Junio C Hamano wrote:

Felipe Contreras [off-list ref] writes:

quoted

quoted

quoted

Is it though?

Yes.

If the proposed log message says "as intended", the author thinks it
is.

The question is not if the author of the patch thinks this is the way
`-s` is intended to work, the question is if this is the way `-s` is
intended to work.

The "author" refers to the author of the "proposed log message" of
the patch in question, i.e. me in this case.  The author of the
patch under discussion thinks it is, so asking "Is it?",

This is the full quote:

====
Let's fix the interactions of these bits to first make "-s" work as intended.
====

If instead you meant this:

====
Let's fix the interactions of these bits to first make "-s" work as I intend.
====

Then that's not a rationale, you are essentially saying "let's do X because I
want".

I am not interested in getting involved in unproductive arguments with you
(or with anybody else for that matter).

This is the way the review process works and all git developers have to go
trough it.

We all have to convince others our proposed change is desirable.

Your patch is implementing a backwards-incompatible change:

  git diff -s --raw master

That command used not produce any output and after your patch it now produces
output.

Your commit message does not provide a rationale as to why *we* want to
implement this backwards-incompatible change.

"This is the way *I* intend `-s` to work" is not a rationale.

And it led to unproductive and irritating waste of time number of times, and
eventually you were asked to leave the development community for at least a
few times.

That is blatantly false. As a member of Git's Project Leadership Committee, you
should know precisely how many times the committee has excercised this power,
and it hasn't been "a few times", it has been one time.

And this is a smoke screen: your commit message still doesn't provide any
rationale as to why `-s` should work the way *you* intend.

Throwing personal attacks at a reviewer for merely pointing out an issue in the
commit message is far from productive.

-- 
Felipe Contreras