Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Modern administration of creation and removal proposals

3 views
Skip to first unread message

Julien ÉLIE

unread,
Dec 6, 2020, 8:04:52 AM12/6/20
to
Hi all,

The last checkgroups sent for the fr.* hierarchy was in 2014, along with
the creation of a newsgroup.
In January 2020, a vote leaded to the creation of a new newsgroup.
There were 8 pros and 3 cons. No control article has been sent yet
because it seems that the public key has been lost since 2014. But
well, that's another point, and we'll deal with that via the generation
of a new modern PGP key and its progressive update in news servers.

In parallel of that, we're discussing in the fr.* hierarchy potential
new rules to create/remove newsgroups, and I reckon it should be
interesting to share best current practices in other hierarchies.

- Do you still use votes by mails?
Or votes directly in an admin newsgroup, in response to the article
asking to vote? or, even more "modern", a Doodle-like vote?
Note that votes are public in the fr.* hierarchy (and not reserved to a
Board like what is done for the Big-8).

- Do you still have a threshold?
In the 2000s, we asked for 80 YES more than NO, which is obviously
impossible today. We only had 11 votes in January 2020...

- How to prevent "unwanted" changes, if the processus is laxed?
I see that de.alt.* allows the creation of a newsgroup if "the protest
was not too violent" (according to Google Translate) after a usual
period of 7 days. But what is the definition of "too violent"? Does
that rule work in practice?

- Do you have a process like the evaluation of the creation of a
newsgroup after 6 or 12 months, with its removal if it is unused?


We would like to make the creation and removal of fr.* newsgroups more
fluid, but defining the process is tricky...
Any thoughts or advices about it according to your experience in other
hierarchies?

--
Julien ÉLIE

« The hardest thing is to go to sleep at night, when there are so many
urgent things needing to be done. A huge gap exists between what we
know is possible with today's machines and what we have so far been
able to finish. » (Donald Knuth)

Adam H. Kerman

unread,
Dec 6, 2020, 1:33:12 PM12/6/20
to
Julien Julien ELIE <iul...@nom-de-mon-site.com.invalid> wrote:

>The last checkgroups sent for the fr.* hierarchy was in 2014, along with
>the creation of a newsgroup.
>In January 2020, a vote leaded to the creation of a new newsgroup.
>There were 8 pros and 3 cons. No control article has been sent yet
>because it seems that the public key has been lost since 2014. But
>well, that's another point, and we'll deal with that via the generation
>of a new modern PGP key and its progressive update in news servers.

>In parallel of that, we're discussing in the fr.* hierarchy potential
>new rules to create/remove newsgroups, and I reckon it should be
>interesting to share best current practices in other hierarchies.

Julien, there has been no change since the beginning of Usenet. With a
mere 8 supporters, even if they were regularly discussing the topic on
Usenet (they probably weren't), that's not enough to sustain a newsgroup.

The only way to infer that a proposed newsgroup might be worth starting
is to look at how and where the discussion of the topic is currently
taking place, and whether those discussing the topic are highly
motivated to change their posting habits in favor of the proposed
newsgroup.

Furthermore, the proponent himself needs to be someone well known for
discussing the topic.

>- Do you still use votes by mails?
>Or votes directly in an admin newsgroup, in response to the article
>asking to vote? or, even more "modern", a Doodle-like vote?
>Note that votes are public in the fr.* hierarchy (and not reserved to a
>Board like what is done for the Big-8).

>- Do you still have a threshold?
>In the 2000s, we asked for 80 YES more than NO, which is obviously
>impossible today. We only had 11 votes in January 2020...

As there is no interest in voting, eliminate voting. Plenty of language
and regional hierarchies have never had voting.

chi.* added newsgroups on consensus. For decades, there was never enough
discussion to start any new groups, and the consensus was pretty much
to use chi.general for most discussions. chi.politics became little more
than crossposted full-text copyright plagarism from political articles
reposted from the Web, with no local discussion. One by one, the other
groups died as there were fewer and fewer Usenet posters.

>- How to prevent "unwanted" changes, if the processus is laxed?
>I see that de.alt.* allows the creation of a newsgroup if "the protest
>was not too violent" (according to Google Translate) after a usual
>period of 7 days. But what is the definition of "too violent"? Does
>that rule work in practice?

>- Do you have a process like the evaluation of the creation of a
>newsgroup after 6 or 12 months, with its removal if it is unused?

There's just no reason to start the newsgroup in the first place if
failure is highly likely.

rmgroup messages as a hierarchy administration management technique
don't work. Given that Usenet is decentralized, there's no control
over News administraition. There are still plenty of News servers that
won't process rmgroup nor checkgroup messages.

It's kind of a bad idea to take the attitude of Let's try it because we
can always send the rmgroup message later! How is the proponent motivated
to avoid failure?

A better idea is to start with a proponent who is highly motivated to
work to promote the group till it's successful, which can take a good
six months if it's possible at all.

Motivated promotion means the proponent spends months looking for
discussion taking place on that topic in other newsgroups and asks those
posters to try the proposed group. He has to let them know that they
must request that the group be created locally if his News administrator
hasn't done so. This is critical in alt.* of course but even with a
managed hierarchy with signed control messages, the proponent can never
sit on his hands assuming that every server allows such control messages
to be processed without intervention. Your small News server with a
small number of users propably creates new groups upon request only.

With your typical proponent who has almost never posted on the topic and
isn't the least bit motivated to get the group going, failure is likely
in the extreme. That's the kind of proponent who should never be encouraged.

The reasons 10s of thousands of proposed newsgroups typically fail to
find an audience is 1) lack of discussion and 2) proponent who could
care less.

>We would like to make the creation and removal of fr.* newsgroups more
>fluid, but defining the process is tricky...
>Any thoughts or advices about it according to your experience in other
>hierarchies?

New groups do not attract new discussion to Usenet. It's more important to
get discussion going on the topic of interest. Once there is sustainable
discussion, then the decision can be made about breaking it off into a
new group. But make sure there is sufficient discussion in the general
newsgroup for fr.* or another newsgroup in which French is used first.

Remember that we recommended to alt.* proponents not to send a newgroup
message if there isn't an average of 10 articles a day in which the
topic is discussed, looking back over 90 days? Crossposting isn't
discussion. Plagarizing an article from the Web isn't discussion.

Discussion of the topic is the important bit. Having a highly motivated
proponent is the important bit. Just discussing the idea of discussing
the topic in a proposed group is the irrelevant bit.

Either the topic is being discussed on Usenet, or it's not.

Nothing has changed.

Julien ÉLIE

unread,
Dec 7, 2020, 3:19:55 PM12/7/20
to
Hi Adam,

> Motivated promotion means the proponent spends months looking for
> discussion taking place on that topic in other newsgroups and asks those
> posters to try the proposed group.
[...]> The reasons 10s of thousands of proposed newsgroups typically fail to
> find an audience is 1) lack of discussion and 2) proponent who could
> care less.

Unfortunately.


>> We would like to make the creation and removal of fr.* newsgroups more
>> fluid, but defining the process is tricky...
>> Any thoughts or advices about it according to your experience in other
>> hierarchies?
>
> New groups do not attract new discussion to Usenet. It's more important to
> get discussion going on the topic of interest. Once there is sustainable
> discussion, then the decision can be made about breaking it off into a
> new group. But make sure there is sufficient discussion in the general
> newsgroup for fr.* or another newsgroup in which French is used first.
[...]
> Discussion of the topic is the important bit. Having a highly motivated
> proponent is the important bit. Just discussing the idea of discussing
> the topic in a proposed group is the irrelevant bit.
>
> Either the topic is being discussed on Usenet, or it's not.

Many thanks for having shared these wise thoughts.
One often forgets that mere fact, when focusing only on the process.

I've mentioned your response in the current discussions in the fr.*
hierarchy. I hope it will inspire us!

--
Julien ÉLIE

« C'est la première fois que je vois traiter une affaire avec autant de
punch ! » (Astérix)

Matthew Vernon

unread,
Dec 8, 2020, 8:23:29 AM12/8/20
to
Julien ÉLIE <iul...@nom-de-mon-site.com.invalid> writes:

> In parallel of that, we're discussing in the fr.* hierarchy potential
> new rules to create/remove newsgroups, and I reckon it should be
> interesting to share best current practices in other hierarchies.

All your questions about uk.* are addressed on our webpage:
http://www.usenet.org.uk/guidelines.html

But, more briefly:

> - Do you still use votes by mails?
Yes

> - Do you still have a threshold?
No

> - How to prevent "unwanted" changes, if the processus is laxed?
We have a fast-track process for uncontroversial changes, but any 6
objections (or 1 that the Committee feels well-founded) stops the
fast-track, and then a vote is needed.

> - Do you have a process like the evaluation of the creation of a
> newsgroup after 6 or 12 months, with its removal if it is unused?
No

uk.* relies on ukvoting volunteers to run our votes, and I don't know
how medium-term sustainable that is.

Matthew
[definitely speaking without any official hat on]
--
`O'-----0 `O'---. `O'---. `O'---.
\___| | \___|0-/ \___|/ \___|
| | /\ | | \ | |\ | |
The Dangers of modern veterinary life

Julien ÉLIE

unread,
Dec 9, 2020, 2:45:32 PM12/9/20
to
Hi Matthew,

> All your questions about uk.* are addressed on our webpage:
> http://www.usenet.org.uk/guidelines.html

Thanks for the pointer. Very well described; I even see a "Quick
creation" request with 11 supporters.


>> - How to prevent "unwanted" changes, if the processus is laxed?
> We have a fast-track process for uncontroversial changes, but any 6
> objections (or 1 that the Committee feels well-founded) stops the
> fast-track, and then a vote is needed.

Do you happen to know how the number 6 was chosen?
The notion of "well-founded" is also easy to define objectively...

--
Julien ÉLIE

« J'oubliais qu'Assurancetourix a une nouvelle corde à sa harpe ! »
(Astérix)

Matthew Vernon

unread,
Dec 14, 2020, 11:37:05 AM12/14/20
to
Julien ÉLIE <iul...@nom-de-mon-site.com.invalid> writes:

> Hi Matthew,
>
>> All your questions about uk.* are addressed on our webpage:
>> http://www.usenet.org.uk/guidelines.html
>
> Thanks for the pointer. Very well described; I even see a "Quick
> creation" request with 11 supporters.
>
>
>>> - How to prevent "unwanted" changes, if the processus is laxed?
>> We have a fast-track process for uncontroversial changes, but any 6
>> objections (or 1 that the Committee feels well-founded) stops the
>> fast-track, and then a vote is needed.
>
> Do you happen to know how the number 6 was chosen?

I'm afraid not - that predates my involvement in uk.* governance.

> The notion of "well-founded" is also easy to define objectively...

Well, yes, that's why there's an elected Committee who is entrusted with
adjudicating.

Matthew
0 new messages