Subject: Re: NetBSD: Certified mom-ready.
To: Ted Lemon <mellon@isc.org>
From: Erik E. Fair <fair@clock.org>
List: netbsd-advocacy
Date: 04/20/1999 13:26:15
At 10:46 -0700 4/20/99, Ted Lemon wrote:
>> If you're an organization trying to do something with/for the project,
>> and you get two widely divergent answers from two alleged
>> representatives of the project, then what do you do?  Pick one?  (not
>> necessarily 'safe' -- what if you base something important on the
>> wrong answer?)  Spend the time to figure out which is right?  (that's
>> the right answer, but it has an up-front cost.)  Punt?
>
>You pick one.   But placing the responsibility to make the choice on
>one or a few individuals hasn't worked very well historically, has it?
>As a whole, I think we are sufficiently egoless and intelligent to
>make good choices.   Individually, we all have blind spots and
>opinions that can override our common sense.   Also, if you make one
>small group of people make all these decisions, they tend to get
>cranky, and their output (criticism) tends to be of low quality (it
>pisses people off rather than motivating people to fix the problems).
>
>Better to structure things so that these sorts of problems don't come
>up.

It would be nice to have a structure that had a Solomon-like quality of
wisdom which would make all such debates vanish before the mutually
unacceptable compromise. In my experience, this is a good goal of any
management, but in practice is not achievable, so it's really not worth
fretting about not being ideal in this way (which is not to say that
continuous improvement of our management is something to be avoided just
because perfection is impossible!).

Some decisions can only be made with experience you don't have yet. In this
case, it is best to simply decide *something*, implement, and measure the
results. If it goes wrong, OK, it goes wrong, you analyze why, and
re-implement the better solution. It's also *key* to own up to your users
that you got it wrong, and are gonna fix it, in public. I think our biggest
problem is that Secrecy breeds Fear, Uncertainty, and Doubt (and
Suspicion!), so, we need to communicate more often, and more completely.

No one is perfect; we are all doing the best that we can within a wide
variety constraints on time & resources. We just need to remember that when
we're sending E-mail to each other.

Fortunately, we're all smart, we don't make so many mistakes that we
alienate our users, and we're pretty quick about fixing *some* things. All
I'm trying to do is spread the work around a bit more, so that the things
that are festering actually get dealt with, make some obvious opportunities
for contribution to the project (I know there are people out there who want
to contribute, but how?), and toss in some individual accountability so
that it's possible to find the place where The Buck Stops for any given
part of the project.

	Erik <fair@clock.org>