Chat Politics

Received: from WEIZMANN by TAMVM1     id 8594; Tue, 26 Feb 85 04:00:31 CST
Received: by WEIZMANN   id 2392; Tue, 26 Feb 85 11:02:14 IST
Date:     Tue, 26 Feb 85  10:51 IST
From:     VSHANK@WEIZMANN
Subject:  Chat server machines
To:       $$$cbfi@nervm, $02@dhdurz2, #013@ddathd21, acsh@uhupvm1,
          acs01043@udacsvm, adi240@dkakfk3, alan@mcgill1, alison@crnlthry,
          amt0@icineca, ansbill@uoguelph, antreasyan@harvhep, armand@rockvax,
          arnold@muvms1, auffret@frmop11, augustin@wuvmd, awcbet@uiamvs,
          azary@mitvma, a19@taunivm, bandini@psuarch, bart@nnomed,
          bates@utdalvm1, bill@crnlcs, bitnet@tamvm1, bitnet@umslvma,
          bjj@psuvms1, borchy@db0tus11, braun@dhdihep1, brian@uoft02,
          bruce@umdb, busch@dearn, butch@nsncc, b04998@anlees, b12431@anlel,
          b17783@anlvm, b25784@anlcmt, b95940@fnalvm, carl@suhep,
          ccbj001@uta3081, ccjcl@bostonu, ccpc@bostciml, ccsa1ob@technion,
          ccsm4nn@technion, cmsmaint@brownvm, cq6adf@irishmvs, crisbh@hlerul2,
          crisrp@hlerul2, crw@psuvms1, csds@biblio31, csyscst@uclavm,
          cvisser@hroeur5, cwhitman@auvm, c0004b@umrvmc, c0033001@dbstu1,
          dae@psuvax1, danny@hujics, david@uchistem, ddj@browncs,
          ddl@uiuchepg, deck@weslyn, delrod@utcvm, dfk@duke, dglasser@yalecs,
          dg9007@sfbsys, diterliz@ibacsata, djg@db0hmi41, dlw@ucbjade,
          dmmartindale@watcgl, dok130@dhddkfz1, earn@imibocco, earn@imisiam,
          earnadm@htikht5, eckcu@cunyvm, ec0n@cmu-cc-te, ed@umuc, ed@yorkuts,
          ed@yugemini, ed@yuursa, ehrhardt@djukfa51, ekonk@dm0mpf11,
          elinsky@yktvmx, endres@ddadvs1, engelke@dkiuni0, eric@watarts,
          fab@crvxdev, farrell@rice, flo@yuorion, fxtbb@bb003, gac2f@uottawa,
          gary@harvarda, gc@bnl, gg.jws@stanford, gg.wsf@isumvs,
          gilbert@yalevm, gonzalo@eearn, greg@fnal, grossen@utkvx1,
          grz027@dbngmd21, grz103@ddagmd11, gschwind@db0tui11, gwb@umd,
          gwd26b@dgogwd01, hantl@dkauni12, hartley@bbadmin, heiss@dkauni13,
          hintz@ucbcmsa, hooper@qucdn, hrz140@de0hrz1a, hstl@dbnuama1,
          hu@cucs20, huberman@fremp11, icsbjb@asuic, ince@watmath,
          ira@pennlrsm, i11c17@dm0tui1s, jacny@ny001, jaf@unbmvs1, jch@clvm,
          jdarling@wiscmacc, jd9014@ucsfcca, jennine@cithex, jerry@nsnccvm,
          jhb@nervm, jjklc@lehman, jm@miteccf1, jm8002@ucsfhc,
          joanne@mcmaster, joel@wisdom, jordan@umass, jpaul@froni51,
          jpm@irishvm, jra@ricecsvm, jxc2f@uottawa, jxh@slacvm,
          kaletka@mitlns, kangaroo@bklyn, karl@sjrlvm1, karl@uscvm,
          kehoe@yalevax5, ken@hamlet, kerry@maine, khaan@asuacad,
          khaat@asuejs, kline@uiucvmd, klocek@yaleads, kutterer@dhnfhs1,
          kyle@vandvms1, k104040@czhrzu1a, lauer@yalehep, ldnqb@qb001,
          lennox@watdcs, lfl@cunyvms1, lhb$@tssnrc00, light@yalastro,
          lindner@dhvrrz01, lingbeek@hwalhl5, lizard@uiucvmc, llopis@mdrvm3,
          maddog@kb001, maint@emdupm11, maint@umab, maint@umcvma, maint@uorvm,
          maintcms@pucc, mar@dm0mpi11, markar@vpivm1, marshal@yorkvm1,
          marty@york, mcdonald@umkcvax1, mci$c3@gen, melanie@crnlgsm,
          messina@anlmcs, mgrnicho@cuccvx, michael@db0tum11, michael@utoronto,
          michau@frors31, mike@bucasa, mikeb@watdcs, morgan@udcvm,
          nbush@sitvxa, neufrode@dkearn, nir@taurus, nitza@yalensl,
          nsm3@uhrcc, ohmdn@cernvm, operator@irmcra, oper2@tamvxrsc,
          oprjff@suvm, oved@techunix, owen@yalevm, owens@umee, owl@ccny,
          p.luck@uor20a, pasch@dhdibm1, pa7280@utkvm1, pc01@itoimgc,
          peter@heithe5, phil@uorvm, phillips@cencol, pmt2f@uottawa,
          p85025@barilan, q0239@pucc, roach.sysmaint@mit-multics, rau@dk0za1,
          rbf@psuvm, rcg@gwuvm, redmond@wiscmse, rens@sunyabva, resmer@vassar,
          reuscher@tamcba, rf@ipiinfn, rick@pennhep1, rick@uogvax2,
          riegner@akron, rimon@israearn, rpcmaint@ifiidg, rpcmaint@ipacuc,
          rpcmaint@irmias, rs08@ddagsi3, russell@browncog, rzs03@dhafeu11,
          rz0@dkafhs1, rz1e@dfvlrop1, rz33@dkauni11, r01moe@dhhdesy3,
          s.boy-boyce-george@crnl20a, salm@dmarum8, sanders@umvma,
          sara@uconnvm, sasaki@harvsc3, sgsct@ttuvm1, sharp@buasta,
          shlomo@techsel, simon1@hbunos, sjyp09@sbhep, smith@umvma,
          snraile@ds0fbd11, sob@harvunxw, sofpjf@uoguelph, sommani@icnucevm,
          spgrjh@uccvma, sp0003@bingvmb, steele@hamlet, stewart@umdnjvm1,
          swb@cornella, sy.ken@cu20b, sy$garla@cuchem, sybarry@cugsbvm,
          syopropr@hmarl5, sys_apv@wiscpsla, sysjcj@uoft01, sysronr@uchivm1,
          sysront@uchivm1, system@crnlns, system@czheth5a, system@jhup,
          system@neuvms, system@wvnvm, systwtz@de0wtz1a, syst8101@ryerson,
          syswork@uordbv, s0004@csuohio, s88y@cbebda3t, teeter@buchmc,
          tfocu@cuvmc, tom@penndrls, trahard@frhec11, trg@watcsg,
          ts0258@ohstvma, tucjfw@tucc, ugasup@uga, unc411@dbnvb12,
          urz02@dmswwu1a, us.travis@cu20b, uworkj@unc, uzr115@dbnrhrz2,
          u001212@hnykun11, u18215@uicvm, u64a0574@jhuvm, vallino@frulm11,
          vax@luccpua, velthuis@hgrrug5, venard@educom, vmsp@ncsuvm,
          vmsp@ncsumae, vmsys3@njecnvm, vm0359@wvnvm, vm4341@earnet,
          vvvcu@cuvmb, wah@dgaipp1s, walsh@irlearn, ward@wiscvm, wgrcu@cunyvm,
          wiegandt@cernvax, willis@ncsuie, wolland@nyspi, wothke@dbnuor1,
          xmaint@haifauvm, yardy@humber, zbzdonm@csu, zdv007@djukfa11,
          zpr@daafht1, zrma0001@dtuzdv1, zrsn@ds0rus1i, zsiscom@eb0ub011,
          zsysdgw@ucccvm1, zsyswjb@ucccmvs, zzz697@db0zib21, 111@uwocc1,
          3crash@utsa4341, 33000001@unfvm, 8rz0209@dczrztu0, 900025@doluni0,
          910002@dosuni
Cc:       oberst@educom, ihfcu@cunyvm

To all,

(Sorry about such a large header but I am only sending to the single
system contact as listed in BITNET NAMES - after having removed
duplications.)

I guess it is time we recognized that CHAT server machines are not
going away and instead will grow in numbers.  Here is a list of the
ones I know about (which may not be complete):

Nodename  Userid    Status
========  ========  ==============
IPIINFN   SPE       Active
WEIZMANN  MCBILL2   In the process of being killed off by me
NCSUVM    ALAN      Being rewritten
BMACADM   BILLY     Recently shutdown by CUNY
PSUVMS1   CHAT      Active
UCHIVM1   FORUM     Active
GWUVM     $G0999    Active
TAMVM1    HELPDESK  Active
WVNVM     VM00D4    Active
WATCSG    WPDIST    Defunct
DS0RUS1I  EAAD      Active

This is double the number that I had known about 6 months ago.

Chat server machines are very different from normal server machines.
In essence, they are message rebroadcasters (similar to a CB system).
But when 20 users are on a CHAT system channel and each user is sending
approximately 4 messages per minute into the server machine, that means
that every minute, the server will have 80 incoming messages and will
have 1520 OUTGOING messages to deliver.  Yes, 1520 outgoing messages
per minute!  Since RSCS gives priority to message buffers ahead of file
buffers, this has a tendency to slow file transfer to a standstill.
The bulk of data being transferred over TP lines becomes a hackers CB
world.  High school students and college undergraduates discuss
everything from dirty jokes to sex to crashing the VM system.  The worst
thing is that a user (or even a systems programmer) has no idea why
file transfer has suddenly slowed down.  Once an RSCS trace is activated,
the problem is determined, but in most cases this is not done even as
a last resort.

Personally, I see nothing wrong with some group of students managing
a server machine for file and digest distribution (example: CSNEWS at
MAINE).  But the CHAT rebroadcasters, present a very large and growing
threat to the BITNET network.

I think all people receiving this mail should examine their system
for CB CHAT systems and inform the author(s) that running such a system
is not allowed.

Henry Nussbacher
Weizmann Institute of Science
Rehovot, Israel

--------------------------------------------------------------------------------

Dear Mr. Nussbacher,
     I recently read your electronic memo addressed to all bitnet contacts
regarding the problems associated with the increasing number of CHAT
machines which are appearing over the entire network. I would like to take
this opportunity to provide you with some feedback. I am a graduating
Senior at Queens College, CUNY and am a communications major although
I have taken a few computer courses. I presently serve as the sole student
representative to the CUNY/UCC Policy committee.
     Although I agree that these chat servers quite probably do slow down
network file transfer, I think your figures are somewhat exaggerated.
First off, 20 users is a rather high example. In the past, the average number
of users on any chat system at any one time does not exceed 15. Secondly,
you state that the average user sends four messages per minute to a chat
server to be rebroadcasted. I suspect this to be, once again, an exaggeration.
I doubt very much that any undergraduate student can type one message
consisting of one to eight words every fifteen seconds, and even if it
were possible, the users are reading the incoming messages and so they are
not constantly typing.
     In addition, you mention a few of the topics which are discussed on
the servers; "dirty jokes to sex to crashing the VM system."  Yes, this
is true. But, you neglected to mention the invaluable assistance that some
users have provided to other users regarding classwork activities, such
as JCL, programming techniques, and brainstorming ideas to further improve
computer usage and education.
     Other typical conversations have included the Bernhard Goetz news story
in New York City, birth control, movie reviews, conventions, career goals,
religion vs. atheism, and much more. These topics are well worth being
discussed by students who have come from completely different backgrounds.
Also, personally, I use the CHAT servers for company on a lonely Saturday
evening and I'm sure that many others do the same. Instead of wasting time
by sleeping or listening to records, I am much more content to logon and
talk to "friends" who also are ill or did not find a date for the evening.
     Let's take the idea of education one step further; do you realize the
number of students around the country who have decided to take a computer
course as a result of hearing about BITNET from friends? I know of quite
a few of these cases at Queens College. Have you ever considered the
benefits such CHAT servers might yield? By allowing students who have
never before spoken to a southerner or a Jew or a citizen from another
country to openly and freely converse, ridiculous prejudices are being
eliminated. BITNET and chat servers are bringing together the populations
of the entire world's educational system. College students possess intiative
and intelligence and are generally eager to help eachother. By bringing
these populations together, it is quite possible that some of the unsolvable
problems of yesterday may be able to be worked on today and lessened by
tomorrow. By generalizing that conversations consist of obscene messages, you
are censoring and undermining any and all productive conversations as well.
     I know that the general concensus of administrators around the country
is to completely restrict BITNET usage from students, but having used
BITNET for the past two years, I must tell you that it has been one of the
greatest pleasures I have experienced in my college career. I have met
a large number of people and have developed, what I consider to be, a
small group of friends hroughtout the country whom I will not lose contact
after I am graduated.
     I urge you to find a solution, WITH the help of all users, so that
BITNET and it's services may continue to positively affect the lives of
the future leaders and computer programmers of our society.
     I also would like to draw your attention to the analogy of television.
The "Boob Tube" has brought products such as Three's Company and other
"non-socially redeeming value" programs, but it has also brought us
Network News shows, Lou Grant, and Cagney and Lacey.
     Should you wish to discuss this further, please feel free to contact me
at MFRQC@CUNYVM. I would certainly like to discuss the matter with you
to a greater extent.
                                              Sincerely,
                                               MARVIN RAAB

cc: CUNY Vice-Chancellor Ira Fuchs  (IHFCU@CUNYVM)

--------------------------------------------------------------------------------

Received: from WEIZMANN by CUNYVM     id 4586; Thu, 28 Feb 85 04:02:48 EST
Received: by WEIZMANN   id 7726; Thu, 28 Feb 85 11:02:31 IST
Date:         Thu, 28 Feb 85 09:32 IST
From:         Henry Nussbacher 
To:           
In-Reply-To:  Your message of 27 February 1985 20:16:37 EST
cc:           Ira Fuchs 

The views expressed below are my own and do not represent the views of
CUNY or BITNET.

Marvin,
Thank you very much for your feedback.  Even though I know that Chatters
throughout the world would like to take a contract out on me (it's
amazing how fast the grapevine is), I am going to stick to my analysis
that Chat systems are very damaging to Bitnet.

Let's back up a step.  My purpose is not to cut off the net to all students
or to make life miserable for the undergraduate.  On the contrary.  One
example - I am presently trying to convince QUCDN of alternate methods
of restricting network use so that students at that node can benefit from
the network.  Presently, they do not allow any students from using Bitnet.

Nussbacher's axiom: "The network should be available to all and that
no one user or group of users should be able to impede the transfer of data
between anyone else".  But when I have a professor here in Weizmann who
is on sabbatical from Case Western Reserve University and who has thesis
students sending their papers to him via the network for correcting and
comes and complains that his files are not getting through and I check
and see a few copies of MONOPOLY and ZORK and ADVENTURE going from Spain
to Germany (via CUNYVM-PSUVM-GWUVM) and then I put a trace on my RSCS
line and see a very large number of message buffers (going to Chat systems
in the United States) preempting the file transfer of legitimate computer
work, I begin to see abuse with a capitol A.

I am not shooting from the hip on this.  These are real problems and the
situation is only getting worse.  You state that only 15 people at once
are ever on Chat.  I am not going to check the ergometrics of how fast
people type but the potential of having hundreds of users on Chat channels
is extremely possible.  If this "service" was announced officially, then
even if only 1% of the logged on users in the network were interested
in using it, then you would have 100 users on (there is usually 10,000 logged
on users currently in Bitnet at any one time).  So your answer would be
don't annnounce it.  What is then the result - a few selected hackers from
around the world benefit from online conversations and only those
fortunate to stumble onto it will benefit (sounds like what is going on
today).

Networks, as computers are run via socialism.  Not the socialism that
censors things and restricts peoples movement, but the socialism that
dictates that *ALL* users are equal and none are more equal than others.
Therefore, if I wish to run a number crunching program and tie up 80%
of the CPU and thereby lock out other users from doing productive work,
the operating system will act in a socialist manner and restrict me from
hurting other users.  Unfortunately, RSCS doesn't have the concept of
'network socialism' (yet) and therefore it is up to individuals to police
themselves and others so that all users can get a fair share of the network
resources available.

Let me show you a typical reaction:

> I think there is a very serious threat amongst us: vshank@Weizmann.
> i've heard of his proposal to shut down all chat machines; it is men
> like these that given the chance would legislate morality, censor the
> press, etc.  His views worry me. I use the chat lines to communicate
> ideas...not to "tell dirty jokes" as he proposes.  The chat lines have
> given me contacts such as yourself, who, nyguy, and many other
> knowledgeable people.  People such as these advance the field of
> computing...not stifle it as vshank would have.

This is typical of the reactions of students who see themselves in a
microcosm of their own.  They do not even begin to understand that what
they are doing is hurting someone else.  Why should they!  VM encourages
the perception that you are alone on the computer and when you type the
computer answers.  Each virtual machine is an entity onto itself and
doesn't have to worry about any other virtual machine.  But that is not
true and is a view of an uninformed person.  If I request a tape drive and
use it for 3 hours, then no one else will be able to use that tape drive.
I have denied a resource to other users.  The same is true for a network
resource.  Sending a very large file (like Adventure or Zork) ties up
the network for the exclusive use of that user.  The same holds true for
Chat.  It is just that many 'little' pieces add up to one very big piece;
the individual cannot see how his 3 or 4 messages a minute can be
damaging the network.

You yourself admit that it is true that discussions in Chat are not always
computer related.  Sometimes they discuss birth control, movies, religion,
etc.  I don't doubt for a minute that Chat serves some useful purpose but
what you are looking for is a computer conferencing system - not a CB system.
Have patience!  I am sure that CUNY/BITNIC/EDUCOM will announce a computer
conferencing system at some future date but I would not want them to
announce a system that was as damaging to the network as Chat has been.
Patience.  This conferencing system will then give you what you envision:

> eliminated. BITNET and chat servers are bringing together the populations
> of the entire world's educational system.


Henry Nussbacher
Visiting Scientist
Weizmann Institute of Science
Rehovot, Israel

--------------------------------------------------------------------------------

Date: 02/27/85 10:05:54 EDT
From: TS0013 at OHSTVMA
To: VSHANK at WEIZMANN
Cc: TS0258 at OHSTVMA
Subject: CHAT server machines

The problems you point out about chat systems are very real.  There
are answers to these problems that should be brought up.

First I want to add my own first hand observations about these systems.
I have used these chat systems many times, especially the one located
at PSUVMS1.  Many topics were discussed using these systems.  The ones
you pointed out were discussed, but were only a minority.  I have had
discussions with other systems programmers about systems programming
topics.  I have met many people through these systems in a way that
mere electronic mail cannot allow.  I now know some other systems
programmers more than I ever would have without this kind of system
being available.  There is a great potential for this kind of system
in a computer network.

The very first question I thought of when I first encountered one of
these systems was the message load.  The messages were being MULTIPLIED
by the server and being re-sent.  The load was a function of the SQUARE
of the number of users on a single chat channel.  Initially the network
seemed to handle the load pretty well, but eventually some slowdown was
noticed.  At one point, the first major chat system placed restrictions
on the number of users of a channel and this helped limit the load it
placed on the network, but it did not eliminate the problem.  Once this
first system, the one at PSUVMS1, was shut off, new and uncontrolled
"pirate" systems cropped up at such places as BMACADM, GWUVM, and TAMVM1.

There are two major problems with these systems that I see:
(1)   The load from a multiplied number of messages can overload the
      network very quickly as the number of users expands.
(2)   Unauthorized users are using excessive resources of the network
      and the various systems in the network.

The solution to the second problem is to implement such a chat, no let's
say CONVERSATION system, that provides means to control access.  There
was an attempt at BITNIC to implement such a system, but it was cancelled
without an explanation.  I suspect it was cancelled because someone at
CUNY confused it with the kind of systems that have been running thus far.
Alternatively, they may have considered the loading problem without having
attempted to address a solution.

The solution to the first, and major, problem is to eliminate the
redundancy of the number of messages being sent.  Once a single user
sends the conversation system a message, that message is duplicated
and sent out to all the other intended recipients.  Since RSCS does
not have the capability to send messages with a list of recipients,
then it is necessary to send a separate message to each users, causing
RSCS to send the text of same message many, many times over.
To eliminate this redundancy requires some capability to send a single
message over one link and have that message redistributed as needed.
This requires having a server machine running at each participating
node and also at each major trunc node.

These servers will consititute a coordinated conversation system.
The users of the system will send message to their LOCAL server only
and must pass through access controls established by their installation.
The servers must reject all messages send from outside their local
installation.  A single server could be usable on one node at a multi-node
installation.  An example for Ohio State is to have the server running on
OHSTVMA but allow users on OHSTVMB to make direct use of it.

These servers will pass on the messages to the servers at the adjacent
participating node only.  Each server will keep a record of the local
users who are to receive messages for a particular channel as well as
what adjacent servers need copies of messages for that channel as well.
Thus, if a channel were to be in use only by a segment of the network,
then the messages would not be needlessly routed all around; they would
stay only within the domain of the network where the activity is taking
place.

I think that if these two problems are properly solved, then Bitnet can
benefit from the advantages of a conversation system.  Additionally, such
a system could provide other advantages.  Problems still exist in the
distribution of mail and other files through Bitnet to multiple users.
It is still necessary to transmit duplicates of each file from the
originating node.  This conversation system could be generalized as
a distribution system and transmit files in much the same way as messages
would be distributed for a conversation system.

One of the things I noticed in the note was that college undergraduates
are being compared to high school students.  I have seen many graduate
students, faculty, operators, and programmers using the chat systems.
Graduate students were once undergraduate students.  Undergraduate
students who are the innovative and creative types, who are our future
graduate students, teachers, systems programmers, and researchers, are
having to put up with enough troubles trying to work through a boring,
uninteresting, and highly discouraging program of undergraduate study
just to get to graduate school where they can start learning something,
without having to put up with people categorizing and labeling them as
high school students.  I am one of these.  The students who used the
chat systems I have seen have shown more motivation to learn than the
average undergraduate.  They are out there trying to learn.  If this
were not so they would be off to some bars somewhere getting drunk,
telling dirty jokes and talking about sex.  I am personally offended
by these remarks.

The students at MAINE who run CSNEWS are recognized as being capable.
Students who write chat systems and run them are not being recognized
for their talents, just because problems are occurring that they have
no opportunity to observe.  This talent should be directed to more
constructive deeds such as solving the problems at hand.

                                    Phil Howard
                                    
                                    Systems programmer, IRCC/OSU