response 51 of 137:

Jul 15 15:40 UTC 1995

And I think the thing you, and everybody else, keep missing, is the fact
that Grex doesn't have the $$$ to migrate to a 486 platform plust buy
the OS it would require.

response 53 of 137:

Jul 15 17:42 UTC 1995

Er, if Grex is such a dinosaur, then why are there so many other
Internet sites that use Sun hardware?  Sun claims in their 
advertisements that their hardware is what the majority (i.e., over
50%) of ISPs use.  This should not be construed as an official
endorsement, but where I work (Merit Network) we have two kinds
of computers: SPARCstations, and Macs.  There are a few PCs
around running NetBSD and Linux, but not more than four or five.
(This, in a place where the computer:person ratio is higher than 1:1.)
CMU uses SPARCstations (along with a few old DECstation machines)
for its fileservers.

Carson, if you're so smart that you can see through all this,
then you must also know that it's sort of a catch-22.  You claim
that Grex doesn't have the money to upgrade because it's running
on old hardware; well, we aren't going to get new hardware until
we get the money to upgrade.

I should also point out that it looks like the Sun-4 project is
going to finally get off the ground -- Greg has told me that it
looks like he'll have some time this Wednesday, so hopefully
in another week or so we will be able to start building software
on the Sun-4.

response 55 of 137:

Jul 15 20:22 UTC 1995

I don't deny that Suns are better than PCs, if you have the money to spend.
If you are running on a shoestring, you are better off in the PC line than
in the Sun line.  Making that transition could be difficult, but I think if
you undertook a fund raiser it would be easily accomplished.

response 57 of 137:

Jul 15 23:33 UTC 1995

        Hey, an HP workstation would be nice.  Why not stick with the Sun 3
and save up for a $50-100,000 HP workstation?  Can't afford it now, but
why assume Grex will never be able to?

response 59 of 137:

Jul 16 02:58 UTC 1995

        But you said an upgrade to a Pentium was impractical because of the
cost of replacing all of the memory.  An upgrade to a Sun 2/5/10 or
whatever isn't, even though the other hardware will cost more, and (re:
Greg above) you'll need as much new memory for it?  Sun memory certainly
will not cost less than PC memory.  It might cost the same, or it might
cost more.
        Mind you, I don't care if Grex runs on a Babbage machine with card
readers and Dixie cup and string networking, if it runs reasonably fast
and doesn't have unbearable amounts of problems.  I readily admit that all
I know is PC hardware.  Even if Grex moves to a PC, I am unlikely to
become a staffer.  However, the problems Grex has are worse than the
hypothetical problems Marcus mentioned (which I've never seen or heard of
anywhere else).
        I've been hearing for many months that Grex was moving to a Sun-4 
which would solve all the speed and hardware and reliability problems, but 
it hasn't happened yet.  After a while, you can begin to understand why 
Mary Remmers is asking how long Grex is going to depend on the efforts and 
time of the one person who can set up the Sun-4 (who appears to be Greg), 
rather than going with something that any random techie can set up.

response 61 of 137:

Jul 16 03:09 UTC 1995

Jan asked why we should worry about "fiendishly clever crackers".  He's
right in one sense; it's unlikely the sort of person who has the time to
reverse engineer "appendix H" is likely to care or even know about grex.
Even so, it would be misleading to conclude all crackers are stupid.
M-net and grex have always had their share of mis-directed bright people
who go and spend time looking for holes in things; I know Jan has spent
much time fixing problems in party; and I know here on grex (and I'm
sure on m-net as well) there are people who look for, and will find,
botched file permissions and holes in newuser, if any exist that would
give them root.  But certainly, the majority of would-be crackers today
seem to have come out of the "McDonald" school of cooking.  Periodically
one of them asks staff for help on compiling and running their programs.
The brighter crowd has always been with us, and is just as numerous as
ever; they're a tiny fraction of today's cracker crowd, only because
rather simple tricks still work on too many systems on the internet.

That's where the real danger lies; because as other systems on the net
get more secure, the crackers that make the cut and still pose a menance
will become progressively "smarter".  The real issue here, too, is how
smart the guys making the tools are; not the guys using the tools; even
with today's tools, it's very probable the guy who came up with the sun
password race program hadn't the least interest in breaking into grex.
I'm not worried about the security of the intel architecture against
today's cookbook guys, but whether the even more complex and vulnerable
intel architecture of tomorrow can stand up against tomorrow's more
sophisicated cracker.  One of the factors that fuels that risk, of
course, is the explosive growth of ISP's offering shell access,
frequently on intel hardware.  Today's virus writer could become
tomorrow's ISP cracker; and we might even see ISP's surreptitiously
funding attempts to compromise each other's reliability.

So far as "on a shoe-string"; I have to disagree; the SUN is still far
more attractive than intel.  Cheap PC's are cheap; but by the time you
plop down for serial ports, RAM, disk, tape, & quality hardware, it's
not really all that cheap.  There is also far stronger demand in the PC
world so used machines aren't nearly the good deal they are in the Sun
world.  The downside is that SUN requires more expertise.

response 63 of 137:

Jul 16 11:15 UTC 1995

Jep, my comments were not directed at any one staff person.  They
were directed at the problem of too few staff for Grex's needs.
There are a number of ways to approach such a state:
   1. Move Grex into a position where it needs fewer (or at least
      not more) techies.
   2. Recruit more staff.
   3. Expect more from our current pool of volunteer staff.
   4. Hope everything stablizes and the above three options 
      are no longer needed prior to our staff burning out.
   5. Deny there is a problem.

It's a hard one.

response 65 of 137:

Jul 17 03:33 UTC 1995

Here are two excerpts from the Jargon file that some up the problems
inherent in that (false)idea:

 :Brooks's Law: prov.  "Adding manpower to a late software
   project makes it later" -- a result of the fact that the expected
   advantage from splitting work among N programmers is
   O(N) (that is, proportional to N), but the complexity
   and communications cost associated with coordinating and then
   merging their work is O(N^2) (that is, proportional to the
   square of N).  The quote is from Fred Brooks, a manager of
   IBM's OS/360 project and author of "The Mythical Man-Month"
   (Addison-Wesley, 1975, ISBN 0-201-00650-2), an excellent early book
   on software engineering.  The myth in question has been most
   tersely expressed as "Programmer time is fungible" and Brooks
   established conclusively that it is not.  Hackers have never
   forgotten his advice; too often, {management} still does.  See
   also {creationism}, {second-system effect}, {optimism}.

 :gang bang: n.  The use of large numbers of loosely coupled
   programmers in an attempt to wedge a great many features into a
   product in a short time.  Though there have been memorable gang
   bangs (e.g., that over-the-weekend assembler port mentioned in
   Steven Levy's "Hackers"), most are perpetrated by large
   companies trying to meet deadlines; the inevitable result is
   enormous buggy masses of code entirely lacking in
   {orthogonal}ity.  When market-driven managers make a list of all
   the features the competition has and assign one programmer to
   implement each, the probability of maintaining a coherent (or even
   functional) design goes infinitesimal.  See also {firefighting},
   {Mongolian Hordes technique}, {Conway's Law}.

response 67 of 137:

Jul 18 02:29 UTC 1995

Remotely experimental?
 
we have an internet connection, don't we?

we allow absolutely anyone to call into or telnet to the system, and sign up
and use the system all they want, with no time limits, and no two week (or
more) validation period, right?

I am not dismissing your reasons for why "you" are losing support, but I am
disagreeing with them.

response 69 of 137:

Jul 18 23:55 UTC 1995

  I think a partial solution to sidhe's complaints would be for
Grex to live up to its open access idealism.  As long as paid members
have different access than non-paying members, there is a two tiered
system.  I dont think anyone buys memberships for grex's crappy
approximation of net access, there are faster and better ways to get
that anyway.  So open up gopher access for everyone.  The people 
who buy memberships will still do so because they care about grex.
That should be the only reason to buy a membership.
  Every time this is brought up, the cry is "oh we cant offer
anonymous net access"  And that is ludicrous.  I can run newuser
under a bogus name, send grex a cashier's check for a one year
membership and I'm anonymous as hell.  Part of what makes grex special
is that it does offer anonmous access, so I think grex should simply
stop the hypocrisy of saying it cant offer the whole deal.  Those of us
who telnet in dont need the access anyway but its the point of the matter. Its
whether grex is true to its promise of being completely open or it isnt.
   It probably doesnt even occur to nonmembers that a sure fire reason
there hasnt been any rush to restore usenet access is that members already
have it.  If you can use outbound services from grex, trn and tin are 
still there.  Im saying that for grex to truly be what it claims to be,
if it really wants to grow, it cant stifle access and it cant be anything
less than user friendly.
    The quality and accessiveness of grex will make it grow, will make people
want to be members.  
    We cant treat grex like  a private club.  Open it up. Live dangerously!

response 71 of 137:

Jul 19 01:32 UTC 1995

Actually, you would still need to be "verified" (see the "Verification
Dangers" item) before you could be a member.  One user has tried to be an
anonymous, paid member, even sent in money, but the money had to be returned
because the user refused to be verified, which means giving ID (or some
equivalent).

> "... Open it up.  Live dangerously."
Heh.  what do you suppose an open newuser program means?  Ask STeve or Marcus
or Valerie or any other long-time staffer about "living dangerously", and
you'll find out we already do.

response 73 of 137:

Jul 19 02:30 UTC 1995

  Put it this way....sure hackers can abuse http and telnet and gopher,
but is it any more dangerous than what we already have.  How much 
stolen code passes through grex in any given month via anonymous
email accounts?  The email access is probably, not even probably it is, 
more valuable than any extra telnet access to hackers.  And unless
you live in ann arbor or can afford long distance, you're going to need
telnet to get here anyway.  
  I am saying that unless you want to take out the current anonymous
access that grex already offers, it is hypocritical to limit it at all
at least from a philosophical perspective.  But that isnt the way
mnet does it and god forbid we should do anything but copy them right?
  Grex is reliant upon a small group of users in ann arbor who keep up
the system maintenance.  If any of you guys leave or move, grex might
die.  Because it is not growing.  It cant grow if there is no place for
it to grow.  It needs new members and bs and hypocrisy dont attract them!