Re: [PATCH] fix many comment typos

From: Junio C Hamano <gitster@pobox.com>
Date: Jun 3, 2012, 21:30 UTC
Message-ID: <7vobp0xeuz.fsf@alter.siamese.dyndns.org>
In-Reply-To: <4FCB4E0A.5080107@lsrfire.ath.cx>

René Scharfe <rene.scharfe@lsrfire.ath.cx> writes:

Show 13 quoted lines

> Am 03.06.2012 09:11, schrieb Jim Meyering:
>> Perhaps a more palatable change, here and in the other two places:
>> s/seeked/cg-seek'd/, i.e.,
>>
>> - * "git bisect" showed mysterious "won't bisect on seeked tree" error message.
>> + * "git bisect" showed mysterious "won't bisect on cg-seek'd tree" error message.
>>
>> Then, if someone does this again, it will be more obvious that
>> it is not a typo.
>
> This change is only valid if the command "git bisect" at some point
> printed "won't bisect on cg-seek'd tree" instead of "won't bisect on
> seeked tree".

Yeah, that is how I should have spelled my original message; thanks for clarifying.

> And even then, it doesn't make now sense to change
> already published release notes (Documentation/RelNotes/1.5.4.4.txt),
> after the fact.

If it makes it easier to touch older release notes at the same time for a mechanical typocorrection procedure, whose primary purpose is to fix the tip of the current tree, I wouldn't mind typofixes to older release notes.

But we would need to be careful to avoid a mistake like this one. Fixing typo in the words I used to describe the changes is perfectly fine and welcomed; changing the exact output string that was copied from the implementation to show historical fact is not.

I have found Jim's patches always carefully crafted and thoughtfully done in the past, especially with the disclosure of the mechanical procedure that was used so that the result can easily be verified independently; the first hunk that showed a change that was totally against the care the procedure claimed to have taken stood up like a sore thumb and that was the only reason I noticed it.

Thanks.