Re: long term goals of debian membership
On Tue, 5 Dec 2000, Ben Collins wrote:
> On Tue, Dec 05, 2000 at 12:23:31PM -0500, Neal H Walfield wrote:
> <snip>
> > winning way. Again, no one forces anyone to be here and when someone no
> > longer want to be here, that person leaves.
>
> No, they just go MIA, and leave their responsibilities and account unused.
> Not very many maintainers (in fact only 3 I can think of in the 2 years I
> have been with Debian) actually take themselves out of the project. This
> is a management problem for accounts, and a security risk (because the
> accounts are not being used, or checked).
We've already been over my solution for this... I take it we've agreed to
disagree?
> We cannot increase this risk/problem. We need to be sure that people are
> going to be around for awhile, and responsible enough to know when they
> need to quit (because of whatever reason), or just take a break for a year
> or so. This cannot be accomplished by making it easier to let people in,
> but by making it easier to determine who is capable/worthy/whatever of
> being a developer.
Again. You say tomahto, I say tomayto.
> IMO, if we curve the number of people going into the NM queue, by setting
> up goals for them before they even get there, then we will have a faster,
> more capable set of NM administrators. Because then they will be able to
> easily determine who is right for the job. They wont have to process every
> joe shmoe that sends an email, since those people will go through a
> "trial period" (sort of a filter, if you will), before hand.
Too easily abusable by a set Cabal that wishes no more
maintainers. Straight objective goals with minimal human interference
save verification is the only way to be fair.
> The trick? Deciding what this trial period is, and how to organize it.
> IMO, most of the "deciding what they do" should be a list of suggestions
> from us, and some initiative on the part of the applicant. We shouldn't
> have to bend over backwards to give them work. At the same time, the
> areas of interest should not be roadblocked (like sponsors who do not
> upload things for an applicant, or CVS access to docs, etc...).
Therein lies the rub. Unless you can guarantee that when a NM meets a
criterion they get their gold star, the NM might as well stay
home. There's no point in playing against a stacked deck. Do you really
want those that make it through such a system? In order to beat a stacked
deck, you gotta beat them at cheating, not at the game, and then all
you've proved is that you're better at cheating... Make the gold star
garner automatically to the NM upon completion and nobody can stack the
deck.
> Ben
>
>
--
Pardon me, but you have obviously mistaken me for someone who gives a
damn.
email galt@inconnu.isu.edu
Reply to: