• all usenet stats are fake

    From PN@noico@noico.net to alt.free.newsservers on Mon Mar 30 06:56:23 2026
    From Newsgroup: alt.free.newsservers

    There is nothing more frustrating to me than the fact we have hundreds
    of usenet servers and probably none of the admins bother to create
    even the most basic statistics about groups. and by "basic statistics"
    I don't mean "MESSAGES ON GROUP" because it tells you literally nothing.
    Is it one guy who is talking to himself for 20 years? Is it just
    spam? Or maybe a few people who just post links on social media?

    So I've created a script which helps me at least somewhat figure out
    which groups are actually alive.

    server: paganini.bofh.team, analyzed 30 days (around feb/mar 2026).
    excluded: *test groups, free.usenet, free.willey, free.pt
    (these groups look like some kind of indexer / thank-you groups,
    related to binary groups somehow)

    Results are at the bottom of this post. but first, some conclusions:

    - we are fucked. i mean look at the unique posters - i assume some
    of them probably post from multiple accounts. barely 20 groups
    have more than 50 active people per month. and i remind you this
    is a service with thousands of servers worldwide. available
    WORLDWIDE. or to put it differently: we are the elite lol

    - national groups are strong

    - political groups, except one national one, didn't make it into
    this list - also a surprise. now i understand why political trolls
    and nolifers keep spamming other groups all the time. because in
    their own groups nobody wants to talk to them.


    HOW I MEASURED IT
    ===================================================
    reply rate = replies / total_msgs
    participants = log(1 + unique_authors) / log(16),
    saturates at 15 authors
    diversity = min(msgs/10, 1) * (1 - Gini),
    0 if single author
    stickiness = authors_last_7d / authors_last_30d,
    0 if single author
    continuity = days_with_posts / 30
    thread depth = min(avg_msgs_per_thread / 5, 1)
    rhythm = max(0, 1 - median_gap_hours / 168)
    volume = 1 - exp(-total_msgs / 50)

    vitality score = reply_rate*0.22 + participants*0.18
    + diversity*0.18 + stickiness*0.13
    + continuity*0.09 + thread_depth*0.09
    + rhythm*0.04 + volume*0.07


    RESULTS
    =================================================================
    groups with at least 10 unique posters, avg thread depth 5+,
    active on at least 10 days out of 30, stickiness >= 0.3 (at
    least 30% of last month's posters showed up in the last 7 days),
    vitality >= 0.3. sorted by unique poster count descending.

    comp.os.linux.advocacy 133 9.78 31 0.7725
    it.hobby.fai-da-te 121 9.12 31 0.7923
    uk.d-i-y 113 5.94 32 0.7985 sci.physics.relativity 112 9.37 31 0.768
    rec.food.cooking 110 7.5 32 0.7597
    sci.math 106 7.74 31 0.7885
    comp.sys.mac.advocacy 102 5.57 31 0.7674
    alt.usage.english 87 8.82 31 0.8134
    comp.os.linux.misc 85 19.49 31 0.8023
    de.sci.electronics 75 5.5 31 0.826
    rec.arts.sf.written 69 6.03 31 0.7915
    ger.ct 68 17.2 31 0.834
    de.etc.fahrzeug.auto 66 19.92 30 0.8053 sci.electronics.design 61 9.58 31 0.7971 alt.comp.software.thunderbird 58 7.34 27 0.8028
    it.sport.calcio.milan 57 7.94 31 0.7822
    de.rec.fahrrad 56 8.63 29 0.843
    de.etc.sprache.deutsch 51 14.17 31 0.8263 alt.folklore.computers 48 10.34 27 0.7886
    uk.legal.moderated 47 12.88 31 0.8333
    de.admin.news.groups 45 7.46 30 0.8538
    uk.railway 44 13.73 31 0.8369 eternal-september.talk 44 8.7 30 0.8029
    it.discussioni.auto 43 6.76 30 0.7982
    nl.politiek 41 5.68 32 0.825 de.alt.folklore.computer 39 27.46 30 0.8479
    de.rec.heimwerken 38 12.23 25 0.842
    de.rec.mampf 35 16.62 25 0.8169
    it.hobby.cucina 34 12.54 30 0.8235
    uk.rec.motorcycles 32 10.86 30 0.84
    uk.comp.sys.mac 32 5.16 28 0.8349
    alt.free.newsservers 32 5.17 22 0.8233
    uk.telecom 32 15.64 25 0.8118
    it.comp.hardware 32 7.43 27 0.7935
    pl.soc.prawo 30 23.33 31 0.8254
    pl.misc.elektronika 29 12.75 28 0.8213
    comp.lang.tcl 29 6.23 25 0.8039
    comp.arch 28 8.12 31 0.8659
    alt.russian.z1 28 7.17 31 0.8059 eternal-september.support 27 7.2 13 0.7735 it.comp.os.win.windows10 26 5.94 22 0.7999 fr.lettres.langue.francaise 26 9.07 27 0.7982
    alt.windows7.general 26 7.17 24 0.7972
    de.alt.comm.android 26 13.83 24 0.7957
    de.alt.talk.unmut 25 21.88 28 0.863
    rec.arts.comics.strips 25 9.35 19 0.796
    alt.flame 25 6.08 29 0.7952
    fr.bio.medecine 24 16.86 31 0.8305
    alt.os.linux.mint 24 14.19 22 0.7903
    de.etc.haushalt 24 8.67 13 0.7875
    alt.usenet.kooks 23 6.6 30 0.8222
    fr.rec.photo 22 11.26 31 0.8103
    comp.theory 22 16.3 31 0.7927 fr.comp.usenet.lecteurs-de-news 22 13.18 19 0.7773
    comp.unix.shell 22 61 10 0.7607
    pl.misc.samochody 21 15.84 28 0.8061
    spline.eisfair 21 9.33 23 0.8058 alt.comp.software.seamonkey 20 5.55 17 0.7763 de.comm.software.mozilla.mailnews 20 5.27 14 0.7683
    de.sci.mathematik 19 21.7 29 0.8215
    de.rec.film.misc 19 10.25 23 0.8114
    fr.soc.environnement 19 11.65 30 0.8057
    uk.tech.digital-tv 19 6.44 11 0.7656
    pl.misc.telefonia.gsm 17 8.06 24 0.8035 de.soc.recht.strassenverkehr 17 13.33 13 0.7864
    comp.lang.forth 16 8.61 28 0.8508
    alt.recovery.aa 16 8.26 30 0.8475 alt.buddha.short.fat.guy 16 7.21 31 0.8323
    alt.sixtyplus 16 7.73 10 0.796 it.comp.sicurezza.varie 16 5.6 14 0.7518
    aus.cars 15 20.8 29 0.8761
    sci.logic 15 20.47 24 0.8306
    alt.os.linux.ubuntu 15 6.38 11 0.737
    it.comp.reti.locali 14 17.67 12 0.7746
    sci.skeptic 13 7.95 29 0.8207 sfnet.keskustelu.varaventtiili 12 8.02 31 0.8648
    fr.sci.maths 12 23.86 25 0.8179 de.comm.software.mozilla.browser 12 7.5 10 0.7263
    lada.talk 11 10.53 30 0.8334
    de.rec.tv.serien 11 6.2 21 0.7875 rec.crafts.metalworking 11 8 27 0.7588
    fr.comp.sys.atari 10 14.38 24 0.8344
    comp.ai.philosophy 10 30.5 20 0.8046
    uk.religion.christian 10 13.79 26 0.7996 it.sport.calcio.fiorentina 10 5.5 23 0.7708
    fr.rec.moto 10 8.17 20 0.7442
    aus.computers 10 8 13 0.7299 alt.comp.os.windows-xp 10 5.33 10 0.6962

    --- Synchronet 3.21f-Linux NewsLink 1.2
  • From Colin Macleod@user7@newsgrouper.org.invalid to alt.free.newsservers on Mon Mar 30 08:34:39 2026
    From Newsgroup: alt.free.newsservers

    PN <noico@noico.net> posted:

    So I've created a script which helps me at least somewhat figure out
    which groups are actually alive.

    Thanks, this looks worthwhile. I had wondered about trying to devise
    some quality index for groups which I could add to my page at https://newsgrouper.org/tops but you seem to have done a comprehensive
    job of this. Could I add your results there? Also do you intend to
    repeat the analysis to track changes?
    --
    Colin Macleod ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ https://cmacleod.me.uk

    FEED HOUSE SAVE FEED HOUSE SAVE FEED HOUSE SAVE
    GAZA GAZA GAZA GAZA GAZA GAZA GAZA GAZA GAZA
    NOW! NOW! NOW! NOW! NOW! NOW! NOW! NOW! NOW!
    --- Synchronet 3.21f-Linux NewsLink 1.2
  • From Kerr-Mudd, John@admin@127.0.0.1 to alt.free.newsservers on Fri Apr 3 09:16:08 2026
    From Newsgroup: alt.free.newsservers

    On Mon, 30 Mar 2026 08:34:39 GMT
    Colin Macleod <user7@newsgrouper.org.invalid> wrote:

    PN <noico@noico.net> posted:

    So I've created a script which helps me at least somewhat figure out
    which groups are actually alive.

    Thanks, this looks worthwhile. I had wondered about trying to devise
    some quality index for groups which I could add to my page at https://newsgrouper.org/tops but you seem to have done a comprehensive
    job of this. Could I add your results there? Also do you intend to
    repeat the analysis to track changes?

    Even so 'a lot' of that traffic will be long-running disputes/flamewars
    with no added value.
    --
    Bah, and indeed Humbug.
    --- Synchronet 3.21f-Linux NewsLink 1.2