PS: Diplomacy (follow-up to information overload)

Clayton Weaver (cgweav@eskimo.com)
Sun, 4 Oct 1998 03:38:10 -0700 (PDT)

Unless we are very lucky, we can't expect the transition to heirarchical
human patch filtering to be painless. Assume that people will accept the
responsibility and found to be unqaulified for it, not based on their
programming expertise but on their inexperience with diplomacy. People
will get high and mighty illusions about their place in the universe and
start throwing their weight around. Not because they are evil, but because
that is their habit when they acquire responsible positions. If they
have to be relieved of that position for the sake of the project, again
it will not be because they are evil, stupid, or can't code. It will
be because their native personality is not a good adaptation to the
demands of that particular job, and they lack the flexibility or
willingness to change their attitude.

What a good diplomat recognizes instinctively and has skills to deal
with is changes in semantic environment:

"Here is my new patch."

"Well, this patch has this problem here, easy to fix, and this other
one over here, not so easy because it interacts with this huge
substructure that would require a huge amount of work to change
and test the changes. So I suggest maybe rewriting it so that
it approaches the problem in a way that has less side-effects on other
parts of the kernel."

"What do you mean I'm stupid?"

This is classic, a shift from the semantic environment of a technical
discussion to the semantic environment of a perceived attack
on one's personal worth and self-image, ie interpersonal warfare. And
the person who makes that last statement isn't evil, isn't stupid, it's
just a habit of behavior. The Diplomat knows how to deal with it:

"No one said you were stupid. Your patch just didn't reflect sufficient
awareness of the entire code context that the code that you propose to
change is interdependent with. Adding this patch would be like moving a
50-story office building a few feet over because someone want's an extra
driveway entrance to the parking lot for the pub beside it. Now that you
know, no doubt you can code up a less chaotic version in terms of the
extent of it's effects on the current kernel source tree."

If the person submitting the patch is still pissed after that, too bad.

Regards, Clayton Weaver cgweav@eskimo.com (Seattle)

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/