Hi Gilles,

as you said: Git is public. As long as it is kept public there's no reason to "duplicate details". My 2 cents.

sincerely, Frank

On Sun, Sep 20, 2020 at 09:41:26PM +0000, Gilles Peskine via mbed-tls wrote:

Hello,

The Mbed TLS team is soliciting feedback on a proposed change of our contributor acknowledgement policy.

Background: our current policy is to acknowledge all contributors who are not Arm employees in the ChangeLog files. This is a tradition that dates back from when the version control history of Mbed TLS (then PolarSSL) was not public.

Proposed change: the ChangeLog file will no longer mention who contributed a patch (“Contributed by …”). A patch will require a ChangeLog entry only if it makes a user-visible changes, regardless of who contributed it. We will still credit reporters in the ChangeLog file, at least for security issues.

Rationale: the ChangeLog file had a dual purpose: informing users, and acknowledging contributors. It will now have the single purpose of informing users. There are better ways of acknowledging contributors nowadays: the Git history is public, and your contributions are visible from your GitHub profile. (Note that we don't squash commits, and attempt to preserve authorship as much as possible when we have to rebase a patch.) Furthermore, writing a good ChangeLog entry often turns out to require some back-and-forth, so lessening this requirement will make it easier to accept contributions and to merge them faster (this is in fact our primary motivation for the change).

We are also thinking of creating an acknowledgement page out-of-tree. It isn't clear yet what its scope would be, but it would include acknowledging contributions in the form of reviews, which we wish to encourage. (Please stay tuned for future announcements and discussions regarding the Mbed TLS review process.)

Question to potential contributors, or people who have participated in other open-source projects: do you think that there is any value in keeping acknowledgements in a file that is distributed with Mbed TLS? Is there any value in maintaining an exhaustive list of contributors in a form other than the Git history?

Best regards,

-- Gilles Peskine Mbed TLS developer

-- mbed-tls mailing list mbed-tls@lists.trustedfirmware.org https://lists.trustedfirmware.org/mailman/listinfo/mbed-tls