Hi all,

As most of you know, we are approaching a feature freeze leading
up to the next stable release (I don't want to get into any
fights - although I personally think 2.0 is better, I don't
actually care what it's called, as long as it gets released).

A feature freeze means we draw a line in the sand and say "no
more features get in here" - after the feature freeze, there
should be no changes accepted into CVS unless they fix behaviour
which should work, but doesn't.

So we have the problem of where (and when) to draw the line. The
when is provisionally the end of camp, giving people the chance
for a feature-filled weekend from the 6th to the 10th. The where
is more difficult. There are currently 361 bugs open against the
GIMP with either a milestone 1.3.x or milestone --. That means
that currently, there are 361 tasks which people would like to
see done before the next release. Many of these are small - of
course, the problem is deciding which can/should be done and
which can't/shouldn't.

And here we come to the reason for the mail - we need help with
this filtering. This is a job that anyone can do, and a few
minutes of your time will allow other people a few minutes more
implementing the features wanted before the freeze. What do you
need to do? Well, go to http://bugzilla.gnome.org, click on the
"Open a new bugzilla account" link, enter a valid e-mail, and you
will receive a mail from bugzilla containing your password, and
the link where you can log in.

Once you are logged in, click on this link...
http://makeashorterlink.com/?S1DF64A55
... which will take you to a list of all bugs which currently
have the milestone "--" (that is, bugs which are not targetted for
any release). And click on one. 

On the bug page, there is a drop-down box called "Milestone". The
milestone of every bug should be set, and the milestone of every
bug with the "--" milestone should be set to future, 2.0 or
1.3.x.

The idea of the feature freeze is to be very brutal with these
bugs. Of the 361 bugs above, 221 are enhancement requests
(feature improvements). Each of these needs to be ranked in terms
of importance and needs to get targetted for a future release, or
closed.

This is a very simple job for any given bug. As an example, bug
#101256 (Improve color blindness filters), the functionality has
been implemented, but has not quite been perfected. Change
milestone to Future - this is not a feature which will block a
stable release.

The only problem is that there are so many bugs to do. So here is
your chance to give a little time to the gimp, and get a warm
fuzzy feeling.

Guidelines for setting milestones:

1) Any bug does not block the feature freeze and should have its
target milestone set to 2.0

2) Any enhancement which is not important gets milestone Future
(where important is something which someone might consider
crucial).

3) Important features should be milestoned 1.3.x.

The problem comes with what's considered important or not. But
don't get too worked up about it. 20 or 30 people receive an
e-mail when a bug is changed, so that if you make a change with
which someone disagrees strongly, they will let you know through
bugzilla.

Once this first filter is done, we will then start cutting more
of the enhancement requests. If you see an enhancement and you're
not sure if it's crucial or not, it isn't. Put it in Future. 

I really cannot overstate how much this will help us get on track
for a 2.0 release soon. The second stage in the Great Enhancement
Chop is to try to associate enhancements with the people who are
going to code them. If no-one is going to code an enhancement, it
gets cut. This is really quite easy when you get the hang of it
:)

So, there we go. Could people who are interested in doing this
reply, just so I know what the take-up is like? Thanks a lot,

Cheers, and happy gimping,
Dave.

-- 
       David Neary,
       Lyon, France
  E-Mail: [EMAIL PROTECTED]
_______________________________________________
Gimp-developer mailing list
[EMAIL PROTECTED]
http://lists.xcf.berkeley.edu/mailman/listinfo/gimp-developer

Reply via email to