XPost: news.admin.misc
Julien Julien ELIE <
iulius@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.
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)