Usenet Proposal and Informal RFC : mod.* Unofficial Hierarchy Brainstorm

[Jingo Kyrma]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Usenet Proposal and Informal RFC : mod.* Unofficial Hierarchy Brainstorm

Noticing the prevalence of Usenet spam, disruption and forum sliding has
impelled consideration of a method of encouraging text moderated
newsgroups not requiring official sanction but having semi-official or
established 'customary' moderation features.

Herein is proposed 'mod.*' a new Usenet top-level hierarchy resurrected
on the ancient 'mod.*' hierarchy of yore. For example some groups under
this hierarchy would be 'mod.newsgroups', 'mod.programming.language.go',
'mod.cryptography', 'mod.bitcoin', or 'mod.protocol.tcp'.

Originally Usenet employed three hierarchies: 'net.*', 'fa.*', and
'mod.*'. The 'mod.*' hierarchy designated moderated groups. Eventually
Usenet was re-configured to allow moderation in any hierarchy and the
'mod.*' hierarchy was deprecated. However what is now missing is a
hierarchy in which all groups are moderated, or in which moderation is
mandatory by default. Since 'mod.*' is no longer used anywhere, Usenet
could resurrect the 'mod.*' hierarchy and put it to its intended
use--limiting noise and abuse and keeping the signal to noise ratio high
in signal.

New 'mod.*' groups would be created similarly to 'alt.*' groups by
sending of a control message, but certain elements would be mandatory: a
valid moderator with cryptographic keys and valid email, the same for
any chosen sub-moderators, and mandatory moderation approval and
signature for any message to appear in the feed.

Like the alt.* hierarchy a mod.* hierarchy would be 'unofficial' but
with different customary rules, chiefly mandatory moderation, rather
than optional moderation. Server admins could carry exclusively mod.*
feeds knowing that moderation would be enforced, hence content-related
administrative problems, less frequently encountered. Under all
circumstances mod.* hierarchies would be mandatorily moderated, and in
case of a absent moderator, held in backlog until someone volunteers to
moderate. A threshold time lapse scheme would be required to sort out
abandoned moderation issues. In short a 'use it or lose it' policy
should apply to the chief moderator of each group. If a chief-moderator
were to disappear, any remaining sub-moderators could vote on who should
become the new chief. If no sub-moderators remained, then server admins
could discuss appointing some volunteer and sending out a new control
message that could be adopted or ignored by other server admins.

Every mod.* group added would be required to be moderated without
exception. If a moderator abandons a group or gets hundreds of messages
backlogged after a certain lengthy time period a compliant server should
mark it as abandoned and backlogged, then available to the first taker
who desires to moderate it. This would be part of the customary 'ethos'
of the hierarchy.

Every mod.* group would be required to have one chief moderator and up
to a dozen cryptographically approved and registered sub-moderators.
Moderators should not be under any pressure to moderate 24/7. Rather
they should have the flexibility to backlog for X number of days if they
wish, and moderate all messages on a particular schedule, if they so
desire. The moderation schedule could be part of each group description.

The chief moderator would have authorization to over-ride any
sub-moderators, issue manual and automated NoCeM / cancels for messages
already approved by sub-moderators, and to remove offending messages
that slip past lazy or rogue sub-moderators. We know it happens that
someone volunteers to help with moderating a forum then drops the ball.
The chief certificate model would ameliorate this, and the impetus of
server admins to collaborate would ameliorate a derelict chief moderator
failure.

The chief proponent of the group would be the chief moderator with the
cryptographic key authorization to approve and remove sub-moderators.
Every sub-moderator can act unilaterally, and it would be incumbent on
the chief moderator to NoCeM or cancel sub-moderator actions.

The chief feature here is that mod.* should remain 'unofficial' with no
official governance over the hierarchy. The Big 8 board and its aims are
currently admirable. However there is no guarantee that in a few years
they won't be taken over by censorship shills or FedCorp apparatchiks
with a penchant for subverting speech freedom, a thing we have witnessed
across the Internet infrastructure. Therefore having no official
governance would be some degree of protection against ideological purges
by censorship czars wheedling into the woodwork.

The second necessary feature is that all mod.* groups would require
moderation, and by default compliant servers would reject groups in
feeds without a proper moderation configuration and signing keys.

The third feature is that it would be necessary and desirable for server
administrators to refuse to carry any mod.* feeds with any appreciable
amount of spam or obvious mindless disruption which has driven away many
people from once useful groups. Each NNTP server administrator would
retain their autonomy on which mod.* groups to approve, synchronize or ban.

The fourth feature would be that the groups should be text groups. Yet
the ability to attach small files such as source code documents, PDF
documents, SVG diagrams, and small compressed archives of source code or
text files should be preserved. Real work is more likely to be shared in
a moderated environment of common interest, without the disruption of
cranks, trolls, pirates, and spammers. And real digital work usually
involves the sharing of work documents in printable formats such as PDF
and postscript, or zip archives of multiple documents. Therefore some
size limit for small files should be maintained that allows sharing
documents but curtails NZB bandwidth hogging and IP piracy nonsense.
This, coupled with the right and inclination of server administrators to
block offenders, should prevent binary abuse of mod.* groups, should a
group moderator be crass enough to enable such abuse.

If a certain group were to be controlled by a censorial, fickle, or
obstreperous moderator the simple solution would be for server admins to
allow other moderators to found a similar group with different
moderation and content policies, rather than trying to force all
discussion related to a subject into one group under one rulership. In
fact this is one of the primary problems of Usenet contributing to its
obsolescence: denying group proposals because some 'relevant' group
already exists. With moderation enforced the only way for a proponent to
get their proposed group is to agree to moderate it, automatically
solving some problems with vanity groups cluttering the namespace.

One of the reasons that people complain about platforms like Reddit is
because the platform is moderated. Yet one of the reasons that millions
of people use platforms like Reddit is because the platform is
moderated. Each subreddit has its own moderation and moderator policies.
Although moderation is not ideal in a vision of a perfect, utopian
worldview, moderation is actually optimal and necessary in the real
world. If Reddit had no moderator controls it would be swamped with
textual graffiti from every crank and malcontent with Internet access,
and totally unusable for real work or discussion. This is in large part
what has contributed to the decline of Usenet--lack of moderation,
over-moderation, and difficulty of creating new moderated groups. The
technical bar to entry is a deterrent to many well-meaning users who
would otherwise employ the protocol.

With these ground rules, users would know that subscribing to a mod.*
hierarchy would provide a low or no noise experience, without any
centralized authority threatening their future use of the forum.

A general group, 'mod.groups.admin' could be first established with a
dozen or more server administrators all established as moderators of
that group, for discussion of administration and strategies for
streamlining. Also this group, or 'mod.groups.moderators' could be the
forum for users who seek volunteer sub-moderators for a new or existing
group. A general group, 'mod.groups.proposals' could be established for
people who want to propose themselves as moderator of a new group. The
general administration group could discuss violations of anti-noise and
anti-piracy efforts by derelict or malicious moderators, and propose
solutions to keep them in line with the ethos of the hierarchy. Perhaps
a group like 'mod.groups.complaints' would serve as a forum for ethos
enforcement discussions.

Along with this proposal admins may wish to consider something like the
'admin.*' top-level hierarchy that works on the same principles as the
proposed 'mod.*' hierarchy, except that only established NNTP server
administrators would be able to propose a 'admin.*' group or moderate
one. The defunct 'net.*' hierarchy could also be resurrected under this
aegis for established administrators with skin in the game to have their
own groups therein. Such separation of concerns would be useful for
people actually getting work done and needing a low-noise way to
collaborate publicly for the record.

I have long since unsubscribed from some alt.* and Big-8 groups that I
once enjoyed reading because the noise-to-signal ratio has made reading
highly unpalatable while filtering spam and crank consumes more time and
effort than I wish to waste. I have no doubt many others have left such
groups for the same reason. This proposal is an attempt to remedy the
unfortunate aggravation of classic Usenet and make it palatable again.
In its current configuration I fear that Usenet cannot regain traction
as a forum for collaboration and tweaks must be employed to remedy this
condition.

I believe this brainstorm proposal is meritorious and should be
considered and discussed as a way to move forward with the shared desire
to slowly re-invigorate Usenet for purposes of legitimate discussion and
more importantly, public collaboration, free of nation-state and troll
disruption campaigns that currently afflict the protocol.

What tools would need to be configured to make a scheme like this
possible? Please post comments, criticisms and suggestions about the
points raised herein and how to make them happen.

Jingo Kyrma (sugarbug 3883)

-----BEGIN PGP SIGNATURE-----

iHUEARYIAB0WIQRhXXeaU0uM3SgnVyVv18MALwf8AgUCY3LZ2gAKCRBv18MALwf8
AmmRAP4mZAjNekg+zly3n0cGjlhniBZkc7zYJhD5OYjvuu9TSQEAiQLqYAAvRvEO
UEHincA+2sI9nktKrUQ01gO4USs/sgU=
=oUQI
-----END PGP SIGNATURE-----

[b...@ripco.com]

Jingo Kyrma <38...@sugar.bug> wrote:

> What tools would need to be configured to make a scheme like this
> possible? Please post comments, criticisms and suggestions about the
> points raised herein and how to make them happen.

Yawn.

If you really want to get rid of 99.9% of the usenet spam, just get news
admins to start dropping posts from *.groups.google.com.

Thats it.

-bruce
b...@ripco.com

[Anton Shepelev]

Jingo Kyrma:

> If you really want to get rid of 99.9% of the usenet spam,
> just get news admins to start dropping posts from
> *.groups.google.com.

That will lose a considerable portion of audience, as well.

--
() ascii ribbon campaign -- against html e-mail
/\ www.asciiribbon.org -- against proprietary attachments