[Top][All Lists]

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

Working on stuff / statement from tschwinge (was: Gnumach clean: Need ad

From: Thomas Schwinge
Subject: Working on stuff / statement from tschwinge (was: Gnumach clean: Need advice, re Header files)
Date: Thu, 9 Nov 2006 13:50:31 +0100
User-agent: Mutt/1.5.6+20040907i

[Cced to hurd-devel.]


On Wed, Nov 08, 2006 at 08:46:36PM -0800, Roland McGrath wrote:
> I haven't looked at the changes myself.  If any files have been added to
> include/, it sounds like there is some inadequate review going on before
> commits.  Folks, don't go hog-wild when you don't really know what you're
> doing.  

Roland, while I in general agree to your given advice, I don't really
think it here is appropriate towards Samuel and me.  Do you dare to
estimate what would have happened if I and the other handful of people
would not have gone hog-wild over the last months?  I wouldn't be
surprised if GNU Mach and / or GNU Hurd and / or Debian GNU/Hurd and / or
the GNU system would be dead and gray by now.  Or would you (and Thomas
Bushnell and Marcus and Neal and Mark Kettenis and Miles Bader and ...)
suddenly have gotten active again?  (Code-wise and --- perhaps even more
important --- guidance-wise towards newcomers --- which I myself still
consider to be on a large chunk, by the way!)  Or would you want to have
Szmidt (alone by himself then by now, I estimate) doing the work I'm
doing with the handful?  I absolutely do not want to praise myself here,
but I nevertheless have the feeling that these last sentences of mine are
not false and not much exaggerated.

Where do you expect that the expertise you're demanding from Samuel and
me (in this example) should come from?  Of course I can read a lot.  I
can experiment with stuff.  I can read old mailing list discussions
(which I do, by the way).  From time to time I dare to ask questions on
the bug-hurd mailing list or --- seldom --- by private mail or I even
call attention to urgent topics on the hurd-devel mailing list.  But if
then my appeals still aren't met with much response, what should I do
then?  Either I go somewhere else --- and yes, I certainly have enough
places where I could spend my time --- or I stay here and strive to make
the best out of it.  I --- so far --- decided to stay.

Now, Roland, this may sound a bit like an accusation towards you (both of
us having in mind the number of unanswered question of mine about glibc
stuff, for example), but I certainly don't want to offend you, but rather
want you (and everyone else) to think about this above comment a bit.
And if one did read it as a bit of an accusation then it is as well one
towards Thomas and Marcus and Neal.  Now, let me explain.  Thomas might
say (again) that he's always been here for us for discussing design
questions and yes, he's even been here recently for rather trivial
code-cleanup questions, but I can say the following: I'd like to discuss
about design issues with you and everyone else, but in the one-and-a-half
years of my work on the Hurd project I as yet didn't get that far.
Basically all I was doing so far was working on keeping the whole thing
running at all!  (And that work still is nowhere near finished, by the
way!)  And apart from that, I was reading and reading and reading.
Mostly alone by myself.  I learned a lot, I can say.  I learned from your
previous works, from others's.  On one side.  On the other, I also
learned how to deal with unpleasant people, I learned and I think I even
managed to motivate others.  So, Thomas, I still hope that one day we'll
get to discuss about different stuff from code clean-ups, device driver
updates, glibc (tls and others) internals and so on.  But if I and the
handful wouldn't care for that mostly-implementational work (compared to
the Hurd's architecture itself) here and now, who would do that then?
Marcus and Neal, you now might say that you've always had an open ear for
my concerns.  This is not untrue.  But also here, your impetus to answer
my questions was lower than my impetus was to ask questions was in the
beginning.  And yes, I of course realize that both of you're also working
on getting your private lifes organized and that you're having a really
busy time with thinking about the future directions of the Hurd, which I
still consider --- and I'll keep confirming that --- a very important
issue.  Roland and Thomas, you're for sure also busy in your lifes, even
without the Hurd I guess.  But who's then left to mentor newcomers like
me and others?  And no, for a project of the Hurd's momentousness one
can't only mentor oneself.  And no, I'm not a porcelain doll that can't
handle a somewhat harsh response.  But _no response_ is different.  And
it happened so often.

I've been meaning to send this email for some time now.  It may even have
been months.  The ``Folks, don't go hog-wild when you don't really know
what you're doing.'' bit was only the trigger to write it now.


Attachment: signature.asc
Description: Digital signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]