response 26 of 62:

Nov 21 06:55 UTC 1996

re #22 ...ummm, the speed of the link has not changed that i know of.
  
the *load* on the TinyLink, however, has increased at what seems to be
an exponential rate. while some staff are creating nifty ways to
so new things and load the link more, and some staff are creating
easier and easier ways for doubling the mailstorm, and some staff are
creating more ways to dissipate grexian resources (not inthe immediate
past, but recently enough to create a problem-in-waiting) it could 
appear that the excesses of the grand experiment have caught up with
the tolerance of others on the staff!?

response 28 of 62:

Nov 22 05:14 UTC 1996

I think that TS is saying that we shouldn't have an 
Internet link . . . 

response 30 of 62:

Nov 22 16:10 UTC 1996

No, I think he is saying Steve and I shouldn't have written Backtalk and
Valerie shouldn't have reset the sendmail configuration, and somebody long
ago shouldn't have spent money on something that wasn't important to Grex.
So now other people on staff are pissed off about the load this has placed
on Grex's link.

At least, that's my guess.  It's all hints and allegations, refering to
things and people by vague and inaccurate descriptions instead of calling
them by name.

I could say "If that idiot politician hadn't done that damn fool thing
then things wouldn't be as fucked up in that great nation as they are now."
That's a statement that is made up of so many opinions and so few specific
references that anyone who wanted could either agree or disagree depending
on which action of which politician they think had which effect on which
nation.  However, whatever your interpretation, neither agreeing nor
disagreeing is very productive because you are unlikely to be talking about
the same thing.

response 32 of 62:

Nov 23 22:22 UTC 1996

(I wonder what would happen if we reduced the number of 
pty's by six?)

response 34 of 62:

Nov 24 03:49 UTC 1996

I'm not necessarily arguing for this -- just seeing how others 
feel . . . 

But if we cut down on the number of people who can use Grex, we
will be cutting down on the number of people who can be sending 
email, the number of people who can be using lynx, etc.  The 
telnet sessions themselves aren't what slows Grex down, but the 
things people do over those telnet sessions.  

Now that we have a queueing and Backtalk, people who want to 
use the conferences are always guaranteed that they can do so.  
For once, we finally have the ability to start guaranteeing a
certain speed to conference-goers.  Is this something we want 
to persue?  

response 36 of 62:

Nov 24 07:29 UTC 1996

OK. A 90 day test. Go for it.

response 38 of 62:

Nov 24 13:38 UTC 1996

Not everybody has PPP yet, or even a computer capable of using it.  I'd hope
that we are able to leave some resources open to those who can't yet get to
Backtalk.

response 40 of 62:

Nov 24 21:08 UTC 1996

Backtalk is *not* a possible replacement for telnet/Picospan anytime soon.
Our policies should *not* be aimed at replacement.  We should offer Backtalk
as an alternative.  It has some advantages, but some very signficant
disadvantages.

response 42 of 62:

Nov 25 09:29 UTC 1996

It's not easy to figure out the answer, because it's very much a matter of
opinion.  My opinion is that for the type of user you described, reducing
ptys would speed up response time when you're on Grex a bit (primarily 
during peak usage periods), while partitioning mail to another machine
would speed up response time most of the time, but would make accessing
your mail a little less convenient.  Nobody can give a precise estimate
of how much of a speed improvement to expect from these or other changes;
"big," "medium," or "small" is probably as precise as anyone can get :-).

response 44 of 62:

Nov 25 18:04 UTC 1996

This sounds like what m-net is planning (cutting telnet ports for 
non-members)  You cant increase membership while decreasing usership.  
That equation doesnt work.  You cant change the basic correlary 
relationship between the numbers of users and the numbers of members.  
Any attempt to limit normal use of Grex will not only lower usage but 
will eventually lower the number of members.  So cutting ptys or cutting 
telnet ports or things of that nature are not solutions.  If you cut 
certain types of usage, you end up unintentionally discouraging other 
types of usage.  

Maintaining all levels of current offerings and access and simply adding 
a second computer and partitioning everything (as Jan suggested) is a 
much better idea.

Also I think people need to be encouraged to use the web page and 
Backtalk.  Many dont know it is there.  That is why I suggested that 
when Grex's normal login routine is turned off so it can process mail, 
people should be directed in the motd to try Backtalk.  Only a few would 
so I dont think it would defeat the purpose of having logins turned off. 
 It is a way of getting people to try Backtalk and they will when for a 
brief time they have no other options.

response 46 of 62:

Nov 25 19:42 UTC 1996

Does having 64 pty's mean that there are (potentially) 64 telnet sessions
coming *into* grex?  Given that grex is a Sun, that sounds pretty damn
"generous" to me!  No one could say that cutting pty's from 64 to 48, thus
still allowing 48 simultaneous users to access grex, is "severely limiting"
access to grex.  It's no good allowing more people to simulataneously access
grex if throughput is dog meat.  It *is* a tuning issue, system wide, which
includes CPU performance...

Re: mail vs. bbs'ing:  I use e-mail almost as much as picospan.  As long as
e-mail continues to be *reliable*, I could live with e-mail sacrificing some
"ease of use" in favor of enhancing picospan "ease of use"...

response 48 of 62:

Nov 25 23:32 UTC 1996

Ok, lets try these things.  Grex seems very slow to me anyway, even with my
antique equipment.  You *know* you'll hear my experiences if it gets too bad.
<big grin>