Stefan Sperling wrote on Thu, 28 May 2020 08:46 +0200:
> On Thu, May 28, 2020 at 06:27:54AM +0000, Daniel Shahaf wrote:
> > Stefan Sperling wrote on Thu, 28 May 2020 08:07 +0200:
> > > +++ releasing.part.html (working copy)
> > > @@ -1300,6 +1300,20 @@ Ensure that your mailer doesn't wrap the URLs over
> > > <p>NOTE: We announce the release before updating the website since the website
> > > update links to the release announcement sent to the announce@ mailing list.</p>
> > >
> > > +<p>It is possible that your message to the ASF-wide announce_at_apache.org will
> > > +be rejected. This generates a moderation notification with a Subject line
> > > +such as: <tt>Returned post for announce_at_apache.org</tt>.
> >
> > Spell out that this is specifically about announce_at_a.o, as opposed to announce_at_s.a.o?
> >
> > > +The originator of such moderation messages is anonymous if the moderator
> > > +neglects to sign the moderation message with their own name, which makes it
> > > +impossible to have an actual conversation with the moderator.
> >
> > Mention that a conversation can be had by emailing announce-owner@?
>
> How about this?
>
Seems to cover everything. I took a stab at a slight reorganization;
WDYT? —
Index: releasing.part.html
===================================================================
--- releasing.part.html (revision 1878220)
+++ releasing.part.html (working copy)
@@ -1300,6 +1300,18 @@
<p>NOTE: We announce the release before updating the website since the website
update links to the release announcement sent to the announce@ mailing list.</p>
+<p>There are two announce@ mailing lists where the release announcement gets
+posted: The Subversion project's announce_at_subversion.apache.org list, and the ASF-wide announce_at_apache.org
+list. It is possible that your message to the ASF-wide announce@ list will be
+rejected. This generates a moderation notification with a Subject line such as:
+<tt>Returned post for announce_at_apache.org</tt>. The moderator who ordered the
+mailing list software to reject the message may neglect to sign their name to
+the rejection message, making the rejection anonymous, and the grounds for the
+rejection may be invalid. Be that as it may, keep calm and forward the
+rejection to the dev@ mailing list so the project can discuss whether anything
+needs to be done about it. (If necessary, announce@ mailing list moderators can
+be contacted via the announce-owner@ handle.)</p>
+
<p>Update the topics in various Subversion-related IRC channels, such as
<tt>#svn</tt> and <tt>#svn-dev</tt> on freenode.</p>
> Index: releasing.part.html
> ===================================================================
> --- releasing.part.html (revision 1878172)
> +++ releasing.part.html (working copy)
> @@ -1300,6 +1300,23 @@ Ensure that your mailer doesn't wrap the URLs over
> <p>NOTE: We announce the release before updating the website since the website
> update links to the release announcement sent to the announce@ mailing list.</p>
>
> +<p>There are two announce@ mailing lists where the release announcement gets
> +posted: The Subversion project's announce@ list, and the ASF-wide announce@
> +list. It is possible that your message to the ASF-wide announce@ list will
> +be rejected. This generates a moderation notification with a Subject line
> +such as: <tt>Returned post for announce_at_apache.org</tt>.
> +The originator of such moderation messages is anonymous if the moderator
> +neglects to sign the moderation message with their own name, which makes it
> +impossible to have an actual one-on-one conversation with the moderator.
> +If you receive such a moderation message please stay calm and forward it to
> +Subversion's dev@ mailing list so the project can discuss whether anything
> +needs to be done about it. (If necessary, announce@ mailing list moderators
> +can be contacted via the announce-owner@ handle.)
> +In the past, this moderation mechanism was abused to alert the Subversion
> +project about ASF-wide policy changes in the way the KEYS files are supposed
I think it's not actually a recent change in policy, but simply
a case where we were never in compliance with the letter of the
self-styled "policy" in the first place. For example, here's one
document from 2011 that describes itself as a "policy" (right at the
top, under "Abstract") and describes the KEYS-file-per-PMC practice:
https://web.archive.org/web/20110716160744/http://www.apache.org/dev/release-signing.html#keys-policy
And here's our practice back then:
https://mail-archives.apache.org/mod_mbox/subversion-dev/201109.mbox/%3CCAJjMeYMcu94n7%3DGSEQ_LLVoPa0acZKPHy-R8sCyZHLpT0AtThw%40mail.gmail.com%3E
> +to be handled. This is inappropriate because moderation is supposed to filter
> +out spam. Policy changes should be discussed on the dev@ list instead.</p>
> +
It's unfortunate that this patch is needed. The Foundation should not
be something we have to work around. I wish the announce@ moderation
policy would change, but until and unless that happens, documenting the
issue is the Right Thing.
Cheers,
Daniel
Received on 2020-05-28 09:59:40 CEST