Minutes of the June 28, 1995 Board Meeting

Jul 2 15:38 UTC 1995

Minutes of the Cyberspace Communications Board Meeting
Wednesday, June 28, 1995    Michigan Union Food Court

Attendees: 6 board members (*) and 3 others.
Valerie Mates <popcorn>  *Chair
Steve Weiss   <srw>      *Secretary
Dan Romanchik <danr>     *Treasurer
John Remmers  <remmers>  *
Mary Remmers  <chelsea>
Rane Curl     <rcurl>    *
Dave Lovelace <davel>
STeve Andre'  <steve>    *
Marc Unangst  <mju>

q. Valerie called the meeting to order at 7:13 PM

w. Treasurer's report (danr)

   We took in $274 in cash from the JCC sale
   We received $150 in donations earmarked for the UPS
   Together with memberships this was a total deposit of $1196
   
e. 501(c)(3) Committee Report (rcurl)

   Rane reports that he has made progress, but that the application is not
   finished yet.
   
r. Computer Rehabilitation Committee (rcurl)

   Rane reported that TS has 4 rehab computers for checkout.
   He is waiting to hear from TS. We have promised a computer
   to a clinic in Pontiac.
   
t. Publicity Committee Report (mta)

   Not given, because Misti was not present.
   
( y. Technical Committee Report (Steve) deferred until later in the meeting )
   
( u. Verification Policy (steve) deferred until later in the meeting )

i. Landlord Report (popcorn)

   Valerie reported that the zoning (use) variance passed.
   This means Grex doesn't have to move again.
   
   We still need to do a little electrical work 
   (a grounding strap and cutback of some insulation)
   
y. Technical Committee Report (Steve)

   ITI donated a Sun 386i. This is a 20 Mhz Sun 386 with
   8 MB ram, and 90 MD HD. It runs unix, and can function as
   a server that does not require much CPU power. It is slower 
   than Grex.
   
   Greg Cronau has the sun boxes in his basement.
   We discussed the importance of getting one of those
   boxes outfitted to be Grex's Sparc replacement.
   
   It is really hot in the dungeon. We discussed fan capacity.
   
   Ever since ICnet reconfigured the IP addresses on their network,
   we have been plagued with routing problems. These stemmed,
   according to ICnet, from a hard route in CICnet for Grex which
   was out of date. Every time the internet link to ICnet broke,
   ICnet's router would throw away the correct route entry for
   Grex, and the CICnet would reseed it with an erroneous one.
   It took manual intervention to fix this.
   
   We are losing 50 to 100 files each time we reboot. 
   
   Dave Martinelli wrote a new update_gloria program.
   
   Jan Wolter wrote "gate" to enhance picospan to word wrap.
   
   The board expressed its appreciation to Dave and Jan for these 
   donated software service efforts.
   
u. Verification Policy (steve)

   STeve reported that this policy will have to wait until next month.
   
o. Switch Internet Carriers (gregc?)
p. WIN Report (srw and others) - these items were discussed together

   Steve Weiss and Valerie Mates met with Arnold Barr and Dave Cahill
   at the last WIN meeting. At that meeting Steve and Arnold described
   the ongoing efforts of HVCN to sign a contract with Isthmus Corporation.
   The contract was expected to be finalized soon, and would provide
   for HVCN to begin testing a system using Isthmus's computers.
   
   Ultimately it calls for establishment of a public access modem pool
   for HVCN independent of WIN, and for public use of the Isthmus/HVCN
   system.
   
   Steve Weiss reported that while representing HVCN Technical Committee
   to Isthmus he had mentioned the importance of WIN, and the desirability
   of establishing an ISDN connection between Grex and HVCN. This would now
   mean an ISDN connection between Grex and Isthmus, and could serve to 
   provide (1) Grex access to a shared modem pool, and (2) Grex access to
   share HVCN's (Isthmus's) internet connection.
   
   Steve believes that Isthmus is interested in supporting WIN, not just
   HVCN, but reported that negotiations were deferred at Isthmus's 
   request, until Isthmus can secure a T1 internet connection. They 
   estimated that it would happen in July.
   
   It was agreed that it would be a good plan to ask each of the internet
   providers in town to provide us with their "best" price for a connection
   using 2 "B" ISDN channels (128K) - including any special discounts they
   might want to make available to a system like Grex which provides its
   services to the public.
   
a. Grex Manual (danr)

   Dan reported that he had finished chapter 1 and started another chapter.
   
s. Grex Newsletter (rcurl)

   The votes have not been counted yet for the newsletter name. When they are
   counted, the top 5 will be voted on again in a runoff election.
   
   3 people have volunteered to work on the newsletter itself.
   They are Misti Anslin (mta) Steve Gibbard (scg) and Steve Ross (hross).
   
   It was agreed that the newletter could appear both in an ASCII and a 
   printed format. The advantage of ASCII is electronic delivery.
   Most would be expected to see the newsletter in a conference, in
   email, or on the web. There are no printing charges for these. One 
   advantage of a printed format is that it can be distributed as publicity
   at local events.

d. New Business

--> JCC Sale

  As reported by the treasurer, we made a significant amount of cash ($274).
  This was despite the fact that the sale was very sparsely attended.
  Apparently there were other sales going on in the area that conflicted
  or occurred the day before. There were about half as many people and
  vendors as usual.
  
  A special "thank-you" to Patti Snyder-Rayle (pegasus) for the donation 
  of books to sell there. These contibuted heavily to the bottom line.
  
  A special "thank-you" to Rob Argy (ajax) for anchoring the Grex table
  for so long.
  
  Valerie repeated her concerns about the difficulties in getting a good
  volunteer to coordinate all the effort required to have a successful
  JCC sale. Valerie proposed replacing Grex's participation in the JCC
  sale with an M-Net-style auction. No decision was made.

--> The AA News has a new computer reporter. The email address
  newsedtr@msen.com was given, but I didn't record any actual names.
  
--> Trainee Progress - the new cfadms are now fully trained. 
  The new party admin is ready to take the reigns when he returns.

f. Final Gavel Banging  (popcorn)
   Occurred at 8:51 PM
   
Respectfully submitted,
Steve Weiss (Secretary)

response 2 of 43:

Jul 4 20:36 UTC 1995

somehow, even as difficult as it might be, "losing files on re-boot"
is exactly the kind of system problem that would REQUIRE (imo) camping
in the damn dungeon until that problem is GONE.
  
I haven't been all that harsh about this LONG TIME system problem, but
it's about time to ADD FUEL to the pilot light priority this "little
problem" creates. 
  
Letting this "little problem" last as long as it has lasted is the
worst possible reputation Grex could create. How long has it been
now, 300-400 days?
  
The only position *I'm* in is to bitch louder, other options having
been severed.

response 4 of 43:

Jul 5 06:11 UTC 1995

  Heh.  If even one staff member could be paid to do things...

response 6 of 43:

Jul 5 15:08 UTC 1995

Re 2: TS: The plan is to move to the Sun 4, in order to fix the disk problem.
Staffers who know more than I do, say that the Sun 4 won't have this problem.
Forcing people to camp out in a hot dungeon with the sun 3 in it won't
help get the Sun 4 running any faster.

response 8 of 43:

Jul 5 17:04 UTC 1995

revert to the architecture wherein the problem did NOT exist. 300-400 days
of +known+ unreliability and +known+, predictable FILE LOSS is unacceptable
almost in the extreme.
  
and, certainly, i've wanted staff around here to get paid for a couple
of years now.
  
paying staff and fixing the disappearing files problem ought not be
coorrelated as was implied above. 
  
this magnitude of problem existing un-patched for such an extreme length
of time would seem to make grex a laughing-stock system quite different
from the direction in which it started.
  
As stated elsewhere, and earlier, grex handles extremes VPoorly.

response 10 of 43:

Jul 6 03:31 UTC 1995

Moving back to the "known good" disk configuration would mean disconnecting
the 1.7GB drive and going back to the two 330MB drives that Grex
started out on when we moved to the Sun-3.  Looking at the disk
space figures, I estimate we could do this by reducing our user base
by 90%.  Will you be the one to volunteer to remove files so we can
do this, TS?

response 12 of 43:

Jul 6 06:07 UTC 1995

Um TS, even if staff does now have access 24 hours a day, staff members have
to sleep sometime.

response 14 of 43:

Jul 6 13:24 UTC 1995

Hm.  While I'll agree that the tone of TS's posts is coming out sounding
very mean and angry, I'd still like to hear what he has to say.  With
enough brainstomring, someone may well think of some forgotten trick that
will fix the file system problem.  If we silence any one person, we lose
their input into finding the solution.

response 16 of 43:

Jul 6 17:51 UTC 1995

I stand behind what I said. TS was not offering constructive ideas, he
was simply complaining. The non-workability of his suggestion should have
been obvious to him if he had simply stopped and *thought* about it for
60 seconds. Instead he choose to just blast away.

As an engineer, I have encountered way too many times, the situation in which
I am struggling with a difficult problem, and then someone else comes along,
aquaints themselves with the merest surface generalities of the problem,
and says "Oh, that's no problem, just do this...." and leaves. I find this
insulting and infuriating. (Even more so when the person in question is
above me in the chain of command and I now have to spend endless hours
unconvincing him about his easy fix, lest it become policy.) I have no problem
with discussing an issue at length with someone I know is really informed
about the problem, and cognizant of all(or most) of the issues. In fact,
this is good design practice to bounce ideas around people to get everyone's
slant on a problem or design. What I *can't* tolerate, is someone who can't
be bothered to get all the facts, and then insists that *they* have the
solution and the rest of us are a bunch of idiots because we didn't see it.

response 18 of 43:

Jul 6 23:18 UTC 1995

  The disk problem is certainly a vexxing one, though I think
people's rudeness has more to do with the humidity than rational
anger :-).  My impression was that after Grex's Move, a couple
staff-folk were going to spend an evening hopefully fixing the
problem.  That plan seems to have been tried or discarded, so
it seemed (at least to me, and probably to TS and others) more
like there was just nothing being done, when in fact there is
work being done (making a Sun 4), it's just a different and more
time-consuming approach.

response 20 of 43:

Jul 7 03:36 UTC 1995

Moving to the Sun-4 (really, the associated move to SunOS 4.1.3_U1,
which is only available for the sun4 architecture) will definitely
fix the disk problem.  I am 100% sure of this.  (I can say this
because I use a sun4 machine with 7GB of disk on a daily basis,
and we have yet to encounter a problem like this.)  Once we can
get the hardware assembled and SunOS loaded, I think this move 
will take 2-3 weeks.  Given that sort of time expenditure, I think
it would be foolish to put any more time into fixing problems on
the Sun-3.  It would probably take at least 1-2 weeks to fix the
problem on the Sun-3, assuming that (a) we had some idea of what was
causing the problem; and (b) we had SunOS source.  Since we have 
neither, I would have to revise my estimate to 3+ weeks.  I don't
think it is useful to ask a staff member to put in 3 weeks of time
to fix a problem that will go away by itself when we upgrade to
the newer version of SunOS.  Yes, it is frustrating to deal with
now, but there is no faster way to fix it.

response 22 of 43:

Jul 7 07:53 UTC 1995

if the machine is "taught" that this 1.7 Gig disk (physically) is
virtually only 900 Meg, the physical size would seem to be
irrelevent.

gregc is simply walking in steve's steps from earlier ... with
a generic " i don't have timme to answer all your questoins/thoughts
that you have collectd from your 30 years of *systems trouble shooting*
and offered for consideration in a public conferencce instead
of the staff conference inside which we +can't+ think about  your thoughts
adn you +can't+ think about ours."

And, beyond that, <<believe it or not>> I only hold grudges against
dishonesty. No one here is (or has been) dishonest. I hold a grudge
aginast ITD, for cause; not agaisnt grex or any board/staff/user
associated thereto.  In case anyone might be confused, VEarly on I
was taught how to recognize and separate useful stuff from unuseful
stuff. And also to separate adn recognize the differences between
time, place, and action(s).


=======dedicated response----
 >
 > TS, what you are doing right now is essentially "engineering in a vacuum".

Well, probably. And what group created the vaccuum for me and others?
Is this engineering, prima facia, unthoughtful, or ill-considered?
 > You don't have all the facts, so your conclusions are worthless.

If you do, why does the problem stil exist 400 days later? It would
seem that *everyones's* conclusions are "worthless."

 > Before
 > you go off making wild-assed accusations like this again,

Please itemize "accusations."

 > why don't you
 > stop and ase ideas hadn't occurred to Steve,
 > mju, popcorn, mdw, srw, and myself. They have, and for reasons that I
 > won't go into right now becuase I just don't feel like spending all the
 > time necasary to type it all in, those ideas were rejected.
 >
 > It's *easy* to play arm-chair quarterback and start saying things like:
 > "Well, why don't you just do XXXXXX? That's no big deal. It should be easy."
 > The point is that when you stop looking at the big warm fuzzy picture, and
 > start getting down to analyzing all the nitty-gritty little details required
 > to actually implement XXXXX, you find out rather quickly that it's NOT easy

 > You don't have all the facts, so your conclusions are worthless.

If you do, why does the problem stil exist 400 days later? It would
seem that *everyones's* conclusions are "worthless."

 > Before
 > you go off making wild-assed accusations like this again,

Please itemize "accusations."

 > why don't you
 > stop and ask yourself if maybe those same ideas hadn't occurred to Steve,
 > mju, popcorn, mdw, srw, and myself.

I presume that something along these lines HAS +more+ then just "occurred"
and to more than just staff. And in my vaccuum, I would also presume
that +many more+ than just Grex-staff has thought about "these things."
At least I would presume so - but it's that "vaccuum" thing again.

 > They have, and for reasons that I
 > won't go into right now becuase I just don't feel like spending all the
 > time necasary to type it all in,

Type in? Why not just clip the pertinent points out of staff.cf? I
would presume that's where the discussion(s) took place.

 > those ideas were rejected.

I can accept that, show the thoughts/reasoning to the unwashed masses.
 >
 > It's *easy* to play arm-chair quarterback and start saying things like:
 > "Well, why don't you just do XXXXXX? That's no big deal. It should be easy."

In a vaccuum it is easey to do lots of stuff, except breathe.

 > The point is that when you stop looking at the big warm fuzzy picture, and
 > start getting down to analyzing all the nitty-gritty little details required
 > to actually implement XXXXX, you find out rather quickly that it's NOT easy
 > and it IS a big deal.
 >

??? "warm and fuzzy" ?? Me? HA! The "picture," ... surely you jest. I
h
can't fantacize a less-warm, less-fuzzy picture than systemic loss
of files "just because."

As for the rst, been there done that already, welcome to the club; it's
tough out here.

 > As an engineer, I have encountered way too many times, the situation in
which
 > I am struggling with a difficult problem, and then someone else comes along,
 > aquaints themselves with the merest surface generalities of the problem,
 > and says "Oh, that's no problem, just do this...." and leaves.

I don't leave.
Further, I play out the "someone else" whether they want to leave or not.

 > I find this
 > insulting and infuriating. (Even more so when the person in question is
 > above me in the chain of command and I now have to spend endless hours
 > unconvincing him about his easy fix, lest it become policy.)

Amen, brother!

 > I have no problem
 > with discussing an issue at length with someone I know is really informed
 > about the problem, and cognizant of all(or most) of the issues. In fact,
 > this is good design practice to bounce ideas around people to get everyone's
 > slant on a problem or design.

gregc and i spent from 10pm to 6-7am one night, after a Grex borg meeting
volunteering on the M-b0x disk-crash, in amazing harmony! I was impressed,
the M-b0x was not. Shit happens. We "watched" some stunningly ignorant
actions as well as a few perceptive moves.

 > What I *can't* tolerate, is someone who can't
 > be bothered to get all the facts, and then insists that *they* have the
 > solution and the rest of us are a bunch of idiots because we didn't see it.
 >

Neither can I although I understand the ^L
 > above me in the chain of command and I now have to spend endless hours
 > unconvincing him about his easy fix, lest it become policy.)

Amen, brother!

 > I have no problem
 > with discussing an issue at length with someone I know is really informed
 > about the problem, and cognizant of all(or most) of the issues. In fact,
 > this is good design practice to bounce ideas around people to get everyone's
 > slant on a problem or design.

gregc and i spent from 10pm to 6-7am one night, after a Grex borg meeting
volunteering on the M-b0x disk-crash, in amazing harmony! I was impressed,
the M-b0x was not. Shit happens. We "watched" some stunningly ignorant
actions as well as a few perceptive moves.

 > What I *can't* tolerate, is someone who can't
 > be bothered to get all the facts, and then insists that *they* have the
 > solution and the rest of us are a bunch of idiots because we didn't see it.
 >

Neither can I although I understand the generic situation rather well.
I ask for the facts, I get a "vaccuum" response.
 
NO CARRIER
ATDP 761-4931
CONNECT 9600/V42

Grex is the Breakfast Food of Champs

New to grex?  Type help at the login prompt

grex login: tsty
  
dammit, i hate that ... let's all praise scroll-back buffers
  
As for "bother," who is bothering whom here? I am being bothered
by the system. How, besides asking questions, can "all the facts" *BE*
gathered? I couldn't possibly claim to "have the solution." I can
claim, along with all the 7857 login lines in /etc/passwd, that there
is no solution +yet+. And that "yet" has encompassed more than a year
in the making. So maybe my probing is 365+ days late, sorry, but the
problem still exists today. Perhaps it's about time for the "unwashed
masses" to start asking some questions. 
  
Where the hell does "the rest of us are a bunch of idiots" come from?
Not from my finger-frothing it doesn't. 

btw, mju, some of us are awake when you are sleeping, and some of
us are asleep when you are awake. That makes for sharing the day.

response 24 of 43:

Jul 7 13:31 UTC 1995

The point *I* was frothing about TS, was your suggestion that, since the 
disk probelem can't be fixed we should simply "revert back to the architecture
in which the problem didn't exist". That can't be done, we have burned that
bridge behind us, it would have been obvious to you if you had thought
about it, and all the information to reach that conclusion was available
to you. In the upgrade to the 2gb disk:
1.) All the partitions on the 2 330meg disks were re-arranged to make them
    larger and more effeciently use those 2 disks. This required 2 10hour
    dyas worth of work. Not only do I not have time to do that again, I am
    not going to put that kind of effort into a project which takes us 
    backwards when there is so much *other* stuff that needs to be done.
    This was all explained in public conferences after the upgrade.
2.) In case you don't remember, we were in a constant, day-in day-out battle
    for disk space before the new disk was installed. We lost as many files
    to disk space exhaustion, as we do now to the disk bug. We made /var/spool
    /var/spool/mail, /bbs, and many other dirs bigger. The system has grown
    since, both in disk usage and number of users. There is no way to take
    those dirs back to their old sizes, any more than it's possible to shove
    a 6 year old into his 3 year old clothes.
3.) And finnally there is /home. Which is currently at 900meg. That should
    be obvious to anyone doing a "df" command. We had 2 330meg disks for the
    whole system. Now /home is at 900meg. That peg just isn't gonna fit in
    it's old hole anymore.

There, I typed it in, are you happy now? The above information was available
to you and should have been obvious.

It's easy to make statements like "The new way doesn't work, so go back to
the old way". Well, the "old way" didn't really work either, it had as many
problems as the new way, actually *more*. You didn't think it out and that's
what pissed me off.