|
Grex > Coop8 > #134: Should Grex get a T1 connection? | |
|
| Author |
Message |
| 25 new of 146 responses total. |
mdw
|
|
response 75 of 146:
| 
Nov 7 15:55 UTC 1996 |
I think we need to *expect* that eventually most members *will* be using
IP, and not modems, to get into grex. The problem with looking at past
& present members, is that they aren't necessarily a good predictor of
future members. That's especially important to keep in mind considering
how fast the technology is evolving.
|
ajax
|
|
response 76 of 146:
|
Nov 7 19:03 UTC 1996 |
I agree with its inevitability. The 30:70 split is already getting
close, and I'd bet member telnet usage is rising over time. But while
the ratio is going up, it's useful from a financial perspective to keep
in mind that the ratio of members to guests for telnetters is still
very much lower than for direct dialers.
Right now, I think it's safe to say that our dial-in line usage,
and hence some of the our local populace, is limited to some extent
by the CPU speed. The phone lines are not busy most of the time.
If Grex were faster when people did get on, it would be a more
attractive system to local dialers, and more of them would continue
using Grex, and would recommend it to other local people. This in
turn would probably yield more donations than if the same resources
are allocated to be less appealing to local dialers and more
appealing to telnetters.
Sorry if this issue seems off-topic, but I think this and other side
issues above really are related to the proposal of getting a T1, as
financing is a central issue of the proposal.
|
dang
|
|
response 77 of 146:
|
Nov 7 19:58 UTC 1996 |
RE 76: I'd have to disagree about the phones usually being not busy. Over
the past two or three months, I routinely try and dial in two or three times
before I give up and telnet, and I very rarely am able to get a dialin. Most
of the time, I have to telnet because the dialins are all full. I don't know
what that has to do with the T1, but it's data.
|
pfv
|
|
response 78 of 146:
|
Nov 7 20:26 UTC 1996 |
Fair Warning: Lose the dialins at your peril.
When a majority of telnetters support your works, you have
arguments of value.
|
kerouac
|
|
response 79 of 146:
|
Nov 7 21:05 UTC 1996 |
Grex has a growing number of users, but its ACTIVE user base, those
who contribute or use it in itns best ways, is not
growing. Its stagnant and many conferences are
under used. Grex's status quo is not good enough to bend over backwards
to maintain over the long term. Even if the T1 doesant pay for itself and
Grex goes under, the gamble is worth taking. When you stop taking
chances, that is when you slowly start to die.
If you dont take risks oncein a while, you may as well be dead already.
Go for the T1. Grex may die anyway, but at least it will have lived!!
|
scg
|
|
response 80 of 146:
|
Nov 7 21:08 UTC 1996 |
We certainly don't have a majority of our dial-in users as members.
While we do havea lot more of our dial-up users as members than telnet users,
it costs a lot more to support dial-up users. 11 dial-up lines cost us $220
per month (that's not counting the dial-up lines that we aren't using yet).
Our Internet connection, which often has 64 users coming over it, I think,
costs us around $40 per month.
|
ajax
|
|
response 81 of 146:
|
Nov 7 21:43 UTC 1996 |
The 70% of our members who mostly use dial-ins contribute around $350
per month, more than the $220 per month cost of the lines, so they're
profitable in that sense. (Dialers also use other resources, like rent,
power, new equipment costs, and CPU cycles, so that's an over-simplified
analysis, but the ratio is high enough that I think each line still turns
a profit). If that ratio of members-per-line could be maintained while
gradually adding more lines, even though the lines cost a lot, they would
produce a profit.
The *current* Internet link is also profitable just counting the 'net
link in this sense, because we get such a good rate, but it will be very
close with the T1 unless it draws several more members. Also, when you
consider that the 'net users use the vast majority of other resources,
if you attribute a greater portion of shared costs to them (equipment,
rent, power), a more expensive 'net link does not look profitable in the
way that dial-in lines do.
|
kerouac
|
|
response 82 of 146:
|
Nov 7 21:52 UTC 1996 |
Does it have to be profitable? Wouldnt the T1 be worth
it even if it was only ever a break even proposition?
|
dang
|
|
response 83 of 146:
|
Nov 8 00:05 UTC 1996 |
Richard has a good point. If Grex were a profit making organization, the T1
would probably not be worth it.
|
steve
|
|
response 84 of 146:
|
Nov 8 14:20 UTC 1996 |
Richard is absolutely right: as long as we can pay for it, its
reasonable. All I care about is that Grex has enough money to cover its
expenses.
|
ajax
|
|
response 85 of 146:
|
Nov 8 14:37 UTC 1996 |
I wasn't suggesting that Grex try and accumulate wealth; the advantage
of self-financing or profitable additions is that they allow losses on
other things. Overall expenses should be in keeping with overall income.
|
mdw
|
|
response 86 of 146:
|
Nov 8 17:21 UTC 1996 |
Nobody, least of all me, is suggesting that we "get rid of the
dial-ins". All I'm saying is that in the long run, this is technology
we should expect to be moving away from. Even as matters stand, there
are aspects of modem technology that Rob deliberately ignored, that
certainly increase our overhead. For instance, *all* of our modems now
online are out of date, & no longer made. Half of them are definitely
antique. We also have 2 modems purchased that are a mistake, a terminal
server with a heat problem, &etc. there is that large a population of
Ann Arbor users who would become members if only grex had more lines.
So modems definitely aren't as good a deal as Rob suggests, and adding a
lot of new lines probably doesn't make sense. Of course, we *do* want
to keep the existing lines, and in fact, we need to do work on them to
upgrade them, so that they continue to be useful.
However, as long as we think of grex as a "computer conferencing
system", and not a low cost ISP, modems are really only an incidental
part of our mission. If a T1 line would provide access to many more
users at a lower per-user cost, then it's a perfectly legitimate means
for us to achieve our organizational goals, & the lower membership rates
is almost completely incidental.
I think, too, it's a mistake for us to be assuming that any sort of
majority of our users *should* be members. PBS stations generally
manage a 3% membership rate. I think, in the long run, that might be a
reasonable goal for us to hope to achieve here. I think it is also
reasonable to hope that the increased speed of a T1 line might well
attract more donations from people who currently telnet into grex, but
currently think it's way to slow.
|
ajax
|
|
response 87 of 146:
|
Nov 8 19:51 UTC 1996 |
I said the "$350 to "$220" comparison was oversimplified, and that
new equipment and other costs have to be factored in. I don't consider
that "deliberately ignoring" the costs, but my point is that they have
a high enough ongoing income-to-expense ratio that even with those other
costs, over time, the lines are cost-effective. The one-time costs to
start a line might make the payoff a year down the road, but they seem
to pay for themselves, and even turn a profit, in the long run.
I agree completely with your last two paragraphs. If we had enough
surplus income right now, I'd certainly support getting a T1, and I
haven't rejected the idea even with our current income - we have been
running at least a little surplus lately. It's an unquestionably
cheaper means of access on a per-user basis, and as you said, the
better service should draw some new members.
Out of curiosity, which way are board members leaning on this issue
right now? The non-Steve board members have been surprisingly
reticent on the topic! :-) (I think the Steves are both in favor of
taking this deal).
|
scg
|
|
response 88 of 146:
|
Nov 8 20:34 UTC 1996 |
This Steve is certainly in favor of it.
|
kerouac
|
|
response 89 of 146:
|
Nov 8 23:02 UTC 1996 |
Currently, the T1 has the support of three of the seven board
members (based on responses in this item)...SCG,SRW, ROBH. If
one of MTA, Aruba,Scott or Valerie support it or if any one of
them misses the meeting or abstains, Grex will get the T1. I'd
say the odds look pretty good...pretty darn good actually.
|
mta
|
|
response 90 of 146:
|
Nov 9 00:35 UTC 1996 |
I'm still extremely ambivalent. I really believe we need to go in this
direction, and as many have already pointed out, this may well be our best
chance. But the money worries me. (Money always worries me.)
But past experience shows that if GREX build a better conferencing system,
the membershiops will come. But I also believe that speeding up the system
to dial-in users is more likely to result in more memberships coming in
faster. (Local users are the vast majority of our members, last I'd heard.
Yes, I'm still supremely ambivalent. Let me waffle some more and I'll get
back to you.
|
steve
|
|
response 91 of 146:
|
Nov 9 07:42 UTC 1996 |
Something to note, and perhaps aruba can give more details on,
is that our membership base has been slowly climbing upwards.
We're at 103 members right now according to the members command.
...We still need 40 more folks to come forward to sustain the T1
forever, but as we grow our services, we get more people. Misti
is quite right about that in #90.
|
scott
|
|
response 92 of 146:
|
Nov 9 13:14 UTC 1996 |
I'm still undecided, but sort-of leaning towards the T1. It's a bunch of
money, that's pretty obvious. It's possible that if we commit to a long term
contract, then have to bail out, that Grex won't have enough to pay off the
early exit penalties. But I hope it doesn't get to that.
The CPU upgrade is sort-of planned, but since it will require all new memory
cards (128Mb!) we're not just talking about waiting for Sun4/400 cards to hit
$150 (or whatever). We're looking about $1000+ for the CPU upgrade.
|
chelsea
|
|
response 93 of 146:
|
Nov 9 21:30 UTC 1996 |
Backtalk would get a lot of exposure and look a whole
lot better through a T1 connection. Possibly.
So this probably doesn't need saying but if one
of the two donors is Steve Weiss, he should probably
abstain from voting on this issue.
|
mdw
|
|
response 94 of 146:
|
Nov 10 23:27 UTC 1996 |
Steve is not one of the two donors. Have you been talking to kerouac?
|
popcorn
|
|
response 95 of 146:
|
Nov 11 01:53 UTC 1996 |
Re the question of which board members are leaning which way: I have some
reservations and uneasiness, but I'm in favor of getting the T1. I'd like
lots of input from aruba-the-treasurer before making a final decision on this.
After Saturday's walk, people were counting up the leanings of the various
board members. The tally was something like 4 board members in favor, two
ambivalent, and one against getting a T1. My impression is that member
sentiments are divided in about the same proportions.
|
dpc
|
|
response 96 of 146:
|
Nov 11 16:52 UTC 1996 |
M-Net upgraded to 56K service this year. After horrendous startup
problems, the service actually *began* on July 31. We pay about $395
per month for it.
There is no evidence of any upswing in memberships or contributions
because of this improved service. All we are seeing is a money drain,
not only because of the extra expense over our former brain-dead 14.4
service through MSEN, but also because a *lot* of people are now telnetting
in without any restrictions on our 64 ptys. The Board voted to restrict
half of these ptys to paying users in June, but the technical people
haven't figured out a smooth way to implement it.
Soo - people have not chipped in bucks, plus those who used
to pay to get a larger number of dialins are now M-Net "guests," telnetting
in for free.
We hope to reduce our Internet connection costs to less than $200
soon.
Just be adviced that the T1 will probably be a net money drain,
not a "draw" for contributions.
|
aaron
|
|
response 97 of 146:
|
Nov 11 17:01 UTC 1996 |
Of course, the circumstances surrounding the implementation, the subsequent
down periods for M-Net and email, and a certain director who was calling for
M-Net to be shut down, didn't help the situation any.
|
scott
|
|
response 98 of 146:
|
Nov 11 17:09 UTC 1996 |
Also, Grex doesn't have the same type of access tiering that M-net uses.
|
mta
|
|
response 99 of 146:
|
Nov 11 20:24 UTC 1996 |
That's a valid warning, Dave. I don't think m-net's experience is directly
applicable to GREX because the incentives for donations to GREX are different
than the incentives to become a paying patron at m-net...but it is worth
keeping in mind.
|