HUMAN-NETS Digest Wednesday, 20 Feb 1985 Volume 8 : Issue 6
Responses to Queries - Computer Ethics Research &
Non-computer user's use of Electronic Mail,
Computers and the Law - MOG-UR update,
Computer Networks - Stargate (3 msgs),
Computers and People - CDs vs Books
Subject: Re: Computer Ethics Research
Date: 09 Feb 85 11:49:54 PST (Sat)
The most important practice I have discovered for supporting good
computer ethics and avoiding computer abuse is maintaining a
cooperative and fairly open social environment, with administrators
and managers in good communication with users and restrictions
minimized and downplayed. I have learned this lesson at both
industrial and academic sites. This is actually no different from
avoiding abuse in other social environments.
On systems where computer management is heavy handed or isolated, some
users become alienated and try to break system security. In general,
the more security is touted, and the more access restrictions are
tightened, the harder some users will try to break them. Many
computer users are very interested in the systems they use. If most
system information is gratuitously off-limits, they will not see any
reason not to try to get at it. They quickly get used to breaking
On more enlightened systems, access restrictions are minimized. Where
restrictions must be applied, the users are notified, the situation
explained, and their cooperation is requested. It is never implied
that the security mechanisms are unbreakable. If anything, it is
implied that they are breakable, but users are requested not to.
Penalties are never alluded to.
Note that this is similar to normal practices of physical security.
If people started putting bank vaults in your office, and warning you
about the heavy penalties of being caught hunting through other
people's desks, etc., it would chill the social atmosphere. Yet with
normal (easily breakable) curtesy locks on private materials, few
people would consider breaking in.
In line with this user oriented viewpoint, the security of personal
files should always be a personal matter, with users able to limit the
access to their own data from other users AND from administrators.
Despite the current lack of legal protection for the privacy of user's
data files, user files that are read protected should be treated as
you would treat letters left in an employee's desk, i.e., as private.
If systems people don't respect a user's privacy, why should a user
respect the privacy of system files?
As part of maintaining an open atmosphere, both personal and system
files should default to being readable by anyone, and users should be
encouraged to browse around in order to learn the system better. This
is typical on most UNIX systems, and is very educational. When
someone accidentally leaves some private matter unprotected, anyone
noticing it is likely to send the owner a note letting them know.
I think that as long as the impregnability of computer security
mechanisms is not touted, it can be a useful exercise to design them
well. Sophisticated users can be invited to try to break them, and to
provide suggestions for their improvement. But they should never be
Date: Saturday, 9 Feb 1985 11:32:18-PST
From: winalski%speed...@decwrl.ARPA (Paul S. Winalski)
Subject: Re: non-computer user's use of Electronic Mail (HNT V8
Subject: number 5)
MCI MAIL allows you to send electronic mail to anybody, whether or not
they have an electronic mail account. If they cannot receive
electronically, or if you elect to send paper mail, MCI MAIL prints
your message ans it is sent by U. S. Snail (but posted locally, so
presumably it will arrive in finite time).
DEC's internal electronic mail system lets you address recipients by
name and site code (using site codes the same as for interoffice paper
mail). Messages to recipients who do not have computer accounts are
printed at the recipient's site and distributed via interoffice paper
Both MCI MAIL and DEC's internal electronic mail system are based on
DEC's Message Router electronic mail product, which uses the draft NBS
standard message format.
Date: Sat, 9 Feb 85 11:47:27 est
From: Larry Kolodney <lkk@mit-eddie>
Subject: MOG-UR update
From: len...@mit-eddie.UUCP (Robert Scott Lenoil)
Subject: The final resolution of the MOG-UR (Tom Tcimpdis)
Date: Fri, 8-Feb-85 22:26:51 EST
I plucked this off of ARPANET's INFO-MICRO bulletin board; spread the
word far and wide - we have won!
PURSUANT TO A TELEPHONE DISCUSSION WITH REGINALD DUNN,
HEAD OF THE CRIMINAL DIVISION OF THE L.A. CITY ATTORNEY'S
OFFICE, I WAS INFORMED THAT THE PROSECUTION BELIEVES IT HAS
INSUFFICIENT EVIDENCE TO CONTINUE THE PROSECUTION OF TOM
TCIMPIDIS, SYSOP OF MOG-UR. THIS DETERMINATION WAS MADE AFTER
I REQUESTED A REVIEW OF THE CASE ON 1/11/85 AFTER THE
DEPARTURE OF CITY ATTORNEY IRA REINER TO BECOME D.A., AND
WHILE THE CITY ATTORNEY'S OFFICE IS BEING RUN BY THE CIVIL
SERVICE STAFF PENDING ELECTION OF A NEW CITY ATTORNEY. MR.
DUNN HAS GIVEN ME HIS WORD THAT THE PEOPLE WILL SEEK DISMISSAL
OF THE CHARGES AGAINST TOM UNDER CALIF. PENAL CODE SECTION
1385, I.E., DISMISSAL IN THE INTERESTS OF JUSTICE. UNDER
CALIFORNIA LAW, SUCH A DISMISSAL IS "WITH PREJUDICE" AND THE
PEOPLE CANNOT REFILE THE CASE SUBSEQUENTLY. TO PUT IT
SUCCINCTLY, A DISMISSAL WILL TERMINATE THE PROSECUTION
AS THE MEMBERS KNOW, THE CITY ATTORNEY'S OFFICE HAS
PREVIOUSLY RENEGED ON REPRESENTATIONS MADE TO ME REGARDING
DISMISSAL OF THE CHARGES....I WISH TO ASSURE EVERYONE THAT I
HAVE KNOWN MR. DUNN FOR 10 YEARS, AND I TRUST HIS WORD
COMPLETELY. IF HE SAYS THE CASE WILL BE DISMISSED, I AM
SATISFIED THAT SUCH AN ACTION WILL OCCUR.
WE WIN. WIN....WIN....WIN....WIN....WIN....MY THANKS TO
EVERYONE WHO CONTRIBUTED TO SUPPORTING TOM AND I IN THE
DEFENSE OF THIS MATTER. I CONSIDER THIS TO BE A MAJOR VICTORY
FOR THE RIGHTS OF FREE SPEECH OVER THE "BIG BROTHER"
MACHINATIONS OF THE PHONE COMPANY.
I WOULD BE GRATEFUL IF YOU WOULD DOWNLOAD THIS MESSAGE
AND PLACE IT ON OTHER SYSTEMS THROUGHOUT THE COUNTRY.....THIS
IS A VERY BIG VICTORY, AND THE BBS AND MODEM COMMUNITIES
SHOULD KNOW ABOUT IT.
AGAIN, THANKS FOR THE SUPPORT. BEST WISHES TO ALL,
ATTORNEY FOR SYSOP TOM TCIMPIDIS
THE CASE OF PEOPLE V. TCIMPIDIS, AKA USE A MODEM, GO TO
JAIL, WAS DISMISSED IN THE "INTERESTS OF JUSTICE" THIS
MORNING, 2/7/85. AS NOTED EARLIER, THIS DISMISSAL IS WITH
PREJUDICE, AND TOM IS NOW FREE OF THE PACTEL SCOURGE. ANOTHER
SMALL STEP FOR SOMETHING RESEMBLING JUSTICE.
Date: Sat 9 Feb 85 12:22:45-PST
From: Ken Laws <La...@SRI-AI.ARPA>
Wed Jan 16 00:58:10 1985
... Let the people who conceived of this know that it
is not appreciated. E-Mail bomb them. Flame them until
they drop. ... Frank Adrian
I can't reply to net.news, so I'll put in my two cents worth on
Frank's paranoia is unjustified. I am a moderator on the Arpanet,
and my AIList is certainly not restricted to just work-related
items. That could change, granted, but so can any network policy.
The people with ultimate responsibility in our society are those
who pay the bills. If Usenetters want a guaranteed-free unmoderated
message stream, they will have to pay for it.
Frank's suggestion of a flame-bomb campaign is counterproductive.
It is exactly such irresponsibility that will make moderation
essential as the net grows. (There are other reasons why moderation
adds value to a message stream, but I won't go into that here.)
Action will have to be taken against any message source that swamps
the system. We cannot allow a few individuals with home computers to
tie up the net with millions of messages of any kind, even
well-intentioned ones such as "Jesus Saves!".
It can be argued that moderation is prior restraint and that offenders
should instead be dealt with through legal sanctions after they have
committed their offenses. A nice theory, but the net is not currently
able to implement such a system. I suggest that a more reasonable
model for net lists is that of the legislative council. There are
small societies where anyone may get up and say his piece at a council
meeting, but as societies become large (or polarized) a need develops
to control who has the floor. The luxury of full communicative
freedom is lost when the listeners have insufficient time to listen
to all that the speakers wish to say; it is then up to a moderator
or chairman to insure that each viewpoint receives a fair share of
the time available.
If all lists must be moderated, power lies in the ability to
select moderators. I am not aware of any current restrictions
on anyone who wishes to become a moderator and compete with
existing lists. Eventually even this priviledge will have to
be restricted, but it's far to early to mourn the death of lists
as we know them.
-- Ken Laws
Date: Sat, 9 Feb 85 19:15 EST
From: De...@MIT-MULTICS.ARPA (Joseph W. Dehn III)
Perhaps someone who knows more than I do about Usenet can answer the
following questions for us:
1) Who are these "backbone" nodes? How did they get to be such?
2) Are all the managers of such nodes agreed on this plan, or only
3) What is to stop other nodes, whose management is in favor of all or
some of the unmoderated discussions, from becoming "backbone" nodes?
4) If the answer to (1) and (3) is "money" (i.e., to pay for machine
capacity and communications facilities), what is so evil about some
nodes deciding that they no longer want to subsidize everyone else?
Date: Sun 10 Feb 85 00:06:13-CST
From: Werner Uhrig <CMP.W...@UTEXAS-20.ARPA>
Subject: Re: [sasw: [bnl art/human-nets]]
it is unfortunate that the article by Frank Adrian was posted here
'out of context' as it's contents is 'totally out of whack', so to say
(and that's saying it mildly). Anyone interested and with access to
USENET should read up on recent messages in net.news.stargate and
net.news.groups (as the topic is kind of "old" already, a lot of
relevant articles may have been expired and deleted on most systems).
I hope, that Lauren or someone else will post a rebuttal here on
HUMAN-NETS, but if not, I will eventually expand on the topic to
Generally, I am not in the habit to simply and briefly make condemning
statements, but too many reasonable people have already spent too much
time and energy on Mr. Adrian's posting, that I try to refrain to add
to all that waste. The topic of STARGATE is most interesting, as is
the future of USENET, in general. To anyone offended by the brevity
of my statements so far, I apologize. Please read beween the lines
and try to understand my reasons.
PS: I have no part in organizing STARGATE or USENET but am but a
simple beneficiary of and participant in discussions on USENET
PSPS: this is NOT an attack on Steven who brought the topic to the
attention of this group (a good idea), nor do I think that
Frank Adrian had any but the best intentions for USENET,
however, something must have set off his "panick-button". But
he probably calmed down quite a bit by now and increased his
knowledge on what is involved with STARGATE and why so many
people found his posting offensive - he certainly had a full
mailbox to help him (-:
hmm, thinking about it, and as a matter of courtesy, I'll send him a
copy of this with an invitation to comment to this group, if he so
desires. Fair enough ??
Date: Sat 9 Feb 85 14:45:07-EST
From: Ken Meltsner <MELTSNER%MIT-C...@MIT-MC.ARPA>
Subject: A great, and dangerous, distribution method!
With recent advances in CD technology, we can predict the time when CD
roms will be commonplace means to store programs. However, how can we
distribute individual pieces of software on disks that store 460MB?
Any new rom costs ~$1000 for the first one, and only $10 for the rest
The solution is simple:
A big computer company or software distributor gets the rights to sell
a huge number of popular packages. They are put onto a disk in
encrypted form. A smart "key" (microprocessor-based decryption key)
is then sold for the use of any package.
The advantages are clear: there are no more distribution costs. A
customer just calls up the distributor, pays a license fee, and
receives a key by overnight mail. All of your software is in place,
and can be sold for a minimum incremental cost. Updates are easy:
every six months (or more often, if necessary) a new CD is sent out.
Corporations can contract for huge numbers of keys, with special
encryption patterns for really proprietary software. Individuals can
get disks with some unencrypted programs or demos of the "locked"
This is all based on the adoption of the ADAPSO hardware "key" system
for copy-protection, but adds in the idea of pre-distribution of
The main danger is that a large company like IBM or DEC or Apple can
lock their customers into their software set. I have a feeling that
smaller producers (i.e. anyone with a less complete line of products
than Microsoft) will have to band together.
1) Cheap distribution of huge amounts of software.
2) Copy protection. (if keys can't be faked -- or the distributor
can have 10000 different patterns to reduce the profit in faking
3) Easy updating methods, and software won't take up your entire
1) Huge anti-competitive possibilities (if you thought bundling was
2) Bug fixes are tougher (patch kits on magnetic media?).
3) Barrier to entry for small firms.
If this idea is original (at least in its combination of ideas) and
anyone out there makes any money on it, buy me a dinner or fly me out
to Hawaii to be a keynote speaker. I'm a grad student -- I work
End of HUMAN-NETS Digest