You are not logged in. Login Now
 0-24   25-49   31-55   56-80   81-105   106-130   131-155   156-180   181-205 
 206-230   231-255   256-280   281-299       
 
Author Message
25 new of 299 responses total.
tod
response 56 of 299: Mark Unseen   Mar 30 00:38 UTC 2005

re #50
I would not want to opt into someone else's interpretation of "annoying user".
I think it would be nice to be able to just type in "ignore russ" if I wanted
to not see his responses anymore in BBS.
(using russ as an example only)
spooked
response 57 of 299: Mark Unseen   Mar 30 01:27 UTC 2005

Two types of filters could be set up:

i) A user-defined filter list, so if I wanted to ignore a, b, c I could
ii) A global staff-defined or conf-admin defined list, defined by those 
respective person/s

Which filter to employ, if any, would be up to the individual user (and 
should be configurable once on first entering BBS, and then re-configurable 
thereafter at the individual user's discretion).



twinkie
response 58 of 299: Mark Unseen   Mar 30 01:27 UTC 2005

Maybe newuser could set that by default if someone enters a birthdate that's
after 1990.

twinkie
response 59 of 299: Mark Unseen   Mar 30 01:27 UTC 2005

Spooked slipped.

sholmes
response 60 of 299: Mark Unseen   Mar 30 02:57 UTC 2005

might as well have a universal ignore then , ignore xyz to ignore xyz on bbs
, party ,telegram and talk requests.
naftee
response 61 of 299: Mark Unseen   Mar 30 03:49 UTC 2005

Yeah, let's ignore all the Jews.
twenex
response 62 of 299: Mark Unseen   Mar 30 06:48 UTC 2005

Re: #55. Whooo! 

You know, you really can't blame stupid people for being stupid.
Unfortunately.
naftee
response 63 of 299: Mark Unseen   Mar 30 07:14 UTC 2005

People who use twit filters are stupid.  Unfortunately.
tod
response 64 of 299: Mark Unseen   Mar 30 16:35 UTC 2005

re #58
Maybe we could advocate throwing Catcher in the Rye into a bonfire, too.
twinkie
response 65 of 299: Mark Unseen   Mar 30 17:45 UTC 2005

I don't think you "got" it. 

scholar
response 66 of 299: Mark Unseen   Mar 30 17:59 UTC 2005

AHAHA THE POINT WAS THAT RUSS IS A PAEDOPHILE
"

AHAHAHA

THAT"S


OH GOD

tod
response 67 of 299: Mark Unseen   Mar 30 18:38 UTC 2005

No, it gives WORMS to ex girlfriends
remmers
response 68 of 299: Mark Unseen   Mar 30 20:16 UTC 2005

The question came up earlier about how many people use the conferences,
or Agora in particular, as a proportion of the total Grex user base. 
Mabye some tools already exist for generating relevant data, but I don't
know where to find them, so I wrote a couple of my own that might be
useful.  I'm sure they can be improved on; any local shell script
hackers are welcome to have at 'em.

The following commands are in the directory ~remmers/bin:

    confstats   -number of items, responses, and distinct responders
                 in a conference
    confusers   -list number of responses by all or a selected set of 
                 users in a conference (Regardless of what you're
                 thinking, the name "confusers" stands for "conference
                 users".)
    at-least-10-logins-in-march
                -a command with a silly name that displays the number
                 of users who logged into Grex at least 10 times during
                 March 2005.

To save you the trouble of running that last command, which takes some
time, I'll report that at this writing the number of users who logged in
at least 10 times in March is 947.  (That might include a small number
of administrative accounts that aren't really people, but I don't think
that's significant.)  I chose a threshhold of 10 because it's the first
2-digit number and also because there are a lot of people who go through
newuser to check out the system, log in only once or a handful of times,
and then never come back.  So they're not really users.  On the other
hand, if someone logs in 10 or more times, I think it's likely that
they've found something to do here.

So anyway, about 1000 real users, give or take a bit.

The "confstats" command gives you statistics on a conference.  You have
to give the actual directory name of the conference as an argument, e.g.
"agora53" (for the current Agora) instead of "agora".  The command

    ~remmers/bin/confstats agora52

(that would be winter agora) gives the output

    # items:       259
    # responses: 12960
    # talkers:     170

The number of "talkers" is the number of distinct logins who entered at
least one response.  It might be reasonable to subtract off people who
made only one or two responses, but I didn't do that.  Estimating the
number of "readers" is problematic due to the fact that no central
records are kept of those, and it's possible to read anonymously.

Maybe this says that about 15% of active Grex users participate in
Agora.  Maybe less, if you subtract off one-time-only responders.

Compare this to an Agora from 6 years ago:

    ~remmers/bin/confstats agora22

    # items:       132
    # responses:  9492
    # talkers:     262

A telling comparison.  Roughly half as many items 6 years ago, but
significantly higher average number of responses per item, and a whole
lot more people participating in the conversations.  The base of Agora
participants appears to have shrunk significantly.

The "confusers" takes a conference and a list of login id's as arguments
and gives you the number of responses by each of those users in the
conference.  Or if you just specify a conference, it lists responses by
all users in alphabetical order.  Example for the current Agora:

    ~remmers/bin/confusers agora53 other remmers scott

    18  other
    8  remmers
    20  scott

Because they're too long for a response, I've saved the all-users lists
in my web directory, sorted in descending order of number of responses.

    http://cyberspace.org/~remmers/winter-agora.txt
    http://cyberspace.org/~remmers/spring-agora.txt

Well, I hope somebody finds something useful to do with these tools in
terms of gauging where Grex is and where it's headed.
scholar
response 69 of 299: Mark Unseen   Mar 30 21:23 UTC 2005

Respond, pass, forget, quit, or ? for more options? !last other
other     ttyq4    pcp05305840pcs.wanarb01.mi.comcast.net Mon Mar 14 17:36
- 17:55  (00:19)
other     ftp      pcp05305840pcs.wanarb01.mi.comcast.net Sat Feb 12 03:37
- 03:37  (00:00)
other     ftp      pcp05305840pcs.wanarb01.mi.comcast.net Sat Feb 12 03:37
- 03:37  (00:00)
other     ttyp2    pcp05305840pcs.wanarb01.mi.comcast.net Sat Feb 12 03:25
- 03:26  (00:01)

wtmp begins Sun Dec 19 14:45 2004


OTHER has logged in only four times this month (and two of those were ftp
sessions!), yet i don't think anyone would consider him an insignificant user.

thus, logging in at least ten times a month doesn't seem like a very useful
way of measuring who's a significant user and who's not.
other
response 70 of 299: Mark Unseen   Mar 30 21:44 UTC 2005

It does appear that wtmp is not logging account usage via Backtalk. 
Remmers, how can your script account for those who are active users but
who primarily (or nearly exclusively) use Grex to conference via http?
remmers
response 71 of 299: Mark Unseen   Mar 30 23:30 UTC 2005

Hmm, weird.  If I recall correctly, http logins *were* logged in wtmp on
OldGrex, but you're right, they're not being logged now.  I suspect
that's a bug that needs to be fixed.

If it's not in wtmp, then offhand I'm not sure where's the best place to
get http login information.  It's probably extractable from the httpd
logs, but I suspect there's a simpler way, since the "laston" command
knows about http logins even if "last" doesn't.
dpc
response 72 of 299: Mark Unseen   Mar 30 23:52 UTC 2005

Wow - thanks for these new tools, John!
scholar
response 73 of 299: Mark Unseen   Mar 31 00:29 UTC 2005

thanks, john!
keesan
response 74 of 299: Mark Unseen   Mar 31 01:29 UTC 2005

Did you take into account the multiple personalities?
other
response 75 of 299: Mark Unseen   Mar 31 05:18 UTC 2005

That then begs the question, if wtmp is NOT logging use of accounts by
Backtalk, then will accounts be reaped after three months of http-only
access?
aruba
response 76 of 299: Mark Unseen   Mar 31 05:20 UTC 2005

Thanks for those tools, John.
scholar
response 77 of 299: Mark Unseen   Mar 31 05:41 UTC 2005

Re. 75:  that's not actually beggint the question.  begging the question is
when an answer to a question is obviously not adequate because it just pushes
the question one step back.

the thing with the elephants holding up earth is the standard example.  what's
holding up the earth?  an elephant.  what's holding up the elephant?  another
elephant.  this last answer is begging the question.
naftee
response 78 of 299: Mark Unseen   Mar 31 06:57 UTC 2005

What!

tod beat me for number of responses in spring agora ?!

that can't be!

http://cyberspace.org/~remmers/spring-agora.txt is not in alphabetical
order, by the way
remmers
response 79 of 299: Mark Unseen   Mar 31 14:24 UTC 2005

Re #74:  Nope, too subjective and difficult to automate.
Re #75:  Good question.  I don't know the answer.
Re #78:  Right.  The "confusers" command outputs in alphabetical order,
         so I piped the output through "sort" to get descending numeric:

                confusers | sort -nr
naftee
response 80 of 299: Mark Unseen   Mar 31 17:33 UTC 2005

ah, whoops.  didn't read that. thanks.
 0-24   25-49   31-55   56-80   81-105   106-130   131-155   156-180   181-205 
 206-230   231-255   256-280   281-299       
Response Not Possible: You are Not Logged In
 

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss