Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > news.groups > #15617 > unrolled thread

List of active (used) groups

Back to article view | Back to news.groups

Contents

  List of active (used) groups Marco Moock <mm@dorfdsl.de> - 2026-06-20 10:51 +0200
    Re: List of active (used) groups Thomas Hochstein <thh@thh.name> - 2026-06-20 11:46 +0200
    Re: List of active (used) groups "Adam H. Kerman" <ahk@chinet.com> - 2026-06-20 15:00 +0000
      Re: List of active (used) groups pschleck@panix.com (Paul W. Schleck) - 2026-06-23 12:31 +0000
        Re: List of active (used) groups Steve Bonine <spb@pobox.com> - 2026-06-23 18:28 -0500
          Re: List of active (used) groups pschleck@panix.com (Paul W. Schleck) - 2026-06-24 03:46 +0000
            Re: List of active (used) groups Steve Bonine <spb@pobox.com> - 2026-06-25 17:56 -0500
    Re: List of active (used) groups gmc@metro.cx (Koen Martens) - 2026-06-20 17:22 +0000
      Re: List of active (used) groups gmc@metro.cx (Koen Martens) - 2026-06-20 17:31 +0000
        Re: List of active (used) groups Marco Moock <mm@dorfdsl.de> - 2026-06-20 21:18 +0200
        Re: List of active (used) groups Marco Moock <mm@dorfdsl.de> - 2026-06-20 21:28 +0200
          Re: List of active (used) groups gmc@metro.cx (Koen Martens) - 2026-06-21 06:59 +0000
    Re: List of active (used) groups Steve Bonine <spb@pobox.com> - 2026-06-22 10:16 -0500
      Re: List of active (used) groups gmc@metro.cx (Koen Martens) - 2026-06-22 17:50 +0000
        Re: List of active (used) groups "Adam H. Kerman" <ahk@chinet.com> - 2026-06-22 18:23 +0000
    Re: List of active (used) groups pschleck@panix.com (Paul W. Schleck) - 2026-06-23 13:02 +0000
      Re: List of active (used) groups Marco Moock <mm@dorfdsl.de> - 2026-06-23 15:12 +0200
        Re: List of active (used) groups Thomas Hochstein <thh@thh.name> - 2026-07-05 12:23 +0200

#15617 — List of active (used) groups

Hello!

Some time ago we created a getting started guide for setting up 
newsgroup access.
https://www.big-8.org/wiki/Getting_Started_with_Usenet

I would like to extend that with a list of active groups, e.g. groups 
that received new posts in the last year etc.

Is there an easy way for server operators to create such statistics?

-- 
Gruß
Marco

Spam bitte an abfalleimer2001@stinkedores.dorfdsl.de

[toc] | [next] | [standalone]

#15618

Marco Moock wrote:

> I would like to extend that with a list of active groups, e.g. groups 
> that received new posts in the last year etc.
> 
> Is there an easy way for server operators to create such statistics?

Probably not out of the box. A bit clunky and work in progress [1]:
<https://code.virtcomm.de/thh/newsstats>

The stats in de.admin.lists and <https://de-usenet.de/files/stats/>
are generated this way.

By default, you'll have to capture "live" data, i.e. the database will
be filled starting with installation and configuration. You can add
single posts by Message-ID with the "addpost" tool (designed to re-add
"dropped" postings e.g. when the database goes down before the
newsserver), so you could grab/grep all storage tokens from your INN
history file and feed them in a loop through sm to addpost. Never
tried, no idea of the performance with really large datasets (you
could take a Message-ID from a post a year ago, get the storage token
with grephistory and process the tokens from history starting with
that token, so you get only a year of posts).

If you don't need to filter out crossposts, don't want to save your
raw data to create more stats without parsing everything again and
don't need more elaborate stats, you could simple do the above without
using NewsStats at all: grep all storage tokens from history (perhaps
starting at a known point in time), loop them through sm and grep for
Newsgroups, split the contents of the header by comma, and count each
component.

-thh

[1] It works, but the code is in dire need of refactoring and removing
code reuse. I never had the time to re-do it right after starting it
in 2010 or so.

[toc] | [prev] | [next] | [standalone]

#15619

Marco Moock <mm@dorfdsl.de> wrote:

>Some time ago we created a getting started guide for setting up 
>newsgroup access.
>https://www.big-8.org/wiki/Getting_Started_with_Usenet

>I would like to extend that with a list of active groups, e.g. groups 
>that received new posts in the last year etc.

>Is there an easy way for server operators to create such statistics?

If I tell you something logical, you will put your fingers in your ears
and ignore it as you've been doing for years.

Such a list might be informative for server utilization but SHOULD NOT
be given as guidance to new users.

Among the most active groups will be *.politics groups. DO NOT, repeat,
DO NOT encourage an inexperienced Usenet user who has no idea what a
flame war that lasts indefinitely looks like. They will scream in terror
and never read Usenet again.

Once again, this is part of your ongoing scheme to purge Usenet of
groups still listed in a server's active and newsgroups files that you
think through useless statistical analysis like this. Instead, new users
should be told the truth about Usenet, that users tend to subscribe to
groups based on topics they are interested in regardless of recent
activity and that, by posting on topic, it is possible to start a
discussion. They should be told to have patience as a followup may not
be posted for weeks. An inactive newsgroup is not useless for
discussion,

You don't know what you don't know.

I have argued that, failing to actually read articles in a group, that a
group you might list as active is literally not useful for discussion
purposes. It's full of articles reposted from the Web. It's full of
crossposts. In truth, an active newsgroup is one in which articles being
posted have not been crossposted and were actually written by the user,
and attract discussion in followup.

You aren't unaware of any of this. You simply choose to ignore how
discussions work on Usenet. You ignore anything and everything that
contradicts the assumptions you already made.

[toc] | [prev] | [next] | [standalone]

#15628

In <11169v6$3t82$3@dont-email.me> "Adam H. Kerman" <ahk@chinet.com> writes:

>Marco Moock <mm@dorfdsl.de> wrote:

>>Some time ago we created a getting started guide for setting up 
>>newsgroup access.
>>https://www.big-8.org/wiki/Getting_Started_with_Usenet

>>I would like to extend that with a list of active groups, e.g. groups 
>>that received new posts in the last year etc.

>>Is there an easy way for server operators to create such statistics?

>If I tell you something logical, you will put your fingers in your ears
>and ignore it as you've been doing for years.


Ah, yes, the "fingers in your ears" rhetoric when accusing the other
side of refusing to "listen."

You know, Steve Bonine uses that one, too.  Perhaps you two have more
in common than you realize?


-- 
Paul W. Schleck
pschleck@panix.com

[toc] | [prev] | [next] | [standalone]

#15631

Paul W. Schleck wrote:

> Ah, yes, the "fingers in your ears" rhetoric when accusing the other
> side of refusing to "listen."
> 
> You know, Steve Bonine uses that one, too.  Perhaps you two have more
> in common than you realize?

Just for the record, I have never "accused others of refusing to listen" 
and I would challenge you to find any Usenet article where I did so.  I 
have stated my opinion for decades, and if people are inclined to take 
my babbling as more than noise, that's their decision.

As for "more in common", Adam and I have disagreed for decades, but we 
have also agreed for decades.

[toc] | [prev] | [next] | [standalone]

#15632

In <111f4rh$2kei5$1@dont-email.me> Steve Bonine <spb@pobox.com> writes:

>Paul W. Schleck wrote:

>> Ah, yes, the "fingers in your ears" rhetoric when accusing the other
>> side of refusing to "listen."
>> 
>> You know, Steve Bonine uses that one, too.  Perhaps you two have more
>> in common than you realize?

>Just for the record, I have never "accused others of refusing to listen" 
>and I would challenge you to find any Usenet article where I did so.


It wasn't in a Usenet article, it was in a complaint e-mail sent to a
newsgroup's moderation team in response to an article rejection:


"I do not agree with your decision to refuse to allow discussion in
the group on this decision.  Obviously you know best and do not want
any input from the people who actually participate in the newsgroup.
Have it your way guys, but you'll have it without me.  The
signal-to-noise was low enough; you've reduced it below my threshold.
Watching you put your fingers in your ears and hum was the last
straw."


The moderators were not refusing to listen, they just did not wish to
allow criticism of moderation decisions in the newsgroup itself.

-- 
Paul W. Schleck
pschleck@panix.com

[toc] | [prev] | [next] | [standalone]

#15633

Paul W. Schleck wrote:
> In <111f4rh$2kei5$1@dont-email.me> Steve Bonine <spb@pobox.com> writes:
> 
>> Paul W. Schleck wrote:
> 
>>> Ah, yes, the "fingers in your ears" rhetoric when accusing the other
>>> side of refusing to "listen."
>>>
>>> You know, Steve Bonine uses that one, too.  Perhaps you two have more
>>> in common than you realize?
> 
>> Just for the record, I have never "accused others of refusing to listen"
>> and I would challenge you to find any Usenet article where I did so.
> 
> 
> It wasn't in a Usenet article, it was in a complaint e-mail sent to a
> newsgroup's moderation team in response to an article rejection:
> 
> 
> "I do not agree with your decision to refuse to allow discussion in
> the group on this decision.  Obviously you know best and do not want
> any input from the people who actually participate in the newsgroup.
> Have it your way guys, but you'll have it without me.  The
> signal-to-noise was low enough; you've reduced it below my threshold.
> Watching you put your fingers in your ears and hum was the last
> straw."
> 
> 
> The moderators were not refusing to listen, they just did not wish to
> allow criticism of moderation decisions in the newsgroup itself.

I challenged you to find a Usenet article.  You found some text, written 
by someone, in some private email, and posted it without attribution.  I 
do not remember writing the text quoted above.  If I wrote it, it is 
accusing someone else of the action you are trying to attribute it to 
me.  If it was written TO a moderation team then it was not written to 
me personally.  At any rate, if the best you can do is dig up 
decades-old text and post it out of context with no attribution, I am 
sincerely sorry for how far you have fallen.

[toc] | [prev] | [next] | [standalone]

#15620

Marco Moock <mm@dorfdsl.de> wrote:
> Some time ago we created a getting started guide for setting up 
> newsgroup access.
> https://www.big-8.org/wiki/Getting_Started_with_Usenet
> 
> I would like to extend that with a list of active groups, e.g. groups 
> that received new posts in the last year etc.
> 
> Is there an easy way for server operators to create such statistics?

I run the below script from cron to create a list of active groups.

https://news.sonologic.net/stats/article_counts.html

I run a variant with the find command modified to:

`find . -type d -links 2 -mtime 7 ! -empty`

to create a list of articles in the last week:

https://news.sonologic.net/stats/article_counts_weekly.html

Quick, dirty, good enough for me.

Cheers,

Koen

```
#!/bin/sh

cd ~/spool/articles

{
    cat << EOF
<!DOCTYPE html>
<html lang="en">
<head>
<meta http-equiv="refresh" content="600">
<title>news.sonologic.net: article counts</title>
</head>
<body>
`date`
<table>
  <tr><td>Group</td><td>Articles</td>
EOF
    for f in `find . -type d -links 2 ! -empty`; do
        echo "$(echo ${f} | tr / . | sed 's/^\.\.//g') $(ls $f | wc -l | sed 's/ //g' )"; 
    done | sort -r -n -k 2 -t\  | awk '{ printf("<tr><td>%s</td><td>%s</td></tr>\n", $1, $2); }'

echo "</body></html>"
} > ~/http/article_counts.html
```

-- 
Software architecture & engineering: https://www.sonologic.se/
Sci-fi: https://www.koenmartens.nl/
Retrocomputing videos: https://retroscandinavian.eu/

[toc] | [prev] | [next] | [standalone]

#15621

Koen Martens <gmc@metro.cx> wrote:
> I run a variant with the find command modified to:
> 
> `find . -type d -links 2 -mtime 7 ! -empty`
> 
> to create a list of articles in the last week:
> 
> https://news.sonologic.net/stats/article_counts_weekly.html

I don't really trust the output of this btw, I should look
into that. Probably the articles are modified after they've
been received and mtime is not a good indicator of post
age.

Cheers,

Koen

-- 
Software architecture & engineering: https://www.sonologic.se/
Sci-fi: https://www.koenmartens.nl/
Retrocomputing videos: https://retroscandinavian.eu/

[toc] | [prev] | [next] | [standalone]

#15622

Am 20.06.26 um 19:31 schrieb Koen Martens:
> I don't really trust the output of this btw, I should look
> into that. Probably the articles are modified after they've
> been received and mtime is not a good indicator of post
> age.

Did you import articles from other spools?
Maybe this updated the filesystem metadata.

-- 
Gruß
Marco

Junk-Mail bitte an trashcan@stinkedores.dorfdsl.de

[toc] | [prev] | [next] | [standalone]

#15623

Am 20.06.26 um 19:31 schrieb Koen Martens:
> I don't really trust the output of this btw, I should look
> into that.

The results look false for me, as various de.* groups got more than 10 
articles this week, but are not listed.

E.g. de.comp.os.unix.linux.misc.

-- 
Gruß
Marco

Junk-Mail bitte an trashcan@stinkedores.dorfdsl.de

[toc] | [prev] | [next] | [standalone]

#15624

Marco Moock <mm@dorfdsl.de> wrote:
> Am 20.06.26 um 19:31 schrieb Koen Martens:
>> I don't really trust the output of this btw, I should look
>> into that.
> 
> The results look false for me, as various de.* groups got more than 10 
> articles this week, but are not listed.
> 
> E.g. de.comp.os.unix.linux.misc.

Yeah, something's not quite right. I'll have to have
a look at that sometime, maybe. It's just an indication
for me, so I'm not inclined to put much time in it.

Cheers,

Koen

-- 
Software architecture & engineering: https://www.sonologic.se/
Sci-fi: https://www.koenmartens.nl/
Retrocomputing videos: https://retroscandinavian.eu/

[toc] | [prev] | [next] | [standalone]

#15625

Marco Moock wrote:
> Hello!
> 
> Some time ago we created a getting started guide for setting up 
> newsgroup access.
> https://www.big-8.org/wiki/Getting_Started_with_Usenet
> 
> I would like to extend that with a list of active groups, e.g. groups 
> that received new posts in the last year etc.
> 
> Is there an easy way for server operators to create such statistics?

There is a basic flaw here. "New posts" does not equate to "useful".

[toc] | [prev] | [next] | [standalone]

#15626

Steve Bonine <spb@pobox.com> wrote:
> Marco Moock wrote:
>> Some time ago we created a getting started guide for setting up 
>> newsgroup access.
>> https://www.big-8.org/wiki/Getting_Started_with_Usenet
>> 
>> I would like to extend that with a list of active groups, e.g. groups 
>> that received new posts in the last year etc.
>> 
>> Is there an easy way for server operators to create such statistics?
> 
> There is a basic flaw here. "New posts" does not equate to "useful".

But it at least tells you which if the tens of thousands of groups you
could check out to find active discussions. Sure, you'll get false
positives, but it narrows things significantly.

Cheers,

Koen

-- 
Software architecture & engineering: https://www.sonologic.se/
Sci-fi: https://www.koenmartens.nl/
Retrocomputing videos: https://retroscandinavian.eu/

[toc] | [prev] | [next] | [standalone]

#15627

Koen Martens <gmc@metro.cx> wrote:
>Steve Bonine <spb@pobox.com> wrote:
>>Marco Moock wrote:

>>>Some time ago we created a getting started guide for setting up 
>>>newsgroup access.
>>>https://www.big-8.org/wiki/Getting_Started_with_Usenet

>>>I would like to extend that with a list of active groups, e.g. groups 
>>>that received new posts in the last year etc.

>>>Is there an easy way for server operators to create such statistics?

>>There is a basic flaw here. "New posts" does not equate to "useful".

>But it at least tells you which if the tens of thousands of groups you
>could check out to find active discussions. Sure, you'll get false
>positives, but it narrows things significantly.

Again: This is would-be advice to new users. These statistics may be of
interest to experienced users and News administrators, it will give new
users the wrong impression.

Again: This will lead a new user to seemingly active *.politics.*
groups, which consist not of worthwhile discussion, but of articles
reposted from the Web (not an example of Usenet discussion), trolling,
and flame wars that have lasted for years. No new user unprepared for a
flame war should be directed to such a group.

This is part of Marco's ongoing purge of Usenet newsgroups by actively
discouraging anyone from even thinking about posting to an inactive
newsgroup. Just the other day, Marco actively discouraged a new user fro
subscribing to and reading *.fan.* groups. The new user saw some group
names that appealed to him.

A new user needs to be told that it can be up to him to start
discussion. He may post on topic in a group that's lacked discussion for
a while, but that act could get discussion going. Inactive groups still
have subscribers.

Attempting to start an on topic discussion is the right thing to do.
Marco's act of discouraging reading and posting is the wrong thing to
do.

[toc] | [prev] | [next] | [standalone]

#15629

In <1115k9u$fnd1$1@solani.org> Marco Moock <mm@dorfdsl.de> writes:

>Hello!

>Some time ago we created a getting started guide for setting up 
>newsgroup access.
>https://www.big-8.org/wiki/Getting_Started_with_Usenet

>I would like to extend that with a list of active groups, e.g. groups 
>that received new posts in the last year etc.

>Is there an easy way for server operators to create such statistics?

>-- 
>Gruß
>Marco

>Spam bitte an abfalleimer2001@stinkedores.dorfdsl.de


The late Alexander Bartolich had a script to accomplish this, used to
help identify inactive newsgroups for removal.  It likely worked as a
client, accessing via NNTP a remote news server that had a long
article history.  To help reduce false positives, it searched not only
for article activity in a given newsgroup, it looked for articles that
had a "References:" header in them, indicating likely followup
discussion, versus just one-way SPAM or flooding.

Roman Racine may still have this script, as he had access to
Alexander's account and files after he passed.  Regardless, this
feature could probably be easily added to any other similar newsgroup
activity report generating script.


--
Paul W. Schleck
pschleck@panix.com

[toc] | [prev] | [next] | [standalone]

#15630

Am 23.06.26 um 15:02 schrieb Paul W. Schleck:
> Roman Racine may still have this script, as he had access to
> Alexander's account and files after he passed.  Regardless, this
> feature could probably be easily added to any other similar newsgroup
> activity report generating script.

Do you have a contact address?


-- 
Gruß
Marco

Junk-Mail bitte an trashcan@stinkedores.dorfdsl.de

[toc] | [prev] | [next] | [standalone]

#15639

Marco Moock wrote:

> Am 23.06.26 um 15:02 schrieb Paul W. Schleck:
>> Roman Racine may still have this script, as he had access to
>> Alexander's account and files after he passed.  Regardless, this
>> feature could probably be easily added to any other similar newsgroup
>> activity report generating script.
> 
> Do you have a contact address?

My most recent contacts to Roman have been over Facebook, as he didn't
reply to my mails.

However, I don't think reaching out to him would be very promising.
albasani.net suffered harddisk damage, and most of the data was lost
in the process. I had contacted Roman years ago to ask if he could
restore Alexander's Subversion repositories; nothing ever came of it,
which is why I uploaded the latest version of those repositories
(though without the revision history) to GitHub two years ago:
<https://github.com/Albasani>

The file Paul referred to may be this one:
<https://github.com/Albasani/Schnuerpel/blob/master/bin/count-posts-per-month.pl>

Regards,
-thh

[toc] | [prev] | [standalone]

Back to top | Article view | news.groups

csiph-web