linux-kernel - Re: [Ksummit-discuss] crediting bug reports and fixes folded into original patch

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <4b973b02352860978331eef110bae69e0d576adf.camel@HansenPartnership.com>
Date:   Thu, 03 Dec 2020 13:13:47 -0800
From:   James Bottomley <James.Bottomley@...senPartnership.com>
To:     Konstantin Ryabitsev <konstantin@...uxfoundation.org>,
        Joe Perches <joe@...ches.com>
Cc:     "ksummit-discuss@...ts.linuxfoundation.org" 
        <ksummit-discuss@...ts.linuxfoundation.org>,
        Vlastimil Babka <vbabka@...e.cz>,
        LKML <linux-kernel@...r.kernel.org>,
        Jonathan Corbet <corbet@....net>
Subject: Re: [Ksummit-discuss] crediting bug reports and fixes folded into
 original patch

On Thu, 2020-12-03 at 14:17 -0500, Konstantin Ryabitsev wrote:
> On Thu, Dec 03, 2020 at 08:55:54AM -0800, Joe Perches wrote:
> > Perhaps automate a mechanism to capture that information as
> > git notes for the patches when applied.
> 
> Git notes have a limited usefulness for this -- they are indeed part
> of the repository, but they aren't replicated unless someone does a 
> --mirror clone (or specifically fetches refs/notes/*). If the goal is
> to improve visibility for contributors, then putting this info into a
> git note will hardly make more difference than providing a Link: that
> someone has to follow to a list archival service. 
> 
> I can offer the following proposal:
> 
> - kernel.org already monitors all mailing lists that are archived on 
>   lore.kernel.org for the purposes of pull request tracking 
>   (pr-tracker-bot).
> - in the near future, we will add a separate process that will 
>   auto-explode all pull requests into individual patches and add them
>   to a separate public-inbox archive (think of it as another 
>   transparency log, since pull requests are transient and opaque).
> 
> We can additionally:
> 
> - identify all Link: and Message-Id: entries in commit messages, 
>   retrieve the threads they refer to, and archive them as part of

>   the same (or adjacent) transparency log.
> 
> This offers an improvement over the status quo, because if 
> lore.kernel.org becomes unavailable, someone would have to have
> access to all backend archive repositories it is currently tracking
> in order to be able to reconstitute relevant conversations -- whereas
> with this change, it should be sufficient to just have the copy of
> the  transparency log to have a fully self-contained high-relevancy
> archive of both individual commits and conversations that happened
> around them.

I don't think this is strictly necessary because there's more than lore
that archive's our lists, but the people at the internet history
project would remind me not to look a gift horse in the mouth, so I
think this would certainly be a useful addition.

The thing which Link: doesn't necessarily track is iterations, so if
you replied to v2 and your feedback got incorporated, there's a v3
iteration which has a different msgid.  Is there a way of getting this
full history, not just the current thread?

> I'm just not sure if this will help with the subject of the 
> conversation, or if it does not serve the goal of recognizing
> developer contributions by making them more visible.

I added Jon to the cc since a lot of managers (community or otherwise)
do use the lwn.net stats as a performance guide.  The real question is
could we get something measurable out of the data?  say number of
replies to an accepted patch counting in the reviewer stats or
something?

James