[U-Boot] [PATCH] M28: GPIO pin validity check added

[Wolfgang Denk]

Dear Robert,

In message <1EC2DDDF-293C-4C6A-A02B-11F45D06205A at delien.nl> you wrote:
> 
> How nice of you to drop a message, I really appreciate it.

What do you mean?  Which message do you claim I have omitted?  And
where?

> In the end, doing things right always pays, but that's not always obvious
> at the beginning. I figured that I could just hand-craft these three

True.  A lesson I have learned is that one should always watch
himself, and as soon as you find yourself repeating something you
already did in the past, you should sk yourself how likely it might be
that you have to repeat the same work a third time, or even more
often.  And then think how such repetition can be avoided or at least
simplified (say, by writing a script, or at least by creating a
detailled log of your actions, so you have the base for a script the
next time you run into that thing).

> patches in three hours and be done with it. Just for that, switching from
> SVN to GIT didn't seem worth the effort.

3 hours?  Hm... I think it should be possible to read enough git
documents and tutorials in one hour to get at least a basic
understanding of git and import and export from / to a SVN repository.
Add another hour for some practical experiments with a simple test
setup. Add another hour for doing it with your real data...

> Now it turns out that whole the mxs gpio subsystem just doesn't work
> (consider the irony of bickering about tabs and spaces), so I'll may send

Do you understand the benefits and importance of a common coding
style in a big project with many, many contributors?

> It's good to get used to GIT, it's just that timing is a bit inconvenient
> right now. In two months from now, when our A-sample is brought
> up and all subsystems are working, things will look differently.

That's exacty the reasoning I hear always in such situations - and it
always and unavoidably leads into the well-known pattern that we never
have the time to do things right, but we always have the time to do
them twice.

> I have searched the web, but Mail really doesn't support it. Apple is
> using format=flow and assumes everybody handles that properly. 

Then use another MUA.  There is tons of them.

> That's very Zen-like. And between you and me; Yes, we've had our
> history there, if you remember. But I have learned from that and tried
> to do things differently, trying to be cooperative, modest and accepting
> the Custodian's authorities. But still I didn't feel appreciated or even
> welcome. I don't need gratitude, but I do deserve the same social
> courtesy as people do in real life. That's what I try to do.

Working in a big project is never easy, especially not when doing the
first steps into a long-runnign community that has (out of necessity)
highly optimized it's communication.  The removal of communication
overhead may easily be (mis-) taken as unkindliness.

It is important to adjust your own expectations: unless you add a
really cool new feature or fix a really nasty, long-hunted bug, you
will probably never receive any explicit praise when you submit a
poatch - even if it is technically perfect and works like a charme.
On the other hand, you can be pretty sure that eveny minor issues like
CondigStyle violations (say, this TAB versus spaces issue) will be
pointed out without mercy, even if you have submitted the same trivial
patch already a number of times before.

This is not meant personally - it is a necessary optimization of the
process that tries to minimize the efforts spent in later stages.
Just have a look at the git commits, how much work we spent again and
again to clean up bad code that slipped in in the past, when we were
not so consequent yet.  And we have to do this cleanup, to be able to
use other tools that help us to work efficiently - like for example
checkpatch.pl

> For example, if somebody - who you have never met before - is trying
> to help you in real life, would you respond like this if the help turns out
> to be a bit clumsy:
> 
> > AARGH! Please add the following lines to the commit message:
> > 
> > Cc: Marek Vasut <marek.vasut at gmail.com>
> > Cc: Stefano Babic <sbabic at denx.de>
> > 
> > !!!!

What exactly are you complaining about here? Marek even wrote
"Please".  The "AARGH!" can be anything. Maybe Marek meant: "Argh.
What an idion am I noticing this only now.  I should have told you
before, in the first round of review. Please excuse that my negligence
causes you additional, avoidable efforts."

Oh - you thought he was referring to you?  This is one of the problems
of communication by e-mail and other electronic media where you cannot
see nor hear the other side.  Maybe he actually was.  We don't know,
and actually we don't care.

Basic rule: never take such comments personally, always stick to the
raw facts.  Expect that the only purpose of the reviewer is to help to
improve the quality of the code.

> Even without politenesses like 'sorries' and 'pleases', this looks a lot
> friendlier and is educational at the same time:
> 
> Next time, include the lines below in the body of your message, and the
> mailing list will take care of the rest:

But that's nearly twice as many characters to type, and probably
much less memorable :-)

> Thank you Wolfgang.

You are welcome.  I really mean it.  And I am sure that many on this
list welcome all comments, bug reports, contributions, questions. I am
sure that Marek does so, too.  It's just that we have _lots_ of
e-mails to read, and tons of patches to process, so we try to optimize
this process to the extreme.  It's unfortunate that we cannot add
polite embellishments and lengthy explanations with every message we
send.  But there is simply no time for it.  This is a highly specilize
technical channel, and the thing we really care about (and are
willing to spend time for) is the technical progress of the project.
Never take this personally. Even if we are terse and sometimes harsh,
we never intend offence (well, usually not ;-)


Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Defaults are wonderful, just like fire.
                  - Larry Wall in <1996Mar6.004121.27890 at netlabs.com>