Post by WandererThe Big-8 newsgroups are being swamped with spam. Please, do something about the spam.
Now comes the part where someone makes excuses and tells me to get a
newsreader that can filter out the spam, so I can that no one posts
here anymore because usenet is a spam ridden hole.
Do something.
I'm not sure if you're addressing the Big-8 Board here? If so, I think
you're greatly overestimating its powers, and assuming a kind of central
administration that does not exist on Usenet.
The Board acts as a trusted source for group-creation (and occasionally
deletion) notices within the "Big 8" hierarchies[*]. It also does some
other minor administrative work such as investigating when a group
moderator has gone missing for a long time, and moderating a small
handful of admin-related groups such as this one. It has power only in
as much as news server administrators think it's worth listening to
about these fairly specific things. It is the news server administrators
themselves who "run" the Big-8, as its newsgroups only exist because a
bunch of news servers on a big distributed network agree that they do.
By the very design of Usenet, there is no central point of control.
[*] Some Usenetters think that even this is too much power concentrated
in one place. The alt.* and free.* hierarchies demonstrate other ways to
handle such matters.
Post by Wanderer1. Makes yourselves the default moderators and implement some simple
generic spam filters. You don't have to get involved in policing off
topic posts and flame wars.
Fortunately or unfortunately (depending on your perspective), this would
not be workable at all. Most groups are unmoderated; that is, there is
no way to moderate the group even if someone wants to. Neither the Board
nor anybody else can simply decide to moderate all the existing groups
in the Big 8.
The only thing way to make something like this happen would be to issue
control messages to change the moderation status of all 2000-ish
unmoderated groups. This would be an unprecedented and ludicrous power
grab that would rightly be ignored by many server administrators: at
best it would have no effect because the attempt would be universally
ignored, and at worst it would fragment all discussion as some servers
would send messages to the moderators and some wouldn't, and some
servers would reject unmoderated messages and some would accept them.
Post by Wanderer2. Find real moderators for groups. Some are easy.
comp.os.msdos.djgpp and comp.lang.python are mirrors of mailing
lists. Make the mailing list admins the moderators.
Converting an individual unmoderated group to moderated (again by
sending a control message to do so) has occasionally been proposed, and
even implemented, in the past. Objections have been raised on the basis
that some servers might fail to implement the change, leading to the
aforementioned problems with the group existing in a different state on
different servers. However, in practice, a bigger problem seems to be
that many moderated groups eventually fail because the moderators
disappear, leaving the group unusable.
This exact sequence of events happened to comp.ai: it was originally
created as unmoderated, later converted to moderated after holding a
vote to do so, and then became unusable for over eight years when the
moderator lost access to their account. In this case, a new moderator
volunteered in 2020 and the group is now active again, but often once a
moderated group is abandoned it stays that way.
Post by Wanderer3. Complain. Contact news servers that inject too much spam.
Complain to Google about Google Groups. They might listen to an
official complaint from the Big-8.
Yes, the more people who complain to Google the better (assuming
anyone's found a reliable way to do so, as they go to great lengths to
avoid any interaction with the general public).
I'm not sure a complaint from the Board would carry more weight than
anyone else's, though. We don't have any contacts there, and they don't
even act on our group-creation messages. If anyone does know how to
contact a human with any influence over Google Groups, please let us
know.
It does, however, look like someone at Google has finally noticed the
spam problem in the last few days. I've seen reports that Google Groups
has entirely disabled posting access for some groups or is requiring
CAPTCHAs to post. This is at least faintly encouraging, and I hope it
will be the start of an ongoing active effort to drive spammers off
Google Groups.
Post by Wanderer4. Issue a statement to news server providers asking them to block
news servers that inject too much spam.
A lot of Usenet providers are already aware of the problem, and there
has been considerable discussion in news.software.nntp and
news.admin.net-abuse.usenet about how to combat the floods of spam from
Google Groups. This has included calls to de-peer Google, and for news
providers and end-users to block any messages originating there. De-
peering and blocking of articles from a rogue provider has been done
successfully in the past, but it requires widespread co-operation among
news providers in order to be effective, particularly if the problematic
provider is a very large and well-connected one.
Post by Wanderer5. Supply spam filters to new servers and issue updates the way
anti-malware software does.
This is a very practical approach. Efforts in this direction already
exist: there are cancelbots and the NoCeM protocol, which exist to allow
co-operative, automated removal of spam at the server level by providing
feeds that identify which articles are spam. I don't know if anyone is
currently sharing lists of spam-filtering rules, but I agree that it
sounds like a good idea.
Post by WandererPlease do something.
Thank you
Broadly speaking, things happen on Usenet by working out a general
consensus among the server administrators who run it. This anarchic
decentralisation is both Usenet's strength and its weakness. It means
Usenet is highly resistant to some classes of problem such as server
failures and censorship, but on the other hand, sufficiently determined
bad actors can cause a lot of harm before they can be dealt with.
There is a parallel here with email, another decentralised messaging
system. Well over half of all email is still spam, despite the fact that
people have been trying to solve the problem for decades with far more
manpower and resources than are available for Usenet.
Despite all this, there are news providers that do a very good job of
filtering out the vast majority of spam. For example, comp.lang.python
on Eternal September (my provider) has only a handful of spam posts,
despite being 99% garbage on Google Groups.
HTH,
Rayner