blob: 45595fd8a66bd92777c6ddad355e0c5a090cc408 [file] [log] [blame]
Mauro Carvalho Chehab609d99a2016-09-19 08:07:56 -03001.. _managementstyle:
2
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -03003Linux kernel management style
4=============================
Linus Torvalds1da177e2005-04-16 15:20:36 -07005
6This is a short document describing the preferred (or made up, depending
7on who you ask) management style for the linux kernel. It's meant to
Mauro Carvalho Chehab8c27ceff32016-10-18 10:12:27 -02008mirror the process/coding-style.rst document to some degree, and mainly written to
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -03009avoid answering [#f1]_ the same (or similar) questions over and over again.
Linus Torvalds1da177e2005-04-16 15:20:36 -070010
11Management style is very personal and much harder to quantify than
12simple coding style rules, so this document may or may not have anything
13to do with reality. It started as a lark, but that doesn't mean that it
14might not actually be true. You'll have to decide for yourself.
15
16Btw, when talking about "kernel manager", it's all about the technical
17lead persons, not the people who do traditional management inside
18companies. If you sign purchase orders or you have any clue about the
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030019budget of your group, you're almost certainly not a kernel manager.
20These suggestions may or may not apply to you.
Linus Torvalds1da177e2005-04-16 15:20:36 -070021
Jiri Pirkoe11e3642008-10-29 14:00:58 -070022First off, I'd suggest buying "Seven Habits of Highly Effective
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030023People", and NOT read it. Burn it, it's a great symbolic gesture.
Linus Torvalds1da177e2005-04-16 15:20:36 -070024
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030025.. [#f1] This document does so not so much by answering the question, but by
26 making it painfully obvious to the questioner that we don't have a clue
27 to what the answer is.
Linus Torvalds1da177e2005-04-16 15:20:36 -070028
29Anyway, here goes:
30
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030031.. _decisions:
Linus Torvalds1da177e2005-04-16 15:20:36 -070032
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300331) Decisions
34------------
Linus Torvalds1da177e2005-04-16 15:20:36 -070035
36Everybody thinks managers make decisions, and that decision-making is
37important. The bigger and more painful the decision, the bigger the
38manager must be to make it. That's very deep and obvious, but it's not
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030039actually true.
Linus Torvalds1da177e2005-04-16 15:20:36 -070040
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030041The name of the game is to **avoid** having to make a decision. In
Linus Torvalds1da177e2005-04-16 15:20:36 -070042particular, if somebody tells you "choose (a) or (b), we really need you
43to decide on this", you're in trouble as a manager. The people you
44manage had better know the details better than you, so if they come to
45you for a technical decision, you're screwed. You're clearly not
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030046competent to make that decision for them.
Linus Torvalds1da177e2005-04-16 15:20:36 -070047
48(Corollary:if the people you manage don't know the details better than
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030049you, you're also screwed, although for a totally different reason.
50Namely that you are in the wrong job, and that **they** should be managing
51your brilliance instead).
Linus Torvalds1da177e2005-04-16 15:20:36 -070052
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030053So the name of the game is to **avoid** decisions, at least the big and
Linus Torvalds1da177e2005-04-16 15:20:36 -070054painful ones. Making small and non-consequential decisions is fine, and
55makes you look like you know what you're doing, so what a kernel manager
56needs to do is to turn the big and painful ones into small things where
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030057nobody really cares.
Linus Torvalds1da177e2005-04-16 15:20:36 -070058
59It helps to realize that the key difference between a big decision and a
60small one is whether you can fix your decision afterwards. Any decision
61can be made small by just always making sure that if you were wrong (and
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030062you **will** be wrong), you can always undo the damage later by
Linus Torvalds1da177e2005-04-16 15:20:36 -070063backtracking. Suddenly, you get to be doubly managerial for making
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030064**two** inconsequential decisions - the wrong one **and** the right one.
Linus Torvalds1da177e2005-04-16 15:20:36 -070065
66And people will even see that as true leadership (*cough* bullshit
67*cough*).
68
69Thus the key to avoiding big decisions becomes to just avoiding to do
70things that can't be undone. Don't get ushered into a corner from which
71you cannot escape. A cornered rat may be dangerous - a cornered manager
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030072is just pitiful.
Linus Torvalds1da177e2005-04-16 15:20:36 -070073
74It turns out that since nobody would be stupid enough to ever really let
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030075a kernel manager have huge fiscal responsibility **anyway**, it's usually
Linus Torvalds1da177e2005-04-16 15:20:36 -070076fairly easy to backtrack. Since you're not going to be able to waste
77huge amounts of money that you might not be able to repay, the only
78thing you can backtrack on is a technical decision, and there
79back-tracking is very easy: just tell everybody that you were an
80incompetent nincompoop, say you're sorry, and undo all the worthless
81work you had people work on for the last year. Suddenly the decision
82you made a year ago wasn't a big decision after all, since it could be
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030083easily undone.
Linus Torvalds1da177e2005-04-16 15:20:36 -070084
85It turns out that some people have trouble with this approach, for two
86reasons:
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030087
Linus Torvalds1da177e2005-04-16 15:20:36 -070088 - admitting you were an idiot is harder than it looks. We all like to
89 maintain appearances, and coming out in public to say that you were
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030090 wrong is sometimes very hard indeed.
Linus Torvalds1da177e2005-04-16 15:20:36 -070091 - having somebody tell you that what you worked on for the last year
92 wasn't worthwhile after all can be hard on the poor lowly engineers
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030093 too, and while the actual **work** was easy enough to undo by just
Linus Torvalds1da177e2005-04-16 15:20:36 -070094 deleting it, you may have irrevocably lost the trust of that
95 engineer. And remember: "irrevocable" was what we tried to avoid in
96 the first place, and your decision ended up being a big one after
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -030097 all.
Linus Torvalds1da177e2005-04-16 15:20:36 -070098
99Happily, both of these reasons can be mitigated effectively by just
100admitting up-front that you don't have a friggin' clue, and telling
101people ahead of the fact that your decision is purely preliminary, and
102might be the wrong thing. You should always reserve the right to change
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300103your mind, and make people very **aware** of that. And it's much easier
104to admit that you are stupid when you haven't **yet** done the really
Linus Torvalds1da177e2005-04-16 15:20:36 -0700105stupid thing.
106
107Then, when it really does turn out to be stupid, people just roll their
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300108eyes and say "Oops, he did it again".
Linus Torvalds1da177e2005-04-16 15:20:36 -0700109
110This preemptive admission of incompetence might also make the people who
111actually do the work also think twice about whether it's worth doing or
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300112not. After all, if **they** aren't certain whether it's a good idea, you
Linus Torvalds1da177e2005-04-16 15:20:36 -0700113sure as hell shouldn't encourage them by promising them that what they
114work on will be included. Make them at least think twice before they
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300115embark on a big endeavor.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700116
117Remember: they'd better know more about the details than you do, and
118they usually already think they have the answer to everything. The best
119thing you can do as a manager is not to instill confidence, but rather a
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300120healthy dose of critical thinking on what they do.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700121
122Btw, another way to avoid a decision is to plaintively just whine "can't
123we just do both?" and look pitiful. Trust me, it works. If it's not
124clear which approach is better, they'll eventually figure it out. The
125answer may end up being that both teams get so frustrated by the
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300126situation that they just give up.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700127
128That may sound like a failure, but it's usually a sign that there was
129something wrong with both projects, and the reason the people involved
130couldn't decide was that they were both wrong. You end up coming up
131smelling like roses, and you avoided yet another decision that you could
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300132have screwed up on.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700133
134
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -03001352) People
136---------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700137
138Most people are idiots, and being a manager means you'll have to deal
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300139with it, and perhaps more importantly, that **they** have to deal with
140**you**.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700141
142It turns out that while it's easy to undo technical mistakes, it's not
143as easy to undo personality disorders. You just have to live with
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300144theirs - and yours.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700145
146However, in order to prepare yourself as a kernel manager, it's best to
147remember not to burn any bridges, bomb any innocent villagers, or
148alienate too many kernel developers. It turns out that alienating people
149is fairly easy, and un-alienating them is hard. Thus "alienating"
150immediately falls under the heading of "not reversible", and becomes a
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300151no-no according to :ref:`decisions`.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700152
153There's just a few simple rules here:
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300154
Linus Torvalds1da177e2005-04-16 15:20:36 -0700155 (1) don't call people d*ckheads (at least not in public)
156 (2) learn how to apologize when you forgot rule (1)
157
158The problem with #1 is that it's very easy to do, since you can say
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300159"you're a d*ckhead" in millions of different ways [#f2]_, sometimes without
Linus Torvalds1da177e2005-04-16 15:20:36 -0700160even realizing it, and almost always with a white-hot conviction that
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300161you are right.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700162
163And the more convinced you are that you are right (and let's face it,
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300164you can call just about **anybody** a d*ckhead, and you often **will** be
165right), the harder it ends up being to apologize afterwards.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700166
167To solve this problem, you really only have two options:
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300168
Linus Torvalds1da177e2005-04-16 15:20:36 -0700169 - get really good at apologies
170 - spread the "love" out so evenly that nobody really ends up feeling
171 like they get unfairly targeted. Make it inventive enough, and they
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300172 might even be amused.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700173
174The option of being unfailingly polite really doesn't exist. Nobody will
175trust somebody who is so clearly hiding his true character.
176
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300177.. [#f2] Paul Simon sang "Fifty Ways to Leave Your Lover", because quite
178 frankly, "A Million Ways to Tell a Developer He Is a D*ckhead" doesn't
179 scan nearly as well. But I'm sure he thought about it.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700180
181
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -03001823) People II - the Good Kind
183----------------------------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700184
185While it turns out that most people are idiots, the corollary to that is
186sadly that you are one too, and that while we can all bask in the secure
187knowledge that we're better than the average person (let's face it,
188nobody ever believes that they're average or below-average), we should
189also admit that we're not the sharpest knife around, and there will be
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300190other people that are less of an idiot than you are.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700191
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300192Some people react badly to smart people. Others take advantage of them.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700193
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300194Make sure that you, as a kernel maintainer, are in the second group.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700195Suck up to them, because they are the people who will make your job
196easier. In particular, they'll be able to make your decisions for you,
197which is what the game is all about.
198
199So when you find somebody smarter than you are, just coast along. Your
200management responsibilities largely become ones of saying "Sounds like a
201good idea - go wild", or "That sounds good, but what about xxx?". The
202second version in particular is a great way to either learn something
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300203new about "xxx" or seem **extra** managerial by pointing out something the
Linus Torvalds1da177e2005-04-16 15:20:36 -0700204smarter person hadn't thought about. In either case, you win.
205
206One thing to look out for is to realize that greatness in one area does
207not necessarily translate to other areas. So you might prod people in
208specific directions, but let's face it, they might be good at what they
209do, and suck at everything else. The good news is that people tend to
210naturally gravitate back to what they are good at, so it's not like you
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300211are doing something irreversible when you **do** prod them in some
Linus Torvalds1da177e2005-04-16 15:20:36 -0700212direction, just don't push too hard.
213
214
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -03002154) Placing blame
216----------------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700217
218Things will go wrong, and people want somebody to blame. Tag, you're it.
219
220It's not actually that hard to accept the blame, especially if people
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300221kind of realize that it wasn't **all** your fault. Which brings us to the
Linus Torvalds1da177e2005-04-16 15:20:36 -0700222best way of taking the blame: do it for another guy. You'll feel good
223for taking the fall, he'll feel good about not getting blamed, and the
224guy who lost his whole 36GB porn-collection because of your incompetence
225will grudgingly admit that you at least didn't try to weasel out of it.
226
227Then make the developer who really screwed up (if you can find him) know
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300228**in_private** that he screwed up. Not just so he can avoid it in the
Linus Torvalds1da177e2005-04-16 15:20:36 -0700229future, but so that he knows he owes you one. And, perhaps even more
230importantly, he's also likely the person who can fix it. Because, let's
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300231face it, it sure ain't you.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700232
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300233Taking the blame is also why you get to be manager in the first place.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700234It's part of what makes people trust you, and allow you the potential
235glory, because you're the one who gets to say "I screwed up". And if
236you've followed the previous rules, you'll be pretty good at saying that
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300237by now.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700238
239
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -03002405) Things to avoid
241------------------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700242
243There's one thing people hate even more than being called "d*ckhead",
244and that is being called a "d*ckhead" in a sanctimonious voice. The
245first you can apologize for, the second one you won't really get the
246chance. They likely will no longer be listening even if you otherwise
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300247do a good job.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700248
249We all think we're better than anybody else, which means that when
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300250somebody else puts on airs, it **really** rubs us the wrong way. You may
Linus Torvalds1da177e2005-04-16 15:20:36 -0700251be morally and intellectually superior to everybody around you, but
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300252don't try to make it too obvious unless you really **intend** to irritate
253somebody [#f3]_.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700254
255Similarly, don't be too polite or subtle about things. Politeness easily
256ends up going overboard and hiding the problem, and as they say, "On the
257internet, nobody can hear you being subtle". Use a big blunt object to
258hammer the point in, because you can't really depend on people getting
259your point otherwise.
260
261Some humor can help pad both the bluntness and the moralizing. Going
262overboard to the point of being ridiculous can drive a point home
263without making it painful to the recipient, who just thinks you're being
264silly. It can thus help get through the personal mental block we all
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300265have about criticism.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700266
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300267.. [#f3] Hint: internet newsgroups that are not directly related to your work
268 are great ways to take out your frustrations at other people. Write
269 insulting posts with a sneer just to get into a good flame every once in
270 a while, and you'll feel cleansed. Just don't crap too close to home.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700271
272
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -03002736) Why me?
274----------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700275
276Since your main responsibility seems to be to take the blame for other
277peoples mistakes, and make it painfully obvious to everybody else that
278you're incompetent, the obvious question becomes one of why do it in the
279first place?
280
281First off, while you may or may not get screaming teenage girls (or
282boys, let's not be judgmental or sexist here) knocking on your dressing
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300283room door, you **will** get an immense feeling of personal accomplishment
Linus Torvalds1da177e2005-04-16 15:20:36 -0700284for being "in charge". Never mind the fact that you're really leading
285by trying to keep up with everybody else and running after them as fast
Mauro Carvalho Chehab7f2b3c62016-09-19 08:07:48 -0300286as you can. Everybody will still think you're the person in charge.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700287
288It's a great job if you can hack it.