response 26 of 146:

Nov 1 23:34 UTC 1996

        Agreed. A 'fix' might as well be an 'upgrade'.

response 28 of 146:

Nov 2 09:54 UTC 1996

no, mdw (#12), not turn grex into an isp. if this faster&wider access
to off-site activity becomes a reality (i don't know any idea for
dissipating grex that hasn't become reality), at least keep the 
improvements *more* available for current users and current levels
of use. 
  
it seems every time there is an improvement, the current users benefit
for about one week until the rest of the planet spreads the word and
then the benefits are redistributed to the new equilibrium at which
point there is about a (example for argument) 10% net improvement 
from 100% *of* the improvement itself. 
  
there might be a technical coup in showing just how little can be
dissipated across so many but that doesn't go very far in generating
supporters, just user/abusers of grex's good will. 
  
i fully realize that the reports of  'the loaves andthe fishes' would
work whether there were 4000, 40,000 or 40,000,000 faces to feed or logins
to maintain. keep it up and grex *will* become an defacto isp - which
might be the goal. 
  

response 30 of 146:

Nov 2 12:00 UTC 1996

I think m-net's problems go far deeper than that, and that there is not
necessarily anything releveant to learn there.  Grex currently has far
more members, & the outlay we're talking about here is far less.  Even
with the current budget, this is an option we can "nearly" afford.
Having several "super-patrons" just means that instead of cutting things
to the bone, we have an extra margin of safety.  This has always been
the way grex has funded expansion - for instance, the founders made a
similar (verbal) agreement to pay for the phone lines & electricity the
first year, because we knew we would have to demonstrate that we could
be up & running for a year, before we could expect the system to be
self-sustaining.  Universities run very much the same way - ever wonder
why Angell hall has that name?

So far as being an ISP - if we *don't* expand pty's, then we very much
run the risk of turning into an ISP.  We *have* to provide more access
for *everyone* if we provide more IP connectivity.  Our niche should NOT
be to provide "high speed" (ISP-like) speed to the few who are lucky
enough to get on the system, but to provide *some* access to as many
people as we can support.  Ie, we should live up to that name
"public-access", and not even pretend to offer the same "performance" as
a local ISP.  I believe there is a fixed relationship between the # of
login sessions we allow on the system, & the # of members we can expect.
Obviously, it's not quite linear, there are other factors, & even if
modems are ultimately doomed to extinction, there is still some premium
there, but on the whole, I think the relationship is inescapable.  If we
do want to learn anything from m-net, we might note that one of the
things that they did that really cheesed off a lot of their users, was
to restrict telnet ports, & to fool around with the telephone dial-in
sequence.

response 32 of 146:

Nov 2 16:44 UTC 1996

Perhaps the question isn't, "Can Grex, as presently situated, afford this?"

Perhaps the question is, "Can Grex, given developments in the computer world,
survive without doing this?"  Or, "Is it better to keep Grex pretty-much as-
is, or to look forward?"  Some people probably will prefer to keep things
as they are, preferring the present state of Grex to any hypothetical future-
Grex.

Grex should certainly get something in writing, at a minimum indicating
the number of months it can rely upon having this donated connection.

With regard to the Ameritech fees, if there is no penalty for a change in
the contract beyond paying the difference in price, it would best serve
anybody to take the cheapest contract.  If there is any chance of having
to reduce the term of the contract, bank the difference so that the money
is available when needed.  If all goes well, you will end up with five
years of cheap service and a healthy savings account balance.

response 34 of 146:

Nov 2 19:20 UTC 1996

M-net's problem was that Arbornet wanted to run it as a private club.  
Grex and M-net play a role in the Cybercommunity.  They ARE free-nets 
essentially and exsist to foster interaction that will lead to the 
betterment of the community.  They are services.  Being a low cost ISP 
is part of that and that should have always been expected.

M-net's plans to restrict half of the telnet ports for paying members is 
the stupidest idea I've heard in a long time.  They will kill themselves 
trying to be a private club.  Boards such as grex and mnet should exsist 
 as a service to the cybercommunity to the absolute extents that they 
are capable of doing so, and the more people they can serve, the more 
people will be around to support it.  

A private club exsists only for its members.  These are public boards, 
not private clubs.  They exsist for everyone to use.

response 36 of 146:

Nov 3 07:20 UTC 1996

Holding the number of ports constant while going to a T1 would be a good 
way of promoting Backtalk use.  Almost by definition, Backtalk users 
will be conferencing participants, not passive anonymous mail 
consumers.
 
I hate managing the Grex load by allowing performance to deteriorate
until the impatient go away;  I'm one of the impatient.
Multi-second character echoes are a real downer.
 
I would be willing to pledge an extra pittance towards the T1
costs -- an extra membership's worth or two.

response 38 of 146:

Nov 3 10:55 UTC 1996

We certainly haven't done as good a job as we could either to attract
internet memberships, or to get them interested in computer
conferencing.  One of the opportunties we missed early off was to get
the indians more involved.  For instance, we probably should have
created an "india" cf, & picked some of the more active indians to run
it (we had one in here.  Somebody should have nabbed him.)  A "madras"
and "bangalore" cf would seem a natural extension of this.

I believe some of our efforts to make grex "easier to use" have also
backfired.  The amount of text we "dump" on people when they join agora
is definitely, um, "impressive".  I very much doubt that it actually
makes grex easier to use however.  The easiest way to see this is to
watch people in party.  If anyone says something "too long" in party,
one likely reaction from the other readers is "Book!" The same thing
happens in computer conferencing.  Many people, when presented with a
response over 10 lines long (like this one!) will skim it, rather than
reading it in detail.  You people reading this sentence in detail are
not at all representative of the unwashed masses out there, and are very
much "the elite", like it or not.

When I did PicoSpan's help, one of the things I tried to do was to break
it down into little "focused" pieces.  My model was a person asking "how
do I do this"?  I tried to break most pieces down into segments no more
than 24 lines (half of them are no more than 12 lines), and I tried to
organize things such that people would get an idea of all the things
they *could* do, as they were finding the thing they *wanted* to do.
Basically, I believe good online help should be like an adventure game -
lots of way to traverse the rooms to learn how to use the program, and
useful clues along the way.  Unlike an adventure game, you shouldn't
*have* to traverse every room & know every clue to "win" - if you know
enough to do the things you want in the program, no reason to know more.
My fundemental driving ambition was to set things up so that people
would only need to know to type "help" to get started, and the rest
would happen.

I wouldn't say the result is perfect, and there were certainly people
who complained about the help.  But I do think the result did a very
serviceable job of training people to use PicoSpan.  I venture to guess
that most of you old-timers reading this are in fact products of that
help, somewhere in dim antiquity.

This is not what people see today.  Indeed, the default prompt we give
new people today, would all by itself get a "Book!" if entered into
party.  The "welcome" message in agora (uh, oops, the bulletin mesage.
Strange,  not "d welcome"?  Anyways) is all by itself 63 lines.  It
doesn't mention PicoSpan's help once.  It does point people to "intro".
That prints out 230 lines, and at the end, invites people to mail carl
if they have any problems.  I guarantee you carl is quite safe.  The
"menumore" program is nice enough to erase the last screen before it can
be read.  It's a perfectly fine paper introduction (and indeed, that's
what it was originally intended for), but as an online document to teach
people how to use PicoSpan, it, uh, is only suitable for mature
audiences.

Now, I could go on at length about how unfriendly "pico" is to users
over telnet, &etc&etc, but the point is, I'm not at all surprised we are
doing a worse & worse job of getting people into the conferences.  We
haven't made any effort to provide interesting conferences for those
newcomers, a *reason* for them to become involved in grex.  Our efforts
to make grex easier to use, have only resulted in a system that is
actively hostile in its efforts to be friendly.

[You may now say "Book!"]

response 40 of 146:

Nov 3 18:25 UTC 1996

If Grex can become more efficient it will attract and keep more 
members.  Anything that represents an inconvenience drives users 
away.  I still maintain for instance, that the countdown que mdw 
installed was, while a technically sound idea, a very bad one on 
the aesthetic level.  If a prospective new user telnets and is 
number thirty or forty in the que, they are going to go away.

So I think faster service is always desireable, even if this 
means more people using Grex as an ISP.  Grex should concentrate 
on doing everything it can do and currently does better and not 
worry about whether users will use it in less desireable ways.  
Of course they will, thats just a necessary evil of tech 
advancement.

response 42 of 146:

Nov 4 01:35 UTC 1996

When I started using Grex I wasl already a DOS user, and was used to a command
line interface.  .  Marcus's help was quite helpful, nad I pretty much learned
Picospan from it.  OTOH, most of my consulting clients are people who are used
to various GUI environments.  If I ask them to do something from the command
line, they tend to become quite confused, almost as if they've never seen a
CLI before.

response 44 of 146:

Nov 4 09:52 UTC 1996

Last I heard, the *average* US citizen read about one book per year.
Which terrifies me when I realize that my 50-book-a-year habit
translates into 49 other people who don't read any.  >8(((

response 46 of 146:

Nov 4 23:19 UTC 1996

T1.  Should we get a T1.  (Yes, Marcus' and Steve's comments are relevent,
but not directly so, and no connection was made.)  Do people think we should
get a T1 andwhy or why not?  Personally, I'm still undecided.  That's a *lot*
of moneymoney, and a significant portion of grexes budget.  Is it worth it?
I don't know.  Can we afford it?  Again, I don't know.  It depends on
projected membership, which I have no past experience on, and can only guess.
How has changes in connection/CPU helped in the past toward memberships?

response 48 of 146:

Nov 5 00:57 UTC 1996

Well, I would certainly *like* to pledge money, but I don't think I'm in any
position to pledge more than my membership.