Would it make sense to add a commit.signOff config?

> I find it interesting that of all the git commands that take a --signoff
> argument, format-patch is the only one that has a corresponding config
> to turn it on for all invocations. I would have thought that git commit
> needs this even more; I find it reasonable for users to decide that they
> want to add a Signed-off-by trailer to all the commits they make.
> 
> But then, what about merge, cherry-pick, and revert? Does it make sense
> to turn it on globally for those as well?
> 
> The background of this question is lazygit, which has a config for
> enabling --signoff for all commits made by it. We recently had a feature
> request to extend the meaning of that config to revert as well, and the
> discussion came up whether that's a good idea, and if so, whether it
> should also extend to cherry-pick and merge.
> 
> Any opinions about this?

> this was discussed recently[0] and the point that was made is that SOA are
> meant to be relevant mainly in open source contexts and should be explicitly
> given.
>
> Carlo
>
> [0] https://lore.kernel.org/git/xmqq4iwvfx8s.fsf@gitster.g/

> Carlo Marcelo Arenas Belón <carenas@gmail.com> writes:
>
>> this was discussed recently[0] and the point that was made is that SOA are
>> meant to be relevant mainly in open source contexts and should be explicitly
>> given.
>>
>> Carlo
>>
>> [0] https://lore.kernel.org/git/xmqq4iwvfx8s.fsf@gitster.g/
>
> Thanks for a spelunking starter.
>
> Perhaps we should resurrect a proposed patch from 2020 (which no
> longer apply cleanly, but quoted to show what the additinoal text
> said).
>
> https://lore.kernel.org/git/xmqqpnfw8gyn.fsf@gitster-ct.c.googlers.com/
>
> to save time from potential contributors?  I do not want to see new
> contributors feeling they wasted their time after putting their
> effort.

> diff --git a/Documentation/git-commit.txt b/Documentation/git-commit.txt
> index ced5a9beab..1909551087 100644
> --- a/Documentation/git-commit.txt
> +++ b/Documentation/git-commit.txt
> @@ -171,6 +171,13 @@ The `-m` option is mutually exclusive with `-c`, `-C`, and `-F`.
>  	the rights to submit this work under the same license and
>  	agrees to a Developer Certificate of Origin
>  	(see http://developercertificate.org/ for more information).
> ++
> +As it makes it harder to argue against one who tells the court "that
> +log message ends with a SoB by person X but it is very plausible
> +that it was done by inertia without person X really intending to
> +certify what DCO says, and the SoB is meaningless." to more
> +publicized ways to add SoB automatically, Git does not (and will not)
> +have a configuration variable to enable it by default.

> Thanks for a spelunking starter.
> 
> Perhaps we should resurrect a proposed patch from 2020 (which no
> longer apply cleanly, but quoted to show what the additinoal text
> said).
> 
> https://lore.kernel.org/git/xmqqpnfw8gyn.fsf@gitster-ct.c.googlers.com/
> 
> to save time from potential contributors?  I do not want to see new
> contributors feeling they wasted their time after putting their
> effort.

> diff --git a/Documentation/git-commit.txt b/Documentation/git-commit.txt
> index ced5a9beab..1909551087 100644
> --- a/Documentation/git-commit.txt
> +++ b/Documentation/git-commit.txt
> @@ -171,6 +171,13 @@ The `-m` option is mutually exclusive with `-c`, `-C`, and `-F`.
>  	the rights to submit this work under the same license and
>  	agrees to a Developer Certificate of Origin
>  	(see http://developercertificate.org/ for more information).
> ++
> +As it makes it harder to argue against one who tells the court "that
> +log message ends with a SoB by person X but it is very plausible
> +that it was done by inertia without person X really intending to
> +certify what DCO says, and the SoB is meaningless." to more

> "brian m. carlson" <sandals@crustytoothpaste.net> writes:
> 
> > I think this would be a fine approach or we could add an entry in the
> > Git FAQ (or both).  I agree that this is something that comes up
> > frequently and writing it down in a more visible way would be kinder to
> > users and mailing list participants.
> 
> Something like this, then?

> On 2025-12-15 at 01:59:46, Junio C Hamano wrote:
>> "brian m. carlson" <sandals@crustytoothpaste.net> writes:
>> 
>> > I think this would be a fine approach or we could add an entry in the
>> > Git FAQ (or both).  I agree that this is something that comes up
>> > frequently and writing it down in a more visible way would be kinder to
>> > users and mailing list participants.
>> 
>> Something like this, then?
>
> Yup, that looks good to me.

>
> "brian m. carlson" <sandals@crustytoothpaste.net> writes:
>
> > I think this would be a fine approach or we could add an entry in the
> > Git FAQ (or both).  I agree that this is something that comes up
> > frequently and writing it down in a more visible way would be kinder to
> > users and mailing list participants.
>
> Something like this, then?
>
> ---- >8 ----
> Subject: [PATCH] document that foo.signoff will not be added
>
> Every now and then we see this coming up on the list.  Let's help
> new contributors who are not aware of past discussions by clearly
> documenting our past consensus.
>
> Helped-by: brian m. carlson <sandals@crustytoothpaste.net>
> Signed-off-by: Junio C Hamano <gitster@pobox.com>
> ---
>  Documentation/gitfaq.adoc         | 15 +++++++++++++++
>  Documentation/signoff-option.adoc | 12 ++++++++++++
>  2 files changed, 27 insertions(+)
>
> diff --git c/Documentation/gitfaq.adoc w/Documentation/gitfaq.adoc
> index f2917d142c..656093047a 100644
> --- c/Documentation/gitfaq.adoc
> +++ w/Documentation/gitfaq.adoc
> @@ -83,6 +83,21 @@ Windows would be the configuration `"C:\Program Files\Vim\gvim.exe" --nofork`,
>  which quotes the filename with spaces and specifies the `--nofork` option to
>  avoid backgrounding the process.
>
> +[[sign-off]]
> +Why not have `commit.signoff` and other configuration variables?::
> +       As it makes it harder to argue against one who tells the court
> +       "the log message of the commit ends with a Signed-off-by
> +       trailer by person X, but it is very plausible that it was done
> +       by inertia without person X really intending to certify what
> +       DCO says, hence the Signed-off-by trailer is meaningless", if
> +       we add more publicized ways to add sign-off automatically, Git
> +       does not (and will not) have a configuration variable to
> +       enable the `--signoff` command line option it by default.

> ++
> +There exists `format.signoff`, but that is a historical mistake, and
> +it is not an excuse to pile on more mistakes of the same kind on top.
> +
> +
>  Credentials
>  -----------
>
> diff --git c/Documentation/signoff-option.adoc w/Documentation/signoff-option.adoc
> index cddfb225d1..0b869dfbe4 100644
> --- c/Documentation/signoff-option.adoc
> +++ w/Documentation/signoff-option.adoc
> @@ -16,3 +16,15 @@ endif::git-commit[]
>  +
>  The `--no-signoff` option can be used to countermand an earlier `--signoff`
>  option on the command line.
> ++
> +As it makes it harder to argue against one who tells the court "the
> +log message of the commit ends with a Signed-off-by trailer by person
> +X, but it is very plausible that it was done by inertia without person
> +X really intending to certify what DCO says, hence the Signed-off-by
> +trailer is meaningless", if we add more publicized ways to add
> +sign-off automatically, Git does not (and will not) have a
> +configuration variable to enable the `--signoff` command line option
> +it by default.
> ++
> +There exists `format.signoff`, but that is a historical mistake, and
> +it is not an excuse to pile on more mistakes of the same kind on top.

>> diff --git c/Documentation/signoff-option.adoc w/Documentation/signoff-option.adoc
>> index cddfb225d1..0b869dfbe4 100644
>> --- c/Documentation/signoff-option.adoc
>> +++ w/Documentation/signoff-option.adoc
>> @@ -16,3 +16,15 @@ endif::git-commit[]
>>  +
>>  The `--no-signoff` option can be used to countermand an earlier `--signoff`
>>  option on the command line.
>> ++
>> +As it makes it harder to argue against one who tells the court "the
>> +log message of the commit ends with a Signed-off-by trailer by person
>> +X, but it is very plausible that it was done by inertia without person
>> +X really intending to certify what DCO says, hence the Signed-off-by
>> +trailer is meaningless", if we add more publicized ways to add
>> +sign-off automatically, Git does not (and will not) have a
>> +configuration variable to enable the `--signoff` command line option
>> +it by default.
>> ++
>> +There exists `format.signoff`, but that is a historical mistake, and
>> +it is not an excuse to pile on more mistakes of the same kind on top.
>
> This feels like it's missing context (it'll take the reader a while to
> figure out why the paragraphs are there and that the two are related),

> Elijah Newren <newren@gmail.com> writes:
> 
>>> diff --git c/Documentation/signoff-option.adoc w/Documentation/signoff-option.adoc
>>> index cddfb225d1..0b869dfbe4 100644
>>> --- c/Documentation/signoff-option.adoc
>>> +++ w/Documentation/signoff-option.adoc
>>> @@ -16,3 +16,15 @@ endif::git-commit[]
>>>  +
>>>  The `--no-signoff` option can be used to countermand an earlier `--signoff`
>>>  option on the command line.
>>> ++
>>> +As it makes it harder to argue against one who tells the court "the
>>> +log message of the commit ends with a Signed-off-by trailer by person
>>> +X, but it is very plausible that it was done by inertia without person
>>> +X really intending to certify what DCO says, hence the Signed-off-by
>>> +trailer is meaningless", if we add more publicized ways to add
>>> +sign-off automatically, Git does not (and will not) have a
>>> +configuration variable to enable the `--signoff` command line option
>>> +it by default.
>>> ++
>>> +There exists `format.signoff`, but that is a historical mistake, and
>>> +it is not an excuse to pile on more mistakes of the same kind on top.
>>
>> This feels like it's missing context (it'll take the reader a while to
>> figure out why the paragraphs are there and that the two are related),
> 
> Very true.  It may be sufficient to leave this part unmodified,
> keep the updates to gitfaq document, and do nothing else.

> On Sun, Dec 14, 2025 at 6:00 PM Junio C Hamano <gitster@pobox.com> wrote:
>> --- c/Documentation/gitfaq.adoc
>> +++ w/Documentation/gitfaq.adoc
>> @@ -83,6 +83,21 @@ Windows would be the configuration `"C:\Program Files\Vim\gvim.exe" --nofork`,
>>  which quotes the filename with spaces and specifies the `--nofork` option to
>>  avoid backgrounding the process.
>>
>> +[[sign-off]]
>> +Why not have `commit.signoff` and other configuration variables?::
>> +       As it makes it harder to argue against one who tells the court
>> +       "the log message of the commit ends with a Signed-off-by
>> +       trailer by person X, but it is very plausible that it was done
>> +       by inertia without person X really intending to certify what
>> +       DCO says, hence the Signed-off-by trailer is meaningless", if
>> +       we add more publicized ways to add sign-off automatically, Git
>> +       does not (and will not) have a configuration variable to
>> +       enable the `--signoff` command line option it by default.
> 
> This feels kind of hard to parse for me.  Maybe it's just the lack of
> sentence breaks, particularly near the end.  Let me take a stab at an
> alternative:
> 
> Git intentionally does not (and will not) provide a configuration variable,
> such as `commit.signoff`, to automatically add `--signoff` by default.
> The reason is to protect the legal and intentional significance of a sign-off.
> If there were more automated and widely publicized ways for sign-offs to be
> appended, it would become easier for someone to argue later that a
> "Signed-off-by" trailer was just added out of habit or by automation,
> without the committer's full awareness or intent to certify their agreement
> with the Developer Certificate of Origin (DCO) or a similar statement.
> This would weaken the sign-off’s value and could undermine its credibility
> in legal or contractual situations. To uphold the integrity of a sign-off,
> Git only adds it when explicitly requested, rather than through automatic
> configuration settings.
> 
> Maybe the last sentence or two are a bit redundant and could be
> stricken.  Anyway, thoughts?

>
> Every now and then we see this coming up on the list.  Let's help
> new contributors who are not aware of past discussions by clearly
> documenting our past consensus.
>
> Helped-by: brian m. carlson <sandals@crustytoothpaste.net>
> Helped-by: Elijah Newren <newren@gmail.com>
> Helped-by: Johannes Sixt <j6t@kdbg.org>
> Signed-off-by: Junio C Hamano <gitster@pobox.com>
> ---
>  * The text is from Elijah and Johannes this round; I am merely
>    playing a copybot.

>  Documentation/gitfaq.adoc         | 19 +++++++++++++++++++
>  Documentation/signoff-option.adoc |  4 ++++
>  2 files changed, 23 insertions(+)
>
> diff --git c/Documentation/gitfaq.adoc w/Documentation/gitfaq.adoc
> index f2917d142c..0e40befb77 100644
> --- c/Documentation/gitfaq.adoc
> +++ w/Documentation/gitfaq.adoc
> @@ -83,6 +83,25 @@ Windows would be the configuration `"C:\Program Files\Vim\gvim.exe" --nofork`,
>  which quotes the filename with spaces and specifies the `--nofork` option to
>  avoid backgrounding the process.
>
> +[[sign-off]]
> +Why not have `commit.signoff` and other configuration variables?::
> +       Git intentionally does not (and will not) provide a
> +       configuration variable, such as `commit.signoff`, to
> +       automatically add `--signoff` by default.  The reason is to
> +       protect the legal and intentional significance of a sign-off.
> +       If there were more automated and widely publicized ways for
> +       sign-offs to be appended, it would become easier for someone
> +       to argue later that a "Signed-off-by" trailer was just added
> +       out of habit or by automation, without the committer's full
> +       awareness or intent to certify their agreement with the
> +       Developer Certificate of Origin (DCO) or a similar statement.
> +       This could undermine the sign-off’s credibility in legal or
> +       contractual situations.
> ++
> +There exists `format.signoff`, but that is a historical mistake, and
> +it is not an excuse to pile on more mistakes of the same kind on top.
> +
> +
>  Credentials
>  -----------
>
> diff --git c/Documentation/signoff-option.adoc w/Documentation/signoff-option.adoc
> index cddfb225d1..9a80d60f1b 100644
> --- c/Documentation/signoff-option.adoc
> +++ w/Documentation/signoff-option.adoc
> @@ -16,3 +16,7 @@ endif::git-commit[]
>  +
>  The `--no-signoff` option can be used to countermand an earlier `--signoff`
>  option on the command line.
> ++
> +Git does not (and will not) have a configuration variable to enable
> +the `--signoff` command line option by default; see the
> +`commit.signoff` entry in the gitfaq for more details.

>
> Elijah Newren <newren@gmail.com> writes:
>
> > I like Johannes' edits to the text I had; this commit looks good to me.
>
> If you liked the final result, it probably should not matter much,
> but I cannot tell if you are (obliquely) pointing out that I failed
> to grab and apply Johannes's edit or if you are doubly agreeing with
> the resulting text in the message you are responding to.

> Every now and then we see this coming up on the list.  Let's help
> new contributors who are not aware of past discussions by clearly
> documenting our past consensus.
> 
> Helped-by: brian m. carlson <sandals@crustytoothpaste.net>
> Helped-by: Elijah Newren <newren@gmail.com>
> Helped-by: Johannes Sixt <j6t@kdbg.org>
> Signed-off-by: Junio C Hamano <gitster@pobox.com>
> ---
>  * The text is from Elijah and Johannes this round; I am merely
>    playing a copybot.
> 
>  Documentation/gitfaq.adoc         | 19 +++++++++++++++++++
>  Documentation/signoff-option.adoc |  4 ++++
>  2 files changed, 23 insertions(+)
> 
> diff --git c/Documentation/gitfaq.adoc w/Documentation/gitfaq.adoc
> index f2917d142c..0e40befb77 100644
> --- c/Documentation/gitfaq.adoc
> +++ w/Documentation/gitfaq.adoc
> @@ -83,6 +83,25 @@ Windows would be the configuration `"C:\Program Files\Vim\gvim.exe" --nofork`,
>  which quotes the filename with spaces and specifies the `--nofork` option to
>  avoid backgrounding the process.
>  
> +[[sign-off]]
> +Why not have `commit.signoff` and other configuration variables?::
> +	Git intentionally does not (and will not) provide a
> +	configuration variable, such as `commit.signoff`, to
> +	automatically add `--signoff` by default.  The reason is to
> +	protect the legal and intentional significance of a sign-off.
> +	If there were more automated and widely publicized ways for
> +	sign-offs to be appended, it would become easier for someone
> +	to argue later that a "Signed-off-by" trailer was just added
> +	out of habit or by automation, without the committer's full
> +	awareness or intent to certify their agreement with the
> +	Developer Certificate of Origin (DCO) or a similar statement.
> +	This could undermine the sign-off’s credibility in legal or
> +	contractual situations.
> ++
> +There exists `format.signoff`, but that is a historical mistake, and
> +it is not an excuse to pile on more mistakes of the same kind on top.

> +
> +
>  Credentials
>  -----------
>  
> diff --git c/Documentation/signoff-option.adoc w/Documentation/signoff-option.adoc
> index cddfb225d1..9a80d60f1b 100644
> --- c/Documentation/signoff-option.adoc
> +++ w/Documentation/signoff-option.adoc
> @@ -16,3 +16,7 @@ endif::git-commit[]
>  +
>  The `--no-signoff` option can be used to countermand an earlier `--signoff`
>  option on the command line.
> ++
> +Git does not (and will not) have a configuration variable to enable
> +the `--signoff` command line option by default; see the
> +`commit.signoff` entry in the gitfaq for more details.

>> +There exists `format.signoff`, but that is a historical mistake, and
>> +it is not an excuse to pile on more mistakes of the same kind on top.
>
> I feel that we should s/pile on/pile/, but I am not a native speaker.
>
> This patch looks good with or without this change.

>[snip]
> ++
> +Git does not (and will not) have a configuration variable to enable
> +the `--signoff` command line option by default; see the
> +`commit.signoff` entry in the gitfaq for more details.

> On Tue, Dec 16, 2025, at 19:54, Junio C Hamano wrote:
>>[snip]
>> ++
>> +Git does not (and will not) have a configuration variable to enable
>> +the `--signoff` command line option by default; see the
>> +`commit.signoff` entry in the gitfaq for more details.
>
> This is in `next` now but:
>
> s/the gitfaq/in linkgit:gitfaq[7]/