reporting requirements [Re: does not rebuild Makefile.in]

2008-08-21 Thread Ilya N. Golubev
As for the original (not of make) issue, confirming my [not a bug]
posted on Wed, 20 Aug 2008 22:49:44 +0400
(<[EMAIL PROTECTED]>).

As for attitude expressed in your reply of Wed, 20 Aug 2008 02:57:21
+0400 (MSD)
(<[EMAIL PROTECTED]>), it is
pretty widespread in many free sofware mailing lists, fora, so
commenting on it.  You wrote:

> Nice way to be helpful

Followed it with assorted what was this, what was that questions,
followed by

> Are we supposed to just *guess*??!

This may seem justified, and is at least understandable.  And
complying with all of this takes in most cases even more work than
isolating (and even possibly fixing) the bug entirely on one's own.
So the posting becomes pointless.

Sadly, with this particular software issue it was exactly what
happened.

The lesson?

In [[The Moon Is A Harsh Mistress]] (by Robert A. Heinlein) even a
machine

> Mike was designed, even before augmented, to answer questions
> tentatively on insufficient data like you do

> Mike was designed to operate on incomplete data.

People are generally able to do that pretty well.  This is reason for
users posting issues (including me) to hope ones replying to exercise
this ability, rather than request users to supply "complete data" ;-)
.

> You'd rather make people

Posted message by itself makes people do nothing.


Also note the detailed bug description (of Wed, 20 Aug 2008 22:43:38
+0400 (<[EMAIL PROTECTED]>)) on <[EMAIL PROTECTED]> (and
proposed fix (of Wed, 20 Aug 2008 22:43:53 +0400
(<[EMAIL PROTECTED]>)) on <[EMAIL PROTECTED]>).  The bug
is shown to be pretty "portable", its manifestation thus does not
depend on (answers to) most of your [what was?] questions.  Trying
random automake input ((source) package) made perfect sense.


___
Bug-make mailing list
Bug-make@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-make


Re: reporting requirements [Re: does not rebuild Makefile.in]

2008-08-21 Thread Paul Smith
On Thu, 2008-08-21 at 23:50 +0400, Ilya N. Golubev wrote:
> This may seem justified, and is at least understandable.  And
> complying with all of this takes in most cases even more work than
> isolating (and even possibly fixing) the bug entirely on one's own.
> So the posting becomes pointless.

Yes.  That's a GOOD THING, and exactly what Philip intended.

People reading this list are volunteers.  No one is getting paid, and in
fact most people are helping here to the detriment of other things they
could be doing for their jobs, families, health, recreation, etc.

As such, it's incumbent upon the people asking for help to not abuse
this generosity.  That means making an honest, _serious_ effort to solve
problems yourself _before_ asking for help.  By asking for all this
information and requiring people to provide it, we are ensuring not only
that we can actually help them, but also that they've made enough of an
effort to fix their own problem that it's worthwhile for us to take time
from whatever other important things we're doing to assist them.  You're
exactly right that half the time the effort they put in to collecting
all this data leads them to find the problem themselves, and then they
can delete the email they're writing to the list.  That's the best
possible outcome for everyone; it's what we hope for.

Yes, of course, we are smarter than machines and we are able to infer a
lot from less data, if we want to expend the effort to do so.  But we
don't want to expend that effort.  We have no reason to make that effort
to help people who won't first spend some effort to help themselves.

It may sound harsh, and some people will put it more bluntly than
others, but that's the way things work in open source software.  You
aren't paying for support with money, but that doesn't mean it's free.
Instead you have to pay for it with effort.


Plus, look at it this way: by making that effort and fixing your own
problem you learned a lot more than you would have if we'd just given
you the answer :-).



___
Bug-make mailing list
Bug-make@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-make