[Mod-discuss] Re: handling held messages we don't want to send to the list

[John C Klensin <[e-mail]>]

Dan, 

--On Thursday, August 20, 2026 18:06 -0700 dharkins
<[e-mail]> wrote:

>    Sorry for the top post, I'm on my phone and my mail app is a
> POS...  
> 
>  
>    It's not really am issue of "censorship", it's documenting what
> happened. In X years there will be an issue and some old timer will
> say "remember what happened on the SSH list back in '26?" and there
> will be nothing for people to refer to and say, "oh, wow, yea, ok,
> understand."  

>    Everyone knows who this person is. If you don't want to read his
> posts on your sshm folder then don't.  

I have carefully not used the term "censorship" in any of my postings
and, fwiw, I think it misses the point.  I'm also trying very hard to
avoid thinking about this in terms of any one person but rather to
think about, and comment on, the range of situations that might
motivate moderation actions, possibly including blocking a particular
posting.

What I think is important about these cases is clarity about the
reasons why postings are blocked.  In the several decades I've been
watching or involved with the IETF and its predecessors, I don't
believe I've ever seen a posting rejected when (i) it contained
substantive content that was largely or entirely new and (ii) it did
not contain comments or writing style the community found
unacceptable and had been clear about that.  For the first case, if
someone have repeated the same position or argument many times, there
is substantially no information value in retaining the redundant
messages, especially if there have been messages to the list noting
the redundancy.   And, for the second, if the person posting the
problem messages believes they contain points that would be helpful
for the community to hear, nothing prevents those authors from
posting a new note that is free of whatever language, comments, or
restrictions that resulted in the message being treated as
objectionable.

So I see very little reason to maintain public copies of blocked
postings.

best,
   john



>> On Aug 20, 2026 at 7:31 AM, Bron Gondwana
>> <brong=[e-mail]>  wrote:
>>  
>>  
>> I'd like to add my +100 to this as well.    We should just reject
>> messages which aren't fit for the mailing list, and discard spam
>> or repeat offenders.    The people who cry "censorship" aren't
>> going to be mollified by a separate list or a transparency log;
>> and our constituency is people working to create high quality
>> documents to improve how the internet works, and they're best
>> served by being able to get on with sane discussions
>> un-interrupted.
>>  
>> 
>>  
>> Bron.
>>  
>> 
>>  
>> On Thu, Aug 20, 2026, at 09:20, John C Klensin wrote:
>>  
>> >  
>> > FWIW, +1
>> >  
>> >    john
>> >  
>> > 
>> >  
>> > --On Thursday, August 20, 2026 13:58 +0200 Eliot Lear
>> > <[e-mail]>
>> >  
>> > wrote:
>> >  
>> > 
>> >  
>> > >  Hi Simon,
>> >  
>> > >  
>> >  
>> > >  On 20.08.2026 10:35, Simon Josefsson wrote:
>> >  
>> > >>  That is an authoritarian view of policys for a community,
>> > >>  and one
>> >  
>> > >>  that I believe is unhealthy if our goal is to keep IETF
>> > >>  relevant.
>> >  
>> > >  
>> >  
>> > >  My experience is that at nearly any other standards body or
>> >  
>> > >  professional organization, the matter is handled exclusively
>> > >  by the
>> >  
>> > >  executive director or governing board without any fuss.
>> > >  Someone is
>> >  
>> > >  simply disinvited.    In this sense, the IETF does IMHO
>> > >  demonstrate
>> >  
>> > >  *extreme*   transparency with its processes.
>> >  
>> > >  
>> >  
>> > >>  
>> >  
>> > >>  Moderation is delicate subjective human process.    Mistakes
>> > >>  happen.
>> >  
>> > >>  The line between acceptable and unacceptable moves
>> > >>  continously.
>> >  
>> > >  
>> >  
>> > >  I 2/3 agree. Moderation is a delicate process; it does
>> > >  require some
>> >  
>> > >  amount of subjectivity, but the line between what is
>> > >  acceptable and
>> >  
>> > >  unacceptable ought not move, or at least not that much.*
>> >  
>> > >  
>> >  
>> > >  And there is a judgment aspect, and there is an aspect of
>> >  
>> > >  forbearance.    By way of example, I think the chairs of the
>> > >  TLS WG
>> >  
>> > >  have exercised *excessive*   forbearance in the face of
>> > >  argumenta ad
>> >  
>> > >  homimem et ad nauseam.    But they know their peeps better
>> > >  than I
>> >  
>> > >  do, so it's *their* call, at least in the first instance.
>> > >  Not
>> >  
>> > >  every crossing of the line may require a response.
>> >  
>> > >  
>> >  
>> > >>  
>> >  
>> > >>  You build trust for that human process by showing what you
>> > >>  are
>> >  
>> > >>  doing in a transparent way, and develop/improve policies as
>> > >>  you go
>> >  
>> > >>  along, sometimes accepting mistakes happen and then
>> > >>  apologies and
>> >  
>> > >>  learn.
>> >  
>> > >  
>> >  
>> > >  I agree, but you must do all of that in the context of a
>> > >  threat
>> >  
>> > >  model, in which the processes themselves would be gamed.
>> > >  For
>> >  
>> > >  example, transparency in a forum taken to extreme is a
>> > >  filibuster,
>> >  
>> > >  and prevents or discourages people from engaging in the work
>> > >  of a
>> >  
>> > >  group.
>> >  
>> > >  
>> >  
>> > >  Eliot
>> >  
>> > >  
>> >  
>> > >  * The one area I think we currently find ourselves struggling
>> > >  with,
>> >  
>> > >  btw, is the verbosity of communications that are AI
>> >  
>> > >  assisted/generated.    I'm not quite sure what to say about
>> > >  that,
>> >  
>> > >  but we are all getting the feel for when we're being
>> > >  communicated
>> >  
>> > >  to by bots.
>> >  
>> > >  
>> >  
>> > 
>> >  
>> > 
>> >  
>> >  _______________________________________________
>> >  
>> > Mod-discuss mailing list --   [e-mail]
>> >  
>> > To unsubscribe send an email to   [e-mail]
>> >  
>> > 
>> >    
>> 
>>  
>>  
>> --
>>  
>>    Bron Gondwana, CEO, Fastmail Pty Ltd / Fastmail US LLC
>>  
>>    [e-mail]
>>  
>> 
>>    
>> 
>>  
>>  
>>  _______________________________________________
>> Mod-discuss mailing list -- [e-mail]
>> To unsubscribe send an email to [e-mail]
>>    
>