[postgis-devel] DMARC/DKIM mitigation on maling lists

Greg Troxel gdt at lexort.com
Wed Oct 25 09:02:54 PDT 2023


Regina Obe via postgis-devel <postgis-devel at lists.osgeo.org> writes:

>> The change would be:
>> 
>>   - Stop adding a [postgis-tickets] suffix in the subject line
>>   - Stop adding a footer to every email
>> 
> I like having postgis-tickets and postgis-development as part of the subject
> line.

Do you really not sort messages to separate folders, so that within a
given folder you know?  Or is this about putting them in the same folder
and then telling them apart?

>> The rationale is that the same information is already included in the
> messages
>> in form of RFC822 headers:
>> 
>>   - List-Id
>>   - List-Unsubscribe
>>   - List-Archive
>>   - List-Post
>>   - List-Help
>>   - List-Subscribe
>> 
> These are hidden metadata no, that I'd then need to setup rules to handle.
> I like my subject to have the prefix, cause I look for that.  I'm a human
> you know. 

I'm a human too, and I expect that when I push the 'r' button, to invoke
the Reply MUA action, it will compose a message to the sender, not the
list.  The From: rewriting breaks this expectation, and I find this
violation far more serious than anything about seeing subject lines.

The headers aren't actually hidden.  It's up to your MUA to show you
what you want to see.  But I sort into separate IMAP folders -- I
honestly don't see how anybody can cope with mailinglists without that
-- and thus in my "lists.postgis" folder I know that everything is from
a postgis list.


Suppose we were having the conversation the other way around, where the
list had no subject munging, no footer, and no false From: and Reply-To:
headers inserted.    What if the question were

  Should we configure the list to modify Subject: by adding
  "[postgis-tickets] ", and then by necessity also

  Replace the correct From: field with one that claims (wrongly) that
  the sender of the message is the list?

  Add a Reply-To: field, that the author did not set, asking that
  replies be sent to the author.

Put that way, it seems much less reasonable.


More information about the postgis-devel mailing list