Grex Coop10 Conference

Item 78: How Can We Get More Users Into the Conferences?

Entered by dpc on Sun Feb 1 22:33:26 1998:

I'd like to see more Grexers participate in the conferences.  I don't
have any firm figures, but I think that we have maybe 150 or so regular
conferencers out of some 16,000 total users.  This is not great.
Two ideas:
        1.  Technical solutions.  For example, make bbs the default
for every newuser.  I haven't been through newuser since 1991, but
maybe people are being given too many choices and never do figure
out that we offer conferencing.
        2.  Publicity solutions.  We could send every newuser an
invitation to join the conferences.

62 responses total.

#1 of 62 by robh on Mon Feb 2 00:24:33 1998:

Hmm, last O jeard every member was being told about the
conferences when they ran newuser...


#2 of 62 by richard on Mon Feb 2 00:42:56 1998:

since so many people use grex primarilyfor email, maybe more would
participate in the confs if they could "subscribe" to them and
have new responses in selectedconfs delivered via email.  Instead
of using picospan or backtalk, theycould read the grex confs as they
read their listservs or other lists they subscribe to.  Couldnt
it be set up so people can opt to have confs delivered via email?

some people simply go straight into email, so this would allow them
to read the confs in their preferred environment, not forcing them to
learn new interfaces.


#3 of 62 by remmers on Mon Feb 2 00:57:07 1998:

There goes bandwidth and disk space down the tubes...


#4 of 62 by krj on Mon Feb 2 02:13:34 1998:

Maybe conferencing is obsolete.


#5 of 62 by steve on Mon Feb 2 04:42:59 1998:

   No conferencing isn't obsolete--the number of web based "forums"
out there in nat land is nothing short of amazing. There are lots
and lots and LOTS of people who've never heard of "computer 
conferencing" but know what a forum is.  This is the new, popular
term for conferencing, unforunately.

   Grex's problem is simply that we don't have good documentation
and pointers to people on how to use things.  Backtalk fixes a
lot of the problems, but it would still be good to have something
that we should show people, who ask for it.

   I'm pretty sure there are more than 150 people using the
conferences here, btw.  Still not enough, but more than 150.


#6 of 62 by gerund on Mon Feb 2 18:18:09 1998:

I wonder if maybe some of this is only perception.
People can actively read conferences, but not necessarily respond and make
their presence known.
(i call that lurking, btw...)   :)


#7 of 62 by richard on Mon Feb 2 18:27:11 1998:

people dont want to run strange interfaces to conf...thats why I was
saying that if email is being at least partially shifted to another machine
when the new SUN is running grex, maybe the time has come to let
people read the confs as mailing lists if they choose.  Bandwidth wouldnt
be the obstacle it would'vebeen in the past.


#8 of 62 by other on Mon Feb 2 19:02:09 1998:

is there a fairly simple way to find out how many people have responded in
a particular conference within a given period?  say if we wanted to see how
much use each conference is getting...  i'd be interested to see a listing
of all the conferences with the number of users responding within the last
week for each one.

gathering that data, or making it readily available, say by running a script
which would gather and format the data, might make it easier to sell the
conferences to people who just don't have any idea that there are things going
on in them....

another idea would be putting together a periodic mailing (say monthly) with
interesting or juicy selections from recent discussions within the previous
month.  this could be sent to either all users (except those who request
omission) or all new accounts within the last three months, or some similar
arrangement.

this  could be put together by group effort.  interested parties could excerpt
and send selections to an editor who could then vet them, and choose some to
include in the mailing.  if this idea has any support, i'll volunteer to take
the first editorial shift.    i'd like to be able to include some conference
use stats like those i mention above.  whaddaya think?


#9 of 62 by other on Mon Feb 2 19:03:44 1998:

oh--  i'd need some assistance from staff in gathering recipient information,
like a newuser list for the relevant period...


#10 of 62 by scott on Mon Feb 2 19:14:15 1998:

Richard, people don't want some strange interface for conferencing, so let's
write an entirely new (and therefore strange) interface?


#11 of 62 by aruba on Mon Feb 2 21:06:55 1998:

I think Richard's idea of showing the conferences within an interface that
users already know is a good one, but it won't work.  In order for a mailing
list to make any sense, people have to quote messages they are responding to, 
and people don't do that in Grex.  If every response came in a separate mail
message, the discussion wouldn't make any sense at all.

(And the last thing I'd like to see is conferencers quoting big chunks of
previous responses - that's why I liked Grex so much better than usenet in
the first place.)

As for mailing out summaries of conferences, I think the main problem with
that might be the disk space it would take up to have umpteen copies of the
excerpt.  If we could set it up so that there was only one copy but lots of
people were given a pointer to it in their mailbox, that might be reasonable.


#12 of 62 by dang on Mon Feb 2 21:30:08 1998:

I'd feel uncomfortable about sending unsolicited mail to people anyway. 
Seeing as I hate it so much...


#13 of 62 by steve on Tue Feb 3 05:05:09 1998:

  Richard, people do seem to be flocking to web based forum systems
quite well.  The web, no matter what the specifics of the interface
(well, within reason, I suppose--you could make a really awful one)
draw people in.  They like the click this, click that style of
interface.  I've now met many people who don't understand what
telnet is at all, but understand how to use places like HomeArts
and such fairly well.  It's really amazing, how things have changed
so rapidly in such a short time.

   Eric, staff could help gather the info, but you need to think
about what it is that you're measuring: simple amounts of entry
into a conference, or amount vs. time, or what.  I've thought
about this and have gleaned information about conferences before
but I've never liked what I've done--it never felt like I had
a good yardstick.


#14 of 62 by other on Tue Feb 3 07:20:14 1998:

specifically, a list of all the conferences, each with a number representing
the number of *users* who have entered responses into that conference within
the last period X.

To my thinking, this would give a good sense of the variety in each
conference, much more so than just the number of new responses...


#15 of 62 by janc on Tue Feb 3 15:46:43 1998:

(One of the functions planned for Backtalk in the future is the ability for
users to choose to receive digests of conference activity in the mail.  You
won't be able to post via email though - you'll have to actually come here
to do that.)


#16 of 62 by remmers on Tue Feb 3 18:45:58 1998:

(Interesting. I don't want to drift too much, but what are the
reasons for disallowing posting via email? I can think of one
reason that may be sufficient in and of itself -- the difficulty
of verifying authenticity of email -- but I'm curious if there
are others as well. Will the "digest" be less than the full
text of the items?)


#17 of 62 by aruba on Tue Feb 3 18:56:23 1998:

And how will the digest make sense, if it refers back to previous responses?


#18 of 62 by richard on Tue Feb 3 19:43:50 1998:

allowing conf'ing via email would allow participation by non-grexers as
well, something that could liven up many confs.


#19 of 62 by mta on Tue Feb 3 20:29:36 1998:

Call me conservative, but I think conferencing via e-mail starts in a direction
very different from what Grex started out to be.  We might become more and more
like Usenet and less and less a community.  I don't think it's necessarily so
-- but it certainly is a risk.

Digests of conference activity could be good, though, for keeping up when you
have to be away for a while.

Eric, are you intentionally ommitting the 80% or so of people in conferences
who lurk but never resond?  It seems to me that in some conferences it would be
quite possible for the interest to be high but for relatively few people to
have much to add to the conference.  (Specialty conferences like the web
conference come to mind.)


#20 of 62 by other on Tue Feb 3 22:23:34 1998:

Yes, but i do so because i cannot see how a measurement of the number of
people who read but do not contribute could reflect the actual activity in
the conference.  It certainly would be interesting to know how many people
*read* the conferences, but it doesn't seem to be of any use other than
determining just how many people *don't* look at the conferences...


#21 of 62 by dpc on Wed Feb 4 01:54:32 1998:

Misti, according to what Valerie has said, that "80% of people are lurkers"
is wrong.  She says (nearly) everyone who enters a conference participates
in it.  While I haven't analyzed Grex' conference usage, I know that
it is true on M-Net.  You might pick a couple of Grex conferences here,
do a "p" at the Ok: prompt, and you'll see a list of "participants"--
people who have joined, whether or not they've ever entered any material.
Then go through the conference and check out if those folks have *actively*
participated.  I suspect the overwhelming majority will have.
        Soo- what you see in the conferences is what you get.  My concern
is that we're not helping Grexers find the conferences effectively.


#22 of 62 by krj on Wed Feb 4 02:36:25 1998:

Back to Steve in #5:
 
>#5 of 21: by STeve Andre' (steve) on Sun, Feb  1, 1998 (23:42):
>    No conferencing isn't obsolete--the number of web based "forums"
> out there in nat land is nothing short of amazing. There are lots
> and lots and LOTS of people who've never heard of "computer 
> conferencing" but know what a forum is.  This is the new, popular
> term for conferencing, unforunately.

1)  So, should we start marketing Grex's conferences as "forums", then?
 
2)  How does Grex compare with some of these other web-based forums?
    Content, network link speed, the lag perceived by the users, etc?
    For better or worse, maybe we need to think of ourselves as 
    competitors with these other sites.


#23 of 62 by remmers on Wed Feb 4 11:11:12 1998:

The term "forum" fits very well with our choice of "agora"
as the name for the default conference.

I hang out in party a fair amount, where there are lots of users
whom one almost never sees in the conferences. Many of them are
quite aware of bbs, have taken a look at it, and simply prefer
live interaction. Although I'd like to see more participation in
bbs by the party users, they can't be forced, and it's not the
case that they need help "finding" the conferences.

So I'm not so sure it's true that the conferences aren't more
active because people don't know how to find them. It's more
likely, I think, that lots of people know how to find them but
aren't really interested. That doesn't surprise me much. Out
there in the offline "real world", just about everybody likes
to chat with folks but a comparative few are into writing
essays or even letters to the editor. Participating in
conferences is akin to writing lots and lots of op-ed
pieces. Some people are into this, but most aren't. Among
the people who use Grex mostly for chatting and mail, one
might be able to find a few who would become active
conference participants -- and this has happened now and
then -- but I doubt that there's a lot of gold to be mined
there.


#24 of 62 by davel on Wed Feb 4 12:38:28 1998:

Re #21: don't do a "p" unless you have lots of time to wait.

And this will give you a somewhat skewed view still.  It will give you people
who joined the conference once, looked, & decided then & there they'd never
want to read *that* one.  As long as they never ran the "resign" command,
they'll still have a participation file.


#25 of 62 by richard on Wed Feb 4 18:35:24 1998:

I never want to take the time to run a "p", but as a conf fw, I wouldnt
mind a faster way to get a list of all the conf subscribers, so I could
mass e-mail everyone once in a while to remind them of what's going on
in the confs.  People *do* forget about confs they've joined, particularly
if they dont include them on their .cflists.  


#26 of 62 by remmers on Wed Feb 4 23:01:29 1998:

Without a redesign of Picospan's record-keeping, there isn't a
fast way to get the participant list for a conference. Picospan
doesn't keep that information in a central place. To generate
the list, it has to search all users' home directories (all
14000 of them or so) for the existence of a participation file.
This is inherently slow.


#27 of 62 by mta on Thu Feb 5 18:33:27 1998:

I'm not sure unsolicited mailings are a good idea.  

People are already overwhelmed with spam and an unsolicited mailing from the FW
of a conference you tried and didn't like would probably fall into that
category.  (Few people think to resign from a conference they tried once.)


#28 of 62 by krj on Thu Feb 5 19:04:19 1998:

Actually I was toying with an idea where Grex would send an automated
mail soliciting participation in the conferences/forums some set period
of time after the user registered: one week, two weeks, a month.
In theory this would be better than throwing more and more 
information into the "newuser" screens.
 
I would disagree with Misti; Grex should be able to send a judiciously 
chosen amount of mail to its own users, even if it is purely 
advertising mail.


#29 of 62 by richard on Thu Feb 5 21:29:03 1998:

I guess this is a marcus question, but couldnt picospan be updated
so it also keeps separate conf participation lists in a central list?
Is this something that would be hard to do?  Who wants to spend two
hours running a "p"!


#30 of 62 by davel on Thu Feb 5 23:18:31 1998:

What Misti said, contra what krj said.  *I* for one would object to the kind
of mailing richard advocated.  Repeated instances would start me thinking in
terms of guerilla reprisals, TBH.


#31 of 62 by scott on Fri Feb 6 01:54:43 1998:

Re: 29:  PicoSpan could keep a list of participants, but at the expense of
being slower to use.  That's a basic programming tradeoff.  I don't know if
modifications to PicoSpan are even possible, given the history of how we get
to use it.


#32 of 62 by valerie on Fri Feb 6 05:35:55 1998:

This response has been erased.



#33 of 62 by scg on Fri Feb 6 06:13:23 1998:

I'm not sure that people would object to getting mail advertizing the
conferences, or other things Grex does.  I work at a somewhat large ISP.  A
while back, we made some fairly major changes to how people had to log on,
so to avoid a situation where nobody could log on anymore, somebody threw
together a script to e-mail everybody in the password file, and sent out a
mass mailing documenting the changes.  Then somebody else started using that
script to send out occasional messages telling customers about new services
we were offering, upcoming changes that would increase performance even though
there was nothing the customers had to do to prepare for them, and things like
that.  I thought people would see it as more unsolicited spam and get upset.
Instead, people started writing back saying they liked being kept informed,
sometimes even suggesting that we should send out messages like that more
often.  The conclusion I've drawn from that is that people really don't object
to mass e-mails if they know where they're coming from and isn't just trying
to sell them something.

In Grex's case, since we don't have paying customers, I wonder how much we
should care if people using Grex object to being told about the things that
those of us who are funding and maintaining Grex think are Grex's important
features.  If people using Grex for free e-mail were to object to an
occasional message telling them what this system they're using is, then I'm
really not all that motivated to keep providing free service to them.


#34 of 62 by rcurl on Fri Feb 6 08:29:49 1998:

I go along with that.


#35 of 62 by remmers on Fri Feb 6 09:39:12 1998:

Re #32: Oh right - forgot about 'locate', which centralizes
everything. Come to think of it, seems not too hard to write
a script which takes a conference name as argument and generates
a fast participant list by looking up the conference directory
in the conflist file, getting the participation file name from
the conference config file, invoking 'locate' to generate the
participant list, and reformatting the output suitably.

This wouldn't give quite as much information as the built-in
Picospan locate command, and accuracy depends on the locate
database being up-to-date, but it'd be pretty fast.


#36 of 62 by dpc on Fri Feb 6 16:07:06 1998:

I'm with Ken and Steve on the idea of sending limited mailings.
After all, if we don't blow our own horn, no one else will!  8-)


#37 of 62 by mta on Fri Feb 6 17:23:41 1998:

I must say that Steve has swayed my thinking on this.

I'd still hate to see it used very often -- but mostly from the PoV of taxing
Grexes resources.  In reading Steve's response I thought that maybe, just
maybe, such mailings occasionally could give people more of a sense that this
is a community they've joined rather than a "cheap AOL clone".

Dunno.  I'd have to think about it some more....


#38 of 62 by valerie on Fri Feb 6 20:45:48 1998:

This response has been erased.



#39 of 62 by carson on Fri Feb 6 20:59:54 1998:

(maybe adding a wrapper of sorts similar to ".partymsg" to the mail
programs?)


#40 of 62 by other on Fri Feb 6 23:49:37 1998:

(side note:  would someone please explain what the .partymsg file is for?
it keeps showing up in my directory with a size of zero...)


#41 of 62 by scg on Sat Feb 7 02:59:36 1998:

There is a script that shows people messages about changes to the party
program when they run it.  To avoid showing people the same message over and
over again, the script has to keep track of who has seen which messages.  I
believe it uses the .partymsg file for that.  If it's a zero byte file, it
must go by the date on the file, rather than its contents.

If I understand Carson's suggestion correctly, he's suggesting a script around
the mail programs that will send a user any system messages they haven't
already seen every time they check their mail.  That's a much more effecient
way of doing things.

In Grex's case, I hadn't been thinking in terms of sending a message to
everybody.  If we are going to do that, what Carson suggested is definitely
the way to go.  If people are thinking of just doing a mass mailing to a bunch
of people, that will work much better in terms of the messages to 30 or so
users people have been talking about than it would for sending a message to
absolutely everybody.


#42 of 62 by rcurl on Sat Feb 7 07:33:44 1998:

Which 30 people was that? I think sending a well worded - and brief - message
to everyone, with that e-mail wrapper, is the way to go. But, won't it
repeat on every mail use while it is there? A couple of days of that is
OK, I would think, for frequent users, but if it isn't left on for a while,
it won't get those once-a-week-ish users. Is there a way to make it appear
just once for each user?


#43 of 62 by scg on Sat Feb 7 08:12:32 1998:

The idea would be to copy the functionality of the .partymsg file, such that
it would check whether somebody had seen the message already, and only send
it if they hadn't.


#44 of 62 by mdw on Sat Feb 7 09:24:22 1998:

I'd certainly find it in poor taste (to say the least) if somebody sent
out a mass mailing to everyone in their conference.  I get quite enough
mail as it is.

Is the quesion here how to drum up more activity in existing conferences
(presumably involving active users), or to attract new people? If it's
just a question of drumming up activity in conferences, there's no need
to send e-mail to users to do that.  It's much more effective to enter
new items and responses, or selectively approach a few people, and send
much more personalized mail encouraging them to participate.  Even a few
people putting some work into posting new things can make a conference
drammatically more interesting.

But I think the original point of this item was not to discuss ways of
reaching existing participants, but *new* participants.  *That*'s a much
more tricky problem.  There are some intrinsic issues that make this
hard: not just anyone is going to be interested, and in fact, the % of
the general population who would really enjoy & use computer
conferencing just to socialize is actually quite small.  You're talking
about reaching people who are literate, enjoy reading, and (hopefully)
enjoy writing, and have sufficient skills at writing to generate postive
feedback.  Right there, that knocks out at least 90% of the general
population.  After that, there's lots of other more subtle stuff that
figures in, and only something like 1-2% of the general population (at
*best*) is really likely to get hopelessly addicted to computer
conferencing in the sense that we, here, all are.

I'm pretty sure that there are technical issues that we aren't aware of,
that dissuade people from participating.  For instance, one thing I
noticed, is the default configuration we give people (pager, terminal
configuration, etc.) really sucks for people who are coming in from an
xterm.  After displaying the last page of text, the pager cheerfully
immediately swaps screens back, which results in replacing the
interesting text on the screen with the prompt & command used to
originally start reading - the effect is that, minus some screen
flicker, "nothing happened".  There's a peculiar combination of being
"helpful" and offering more functionality that causes this to happen -
"less" is "supposedly" more friendly than "more", some terminal
emulators (including X) support multiple screen buffers, & etc.
Unfortunately, however helpful this all is, it doesn't really help a new
person just coming in who experiences this, who probably has no clue
even that something went wrong, only that it's not interesting.  To a
lesser extent, this is a problem with any terminal emulator that's
running at anything but 24x80, and there are of course a lot of terminal
emualtors out for windows these days.


#45 of 62 by valerie on Sat Feb 7 13:48:26 1998:

This response has been erased.



#46 of 62 by mary on Sat Feb 7 14:25:33 1998:

Someone, way back there, suggested putting an encouragement in
the MOTD rather than mail.  I didn't see, but I may have missed,
why this wouldn't be a better way to go.


#47 of 62 by dpc on Sat Feb 7 16:54:13 1998:

Marcus, my main interest is, as you say, to attract new people who
are already on Grex into the conferences.  If my estimate of 150
conferencers out of 16,000 users is right, we are only getting 1%
of the users into the conferences.


#48 of 62 by krj on Sat Feb 7 21:36:26 1998:

How did M-net's experiment of putting short-term promotions for 
various conferences into the MOTD work out?  


#49 of 62 by jared on Sat Feb 7 23:07:24 1998:

You can use the msgs program to do that.


#50 of 62 by remmers on Sun Feb 8 00:04:26 1998:

What is msgs?

        !man msgs
        No manual entry for msgs.



#51 of 62 by valerie on Sun Feb 8 15:31:09 1998:

This response has been erased.



#52 of 62 by valerie on Sun Feb 8 15:33:05 1998:

This response has been erased.



#53 of 62 by dpc on Sun Feb 8 16:26:22 1998:

Nice plug, Valerie!
        Ken, M-Net's program of advertising particular conference did
rather well.  Each time we would pull from 20-50 new people into that
conference.  Some conferences were re-invigorated; some not.
        Grex could easily have 300-500 people who participate in our
conferences; if so, then based on my comparisons between Grex and
M-Net conferences, M-Net probably has double Grex', or 600-1000 folks
altogether.
        Taking the mid-range for each (400 Grex; 800 M-Net), then
since Grex has 16000 users and M-Net around 5500, about 3% of Grexers
use the conferences and 15% of M-Netters.  L-o-w for both.  Maybe
M-Net's promotion of its conferences explains the difference, 
maybe not.


#54 of 62 by dang on Sun Feb 8 16:50:33 1998:

(I'd guess the difference is outside "noteriety" and m-net's tiered access.
Grex made it onto some lists of places to get shell accounts and email, and
everyone gets the same access, so more people show up for shell accounts and
email.)


#55 of 62 by jared on Sun Feb 8 19:28:30 1998:

I've put the man page for the msgs program on grex in my home directory.
to view it do this:
!more /a/j/a/jared/msgs.1   or for those folks who use a 'real' pager,
!less /a/j/a/jared/msgs.1   it's standard software on FreeBSD, and
i know folks using it with sunos also, so it must be easy to port.


#56 of 62 by jep on Thu Feb 12 15:10:54 1998:

My ISP sends a message or two to each subscriber, telling people of 
changes and improvements to the system.  It's a small ISP based in 
Adrian, but it must have at least a few thousand subscribers.  The 
messages are usually several screens long.  I'm not annoyed by them at 
all; I read every one.  Grex could do this, if Grex has anything to tell 
the users.  I'll bet most people wouldn't mind, and a lot would 
appreciate it.


#57 of 62 by aruba on Thu Feb 12 16:24:16 1998:

I certainly get the feeling that there are a lot of people using Grex for mail
who don't have a clue what we're all about.


#58 of 62 by rcurl on Thu Feb 12 19:58:24 1998:

Send them mail telling them. We will get more conferencers, and maybe 
some new members. The motd is not motivating.


#59 of 62 by jared on Fri Feb 13 04:48:10 1998:

<insert a newuser welcome-to-grex e-mail here>


#60 of 62 by valerie on Sat Feb 21 05:48:35 1998:

This response has been erased.



#61 of 62 by srw on Wed Mar 4 03:25:02 1998:

I like that idea, too.


#62 of 62 by lilmo on Mon May 11 23:13:58 1998:

Re #60:  s/each user/each new user/

 :-)


There are no more items selected.

You have several choices: