Subject: Re: From PR to pkgsrc
To: Todd Vierling <>
From: Greg A. Woods <>
List: tech-pkg
Date: 12/30/2004 13:54:13
[ On Thursday, December 30, 2004 at 09:27:00 (-0500), Todd Vierling wrote: ]
> Subject: Re: From PR to pkgsrc
> On Wed, 29 Dec 2004, Greg A. Woods wrote:
> > I do agree pkgsrc-wip is fine and good for those able to use it.
> >
> > However as my own broken record keeps saying, until and unless the "pkg"
> > category of GNATS is officially deprecated then _someone_, be they
> > pkgsrc or pkgsrc-wip committers
> pkgsrc-wip committers are not responsible for GNATS.  Stop trying to imply
> offloading responsibility in this way.

I'm _not_ trying to offload responsibility -- certainly not from those
who wish to accept it!

I am trying to show you that by your own admission there's an apparently
vast untapped resource that could be brought to bear on the fruits of
GNATS should anyone "officially" be so inclined to encourage such
activity.  How much more simple can it get!?!?!?!?!?

I'm also suggesting that the "encouragement" would be one hell of a lot
stronger if it had some "privilege", even if only implied, to go along
with it.  Would you and other developers rather be clerks transcribing
the efforts of others into GNATS, or would you be better off delegating
that responsibility to those doing the original work?

GNATS provides more than adequate audit trails and other such mechanisms
to make it much safer to provide access to than, say, CVS does.

> pkgsrc-wip is a working tree for contributors to use *directly*, period.

You have also been emphasizing and repeating the point that pkgsrc-wip
is also a training ground for those contributors, and also the point
that it is a mechanism of collaboration which facilitates use of the
resources those contributors bring to bear such that the effort required
by official pkgsrc committers to move a pkgsrc-wip module into the main
tree is as low as possible.

Does it not make sense to also train pkgsrc-wip committers in the ways
of dealing with the oh-so-apparently-terrible burden of handling PR
submissions so as to ease the pain of it on all?  Many hands make light
work -- but you do have to provide them direction and incentive.

> You're just refusing to use it yourself

Well, regardless of my reasons I'm clearly not the only one using GNATS
to submit new packages, as per the official recommended way of doing so
(or at least minor variants of it).

You can bash me all you want but all you're doing is further alienating
all those who like me would prefer to continue using GNATS for pkgsrc.

You can continue to ignore the offerings already in GNATS and use as
your excuse that you're unable to find some way to use already available
human resources to bring those submissions into form for pkgsrc.

I'm just trying to show that until and unless the GNATS pkg category is
officially deprected for new submissions then _SOMEONE_ _MUST_ keep an
eye on new submissions arriving by way of GNATS!

						Greg A. Woods

H:+1 416 218-0098  W:+1 416 489-5852 x122  VE3TCP  RoboHack <>
Planix, Inc. <>          Secrets of the Weird <>