Eighteen months of Bamby failure
Peter J Ross
results of the Bambies' first 18 months of "group creation" (i.e.
newgrouping) and the results of the 18 months that preceded them.
30 Nov 2004 to 29 May 2006 (last 18 months of "Troika" management):
Total number of groups newgrouped: 5
Aggregate number of posts to these groups during past 10 days: 659
30 May 2006 to 29 Nov 2007 (first 18 months of "Bamby" management):
Total number of groups newgrouped: 32
Aggregate number of posts to these groups during past 10 days: 549
Despite newgrouping at more than six times their predecessors' rate,
the Bambies have attracted only 549 posts to their 32 new groups,
compared with 659 posts in their predecessors' five groups during the
same period of time. While the old system was not unreasonably
criticised for resulting in the creation of too few groups, even the
tiny number of new groups it produced contain a larger aggregate
number of recent posts than the aggregate number for the groups
newgrouped under the new system.
It seems clear to me that, despite newgrouping almost everything that
has been proposed as well as many groups that haven't been proposed
but "seemed obvious", the Bambies have failed to demonstrate that
their system works better than the "broken" system it replaced;
indeed, it seems to work slightly worse, even without dividing the
aggregate numbers of posts in new groups into totals for the
individual groups or otherwise adjusting the figures.
I wonder whether the Bambies will try to explain this evidence of
their failure away or merely ignore it.
Here are the details for each group in case anybody's interested. MI5
spam is excluded; all other posts seem by their Subjects to be
genuine.
1. Dec 2004 to Apr 2006 (Created after CFV)
NEWSGROUPS RECENT POSTS
misc.writing.screenplays.moderated 416
comp.databases.mysql 221
comp.os.os2.ecomstation 19
comp.databases.postgresql 3
soc.religion.satanism 0
2. May 2006 to Nov 2007 (Created after RFD)
NEWSGROUPS RECENT POSTS
comp.lang.xharbour 200
rec.ponds.moderated 74
sci.physics.foundations 60
news.groups.proposals 33
talk.current-events 28
rec.radio.amateur.moderated 26
comp.lang.haskell 9
soc.religion.asatru 9
comp.sys.laptops.thinkpad 8
soc.men.moderated 6
soc.support.stroke 2
news.admin.moderation 1
comp.soft-sys.octave 0
rec.arts.sf.tv.jericho 0
rec.motorcycles.victory 0
soc.support.vision-impaired 0
3. Sep 2007 to Nov 2007 ("Obvious" groups)
NEWSGROUPS RECENT POSTS
rec.games.video.microsoft.xbox360 19
comp.internet.services.google 18
rec.media.players.portable.ipod 16
rec.arts.tv.news.oreilly-factor 11
rec.games.video.nintendo.wii 11
rec.arts.tv.comedy.the-office 8
rec.games.video.sony.playstation3 6
misc.phone.mobile.iphone 2
comp.file-sharing.bittorrent 1
rec.arts.tv.comedy.colbert-report 1
comp.internet.services.blog.livejournal 0
comp.internet.services.social.myspace 0
comp.internet.services.video.youtube 0
comp.internet.services.wiki 0
misc.phone.mobile 0
rec.media.players.portable 0
--
PJR :-)
Raving
> I thought it might be interesting to make a comparision between the
> results of the Bambies' first 18 months of "group creation" (i.e.
> newgrouping) and the results of the 18 months that preceded them.
>
> 30 Nov 2004 to 29 May 2006 (last 18 months of "Troika" management):
> Total number of groups newgrouped: 5
> Aggregate number of posts to these groups during past 10 days: 659
>
> 30 May 2006 to 29 Nov 2007 (first 18 months of "Bamby" management):
> Total number of groups newgrouped: 32
> Aggregate number of posts to these groups during past 10 days: 549
Does this mean that I am unlikely to have your support for my upcoming
RFD to create rec.food.lutefisk ?
http://cache.aftenposten.no/multimedia/archive/00076/null_76082a.jpg
"The most traditional lutefisk presentation: with unpeeled potatoes,
bacon, stewed peas, and mustard sauce - and with beer and aquavit to
keep things moving. Skål!"
Daedalus
wrote:
>I thought it might be interesting to make a comparision between the
>results of the Bambies' first 18 months of "group creation" (i.e.
>newgrouping) and the results of the 18 months that preceded them.
>
>
>30 Nov 2004 to 29 May 2006 (last 18 months of "Troika" management):
>Total number of groups newgrouped: 5
>Aggregate number of posts to these groups during past 10 days: 659
>
>30 May 2006 to 29 Nov 2007 (first 18 months of "Bamby" management):
>Total number of groups newgrouped: 32
>Aggregate number of posts to these groups during past 10 days: 549
>
>
>Despite newgrouping at more than six times their predecessors' rate,
>the Bambies have attracted only 549 posts to their 32 new groups,
>compared with 659 posts in their predecessors' five groups during the
>same period of time. While the old system was not unreasonably
>criticised for resulting in the creation of too few groups, even the
>tiny number of new groups it produced contain a larger aggregate
>number of recent posts than the aggregate number for the groups
>newgrouped under the new system.
And remind me, what group were all of those five groups discussed and
created on?
>
>It seems clear to me that, despite newgrouping almost everything that
>has been proposed as well as many groups that haven't been proposed
>but "seemed obvious", the Bambies have failed to demonstrate that
>their system works better than the "broken" system it replaced;
>indeed, it seems to work slightly worse, even without dividing the
>aggregate numbers of posts in new groups into totals for the
>individual groups or otherwise adjusting the figures.
>
>I wonder whether the Bambies will try to explain this evidence of
>their failure away or merely ignore it.
>
Maybe they can get to that after I finish locating more than one
oppressed soul whom ngp was created for because they felt too
intimidated to post to ng.
Excellent work, Abusive asshole sock of mine.
Jade
Dave Sill
> I thought it might be interesting to make a comparision between the
> results of the Bambies' first 18 months of "group creation" (i.e.
> newgrouping) and the results of the 18 months that preceded them.
It is interesting, thanks.
> 30 Nov 2004 to 29 May 2006 (last 18 months of "Troika" management):
> Total number of groups newgrouped: 5 Aggregate number of posts to
> these groups during past 10 days: 659
>
> 30 May 2006 to 29 Nov 2007 (first 18 months of "Bamby" management):
> Total number of groups newgrouped: 32 Aggregate number of posts to
> these groups during past 10 days: 549
One problem with this comparison is that the "Troika" groups are more
mature than the "Bamby" groups. Some new groups have an initial flurry
of activity and taper off. Other new groups start slowly and build
gradually. One could look at the Troika groups at the same point in
their lifespan as the Bamby groups, but that assumes that the overall
level of Usenet activity over the last 18 months has been constant, and
I think it's pretty clear that it's on the decline.
--
Dave Sill, not speaking for the board
Aratzio
<da...@sill.org> bloviated:
Lovely reasoning, if you are a moron.
The periods are exactly the same in their life cycles, so his primary
refutation is bogus.
Then comes the "But usenet is dying so our results are worse"
Noting that Sill (like Lee) uses very imprecise terms to refute
precise data.
Rather than do his own research to refute actual hard data, he just
waves his hands and dismisses the data as not important.
Peter J Ross
<da...@sill.org> wrote:
> Peter J Ross wrote:
>> I thought it might be interesting to make a comparision between the
>> results of the Bambies' first 18 months of "group creation" (i.e.
>> newgrouping) and the results of the 18 months that preceded them.
>
> It is interesting, thanks.
Please try to put it out of your mind. Taking an interest in factual
evidence could damage yourBamby career prospects.
>> 30 Nov 2004 to 29 May 2006 (last 18 months of "Troika" management):
>> Total number of groups newgrouped: 5 Aggregate number of posts to
>> these groups during past 10 days: 659
>>
>> 30 May 2006 to 29 Nov 2007 (first 18 months of "Bamby" management):
>> Total number of groups newgrouped: 32 Aggregate number of posts to
>> these groups during past 10 days: 549
>
> One problem with this comparison is that the "Troika" groups are more
> mature than the "Bamby" groups.
Most of the "dead" groups are more mature than either. How old does
a newsgroup have to be before you take the trainer wheels off its bike?
In any case, this herring smells red to me. Please focus on *posts*
that wouldn't have existed or would have existed elsewhere if the
groups hadn't existed.
> Some new groups have an initial flurry
> of activity and taper off. Other new groups start slowly and build
> gradually.
And some, such as soc.men.moderated (to which I've intermittently
tried to apply artificial respiration since the day it was newgrouped)
and most if not all of the "obvious" groups show little signs of
independent life despite their sponsors' interference - not at first
and not later.
> One could look at the Troika groups at the same point in
> their lifespan as the Bamby groups, but that assumes that the overall
> level of Usenet activity over the last 18 months has been constant,
> and I think it's pretty clear that it's on the decline.
If so, perhaps it's declining faster in your new groups. But at
present, I have only one snapshot of current activity. In order to
turn the screw tighter, I'll need to post such snapshots at regular
intervals.
--
PJR :-)
Brian Mailman
> I thought it might be interesting to make a comparision between the
> results of the Bambies' first 18 months of "group creation" (i.e.
> newgrouping) and the results of the 18 months that preceded them.
>
>
> 30 Nov 2004 to 29 May 2006 (last 18 months of "Troika" management):
> Total number of groups newgrouped: 5
> Aggregate number of posts to these groups during past 10 days: 659
>
> 30 May 2006 to 29 Nov 2007 (first 18 months of "Bamby" management):
> Total number of groups newgrouped: 32
> Aggregate number of posts to these groups during past 10 days: 549
>
>
> Despite newgrouping at more than six times their predecessors' rate,
> the Bambies have attracted only 549 posts to their 32 new groups,
> compared with 659 posts in their predecessors' five groups during the
> same period of time.
You're being generous in counting the cut'n'paste announcements in
soc.support.stroke, and the meta-threads in smm discussing moderation
policy.
B/
Fjiluk The Hogfish-Beater
> On Nov 29, 11:40 pm, Peter J Ross <p...@example.invalid> wrote:
>> I thought it might be interesting to make a comparision between the
>> results of the Bambies' first 18 months of "group creation" (i.e.
>> newgrouping) and the results of the 18 months that preceded them.
>>
>> 30 Nov 2004 to 29 May 2006 (last 18 months of "Troika" management):
>> Total number of groups newgrouped: 5
>> Aggregate number of posts to these groups during past 10 days: 659
>>
>> 30 May 2006 to 29 Nov 2007 (first 18 months of "Bamby" management):
>> Total number of groups newgrouped: 32
>> Aggregate number of posts to these groups during past 10 days: 549
>
> Does this mean that I am unlikely to have your support for my upcoming
> RFD to create rec.food.lutefisk ?
I thought it was Ljutefisk?
Raving
wrote:
See http://photojournal.jpl.nasa.gov/figures/PIA07018_fig1.jpg
Kathy Morgan
> at present, I have only one snapshot of current activity. In order to turn
> the screw tighter, I'll need to post such snapshots at regular intervals.
Please do, because it is interesting.
--
Kathy
Peter J Ross
<p...@example.invalid> wrote:
> 30 Nov 2004 to 29 May 2006 (last 18 months of "Troika" management):
> Total number of groups newgrouped: 5
> Aggregate number of posts to these groups during past 10 days: 659
Update:
Aggregate number of posts to these groups during past 10 days: 813
> 30 May 2006 to 29 Nov 2007 (first 18 months of "Bamby" management):
> Total number of groups newgrouped: 32
> Aggregate number of posts to these groups during past 10 days: 549
Update:
Aggregate number of posts to these groups during past 10 days: 685
> Despite newgrouping at more than six times their predecessors' rate,
> the Bambies have attracted only 549 posts to their 32 new groups,
> compared with 659 posts in their predecessors' five groups during the
> same period of time.
Raw data would suggest improved performance by the Bamby groups during
the current 10-day period, but closer inspection shows that the
difference is largely the result of a crossposted flame war in
misc.phone.mobile.ipod.
<...>
> I wonder whether the Bambies will try to explain this evidence of
> their failure away or merely ignore it.
So far they've offered a mixture of the two. It remains true that, if
the old newsgroup creation system was broken, a system that performs
worse than it did (as seems to be the case with the current system)
can only be described as "more broken" and is therefore a failure.
> Here are the details for each group in case anybody's interested. MI5
> spam is excluded; all other posts seem by their Subjects to be
> genuine.
Previous totals and current totals are listed side by side this
time. Note that rec.arts.tv.comedy.colbert-report was
previously mistakenly listed among the "obvious" groups.
1. Dec 2004 to Apr 2006 (Created after CFV)
NEWSGROUPS POSTS
misc.writing.screenplays.moderated 416 577
comp.databases.mysql 221 215
comp.os.os2.ecomstation 19 14
comp.databases.postgresql 3 7
soc.religion.satanism 0 0
--- ---
TOTAL 659 813
2. May 2006 to Nov 2007 (Created after RFD)
NEWSGROUPS POSTS
comp.lang.xharbour 200 164
rec.ponds.moderated 74 85
news.groups.proposals 33 64
rec.radio.amateur.moderated 26 34
comp.sys.laptops.thinkpad 8 32
sci.physics.foundations 60 31
rec.motorcycles.victory 0 25
news.admin.moderation 1 17
talk.current-events 28 12
soc.men.moderated 6 10
comp.lang.haskell 9 4
soc.support.stroke 2 2
rec.arts.tv.comedy.colbert-report 1 2
soc.religion.asatru 9 1
soc.support.vision-impaired 0 1
comp.soft-sys.octave 0 0
rec.arts.sf.tv.jericho 0 0
--- ---
TOTAL 457 484
3. Sep 2007 to Nov 2007 ("Obvious" groups)
NEWSGROUPS POSTS
misc.phone.mobile.iphone 2 115
rec.games.video.microsoft.xbox360 19 26
rec.games.video.nintendo.wii 11 14
comp.internet.services.blog.livejournal 0 13
rec.media.players.portable.ipod 16 9
rec.arts.tv.news.oreilly-factor 11 9
comp.internet.services.google 18 7
rec.arts.tv.comedy.the-office 8 5
comp.internet.services.video.youtube 0 2
comp.internet.services.social.myspace 0 1
rec.games.video.sony.playstation3 6 0
comp.file-sharing.bittorrent 1 0
comp.internet.services.wiki 0 0
misc.phone.mobile 0 0
rec.media.players.portable 0 0
--- ---
TOTAL 92 201
--
PJR :-)