10
> year old piece of software. Nobody wants that.

Actually, if I were given the choice between more features (with more bugs) 
versus less features (and less bugs) for mission-critical work, I will choose 
the less bugs every time.  If a feature is delayed by bugfixing, that is a 
good thing.  If fixing a bug is delayed due to feeping creaturism, then 
somebody's priorities are askew.

Yes, features need to be worked on.  No, developers already pressed for time 
shouldn't be always in bugfixing mode.  But data destroying or serious 
usability bugs should always trump feature additions, which bring in their 
own new bugs, creating a vicious cycle of where developers feel like they're 
drowning in a sea of bugs.  Is reducing the number of bugs that cause people 
to use something else not a worthy goal?  Use some moderation.  Or take a 
page from the developer book of Tom Lane, who has nearly singlehandedly made 
PostgreSQL one of the least bugridden databases on the planet.  Yet he finds 
time for feature work.  It's all about balance.

Assuming that new features is better than less bugs is the Microsoft Way.
-- 
Lamar Owen
Director of Information Technology
Pisgah Astronomical Research Institute
1 PARI Drive
Rosman, NC  28772
(828)862-5554
www.pari.edu