[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: -= PROPOSAL =- Release sarge with amd64



Matthew Palmer <mpalmer@debian.org> writes:

> On Fri, Jul 16, 2004 at 04:58:45PM +0200, Goswin von Brederlow wrote:
>> Matthew Palmer <mpalmer@debian.org> writes:
>> > On Fri, Jul 16, 2004 at 02:27:38PM +0200, Ingo Juergensmann wrote:
>> >> People are mostly more interested in contributing and learning if you
>> >> guide them during real work. Sure, they must show some serious interest in
>> >> that task. A simple "here i am, what can i do?" is not enough.
>> >> But with regards to your bicycle example: a child will not learn riding
>> >> bike, when you just tell it: "look, there's a bike, there's a somewhere
>> >> some howtos.. .have fun!" Instead you'll need to guide it while it sits on
>> >> the bike, you'll need to catch it, if it might fall, etc.
>> >
>> > What a perfect analogy.
>> >
>> > It takes time to teach someone how to ride a bike.  If it's your own child,
>> > you'll probably want to spend time doing it.  How many random strangers have
>> > you taught to ride a bike?
>> >
>> > Furthermore, note how incredibly difficult it is to teach someone how to
>> > ride a bike whilst simultaneously riding yourself.  You have to stop riding,
>> > hop off, and go help the other person.
>> >
>> > Now, ask yourself this -- would you rather have ftp-masters, listmasters,
>> > project secretary, and a million other things in Debian riding their bikes
>> > to the benefit of the project, or spending all their time herding newbies
>> > who are just as likely to decide it's not their cup of tea after all and
>> > wander off and do something else as stay and help?
>> 
>> The later. Some are bound to stick to it and, to continue the bicicle
>
> And many will not.  Look at AMs for an idea of how frequently that happens. 
> And upfront there's bugger all idea of who's going to stick to it or not. 
> If you get ten people saying "yeah, I'll do that" when you put out a general
> request for assistance, which of them is going to stick to it?  Sometimes
> it'll be none of them.  Look at the d-i install report processing.

Sooner or later the extra people will be needed. Could be due to
growth, accident or resigning. It will happen.

Would you rather get 10 volunteers without anyone around to teach them
that give up after a week then?

Weeding them out now and getting that fraction of persons that sticks
with it and is good at it sounds far better.

>> analogy, as the RM process shows giving persons a bike with support
>> wheels first so they don't fall is a great way to teach them how to
>> ride the bike.
>
> But that was with people who had rolled up to the RM with their bike, shown
> that they were willing to jump on, try for a bit, collapse in a heap every
> now and then, and get back on again.  None of the RM assistants got their
> stripes by just sticking their hand up when the RM yelled across the room
> "hey, anyone want to be an RM assistant?".

But that seems imppossible for several jobs in which James and Ryan
are involved, like helping with a mips buildd.

The attidutes have to change from being a one man army to the more
cooperative way the RM works now. That would be a great improvment.

>> Remember debian is growing exponentially. So at some point more people
>> are needed to handle the extra work and the longer you wait the less
>> time there is to teach them (or the more harm will come to debian
>> because they need to be teached).
>
> Teams may need to be enlarged, but we obviously disagree on the method of
> making the enlargement.

When offers of enlarging teams in desperate needs of enlargement are
just rudely rejected something is wrong. And that was what happened in
the buildd case back then.

> - Matt

MfG
        Goswin



Reply to: