+++ plato iv group notes +++

Public Notes

notes beginning feb 19, 1976

file pbnotes5 printed at 2:51 pm on june 12, 1976

---------- note 0 seminar
  02/19/76     11.20     golden       s

PLATO SEMINAR

Pual Tenczar
Head, System Software, CERL

Title: An Account of how the PLATO/TUTOR
       Man-machine Language Works

Wednesday, February 25
Room 198, CSL, 3:00 pm

Abstract:
--how the structure of TUTOR came about
--how TUTOR statements are changed into binaries
--how PLATO unites a person and a binary
--how to strike a balance between the constraints of space
  and time
--how a new statement is added to TUTOR
--playing with constraints

---------- response 1
  02/19/76     11.51     maggs       law

Will anyone be taping or taking notes on this and other
Wendesday seminars for the benefit of those like myself
who have a class on Wednesday at three?

---------- response 2
  02/19/76     12.44     marty smith       orl

or for those, because of their home locale, can't make it?

---------- response 3
  02/19/76     13.54     fritz       ames

Or, for those of us who will be coming in two days later?

---------- response 4
  02/19/76     13.58     b sherwood       s

The software group is attempting to write a paper on
the basic aspects of the condensor and executor, which





would cover some elements of Paul's talk. In fact, we
are hoping that his preparation of the talk will help him
write the paper! So watch a few months from now for
an expose¬e of such PLATO machinery....

---------- response 5
  02/19/76     16.44     simons       cs196

      i am free at this time, and would be willing to
      tape the lecture (if there are no complaints
      about it). anyone wanting to listen to it,
      just send a pnote asking to borrow the cassete
      (i'll assume you have a cassete recorder).

---------- response 6
  02/19/76     17.37     mike b       cornell

      It's things like this that make me sorry I'm on
a remote site. Sigh....

M¬ ike≤≤≤≤٧B

---------- response 7
  02/20/76     00.47     kent       unl

      ...I think this one should be filmed or video-taped
and made available for distribution to all of us PLATO-
freaks our here in the hinterlands!

---------- response 8
  02/20/76     08.59     bonetti       ced

I'll second that. A video tape would be much better than
just a cassette tape recording. How about it?

---------- response 9
  02/20/76     11.05     berger       mfl

Isn't that getting just a little bit ridiculous?
After all, they're not putting on a variety show,
and how do you all expect to have the same format
video tape?

(I want mine compatible with an IVC 800 series)

---------- response 10
  02/20/76     23.03     kent       unl

      Well, film it then, for historical and archival
purposes at least!

---------- note 9  exec. err
  02/19/76     13.49     errol       kka





Got an execution error when trying to look at
module design for course kb111gc
lesson: mredit
unit: modsho.....regular state...last command, showt
12th one........error type: index err., join seq. ddisp

Please help.

---------- response 1
  02/19/76     13.59     b sherwood       s

Will look into right away.

---------- response 2
  02/19/76     15.50     b sherwood       s

Fixed.

---------- response 3
  02/19/76     14.39     wickham       unidel

I need an animated drawing of a beating heart.

repond here or to wickham of unidel
Chas.

---------- response 1
  02/19/76     15.01     marty smith       mtc

lesson -fay2- has a beating heart. don't know who to
contact, though.

---------- response 2
  02/19/76     15.09     boggs       biocc

Just what the doctor ordered:

A beating heart, complete with pulsating slide selector,
resides in lesson "heart" (the same animation as in "fay2").

Contact Mits Yamada, "yamada" of "mxc" for more information.

---------- note 17 WHERE
  02/19/76     15.07     oi       cerl

Does PLATO have a terminal located in the
Rochester, N.Y. or Buffalo, N.Y. area ????

---------- response 1
  02/19/76     17.40     mike b       cornell

        This note is coming to you from a PLATO terminal
at Cornell University in Ithaca, New York. Lesson "network"
lists PLATO terminals in Albany and New York City. See





lesson "network" for details.

M¬ ike≤≤≤≤٧B

---------- note 19  vtoa lib!
  02/19/76     15.15     berger       mfl

Has any progress been made regarding the "vtoa" command
proposed some time ago? A systems response said that
this was definitely going to be done sometime.

---------- response 1
  02/20/76     08.51     fay       o

PLEASE! Among other things, such a command would GREATLY
simplify the use of the copy key with floating numbers
at an arrow!

---------- response 2
  02/20/76     11.05     berger       mfl

Not to mention formatting things for printer output.

---------- response 3
  02/20/76     18.33     blomme       s

   This has most recently been discussed in the more
general form of permitting embedding in a pack command.
However it is done finally, it nearly certainly will not
be by adding more and more commands of the itoa type.

---------- response 4
  02/20/76     22.10     dave       infe

Great!
The -pack- extension sounds like the best possible
way.

---------- note 23  ameschange
  02/19/76     16.37     daleske       ames

There will be a redistribution of courses and signons
at the ames site within the next two weeks. Many authors
presently under course ames will be assigned to a course
which is more closely associated with the area in which
the individual authors is working. A new course, amesrad,
is being created for the ames authors working in research
and development of software.

A lesson space will be set up and announced soon that will
detail the changes to help you find the signon of the
relocated authors.

By Direction





John

---------- note 24 anscnt→0
  02/19/76     16.43     michael       english

For heavy concept users:

arrow
join    concept1('group 1')
ok
judge   again
join    concept2('group 2')
ok
judge   again
join    concept3('group 3')
endarrow
write   group=¬0a,group¬1;anscnt=¬0t,anscnt¬1

The -judge again- would retain the -specs- , -bump-s, and
-put-s, like -judge continue-, but would reset *anscnt*
to zero. Thus it would be possible to maintain independent
concept lists to which entries could be added without throw-
ing off the relative addresses of all later concepts. The
effect can be gotten by juggling a parallel counter with the
value of *anscnt*, decrementing after the -join-: messy.

---------- response 1
  02/19/76     17.26     tenczar       s

better still...

arrow   1010
join    concepts1
anscnt  0 $$ or any value of n
join    concepts2
anscnt  0
etc.

which would save the ok,judge,etc structure...
will put on my list...but will be some time...next job I
intend to do is to "float" the vocabulary out of particular
lessons so that it becomes like charsets,micros,etc. and
thus just one copy is present in ECS no matter how many
lessons use it

---------- response 2
  02/19/76     18.33     michael       english

Thanks, Paul!

---------- response 3
  02/20/76     03.40     daleske       amesrad

Seems to me, the project you have on your list is the
one that Gary was saying would be the thing he most
wanted to see!!! Gee, now sysprogs are reading our
minds! A sincere thankyou....
John





---------- response 4
  02/20/76     09.05     silver       ve

Yes, indeed!

---------- note 31  *unit
  02/19/76     18.55     buz       phar

It sure might be nice if..................dept't.

Would it be feasible to have a command -*unit____tag-
which would give the effect of having a dummy unit "tag"
in the lesson. This would be useful when you're editing
a lesson and want to remove some time consuming displays
and/or animations. The *unit command would keep track
of the unit name so the 'unit XXX not found' messages
would stay in their caves, but would not execute the unit
itself.
This article would be most useful in cases where the units
to be *ed are interspersed with other needed units and are
so many that the cstop-cstart combinations would be a pain.

Just a thought                                   *buz*

---------- response 1
  02/19/76     19.01     frankel       p

Presumably these "dummy units" are do-ed or join-ed; why
not just insert a "goto    q" right after the unit command
(you could mark it with a comment so you could easily
find it when your testing is done)?

---------- response 2
  02/19/76     21.03     buz       phar

It wouldn't necessarily be a do-ed or join-ed unit...some
might be regular units done in succession...
*
*
unit    a
        write explanation
inhibit erase
*unit    b
        some graphical additions to expl
inhibit erase
*unit    c
        mark up of points on graph
unit    d
        next bunch of stuff
*
*
where units b and c take a lot of display time, and just
a and d are the ones to be modified

---------- response 3
  02/19/76     21.28     mike b       cornell





        That would be _very_ dangerous, unless you were
extremely careful about which units were *ed out. So star-
ing out working units should probably be avoided. I have
suggested that you be able to specify at condense time
whether you want to see errors and/or warnings. The initial
response to this suggestion was favorable, but apparently
no work has been done on it as yet. Someday, maybe....

M¬ ike≤≤≤≤٧B

---------- response 4
  02/20/76     10.23     judy       pso

*-ing out working units (or portions thereof) can be
dangerous, TRUE! However, I find that "signing" my
*-ed lines workes quite well:

*do     bigone    $$,jns

When I want to put everything back together, I just do a
search for "jns" to find all those lines.

---------- note 33 -showa-
  02/19/76     19.24     dave       cornell

    Does anyone know of any limits on the length argument
of the -showa- command? When I do a -showa- with a
length greater than 2772 characters, the number of
characters actually shown isn't predictable. I have
tried using several consecutive embedded -showa-s but
the results are no better. I would appreciate any
advice as to any other way that 4224 chars could be
displayed at one.
Dave T.

---------- response 1
  02/20/76     08.29     avner       s

As a point of curiosity, since there are only 2048 character
spaces available on a single display, why do you want to
display that many?

---------- response 2
  02/20/76     09.11     parrello       uimatha

Well, for one thing, the number of character spaces is
usually a lot smaller than the number of 6-bit character
codes, as anyone who has tried to write a line of caps
in lesson "pad" knows.
I found when I had this problem that putting a -catchup-
before the -showa- (and AFTER the -at-) helped a lot.

---------- note 34 L lang
  02/19/76     19.35     dave        nfe1





        Does anyone know what's become of the L language?
It used to reside in lesson L back when such things could
exist (I think).

Dave

---------- response 1
  02/19/76     21.24     nate       iumusic

   The lesson still exists, but because 1-letter things
aren't liked, the name got changed to -lister-.

Nate

---------- response 2
  02/20/76     22.20     dave       infe

        Thanks for the response, but...
        Lister may have been 'l' in '74, but something
else was more recently (I think)...
        What I'm looking for is a language that manipulates
'strings' of bits (actually stacks, since only the rightmost
bit is accessable). It was on system in the last few months

Dave

---------- response 3
  02/22/76     01.54     trueblood       unidel

Unless I'm mistaken, shapiro of iucs might well be able
to tell you . . . if not. you might try writing to
Dr. Mitchell Wand, Comp. Sci. Dept, Lindley Hall,
Indiana University, Bloomington, IN 47401, who (I'm
pretty sure) supervised its writing by an author whose
name I don't remember.
jjt
P.S. If you try the latter, phrase your letter to
cover the possibility that I'm mistake, please...

---------- response 4
  02/25/76     22.17     broadus       css

As I remember from way back when, the name is languagel

Bob, Broadus

---------- note 37 befudgedle
  02/19/76     21.02     hody       med

i am really not sure if this should be here or in repair-
but i dialed up just now and ended up in someone else's
records... i determined who it was by sending "myself"
a p-note and then sent my alter-ego a note describing the
situation. perhaps (s)he forgot to sign off ...?





i didn't realize this could happen...
*
if anyone from the staff is interested in further details
or name/course, please let me know by p-note.
*
ps: this is my REAL sign on!

---------- response 1
  02/19/76     21.09     woolley       p

This will happen if a person using a dial-up hangs
up without signing out.

---------- response 2
  02/19/76     21.34     blomme       s

   Pressing SHIFT-i on the AUTHOR MODE display is
a simple way to find out the name and course of the
records.

---------- response 3
  02/20/76     09.54     hody       med

thank you. perhaps someday you might want to add
something which presses shift-STOP twice for terminals
coming in from dial up?

---------- note 39  $or$??
  02/19/76     21.03     mcneil a       uicc

why is [1$or$0]=-1??????? AL9000

---------- response 1
  02/19/76     21.12     buz       phar

It seems as if the true/false judger assumes if x¬=0 (false)
then x=-1 (true)... the $or$ operator is a true/false or
not a bit or (a$jnion$b). (0$or$1) ¬=0 therefore it's =-1.

*buz*

---------- response 2
  02/19/76     21.32     blomme       s

   Most safely you should simply assume that the result of
applying a truth value operator to non truth values is
undefined--our logic here is still of the primitive
two valued kind!

---------- response 3
  02/20/76     10.29     judy       pso

To rephrase Rick's comment, for $or$ and $and$,
your expression should always have the form:





(truth equation)$or$(truth equation)

i.e.

(x=4) $or$ (y¬>17)

---------- response 4
  02/23/76     13.03     agg       iowa

I believe that if there are any bits on in the word that
results form the expression, the logical value of the
expression is true (except -0).

---------- response 5
  02/24/76     15.57     obrien       uicc

Bitwise "our, is $union$.

---------- note 41 near full
  02/19/76     21.10     keith s       matha

What determines when a notes file is "almost full"?

What kind of mileage can be expected when that message
appears?

---------- response 1
  02/19/76     21.16     woolley       p

The message appears when the number of possible notes
left is down to about 10, or when the number of blocks
available for note storage is down to 1. It's sort of
arbitrary.

---------- response 2
  02/20/76     10.44     thiher       ames

Is this also true in gnotes files? The almost full
display I mean.

jim

---------- response 3
  02/20/76     11.10     berger       mfl

That is what he was referring to.

---------- note 46 nit?
  02/19/76     21.29     hody       med

can we have a response to the note which follows:
********
libraries  02/16/76     2234     hody       med





neither linesets not microtable libraries are listed
under "library" in "aids"... is this an intentional
ommision? (charsets ARE there of course).
*
also: the library lesson for charsets is "charsetx!...
the one for linesets is "linesets" and the one for
micros is "microtable". would it not be more mnemonic
to call them all by the name of the item followed by
some consistent suffix?
********
i am really not trying to nitpick-- changing to a consistent
approach should be easy and would be very helpful!

---------- response 1
  02/19/76     21.34     hinton       ssu

It really would be nice, and would have saved me a lot of
searching around, note-writing, etc. a few days ago (a) to
have some regularity to these lesson names and (b) to have
AIDS reference them.....
ndh

---------- response 2
  02/20/76     09.08     jmk       pso

We are working on this stuff now.

---------- response 3
  02/20/76     09.53     hody       med

thanks!

---------- response 4
  02/20/76     12.12     hinton       ssu

Thanks from me too.

ndh

---------- note 51 fixes
  02/20/76     02.05     blomme       s

   The c type search in the editor has been fixed so
that it finds an occurrence in the first line of a new
block when the full lesson (SHIFT-NEXT) search is in
effect. Note: the x and u searches have been correct
in this regard all along.
   The latest version of the system also has a fix
so that the personal notes flag gets set properly
even when the receiver was in a "system term" (like
talk or calc) when the note was sent.

---------- response 1
  02/20/76     08.43     friedman       csa





Wonderful! Thanks.

---------- note 52 code gens
  02/20/76     02.29     d zweig       iu

Is there a listing anywhere of some of the "code generators"
currently available on the system?

At one time I knew the where-abouts of code generators for:
        Music(Staves and Notes)
        Chemistry(Molecular Chains and Rings)
        Electronics(Skematics)
        General(of course, Service,etc.)

DZ

---------- response 1
  02/20/76     10.46     judy       pso

Interesting question!

References:
music   peters/music
chem    wegner/phar (fancy generator)
        stan smith/chem (a different, simpler approach)

I don't know about the schematics.

Another interesting code generator is "csdesigns" by
Axel Schreiner. I have not made use of the "code generator"
portion of the lesson. The design part of the lesson is
extremely handy when you want to "talk" to someone about
lesson structure, because you can make flow charts very
quickly.... complete with NEXTs, BACKs, LABs, imain units,
and multiple branching. It is much faster than drawing
a flowchart using SD.

---------- response 2
  02/20/76     11.00     michael       english

Also try notefile "printer", devoted to a discussion of
code generators.

---------- response 3
  02/20/76     11.51     nate       iumusic

   Judy, you forgot me about who has a music-generator.
(plug)

Nate

---------- response 4
  02/20/76     12.03     judy       pso

Sorry, Nate! I was trying to list who could give
more information in that specific area, rather than
actual authors. I hope the UI and IU music departments





are speaking to each other??

---------- response 5
  02/20/76     18.03     thomasson       music

No, sorry, the UI music department, the IU music department
and the Goochbox people are NOT on speaking terms.↓
As I understand it, each of the three have their own set
of editors, compilers, interpreters, generators, executors,
¬+c. ¬+c. ¬+c.↓
In some effort to establish communication, there is a
group-notes file named gswnotes for this and other
things.↓
   Jimmy

---------- note 55 huff bye
  02/20/76     08.30     huff       ve

I would like to say goodbye and express my appreciation to
all of you who have helped me over the last few years.
PLATO has made great strides forward, especially at the
College of Veterinary Medicine, and I am pleased to have
been associated with it. I am accepting a job at the School
of Assoociated Medical Science as an instructor in medical
art. I will teach medical photography and do service
photography for individualized instruction.
well.. thanks, keep looking ahead, bye.
jim huff veterinary medicine

---------- response 1
  02/20/76     08.45     olson       ced

Good Luck and don't forget to take the lens cap off.

D٧O

---------- response 2
  02/20/76     08.54     helper       park

thanks Jim for the help you have given pat and I in our
work on plato..particularly with the audio taping we
did in case of system failure for the demo in New orleans
good luck in your new endeavor! jean helper

---------- response 3
  02/20/76     09.48     bator       kka

I am afeared we may need a whole "goodbye" file
soon. I hope not. Glad to see people tell us what they
 are up to: the human side of PLATO is not dispensable.

---------- note 56 search
  02/20/76     08.42     olson       ced

I'd like to write a small program that does what 'X-search'
does. only on a smaller scale. It only needs to search a





few lessons <b. Is there anyone who would like to
help me with this code or is there already a lesson that
I could copy or -use-? Thanks much...

Dennis

---------- response 1
  02/20/76     10.15     mike b       cornell

        Good luck! Without the ability to accept source
code as data, I don't see how you can do this. Why can't
you just use X-search?

 M¬ ike≤≤≤≤٧B

---------- note 63 nsf flag
  02/20/76     09.08     roper       siu

Suggestion:

Could a flag, similar to the arrow flag for personal notes,
be set so that an author, when glancing at the note index,
could immediately tell that there had been something added
to New Systems Features Notes?

After the recent avalanche of New Systems Features, I've
been backing in and cut of the file every day. This little
flag would mean that I would only have to look when there
was actually something new.

By the way, I like all the new systems features.

Thanks,
Quetzal

---------- response 1
  02/20/76     10.17     k mast       p

        We plan to eventually keep track of the last time
you read system features notes. Then if something has been
added since that time you will be notified.

---------- response 2
  02/22/76     14.29     roper       siu

Thanks.

---------- note 66 m.a.x.help
  02/20/76     09.40     bator       kka

During drop in periods (at least at City Colleges of
 Chicago) only fortunate few early students can get
lessons. Others get "memory allocation exceeded" which
drives them to despair, confusion, etc.
I noticed that some games lessons as in "sports"





tell the would -be signer on that ecs is high and that
s/he cannot pick freely from the list but may pick
the same lesson someone else is using.

Suggestion:
If we do this for games why can't we show/tell the
student what the lucky few are using so s/he can
 access same non-game lesson someone else is using?
Rationale: A circumscribed choice is better than no
 choice at all.

---------- response 1
  02/20/76     09.48     buz       phar

At Purdue, the router we're using shows which lessons are
in memory already, you might want to contact S.Deiss if you
are interested in using this type of router.

---------- response 2
  02/20/76     10.50     chabay       chem

If a student using the system router fails to get a lesson
because ECS is not available, he will be shown which lessons
on the index are currently in ECS at this site.

---------- response 3
  02/20/76     12.01     warner       iu

        This type of code is very easy to write in your own
router. It uses the "lessin" command (please note in aids)
which tell you if the lesson is on at your site.
        The ECS crunch status can be determined by using the
reserved words "muse" and "mallot".

---------- note 71 touchcalc
  02/20/76     09.55     rlr       catid

There was a lesson "advertised" in notes several months
ago that gave a touch calculator. The lesson name was
perc37 or some other number behind perc. I thought it
was a geeat calculator, although there were no instructions
with the lesson. Now the lesson is gone and I was wandering
where it is and if it might still be available.

R↑٧R

---------- response 1
  02/20/76     09.58     jim g       reading

Sorry about the confusion but we have to rearrange files
every once in awhile so it has moved to perc31. I will try
to write a help page soon.
Jim Gilfillan

---------- response 2
  02/20/76     12.16     friedman       csa





You might also like to try lesson "rpcalc".

---------- note 84 xmaspast
  02/20/76     11.13     hody       med

there is insufficient ECS to run lessons on site network
at this time but someone, and i won't say who now, is
using "trek" ... do we have to go through this whole
business again... where are you oh great equal opportunity
Father (site director?!)??

---------- response 1
  02/20/76     11.33     woods       med

let's do go thru it again, george...if we squeak loud
enough and often enough, sooner or later site directors
will take the action required to shut us up....

---------- response 2
  02/20/76     14.52     j thiher       amesrad

What is necessary is your authors of "game" type lessons to
use daycheck screening processes on the users. This is
the practice I've seen several people use which
effectively inhibits purely recreational lessons from
general use at hours of high ECS crunch. Usually this is
between 8 and 5. If this were done by authors or such
lessons, the problem wold virtually disappear.
jim

---------- response 3
  02/20/76     15.12     berger       mfl

Jim, let's put your suggestion in the proper perspective.
It is up to the individual site director to decide how
his site is used - if the director of a site allows trek,
that is his prerogative.

Since site 0 is unique, it may be that special checks
should be made for "network", but to call for blanket
time check on all recreational lessons is unreasonable.

---------- response 4
  02/20/76     20.26     hody       med

not exactly, sir berger of mfldom... if you check the
note on 2/15 keyworded "resources" you will a discussion
in which "systems" admit that the excessive use of
computer resources by some users affects all others
users. somewhere back, there is a note in "mednotes"
by golden is which he says and i quote more or less
loosely: "it is unconscionable for people to play
games in prime time... when ecs is tight..."
*
the correct solution is for the system to flag lessons





which are essentially recreational in nature-- not
game type simulations used in teaching-- and remove these
automatically from ecs when ecs available is less than
a specified portion of the available total-- say 75:.
*
another possibility, though a tedious one, is the
publication is these notes of the names and courses
which abuse ecs in prime time.

---------- response 5
  02/20/76     23.33     frye       mfl

It strikes me as being even MORE tedious to sit here and
bring up this whole can of worms about games/nongames
once more! Go ahead and publish those names and courses
-- perhaps someone might consider a "gripe"notes file for
general-purpose nagging and other worthwhile activities.
If enough people get embarassed by public criticism THEN
you could get results.

The best solution in my mind is to finally come up with
a site-dependent lesson _i__n__h__i__b__i__T_ function similar to the
"reserved lesson" option. Is that really so difficult?

Not trying to offend anybody,

G. David

---------- response 6
  02/21/76     11.52     hody       med

good idea, frye
*
but you would also need a really effective assigment
of all resources (including CPU and condensOr) to
sites and that doesn's exist now.

---------- response 7
  02/05/76     22.31     broadus       css

Is flyboy wiggy on your list?

---------- note 85 stats
  02/20/76     11.15     barr       econp

        I would like to look at some statistics lessons.
Especially lessons which teach stats.

        I am also interested if anyone has code which will
1) generate random numbers with a normal distribution.
2) gemerate z.t,or F tables.
3) generate random numbers with other distributions.

        Can some one help me? Answer here or pnote.

mike barr

---------- response 1





02/20/76     11.31     fay       o

Try contacting Al Avner (avner/s) or Jim Kraatz (jmk/pso).

---------- response 2
  02/20/76     12.01     judy       pso

or cohen/psych (Jerry Cohen or Psychology)

I believe Avner's and Kraatz's lessons are more service
packages, while Cohen's are more teaching lessons.

---------- response 3
  02/20/76     14.10     avner       s

Also contact Kikumi Tatsuoka (kumi of peer) for stat
lessons.

For questions 1 and 3--see lesson "library" (inspect mode)
--block "random"......
For question 2--see any math or CS library for books on
numerical approximations. If you just want specific
values for these distributions, see lesson "star" (select
the probability distribution option). Actually, a table
is a device which is required for another medium (the book).
Usually all you really want are a selected set of values.
Lesson "stat" provides more values with higher accuracy
than any published table could possibly do (not bragging-
it comes with the medium). Al Avner

---------- note 92 BUGS
  02/20/76     12.04     buz       phar

Did you know that..........................dept.
the following code:

write   ¬0at,x1¬1Some stuff to say
        some more stuff
        stuff line 3
        ¬0size,1¬1 Emphasize this line
        this line too
        last to emph¬0size,0¬1

"this line too" will be plotted on top of "some mor stuff"
and "last to emph" will write on top of "stuff line 3"

I was told that this is/was a know (but possibly forgotten)
bug

*buz*

---------- response 1
  02/21/76     16.28     putch       arizona

     I don't know what causes it, but the following will
work the way you want:





write   ¬0at,x1¬1Some stuff to say
        some more stuff
        stuff line x
        ¬0size,1¬1 Emphasize this line
        ¬0at, where+300¬1 this line too
        last to emph¬0size,0¬1

---------- response 2
  02/24/76     20.50     rader       s

Ye gads. Thanks for the info. I am not aware of our
being aware of this one....(is somebody feeding these
bugs?).

---------- note 95 1bit→-0
  02/20/76     14.05     mcneil a       uicc

define  segment,mt=n10,1
calc    mt(1)‎-(2<3)
        mt(2)‎-(3<3)
write   ¬0s,mt(1)¬1
        ¬0s,mt(2)¬1
*
shows the values as both 1. UGH.
Ah the horrors of ones compliment!
To get around this what will be fastest??
1) abs(-(logical))
2) -sign(logical)
or something else

---------- response 1
  02/20/76     14.32     marty smith       mtc

add 0 to it.

---------- response 2
  02/20/76     14.38     putch       arizona

     Or add a third argument, s, to your segmented define.
This allows space for a minus sign where one is needed.

---------- response 3
  02/20/76     14.45     clark       lawyer

The segment size is only 1 so a signed segment will be
of no help. Of course, increasing the segment size to 2
solves the problem.

---------- response 4
  02/20/76     14.49     putch       arizona

Sorry, Meant to include that.

---------- response 5
  02/20/76     17.04     warner       iu





        In short, fastest version is
calc    seg‎0-logical
        Systems, this is unbelievably messy. Can you please
change the operation of segments so that a 1-bit signed
segment can have only the values 0 and -1? It would make
things a lot easier on may people!

---------- response 6
  02/20/76     17.23     mike b       cornell

        I'd like to know by what magic a 1 bit segment can
have a value of -1. To store -1 requires _two_ bits--the
number and a sign bit.

M¬ ike≤≤≤≤ˇB

---------- response 7
  02/20/76     18.02     clark       lawyer

calc seg‎(logical)ˆ2 also takes care of any problem
and can be used with any size segment

---------- response 8
  02/20/76     21.15     warner       iu

        By making the sign bit the _only_ bit stored, and
setting the lsb to 0, you can store the proper values.
In this scheme an n-bit signed segment is converted to an
then $cls$ing it 59-n places to the lsb of the integer,
the $ars$ing the integer 59-n bits.
What I am suggesting is that for 1-bit signed segments
this process be used exactly as is. Then in conversion.
o0$cls$59$ars$58=o00000000000000000000=0 or
02$cls$59$ars$58=o77777777777777777776=-1
At present, a 1-bit signed segment is stored the same as
a 1-bit regular segment. To pull regular segments out
they are simply put in the lsb's of the destination
integer.

---------- response 9
  02/21/76     05.07     dave       infe

        I like Silas's suggestion. Can't see how it could
affect any existing lessons; would make future ones simpler.

---------- response 9
  02/21/76     19.05     mike b       cornell

        Yes.  i hadn't thought of just storing the sign bit.
Given TUTOR's logical conversions, this would indeed by
useful.

M¬ ike≤≤≤≤ˇB





---------- response 11
  02/22/76     11.44     midden       s

Uh, I somehow believe that it will effect existing lessons.
There are several ways to check for -0 with a signed segment
define segment,seg=n1,1,signed
a)     bitcnt(seg(1))
b)     seg(1)$mask$_1_

If signed segments were changed as suggested, case a and b
could be affected.
One other point in favor of not changing is "consistancy".

All in all, one bit signed segments seem rather useless.
Marshall Midden

---------- note 99 discussion
  02/20/76     14.43     carter       comm

A discussion concerning bicycle transportation has been
started in the gnotes file "commnotes". The file is
unrestricted--everyone is invited to participate. This
beautiful spring weather has evoked two of my passions,
one of which is the current topic of discussion.

P.S. Hasn't anyone designed a spring fever welcome to
plato page?

---------- response 1
  02/20/76     23.10     kent       unl

        ...I'm just shuddering in my shoes to see what
happens to the plato page on April 1......

---------- response 2
  02/20/76     23.24     frye       mfl

Any guesses as to the second passion, George? If the
topic of discussion is the first, my guess is that dis-
cussing it is the second. Or discussing anything, for
that matter!

G. David

---------- response 3
  02/21/76     17.02     carter       comm

Thanks Dave, but I see computer-based discussion more as
a professional commitment than a passion. I am far less
of a blabber-mouth in person.

I think if PLATO bases its existence too strongly on
traditional instructional lessons, as opposed to computer
conferencing applications, then it will only have a
transient importance in the field. Within five years it
should be possible to build stand-alone plato terminals





that will do everything the current architecture does
without the high communication charges for isolated
terminals. Lessons will be stored on video disks and
executed in microprocessors connected to 64k RAM's.
If these technologies come through, and there is every
reason to expect they will, the PLATO gospel claiming
longer hold any water. The current emphasis puts all our
eggs in one basket in my view.

---------- note 104 E forgets
  02/20/76     15.02     guerra       uimc

Could it be arranged for the author mode page not to
"forget" which lesson you were editing when you go into
the ECS display ("E" at the author mode arrow)? Very often
I go to that display after finishing a bunch of editing
in order to see if there's room in ECS at my site to
condense it, and it would be much more convenient if I
could go right back to the lesson by pressing NEXT or DATA
on the author mode page instead of having to retype the
file name.

---------- response 1
  02/20/76     15.32     parrello       mfl

Same deal would be nice for the 'M' option, which at one
time _did_ remember the lesson, but has now forgotten.

---------- response 2
  02/20/76     20.28     hody       med

same deal would be nice for the L option after you leave
a lesson (might be nice?)

---------- response 3
  02/23/76     10.53     p cohen       med

Same deal would be nice for the 'C' option. (But just
wishful thinking, I assume--since each charset is at-
tached to a lesson.)

--paul

---------- note 111 new search
  02/20/76     18.30     fritz       amesrad

I don't know if this has been mentioned anywhere yet,
but I like it anyway! The new changes to "search" are
really nice! Sure makes it easier to read without all
that redundant garbage.

---------- response 1
  02/21/76     04.07     sellers       arizona

Thanks to Fumento!!





Just wait until the X-search sorts out the objects of the
search!!! Then my time in Heaven will be called:

Total hours on system...

---------- note 116 jumpout
  02/20/76     21.38     dave       infe

        How much does a -jumpout resume- cost when executed
in the restart lesson?
       (The question would be unnecessary if command
-jump (rstaptu)- were legal...)
       Thanks/

Dave

---------- response 1
  02/21/76     12.38     white       p

Currently, a -jumpout- to the lesson you are in
does a -jump- to the specified unit.

---------- response 2
  02/21/76     13.51     al       mfl

Remember that when you do a jumpout to your own lesson (the
one that contains the "jumpout, command) the ieu isn't
executed. If you want to start a lesson over again AND
execute the ieu you must do a "jumpout kludge".

---------- response 3
  02/26/76     22.39     walter winchell       infe1

Thanks.

---------- note 119 iumed=iusn
  02/20/76     22.22     church       iumed

******** church iumed  02/20/76  22.19.22

Effective sometime Monday:

          Course 'iumed' will be changed to 'iusn'.

Incorporate this information in your signon proceedures
as necessary.

chris church

Any ideas for a new corporate symbol?
********

---------- response 1
  02/21/76     04.18     bonnie       matha





What does "iusn" stand for?

Knowing this will help use remember your new course name.

---------- response 2
  02/21/76     12.23     kimble       ustaf

Indiana Univ. School of Nursing?

---------- response 3
  02/21/76     14.55     nate       iumusic

   Well...that's one way of looking at it...but the idea
is "Indiana University Student Nurses". Chris I still
think you could have come up with something better than
that. (snicker)

Nate

---------- response 4
  02/21/76     18.15     church       iumed

CK Nate... the new name stands for:

Indiana University School of Nursing

---------- response 5
  02/23/76     12.26     stephanidis       iusn

        What type of corporate symbol
are you exactly talking about?

---------- note 127 congrats.
  02/21/76     08.09     green       forestry

Many congratulations to Marshall Midden.

---------- response 1
  02/21/76     08.12     matthews       css

WHY?!?!?

---------- response 2
  02/21/76     08.36     loitz       o





Try sending him a pnote and you'll see.

---------- response 3
  02/21/76     11.46     matthews       css

sorry 'bout that marshall may your days
as a system prog be many and pleasurable

---------- response 4
  02/21/76     19.33     midden       s

thanks...

---------- response 5
  02/22/76     15.53     buz       phar

congratulations...Mr. Midden/S.

feels like christmas all over again...doesn't it.

---------- note 130 healthauth
  02/21/76     09.45     levy       mcl

   A group notes file is available to those authors engaged
in the writing of, or otherwise seriously interested in
health science material. Their comments are welcome.

The gnotes name is " healthauth". It is unrestricted to
authors. Comments on specific health science lessons can
better be inserted in mednotes;general philosophical discus-
sion in medtalk. Healthauth is for problems relating to
matters of specific interest to health science authors.

---------- note 136 routerstat
  02/12/76     14.09     berger       mfl

A router is associated with a routed student while he
is running.

Do the "account" statistics on lesson usage for routers
reflect the time student spends in the router, or
does it also reflect the entire time that the student
was signed on?

---------- note 139 sysvar ‎ n
  02/21/76     16.47     fritz       amesrad

  Re note ¬$24 (anscnt‎0):
Rather than have a special command to set *anscnt* (and a
special command to set *lscore*, and a...), why not cut
down on the plethora of commands (¬+ add several handy





features!) by combining it all into one command? I'm
proposing a -reset- command (or something) which would
effectively be a -calc- on systems defined vars. This
would obviously have to be restricted somewhat (I can see
it now: "reset mallot→2ˆ6ˆ0-1"), but think how handy it
would be with vars like

ntries (!)       key(!!!)     anscnt     jcount (marginal)
wcount          nhelpop        fromnum     lscore

etc., etc.  Possible?

---------- response 1
  02/21/76     16.59     frye       mfl

Definitely not "key".

---------- response 2
  02/21/76     17.23     clark       lawyer

Why not key?

I would even be satisfied if there could just be an easy
way to zero some of those vars (though not quite as
handy of course)

---------- response 3
  02/21/76     17.36     putch       arizona

     Why not just let authors redefine any sys variable
they want to. It is their problem, not the system's, if
they louse up one of their routines that depends on one of
the redefined variables

---------- response 4
  02/21/76     18.39     berger       mfl

That isn't true at all - the system guarantees the integrity
of those variables. If you could modify them, the system
might just as well withdraw them altogether.

---------- note 144 modifycont
  02/21/76     17.43     putch       arizona

     Is there any possibility of modifying the mechanism
that prevents editing of notes and/or responses that have
received responses so that you could modify such an item,
provided you are the only one to have responded to it?
This would be a great to those who write long, continued
notes and then find a mistake back in the first part.

---------- response 1
  02/21/76     20.53     woolley       p

That might be reasonable in theory, but in practice it's
messy and not likely to get done.





---------- note 150 HELP
  02/21/76     23.52     j mayer       etsu

HELP!!!!!!!!!!!!

val(cor)‎val(cor)-val1+val2

this -calc- line is giving me an -index- execution
error but when I check my variables are as follows:

val1=1
val2=0
cor=13

this seems simple but val(13) is within range!...In fact
the execution error may be prevented by commenting off
the end of the line (-val1+val2) and may be caused
by replacing the end with (-1)!?!!/!!/!?
   None of the variables are segmented...val is a system
defined array but the error happens when it is defined
as     [ val(a)=n(43+a) ] also.
                 thanks...Jim Mayer (j mayer/etsu)

---------- response 1
  02/22/76     01.22     thomasson       music

We found what was wrong, This line was ok.↓
It was one of those frustrating errors which you can
stare at for hours before finding it.↓
Jim

---------- response 2
  02/22/76     18.51     kent       unl

        What was wrong? I've got something like that
bothering me with an array in common.

---------- response 3
  02/23/76     13.24     al       mfl

Seems to me it is one of those times when the implied
multiplication "feature" is causing more trouble than it
was ever worth. Possible that val1 was being read by tutor
as val(0)≠1?

Also notice such code as:

charset cslessons,"keytops"

as keytops without the quotes resulted in "tops" multiplied
by the system variable key, I believe.

---------- response 4
  02/23/76     23.20     kent       unl





I found my problem too (with some help). Stupid mistakes!

---------- note 153 note resp
  02/22/76     11.39     woolley       p

Some new features in notes to make it easier to wade through
lone chains of responses:

      "+" forward one response
      "-" back one response
shift "+" jump to last response
shift "-" jump to base note

All other keys still do the same things as before.

---------- response 1
  02/22/76     11.51     midden       s

Let me be the first to thank you Dave.
THANK YOU!!!
your humble roommate

---------- response 2
  02/22/76     12.22     hody       med

dotsanice

---------- response 3
  02/22/76     18.54     kent       unl

↑٧SAH'RIGHT≤≤≤≤≤≤≤≤≤↑٨_________

---------- response 4
  02/23/76     13.36     gilpin       peer

Thanks for this feature!

And thanks for the nice key choices.

---------- response 5
  02/23/76     18.45     snellen       medneta

Am I overlooking something or is "+" (for moving ahead one
response) actually redundant? It seems to me that pressing
LAB does exactly the same thing.
--JES.

---------- response 6
  02/23/76     19.10     parrello       uimatha





Yeah, but "+" makes things more symmetrical.

---------- response 7
  02/26/76     22.34     wong       cs317

However,it does nothing at the last response...

---------- note 158 U options
  02/22/76     15.19     clark       lawyer

Nice feature for the U option...thanks goes to whoever
did it!!

In the realm of small mistakes: When you view all
of the sites, it still says at the bottom "NEXT for
next site"

---------- response 1
  02/22/76     16.36     copland       csa

        What about the possibility of making the NEXT key
go to the next site at which there are terminals currently
in use by people with their users flag set, as opposed to
simply the next site. This would save on having to display
sites at which there are apparently no terminals currently
in operation.

---------- response 2
  02/22/76     16.45     harkrader       o

       The new 'user' is a product of Dave Frankel.
There are a few ideas still in the hopper, so I think
more changes may be coming. Thanks again, Dave.

---------- response 3
  02/23/76     11.15     t obrien       politics

One thing I saw about the new 'U' option. While
going through it, I saw a blank line for site-station
9-28 (just showing the number, but no name). Was this
an error, or just catching someone signing off? I think
this didn't happen in the old version, and should not
happen here. I saw it for about 20 seconds, or 3-4
replots.
Just thought someone should know.
Terry O'Brien

---------- response 4
  02/23/76     22.37     koning       csstaff

Signing on, I would guess..... Similarly, the new version
will show a blank line for 0-0 if that terminal is in use.





---------- response 5
  02/23/76     22.38     fisher       conn

This problem was noticed on several stations tonight.

Especially terminal 0-0.. that has been just numbers for
at least ten minutes...
Have also seen (not seen?) the names on at least five
terminals for at least 3-4 replots of them...

fishswine

---------- response 6
  02/24/76     09.21     michael       english

I noticed it a lot this morning around 8:00, and I doubt it
was people signing-on since it lasted several minutes for
some terminals.

---------- note 159 branch
  02/22/76     15.33     anderer       ee

     Why, in the following code, does the code in the
-doto- get done? To my way of thinking (and obviously
not PLATOs), the branch should skip around the -doto-.

unit   test
branch 1end
doto   1end,n1‎1,10
at     202+(100≠n1)
write  In the loop.
1end
end    lesson

  However, this isn't the case. The above code produces
a nice group of 'In the loop's'.

---------- response 1
  02/22/76     15.51     jim g       reading

If you read aids on branching and doto all will become
clear to you. What you are doing is a bad programming
practice and unpredictable. What you are doing is branching
into the doto loop because the end label of the doto loop
does more than act as a normal branching label.

P.S. In aids ask for info on "branch" (DATA key), choose
choice "e" and what you need is on the 5th page of writeup.

---------- response 2
  02/22/76     17.26     friedman       csa

In other words, the end label of a doto loop is WITHIN





the loop. So by branching to it, you are getting into
the loop.

---------- response 3
  02/23/76     08.33     clson       ced

If you change your last '1end' to '2end' and change your
branch to '2end', it should work!!

Dennis

---------- response 4
  02/23/76     09.08     koning       csstaff

No, because then there is no label for the -doto- to refer
to. Adding '2end' would be correct.

---------- response 5
  02/24/76     01.20     anderer       ee

  I understand the multitude of ways to get around this...
I just didn't think it should work that way.
dg

---------- response 6
  02/24/76     09.34     t obrien       politics

One other thing. I noticed several students listed
on the page, but not on others. One case in point,
674/iudemo was listed, but not student/pfw. Is the
new page supposed to do this? If so, are you combining
both the 'E' page and the 'U' page? It seems like it.

Terry O'Brien

---------- note 164 -used- ieu
  02/22/76     17.01     hecht       ed

       How are defined variables treated when one
       lesson -uses- another? In particular, is
       it necessary to redefine all variables in
       an ieu for the -used- lesson, providing
       they have the same names as those in the
       -user- lesson?

       I have several lessons that all -use- a
       single lesson which contains various
       subroutines, etc. I am wondering whether
       it is really necessary to have -defines- in
       the -used- lesson, since it is never run
       by itself.

---------- response 1
  02/22/76     17.29     friedman       csa





When one lesson -use-s part of another, everything acts
just as if the code in the -use-d block were inserted
into the -use-ing lesson in place of the -use- command.
So any -define-s in effect before the -use- are still
in effect during and after it, unless there are -define-
commands in the -use-d block which change this.

---------- response 2
  02/22/76     17.59     hecht       ed

Thanks for the information. This will save about 150 words
of condense space for me.

---------- response 3
  02/23/76     08.38     olson       ced

I have several lessons which all-use- a block of defines
from one lesson. As Mr. Friedman mentions, it isn't
necessary to insert defines in all your lessons as long
as the defines aren't being overridden in your other
lessons.

DˇO

---------- response 4
  02/24/76     09.19     shirer       vu

   The other way around is just as convenient. In one
series of related lessons, all the defines are in a
common lesson -used- by all the others.

---------- note 169 Getword!
  02/22/76     18.25     frye       mfl

Many thanks to the systems staff for "getword"!

One more feature (or extension) I would like to see is
for -getword- (or something similar) to return EVERY
scannable "entity" -- i.e. the sentence "The options are
a,b, and c." would return the words (the,options,are,a,b,
and,c) and _a__l__s__o_ the delimiters (,,.). Perhaps a -specs-
option?

G. David Frye

---------- response 1
  02/22/76     19.19     fritz       amesrad

Thank you, thank you, thank you, to all parties concerned!

(And I'd like to be able to pick out delimiters, too...)





---------- response 2
  02/22/76     19.22     clark       lawyer

A great improvement!!!

by the way, note the overwriting which occurs in that
system note: NOT YET AVAILABLE etc. and "2 responses"

---------- response 3
  02/22/76     20.15     hinton       ssu

I could use the delimiters too...but I can hardly wait for
this to come up in prime time !!!! Thanks/thanks/thanks.
ndh
Will there be a -getnum- ? Or will numbers be "gotten" as
words are ?

---------- response 4
  02/22/76     22.43     tenczar       s

the overwrite error is known...

and final testing is being done on the option that will
let you completely define all your characters as far as
"judging behavior"...one option allows you to define
a "puncword" which is to be considered a punctuation and
a word...hence, you can finally put punctuations in
answer-type commands and obtain the "puncword" via
the new getword command

why, you can even define the letter "o" to be equivalent
to "0"...and then "10≠10" will yield "100"

---------- response 5
  02/22/76     22.54     fritz       amesrad

Amazing...
  We will indeed be indebted to you when this is working!

---------- response 6
  02/23/76     13.28     al       mfl

Hinton: for "getnum" read aids on "storen"

---------- response 7
  02/23/76     17.03     jones       mcl

Thanks.

Getword fills a long existant gap in TUTOR judging.
It is much appreciated.

---------- response 8
  02/24/76     13.03     hinton       ssu





Al..as I understand what Tenczar is saying about -getword-,
it looks more flexible and efficient than -storen-.
ndh

---------- note 177 tenure
  02/23/76     08.01     luke       cornell

       Could anyone possibly get me in touch with
the person responsible for lesson -tenure-?
       I am in the process of writing a lesson
using many of the same ideas, but relating them to
language instruction through such trial and error
learning, i.e. being placed in a foreign country,
and having to adapt to gain surviva by learning the
language. I felt that a lesson on the order of
-tenure- was a possibility to explore.

Any ideas, comments?

Sincerely,
Luke Kaven

---------- response 1
  02/23/76     08.16     maggs       law

When you do get in touch with him, tell him that his
copyright notice is no good, since it lacks a date.

---------- response 2
  02/23/76     08.41     olson       ced

Contact tebby/pso for information about authors. She
knows who wrote everything.

DˇO

---------- response 3
  02/23/76     12.46     gaede       ed

You can get in contact with me via pnotes:
gaede of course ed
Owen Gaede

---------- note 184  units
  02/23/76     11.00     kane       phys

A problem seems to have developed over the weekend with
-define-ing units. Units whose definitions appear near the
end of the student define set are not being handled
correctly. If these same definitions are moved to near the
beginning of the define set, they work ok. The problem is
that the dimensionality is not recognized (for instance by
-storeu-). The numerical conversion is being done right,
however.

---------- response 1





  02/23/76     14.44     golden       s

Should be working correctly now.

---------- note 186 monitoroff
  02/23/76     11.53     clark       lawyer

SHIFT-ERASE did not work for me in monitor mode. Normally,
it erases the entire line; today, it was just erasing
one word at a time. I kept having to press NEXT and then
SHIFT-TERM again. Am I missing something?

---------- response 1
  02/23/76     11.59     fritz       amesrad

  I would LOVE it if ERASE1 worked in monitor mode!!
It would be nice, however, to be able to erase the line
without losing the arrow.

---------- response 2
  02/23/76     12.04     koning       csstaff

The same change shows up in the term arrow (it being the
same thing, presumably). Looks like it could be a
side-effect of the changes to term handling.
Perhaps TERM could erase the entire line, leaving ERASE1
as it is?

---------- response 3
  02/23/76     12.30     fay       o

I had the same experience with monitor mode over the weekend
and it is MOST inconvenient! I hope this is fixed soon.

---------- response 4
  02/23/76     12.38     celia       pso

It's better than before! The consultee, or monitoree, can
clear his line with LAB, and not lose his arrow. Shift-
erase now works like it does at an arrow.

---------- response 5
  02/23/76     12.45     clark       lawyer

thanks for the change...

---------- response 6
  02/23/76     13.10     j thiher       amesrad

Will take a bit of getting use to but I think
the change is for the better. Don't "fix" it.
jim





---------- response 7
  02/23/76     14.45     temp       phar

It would be nice if the monitor mode could be ended
from both ends with a BACK1 or DATA1 so that you would
not have to SHIFT-STOP out to get rid of an unwanted
monitor.

$ *≤ *≤ *≤ *≤ *≤ *≤ *≤ *≤ *≤ *≤ *≤G*≤r*≤e*≤g*≤ *≤C*≤o*≤r*≤s*≤o*≤n*≤ *≤≤ *≤ *≤ *≤ *≤ *≤ *≤ *≤

---------- response 8
  02/23/76     16.06     artman       uimatha

NO. PLEASE!

Because I may want someone to follow me through a lesson
where I may need to hit every concievable function key.

Unfortunate, but it seems that the only thing that the
system can be sure that you do not want your monitor to
see is whatever is done after you leave the system.
Otherwise, you would be restricting the use of one of the
function keys to non-monitor mode, which would defeat it's
whole purpose.

eric artman

---------- response 9
  02/23/76     16.48     michael       english

Possibly a TERM-reject while being monitored would be a
logical alternative.

---------- note 200 mucho doto
  02/23/76     14.38     loitz       acoustic

How come I received condense error message

To many doto's in unit

when I only had one doto loop? Could the large amount
of calcing in this loop have anything to do with it?

---------- response 1
  02/24/76     09.27     judy       pso

Could we see an example of the code? I've never seen
this error before.

---------- note 201 -return?-
  02/23/76     14.39     warner       iu

The -return- command is a source of confusion to





new TUTOR authors. It "returns" control from the user's
lesson to the PLATO monitor, thus forcing an end to the
timeslice. Its most useful use is in waiting for common:
1wait  branch comuse(him),x,1ready
branch $$ allows others to process
branch 1wait
1ready $$ go on with stuff
       The problem is that new PLATO users almost always
use some other command there, i.e. -pause- (does the work
but slow) or -delay- (fills up TUTOR output buffer but does
not halt processing). The -return- command they almost
always confuse with -goto q- or -exit-.
       A suggestion: change the name of this command to
-wait-. It more nearly conveys the use to which the command
is put.

---------- response 1
  02/23/76     14.59     bowery       comm

Perhaps a -pause- with a tag of 0 could work in this
capacity (no -timeup- involved).

---------- response 2
  02/23/76     21.06     david       cerlcc

Using -pause 0- for this clearly independent function
only further obscures tutor code. -return- should probably
be renamed though. How about -endslice- , with no tag.

---------- response 3
  02/24/76     09.36     friedman       csa

I agree the command name is misleading. How about
-newslice-? I'm not crazy about it, but can't think of
a better suggestion...unless....
How about -break-? [Remember the OLD -break- command?]

---------- response 4
  02/24/76     10.39     warner       iu

       -break-?
       I always thought that sounded like a non-computerese
version of -bug-.

---------- response 5
  02/24/76     14.57     b sherwood       s

Even better: teach people to use -reserve- and -release-,
in which case the -return- command has no purpose.

---------- response 6
  02/27/76     17.06     obrien       uicc

"newslice" sounds to me like a problem for Parrello.

---------- note 205 punctuat?





  02/23/76     15.09     mike b       cornell

        After spending some time kludging judging code
to accept +,≠,-,≡, etc, at a -match- command (including
use of the venerable -put- command) I find myself wondering
as to the status of punctuation handling by judging
judging commands. Will we soon be getting the ability to
specify what should be treated as punctuation and what
shall not?

M¬ ike≤≤≤≤ˇB

---------- response 1
  02/23/76     15.22     golden       s

Yes. Quite soon.

---------- note 206 $$....??
  02/23/76     15.10     deiss       phar

Is there expected to be any decreases or increases in the
cost of CERL PLATO service over the 1976-77 fiscal year
for remote sites? Please comment on all major charges:
1) CPU/IO/DISK, 2) Repair and parts replacement, 3) other.
Especially interested to hear whether the increase to
1000 terminals will drop total charges for 1 above, or
whether the new accounting scheme will affect different
users differently. Also interested on what the policy will
be for replacement of burned-in (or out) panels.

Steve Deiss

ps: Would appreciate a semi-official response.

---------- response 1
  02/23/76     15.41     golden       s

The decision isn't official yet, but most likely your costs
both for PLATO service and repairs will be the same next
year as they are this year.

I don't know about panel replacement.

---------- note 208 packing?
  02/23/76     15.29     bowery       comm

  The ideas about what to do with the new -packc- are
real neet. Here is a suggestion which I stated previously
and would like to have criticized by anyone who gets
the urge (preferably the people working on -packc-):

  Instead of the -pack- and -packc- commands with the embed
stuff, simply allow a person to divert the output of the
internal code generating commands (-write-, -showa-, embeds
etc...) to variables.





Something like:

outvar on,n1,100 $$100 chars available from n1
writec var¬,bloop¬,¬0a,stuff,21¬1 ¬0e,exp¬1
showo  octal
outvar off,n11 $$put char count in n11

It seems that this would be a somewhat more elegant way
of doing things. Any reasons why if would be unwieldy?

---------- response 1
  02/24/76     14.47     michael       matha

This sounds like a very good idea to me.

Why _couldn't_ something like this be done?

---------- response 2
  02/24/76     14.59     b sherwood       s

There have been arguments pro and con over both methods.
Probably your scheme will not be implemented, since it
has altogether too many repercussions. For example,
what should be done with the output generated by
-draw- and -circle- statements?

---------- response 3
  02/24/76     15.56     bowery       comm

Ooops, I didn't mean "internal code" in the system
programming sense. I meant it in the "character code"
sense (ala AIDS).

Output from line, char, ext, audio, etc... commands
should simply be sent as usual.
This would make programs that format and send printer
output much simpler and easier to read.
There are many applications like this

Of course this DOES present a minor problem with -plot-
and -codeout- but it would seem that these commands
could generate the "internal code" and zap it into the
variables with the -outvar- flag on. Such a conversion
would be very useful.

---------- response 4
  02/27/76     17.12     obrien       uicc

Every other system I've ever seen has either had, or has
been working hard on obtaining, a fairly complete and
general format for I/O redirection. One of the most general
I've ever seen (that of UNIX) allows any process to send
output to, or get input from, any file or process as well
as the terminal.
Example:
(date; ls)¬Ipr -4
would output the date and a list of all one's files to





another program that would print the stuff in 4-column
format on the terminal. Adding "¬Iopr" to the chain would
send the 4-column output to the line printer via a spooler.
I speak from experience when I say that this degree of
flexibility adds an unimaginable degree of power to a
system!

---------- note 209 hiddenchar
  02/23/76     15.33     clark       lawyer

It would be nice if in the common editor you would see
the hidden characters. I am going blind trying to count
over in the octal the correct numbers of spaces to read
the correct octal. Something like the H option would be
nice.

---------- response 1
  02/23/76     17.26     straayer       telcomc

     Would converting the common to source, then using the
"hidden character" option help? You can then convert the
source back to common, after the "hidden" check.

---------- response 2
  02/23/76     18.13     clark       lawyer

That is quite messy. The format of the editor now is that
whenever there is a hidden char in the variable, a space
is printed in that spot of the alpha representation. Why
not put in a symbol to show what that char is????

---------- response 3
  02/23/76     19.16     parrello       uimatha

Extreme slowness of plotting, for one thing. Such an option
would most certainly have to be just an option.

---------- response 4
  02/24/76     09.32     judy       pso

Attacking the problem from the other end---would it help
yur counting to have something like this in the octal
display? This should be an easy additional to make.

          ¬X    ¬X    ¬X    ¬X
      00000000000000000000
      00000000000000000000
      00000000000000000000
          ¬W    ¬W    ¬W    ¬W

---------- response 5
  02/24/76     10.24     gilpin       peer

¬          ¬≤0000¬≤¬I≤ˆ¬I≤ˇ¬I≤0000¬≤¬I≤ˆ¬I≤ˇ¬I≤0000¬≤¬I≤ˇ¬I≤ˆ¬I≤0000¬≤¬I≤ˆ¬I≤ˇ¬I≤0000
         0000¬≤¬I≤ˆ¬I≤ˇ¬I≤ˆ¬I¬0000¬≤¬I≤ˆ¬I≤ˇ¬I≤0000¬≤¬I≤ˆ¬I≤ˇ¬I≤ˇ0000¬≤¬I≤ˆ¬I≤ˇ¬I¬≤0000





0000¬≤¬I≤٨¬I≤٧¬I≤0000¬≤¬I≤٨¬I≤٧¬I≤0000¬≤¬I≤٨¬I≤٧¬I≤0000¬≤¬I≤٨¬I≤٧¬I≤0000

---------- response 6
  02/24/76     12.10     fritz       amesrad

Sloooowwww.... why not

         0000 0000 0000 0000 0000
         0000 0000 0000 0000 0000
         0000 0000 0000 0000 0000
         0000 0000 0000 0000 0000

Like it's formatted for the COPY key?

---------- response 7
  02/25/76     14.48     mcneil a       uicc

Drawing vertical lines between the digits (every 5 or what-
ever) would be fast than all of the above.

---------- note 212 nested
  02/23/76     16.15     bankss       mfl

How about a system reserved word like the baseu ¬+ mainu to
find out what unit a suroutine is called from, if we are
nested 4 or 5 deep in subroutines?

---------- note 213 rochester
  02/23/76     16.26     oi       cerl

Are there any PLATO terminals closer to Rochester, N.Y.
than Albany or NYC????

---------- response 1
  02/23/76     16.49     michael       cornell

I was given to understand that lesson "network" was the
final authority on all but temporary (demo) terminals...
and there are only three entries for New Your State.
(Now watch somebody complain about forementioned lesson
not being updated properly)

M↑٧i↑٧k↑٧e O↑٨l↑٨t↑٨z

---------- response 2
  02/25/76     17.16     oi       cerl

Righ, Mike I heard the network may not show
all PLATO terminal locations.

---------- note 214 New do?
  02/23/76     16.30    cole       langlab

When I use the -do- command I find it inconvienent





that I can't branch back to my original unit. Would it
be possible to make the code below possible?

----------
unit   qwerty
do     uiop
code   ....
code   ....
1br
more tutor code
unit   uiop
branch 1br
----------

D. Cole

---------- response 1
  02/23/76     16.45     celia       pso

What you suggest is impossible, because unit uiop could
be attached from several different main units, and the
statement label would presumably have to appear in all
those main units.

What you want can be acheived with -goto q- or -exit-.

In the example you've given, no explicit statment to return
to the main unit is needed; such return is automatic.

---------- response 2
  02/23/76     17.05     clark       lawyer

But then he would not skip all of the code between the
do and the statement label. I usually but a marker in
the unit -do-ne and then put a branching command after
the do to skip the code if the unit has been done. It
is also possible to change the statement label to an
entry command and -goto- if from the -do-ne unit.

---------- response 3
  02/24/76     09.43     friedman       csa

Cole, what you want, even if possible, would result in
code which would be a nightmare for anyone else to ever
read or debug. And the chances of it working properly
are small. You really should appreciate the fact that,
if you're writing a program or lesson for some group,
the time WILL (not may) come when that program is going
to have to be changed. Weird little obscure coding tricks
make working on someone else's program a nightmare.
Stick to writing programs that humans can read, as well
as computers, and save the sanity of those who come after
you!

---------- note 215 talk bug
  02/23/76     16.34     marty smith       mtc

just a minor bug, but if you have your talk options set





for everything off except let your own course talk to
you; and someone outside your course pages you while
you are term-talking, instead of them getting the message
"you cannot talk to marty smith", they get the message
"marty smith is busy, but has been told you called."

and i, the pageE, get the message "somebody-anycourse wants
to talk to you". this is definitely not right.

---------- note 216 name cbomb
  02/23/76     16.36     roper       siu

I had some problems when I changed the names of some lessons
today that were linked by jumpouts and use commands.

The first thing I did was to go through and change all
the linking jumpouts and uses. Then I changed the names
of the lessons from newbosnia, siu0, roper, and guanogap
(where newbosnia was the -used- lesson) to guanogap,
guanogap1, guanogap2, and guanogap3 respectively.

Then I went to test the jumpouts. I found that while
the jumpouts to guanogap1, 2, and 3 worked properly,
attempts to jump to guanogap produced, in order, missing
unit story in lesson guanogap, No Such Lesson, missing
unit story in lesson guanogap, and No Such Lesson.

In desperation, I recondensed guanogap, bouncing the
three authors who were in there at the time. (My
apologies to them.) Then all went well.

Anybody had anything like this happen to them?

---------- response 1
  02/23/76     16.53     woolley       p

Changing the names to guanogap1,2, and 3 allowed you
to recondense those lessons without the old copies of
the lessons (under the old names) necessarily being
deleted from ECS. However, when you changed the name of
one lesson to a name that had been in use previously
(guanogap), jumping cut to that lesson took you to the
old copy in ECS, which was actually the lesson that had
previously had that name. It therefore didn't have the
units you expected to be in guanogap.

I don't know why you would get "no such lesson"...

---------- note 224 oh come on
  02/23/76     17.34     al       mfl

In my lesson I have the define:

define board(qwerty)=n(qwerty+15)

The student is to type in the number of a square on the
board. If the number is greater than 9 (it is a ttt game
for esl students) or the square is occupied I want to judge
the response no. I used the following judge command:





judge   placex>9$or$board(placex)¬>0,no,ok

On typing "35823582" at the arrow I got an index error on
judge statement.

The question is, since in the judge statement either
condition will satisfy the conditional, why after being
satisfied by placex being > 9 does it go on to waste all the
computer time needed to evaluate the second possibility?

---------- response 1
  02/23/76     19.00     midden       s

Take for example:
judge   something=somethingelse $or$ (var‎var+1)=whatnot....

In this case you would want var‎var+1 to be done.

---------- response 2
  02/24/76     09.43     judy       pso

How about using a -long 1- for the response?

---------- response 3
  02/24/76     10.21     ahasic       med

    Instead of using on "ok" command to switch to regular
state try using "ansv 5,4". This will limit the responses
you have to deal with to 1-9. All other responses can be
picked up with a following "no".

---------- response 4
  02/24/76     10.28     frye       mfl

Al, you'll find that almost EVERY compiler that handles
simple logical statements like the one you mention will
evaluate the entire expression, then do or not do the
branch. This is uslally because it is far easier to
break the expression up into simple comparisons, then
OR or AND or <whatever> them together, THEN branch on
that result. Generally the only exception (usually) is
on one-accumulator machines, where it is more difficult
to store intermediate results.

G. David

---------- note 230 no dice
  02/23/76     20.54     schramm       cs317

I've got a real strange bug in my program. I have a simple
-unit- statement that reads

     *** previous unit ***
     unit    3a
     *** the unit ***
     *** next unit ***





but when I condense the lesson, I get an error message that
says "unit 3a" is an uninterpretable line. What's wrong?
It's sure screwing up execution of the lesson.

---------- response 1
  02/23/76     20.58     midden       s

you already have unit 3a, most likely.

---------- response 2
  02/23/76     21.34     schramm       cs317

no, there wasn't any duplicate, there weren't any hidden
chars, the typing was aok (i retyped 3 times), etc.
i think it's just one of those things that happens every
once in a while. anyway, i changed the unit name and now
everything works, so thanks to all, and if anyone gets a
revelation, please let me know. this is the kind of thing
 i'd like to avoid in the future.

---------- response 3
  02/23/76     22.28     schramm       cs317

the error was discovered thanks to monsieur midden of s.
it _was_ a dupe, and i guess that makes me a dope. I guess
this teaches me the true value of an x-search!
happy evening to all, and to all a good night!

JˇRˇˇS

---------- note 235 rewrite
  02/23/76     21.57     mailman       ustaf

        Where in "aids", if it is in "aids", is there
info on how to do that fancy little rewriting stuff
in writing notes like these? I recall it's some combina-
tion of font and access and something else, but darned
if I can ever remember what.

---------- response 1
  02/23/76     22.20     snellen       medneta

Take a look in lesson 'discuss1'; I think you will find
what you are looking for there.
--JES.

---------- response 2
  02/23/76     23.16     warner      iu

        This is an 'unannounced feature' of TUTOR:
that is, you are warned that systems can change or
remove it any time they feel like it.
        It is very handy, but not to be counted on!
Hence, it is not in aids.





---------- response 3
  02/23/76     23.40     dave fuller       uimc

        Also, it is not to be used in notes, because it
makes for unreadable printed output. Since gnotes files
will be eventually printable, the extraneous junk done by
backspacing and overwriting will look gruesome.

---------- response 4
  02/24/76     08.33     mailman       ustaf

        Discuss1 is ok; what I was looking for.
        Will SHIFT ‎ 1 or some other combination get you
_out_ of these crazy writing modes?

---------- response 5
  02/24/76     08.50     berger       mfl

"Normal" mode is mode write, therefore, SHIFT '‎' '3'
will put you back in "normal" mode write.

---------- note 236 micro drop
  02/23/76     22.00     koning       csstaff

There seems to be a very intermittent bug in the handling
of micro tables.... As a result, the microtable sometimes
does not get loaded by a -micro- command even if the
microtable in question exists.

---------- response 1
  02/23/76     23.56     blomme       s

   Is it the case that "zreturn" is being tested and
shows the microtable loaded, when in fact it is not?
(Or isn't "zreturn" being tested to confirm loading?)

---------- response 2
  02/24/76     09.41     little buddy       matha

in our experience, zreturn _is_ set to indicate that the
micro table did _not_ get loaded. It is not repeatable and
probably due to lack of space at the moment of execution.

LB

---------- response 3
  02/24/76     11.24     koning       csstaff

Well I don't know about lack of space. There is no test for
zreturn but the micro command says -micro cslibrary,fortran-
and that microtable is there and has been there for a long
time. This note was prompted by a report from a student
in cscomments, so I can't tell how ecs was. I have had
the problem once last semester shortly after a message
'system test-may crash' was sent, and rumors were that this





tes was the running of 800 pseudo-terminals, so I can see
that there would have been a shortage of ecs then...

---------- response 4
  02/24/76     18.23     blomme       s

  If you can tell me of a positive case where you
checked "zreturn" and it said the micro table WAS loaded
yet you found that it was not, then I will look at the
code. If you don't bother to check if a micro table
has been loaded, then that is YOUR error.

---------- response 5
  02/25/76     12.52     friedman       csa

But see Rader's response to note ¬$267.

---------- note 246  CR¬+SS
  02/24/76     01.51     platt       cornell

There seems to be a bug in SS. When you use "t" to enter
a line with a CR in it, and press SHIFT-BACK, you get
an execution error in unit toedit 4th command -- a findall.
Is there a way around this?

---------- response 1
  02/24/76     02.15     midden       s

Thanks to Tim, fixed.

---------- note 249  seminar
  02/24/76     08.45     golden       s

PLATO  SEMINAR

Pual Tenczar
Head, System Software, CERL

Title:  An Account of how the PLATO/TUTOR
        Man-machine Language Works

Wednesday, February 25
Room 198, CSL, 3:00 pm

---------- response 1
  02/24/76     09.21     deiss       phar

For the benefit of those of us remote (or at least not
within walking distance), I request that lectures of this
type be taped and made available through the PLATO pub-
lications catalogue.

---------- response 2
  02/24/76     09.57     hody       med





hear hear!

---------- response 3
  02/24/76     09.57     artman       uimatha

See responses to note ¬$0, this file.

hopefully helpfully,

eric artman

---------- response 4
  02/24/76     23.13     kent       unl

...I only wish I could hear here....

Kent Wendler

---------- note 251 t,nelson?
  02/24/76     09.14     churches       iu

        Can anyone tell me how to get in touch with Theodor↓Nelson? Apparently he was at the Chicago circle in
1974.....Maybe The Chicago people have their own note file ↓...what's the name of it??
DC

---------- response 1
  02/24/76     09.31     berger       mfl

Ted Nelson is on sabbatical now, I think.

---------- response 2
  02/24/76     12.37     lieber       m

        Try writing to him at:

        University of Illinois at Chicago Circle
        P.O. Box 4348
        Chicago, Illinois 60680

        I've seen him around but he is seldom on Plato.

---------- note 253 condit'nl
  02/24/76     09.20     judy       pso

specifically to nate/iua202,  but in general to everybody:

We (pso) keep getting questions and comments about
the tags of conditional commands, and we do try to clarify
the AIDS explanations ...... but:

¬D ¬D  ALL  ¬A ¬A conditional commands have the form

command (expression),what to do for -1,what to do for 0,





what to do for 1, etc.

The commands -match- and -keytype- are not really
conditional commands because there is NO expression eval-
uated to decide "what to do". Instead, something in the
extended tag is evaluated (the word or the key value)

in order to decide what to put INTO the variable. The
is just the reverse of the way a conditional command works.
"-1" has to be reserved for the "not found" case.

match (variable),cow,horse,pig,abracadabra

---------- note 264 nookno
  02/24/76     09.54     eyl       mtc

   I have an arrow preceded by a -size- command and the
student's response appears in sized writing as I desire.
There is also a -specs-nookno- in effect, and the markup
does NOT appear as you would expect. Upon entering an
incorrect response and pressing -NEXT-, the student's
response is erased (so far, so good). However, at this
point the markup, "no", is written in mode erase in sized
writing just after the students response. Apparently, PLATO
forgot that it did not do the markup in the first place,
and does not now need to remove the "no". This apparently
happens only when using other then -size-0- at an arrow.↓
   I only noticed this because I had other writing
displayed at the location where the -no- usually appears
and it was being messed up. Any chance for a fix?
Thanks,
A.W. Eyl

---------- response 1
  02/24/76     13.11     judy       pso

You might try using
-noword     -(blank tag)
instead of the -specs nookno-

---------- response 2
  02/25/76     10.17     guerra       uimc

Yes, there are ways around it, but if it IS an error, it
SHOULD be fixed, regardless of whether you can sneak your
way around it or not. After all, specs nookno and noword
do two different things: specs affects just the arrow you're
at, but -noworo- remains in effect unit you either
encounter another -noword- or leave the lesson, so that
if you have something like this:
back    review
arrow   anywhere
noword
answer  whatever
endarrow
noword  no
...and the student hits BACK before answering the question,
the noword is set to blank for every arrow in the whole
rest of the lesson, unless the student is fortunate enough
to encounter another-noword- somewhere. Thus, there is a





very important difference between specs nookno and noword
and I think care should be taken in advising people to use
the latter in place of the former.

---------- note 267 micr no ld
  02/24/76     10.10     friedman       csa

With reference to note ¬$236 about a micro table not being
loaded, I'm looking for suggestions:↓
What does a lesson DO when zreturn after a -micro-
command indicates the micro table not loaded?
Is the following code reasonable?
1try
micro   lesson.block
branch  zreturn,1ok,z
return
branch  1try
1ok

By the way, it never occured to me before to test zreturn
after a -micro- command; I know the micro block exists,
so it will be loaded. Perhaps with ECS the way it is,
that assumption is invalid.↓
Also, Elaine Avner's "Summary" (2nd edition) does not
mention zreturn (or error) re. -micro-. Is this a new
feature of the -micro- command?

---------- response 1
  02/24/76     10.41     e avner       pso

George,
Regarding your last point (about "error" not being mentioned
with -micro- in the second edition of the Summary of
Commands): it is in the third edition. It may have been a
new feature, but more likely I just missed it. I picked it
up with the conversion to "zreturn". Sorry if the omission
caused any problems for you.
Elaine

---------- response 2
  02/24/76     11.08     friedman       csa

No real problem, elaine. Thanks for the note. I was just
trying to pin down a historical fact, but it's not an
important one.↓
Anyone got any good ideas on the question what to do
with a zreturn ¬= -1?

---------- response 3
  02/24/76     11.44     judy       pso

According to marty/mfl (who has a BIG interest in micro-
tables), the non-loading of microtables was a real problem
which Blomme fixed last night. Any new problems should
be reported!

I don't know about using zreturn to try a reload.





---------- response 4
  02/24/76     18.31     blomme       s

   No, the problem with micro tables mentioned by
Judy has NOTHING to do with this. The new term handling
temporarily had an adverse affect by clearing a flag
it shouldn't have--this was fixed immediately.
   The error return indicating loading/non-loading of
a micro table has been present in the system ever since
the micro command existed I think--however, I believe it
is the case that this particular error return was not
mentioned in aids for some time. I seem to recall finding
that omission when I was adding "zreturn"s everywhere.
There are several reasons why loading might fail, the
most common/obvious being the non-existence of the
specified lesson or of the micro block. I would think
looping to re-try would be appropriate, provided some
sort of indication of the failure and some means of
"giving up" were provided.

---------- response 5
  02/24/76     22.37     rader       s

After a brief check through the code, it would seem that
looping on "zreturn" is not the best idea--the shortage
of ecs situation produces internal looping (which you
could escape with a stop1), not a zreturn value. The
other error exits appear to be of a nature that should
be reproducible--such as microtable not found. All I can
suggest at present is that instances of microtables not
being found when they are in fact there be reported,
including circumstances. A system/pso person should
be contacted to look at the problem while it is still
fresh, if possible.

---------- response 6
  02/25/76     12.51     friedman       csa

Thank you both for your responses.

---------- note 273 join depth
  02/24/76     10.42     frye       mfl

This sounds like a simple one:↓
I have several places where it would be handy to know
just how deep I'm nested in -join-s or -do-s. A system
reserved word that would describe join stack depth would
be greatly appreciated. Anyone else need it? One exam-
ple of use would be:↓
unit nest $$ can either be do/joined, or by itself
branch jsdepth-1,z,0ncplot $$ if joined then skip writing
write Some Text
0noplot
do nest $$ recursive..
*    other code↓
What are the odds? A reasonable substitute would be a
logical true/false that describes whether this particular
unit was join/do-ed or is the base.





G. David Frye

---------- response 1
  02/24/76     14.03     berger       mfl

It would be extraordinarily helpful.

---------- response 2
  02/24/76     18.33     blomme       s

Simple enough to do, but helpful for what?

---------- response 3
  02/25/76     09.02     jmk       pso

Aside from Frye's example, it would be very helpful as
a debugging aid.

---------- response 4
  02/26/76     20.48     fritz       amesrad

Indeed it would! I (along with several others) spent a
looong time trying to trace down a bug in one of my
lessons. It turned out that the join stack was getting
zapped after a specs. We finally figured out that something
funny was happening to the stack, but had to add in a whole
bunch of purposeful exec errors to check the status of the
stack. Messy.

---------- note 275 U list?
  02/24/76     11.20     parkep       arizona

I was just looking at the 'users' listing and noticed
something really weird. It seems that the users listing
lists only SOME students and multiples, not all. And
also some stations have blank entries.

What is going on? Is the users listing going to change
so that all students and multiples are listed?

Mike Parker

---------- response 1
  02/24/76     11.37     artman       uimatha

I also noticed that 0-0 was listed. No big deal, but
can you imagine trying to talk to '' of ''?

Just so I don't see a 'suppeme being' of 'p' on system,
I feel fine.

just another dummy,

eric artman





---------- response 2
  02/24/76     12.57     harkrader       o

Seems to be a bug in user...

DPF, where are you...

---------- response 3
  02/24/76     13.03     golden       s

Students (including multiples) should not show up in the
user lists.

---------- response 4
  02/24/76     13.08     frankel       p

fixed.

---------- response 5
  02/24/76     13.23     parker       arizona

Thank you....

---------- response 6
  02/24/76     14.04     berger       mfl

It was more interesting the other way.

---------- response 7
  02/25/76     09.19     railing       iu

  Yeah...especially when I saw stud/playdo listed
the other night. Gee...imagine changing it to an
author record, and leaving the password the same.
  The mind boggles at the effect.

Malkin

---------- response 8
  02/26/76     20.48     fritz       amesrad

At least then there would only be one of them...

---------- note 281 crash note
  02/24/76     12.15     roper       siu

Just before the system crashed, I was reading one of my
notes. Part of the text (about 3 lines) started repeating
at various screen locations, filling up the screen.
Shift-STOP wasn't active, nor was anything else. I
finally powered off the terminal to escape.

Hope this information may help.
Quetzal





---------- response 1
  02/24/76     12.38     curly       iu

  Same thing happened to me the last two times the
system crashed. First time i was reading a note
in a gnotefile. second time I was in the editor.
Both times bout 5 lines of writing repeated themselves.
You arent alone Quetzal.

curly≤≤≤≤≤¬↑ˇ↑ˆ↑ˆ↑ˆcurly

---------- response 2
  02/24/76     12.40     fritz       amesrad

  This happens when the program which sets up output for the
terminals dies, and the program that sends it to you doesn't
know about the untimely passing. It thinks it's sending out
all kinds of nice new goodies for you, when actually it's
the same old stuff all over again.

---------- response 3
  02/24/76     15.12     sacco       conn

Yes, but has this kind of crash become more common as of
late? I've had the same thing happen several times
recently, and suddenly many people seem to be noticing
this sort of thing.. Perhaps this indicates a recurring
glitch??

jsac

---------- response 4
  02/24/76     22.41     rader       s

This has been our only sort of software crash recently...

---------- note 293 variancopy
  02/24/76     14.13     harkrader       o

Attention: CERL VARIAN COPIER users

    You may now reserve time at the CERL Varian copier
through lesson Copy'. At the moment, you may schedule
time in 15 periods during the next 7 days. You may
enter the lesson directly or through the print request
lesson. Thanks to Tim Fay for polishing up the rules.
More information in the lesson.

---------- response 1
  02/24/76     14.27     stone       research

        maybe now would be a good time to include some
recommendations for use of the copier to help speed up
the process. for example:





       don't make multiple copies of a single page, as it
is cheaper (and faster) to use a xerox or equivalent on the
first copy.

        i've seen a user copy pages and pages of solid text
in size 0 rotated 90¬o so it would be written on the page the
long way. if it really matters, it can be cut, rotated and
recopied using more conventional methods.

        debug the program that produces the display before
sitting down at the copier.

        anybody else have some favorite recommendations?

---------- response 2
  02/24/76     14.30     mike capek       medsiuc

it seems that one cannot remove his own reservation in
program 'copy'...it says: sorry you cannot do that

---------- response 3
  02/24/76     16.30     harkrader       o

Fixed the bug...

---------- note 302 rorigin
  02/24/76     16.16     white       p

A brief word of warning:

-rorigin- with a blank tag currently sets the origin for
relative commands to (0,0). Soon this will change to
set the origin to the current (wherex,wherey).

(More additions and changes to the display commands
are planned... this is the only change that affects
current commands, and thus needed to be mentioned.)

---------- note 303 day=date
  02/24/76     16.24     michael       english

Considering the number of such routines in existence, is
there any chance that the system could provide a comple-
mentary set of commands to do the day ¬A¬D date conversion?

daydate v1,n2   $$floating day in v1 ¬D alphameric date in n2

dateday n3,v4   $$alphameric date in n3 ¬D floating day in v4

The task is too complex and costly ECS-wise to code into a
lesson unless absolutely needed; many people just store
both values instead.

---------- response 1
  02/24/76     16.36     clark       lawyer





I don't see that it is too much trouble to just have
both a day and date command. Combining those two into
one would only cause more confusion.

---------- response 2
  02/24/76     17.43     frye       mfl

What gary is asking for, clark, is NOT a combination
command (one that does both functions), but a conversion
function -- takes *date* and converts it to *day*, and
vice versa.

G. David

---------- response 3
  02/24/76     19.15     clark       lawyer

Then what savings is there in space if, as Gary says, one
just stores both values??

---------- response 4
  02/24/76     19.28     halvorsen       pcp

I'd love to have something like this. I have several
lessons that have routines to do both conversions
from date to day and day to date. It seems strange to
have a command like -day- and not provide something
so that we can get a readable date out of it without
having to write some big hunk of code which handles
leap years, etc.

---------- response 5
  02/24/76     19.30     halvorsen       pcp

Gary didn't say there would be a savings in space to
save both! There would be a savings when those users
who now store both values will only have to store one.

---------- response 6
  02/24/76     22.44     rader       s

This is an idea whose time was ripe a few years ago; it
just has never been done. When will it be?...Hard to
say!

---------- response 7
  02/25/76     08.44     avner       s

For those who wonder WHY one would want the v-format
("julian") date instead of the alpha format, consider that
the julian day permits one to specify the date and time
(to a fraction of a second) all in one word and thus
allows easy computation of elapsed times with special
checks to prevent errors that can occur when data are





lost (e.g.--a "stop" time that is one hour earlier than
the "start" time--because it is actually from the following
day and the other times were lost). Even knowing that a
student has not been present for 7.4 days, it is nice to
also know what the last calendar date of attendance was.

---------- response 8
  02/25/76     08.52     judy       pso

Such a -daydate- conversion would be very handy.

A thought... should -daydate- require a "v" type variable
for the "day"? I have had consultees with problems because
of round-off in "day"; if "day" is allowed to go into an
"n"-type variable. should it be rounded or truncated?

---------- response 9
  02/25/76     11.21     alan       reading

I should think anyone storing day in an n variable would
want it truncated.

---------- note 306 exec error
  02/24/76     16.41     cole       langlab

    Why do I get a execution error when I do a
randu n1,-10. There was no mention of this in
aids under the randu command. Should this happen?

D. Cole

---------- response 1
  02/24/76     17.05     clark       lawyer

It says in the third paragraph in the write-up on
randu in AIDS that an execution error will result
if the second tag is negative.

---------- response 2
  02/25/76     08.45     avner       s

Use a calc to convert the results to negative values after
the randu, if this was desired.

---------- note 308 varloc
  02/24/76     17.39     bowery       comm

define test
       F=n10
       Funct(i)=(i≠5)
unit   test
at     1010
show   varloc(Funct(F))
*
*
*





*this unit shows "10" at 1010 ..... perhaps
*a condense error is in order here instead?
*
*it might help debugging.....

---------- response 1
  02/24/76     19.06     ian       conn

reason is that:

varloc(n10*2)=10
varloc(2*n10)=0 !?

 and therefore:

varloc(n10+2)=10
varloc(2+n10)=0

now, HERE(¬W) is something to be explained!!

---------- response 2
  02/24/76     19.32     clark       lawyer

A very subtle error! Apparently PLATO does not like to
have operators in the parens of the varloc function. It
treats the operator as if it were a paren! That is why
varloc(4≠n10)=0 and varloc(n10≠4)=10: in the first one,
PLATO takes varloc of 4 (which = 0)...in the second one,
PLATO takes varloc of n10 which is 10. NOW, in your
define, you have Funct(i)=(i≠5). Then you have in
the varloc, Funct(F), which is, rewritten, n10≠5. PLATO
sees the n10 and quits. If you change your define to
read Funct(i)=(5≠i), then, varloc(Funct(F))=0! Perhaps
there should be a condense error when operators are put
into the varloc function.

---------- note 319 user list
  02/24/76     21.20     larry       mathw

    I hate to bring up the user list again, but why force
the choice of site on the first page? I like the option of
looking at individual sites, but the way it is now, it is
slow and can be accessed once in the user list anyway.
Maybe I am just used to the old user list, but perhaps
others feel this way too. Responses?

Larry Stinson

---------- response 1
  02/24/76     22.13     hagerman       ames

    I like it this way...personal preference, I guess.
    Sauron

---------- response 2
  02/24/76     23.00     luke       cornell





       I agree that its a an annoyance to have to
make that choice before getting into the users list.
       I dont really see the advantage to this manner
anyway

Luke Kaven

---------- response 3
  02/24/76     23.41       daleske      amesrad

The change to this set-up is twofold better. Allowing
the choice of sites to be displayed allows one to skip
the uselessly large amount of listing of author-users.
This cuts down a lot on system's overhead. The other
is that a person does not necessarily want to see who
is at site 0 out would only like to see those on at site
27. This saves on both user time and system time(re-
sources).

Now, having the choice directly available upon entry to
the U option skips the unnecessary display of the first
page of users here the option to not be implemented until
after the entire first page had been displayed.... It is
a simple thing to choose a complete list by one keypress.
John

---------- response 4
  02/25/76     00.01     artman       uimatha

Yes.

The one NEXT press needed to get to the old type listing
is much more convienent than having to wait for one page
to plot. The greatest good for the greatest number,
I guess..........

eric artman

---------- response 5
  02/25/76     10.14     mike b       cornell

        Personally, I would prefer the ability to the
start listing users _from_ a particular site.

M¬ ike≤≤≤≤ˇB

---------- response 6
  02/25/76     10.29     rowlen       mflu

       I like it the way it is now. Surely we aren't
       so lazy that we can't press -NEXT- an extra time
       to get to the original listing.

       The current method is much more convenient when just
       wanting to check a given site to see who's there.





Jim

---------- response 7
  02/25/76     14.15     hinton       ssu

I would like the new way better if there were a handy site/
course directory, so one would know what sites to access:
as it is, I have to run all the sites to find out if some
particular person is on, since I have no idea which sites
are likely to contain which courses.

---------- response 8
  02/25/76     15.33     copland       csa

        I agree with Mike (see responce 5), I too would
like the idea of specifying an initial site number, from
which, the list would begin, but instead of only listing
that site, continue to list for the entire length of the
screen. Also, I think it would be nice if when going
through the list one site at a time, if when you hit
next to go to the next site, you were taken to the next
site with terminals in use by users who have their users
flag on. This would stop the displaying of totally blank
screens.

---------- response 9
  02/27/76     12.45     saltan       uicc

  I agree with the last response. After all why not!

---------- note 328 piecewise-
  02/25/76     01.13     kent       unl

        I am interested in the situation where an author
wishes to provide a student with the ability to enter a
piecewise continuous function. Now the only way I know of
doing this at the present is a rather messy structure of
conditionally executed -compute-s. Therefore I would like
to propose the following new command and modification of
an old command which I think would greatly facilitate this:

        First I would like to propose an -index- command:

index ndex,xval,vector

Here, vector is a vector containing the values of the
boundaries of each sub-interval of the piece-wise contin-
uous function. The command would "search for the location"
of xval in vector and store the result in ndex: ndex=-1 if
xval not in range; ndex=0 if vector(1)¬<xval¬<vector(2); etc.

(continued in first response)

---------- response 1
  02/25/76     01.22     kent       unl





       Next I would like to propose the following modifi-
cation to the -calcs- and -calcc- commands:

calcs ndex,y‎,,<pointr1>,<pointr2>,<poin...

Here, pointr1, pointr2, etc., are the pointers set when
the -compute- command is first executed (the fourth argu-
ment) and the brackets, <>, indicate that the condensed
code indicated by these pointers is to be executed; rather
than the usual meaning.

         Of course these are only suggested formats, but
I think something along these lines could be _very_ conven-
ient under the circumstances mentioned.

        Any comments, official or otherwise?

---------- response 2
  02/25/76     11.20     b sherwood       s

Sounds bizarre to me, but maybe I don't understand what
problem you are trying to solve.

In case it is related to your problem, I might mention that
one way to code a piece-wise function is like this:

-(x<3)[3xˆ2-5sin(x)]-[(x¬>3) $and$ (x<5)][-4x]-(x¬>5)[exp (x)]

---------- response 3
  02/25/76     23.52     kent       unl

        But that's not going to help in this situation.
I want to let the student be able to key in a piecewise
function: He would have to give the number of sub-intervals
(pieces)--this+1 would be the length of vector. Next, he
would key in the value of each sub-interval boundary. Then
he would key in the algebraic expressions denoting the
functional values in each sub-interval. These expressions
would be condensed by a -compute- statement.
        As it stands now, evaluating such a student-given
piecewise function is messy but with my suggestion, the
following code would be possible given that the values of
vector and the pointers had been specified as implied above:

doto  1loop,x‎val1,val2,dx
index ndex,x,vector
goto  ndex,errunit,x
calcs ndex,y‎<pointr1>,<pointr2....
      (other code)
1 loop
      Now just try to do that with current TUTOR!

---------- response 4
  02/26/76     15.29     shirer       vu

   Just try doing that with some other language!

Can't really see that such a specialized situation would
be of general interest. Remember that PLATO doesn't





promise to do everything with one line of code...but we
try to make the most used features available in the
simplest manner. I think that the multiple computes are
probably as simple a way as any other in this case.

If the types of functions are restricted to just a few
general cases...sines, exps, powers,etc....I can think
of an easier way to do it...p-note me if interested.

---------- response 5
  02/26/76     23.30     kent       unl

        You _can_ do that in at least one other language that
I know of: APL/CMS--a language which runs on IBM's VM/370.
And it _can_ do it on _just_ _one_ _line_ _of_ _code_!
        The "execute operator", T≤¬o, is a handy little
critter: it will take any input character string and
execute it as APL code--and that makes it VERY powerful!
        And as for limited usefulness; well, TUTOR is
loaded with commands--many of which are not used by very
many authors. My personal opinion is that this would be
sufficiently useful to be justified. Notice that this
isn't the only use of my ideas: They do serve to expand
the power and versatility of TUTOR.

---------- response 6
  02/27/76     15.22     mailman       ustaf

        Sigh. The philosophy of PLATO seems to be: if you
want to do something, invent a command to do it. We can
all see the results: how many commands are there now in
TUTOR? Is there any hope for a language for PLATO which is
based upon a small number of very powerful commands rather
that kludging up a new command to meet every new contin-
gency?

---------- response 7
  02/29/76     03.06     kent       unl

        It seems there is already something close to my
-index- command in TUTOR (see note no. 531). With that
suggested extension to the -find- command and my suggested
extension to the -calcs- and -calcc- commands, we would
be making TUTOR a more powerful language--without "kludging"
up any new commands, just making the old ones more powerful.

---------- note 336 more micro
  02/25/76     09.27     marty       mfl

To continue note 267 about micro tables not loading.
An occasional cause is that the student hits the STOP
key by accident while the table is loading. The lesson is
then entered with the table not loaded. In this case,
no check can be made (at least, at present) since the
value of zreturn in that instance is -1.
zreturn should be set to -1 only when the micro table
is completely loaded.





---------- response 1
  02/25/76     11.21     b sherwood       s

The STOP key has no effect on the loading of micro tables,
which are not sent to the terminal itself. The STOP key
aborts output sent to terminals, not internal operations.

---------- response 2
  02/25/76     20.37     blomme       s

   I am not quite so positive in my judgment, although a
fairly careful examination of the coding which handles the
micros and charsets, together with some testing leads me
to believe that:
(1) STOP should not affect the loading of micros--i,e.,
  that key should be ignored and any micro failure should
  be due to some other reason (and should be flagged as
  a failure).
(2) STOP _does_ stop the loading of a charset--however, the
  value for "zreturn" in this case is 1 (an error) and
  not -1 (loaded ok). Also, a chartst command executed
  after such an abort of a charset load will indicate that
  the charset is NOT loaded.

---------- note 339 no-respons
  02/25/76     10.23     kehoe       shs

...
I left a note before, but....

Is it common that plato doesn't catch many keypresses
from the editor mode?
We have experienced that problem quite often, lately,
and it now seems to be spreading to when we execute
lessons, too.

We are curious as to whether this problem is system-
wide, or only our site. or just our keyset.

Thanks,

Dave Kehoe
Springfield High School

---------- response 1
  02/25/76     11.22     b sherwood       s

At an -arrow-, or in other places?

It certainly is possible to lose control keys when
system load is heavy, if you have been doing a lot
of processing, but I would be surprised (worried?)
if you lose keys when responding to an -arrow-.

---------- response 2
  02/25/76     12.33     kehoe       shs





The keys that we lose are _all_ keys and at different
times. We find that the problem is especially disturb-
ing when we are in the editor and press 'R', 'I', etc.,
to edit and then find that no key presses are recorded.

Dave Kehoe

---------- response 3
  02/25/76     13.02     olson       ced

Your problem is not confined to your site. I'm having
the same situation occur when I'm in the editor. It's
frustrating to type F26, hit next and only have F2 or F6
done. Glad?, to hear someone else is having that problem.

Dennis

---------- response 4
  02/25/76     14.37     bowery       comm

Do you (by any chance) have the spiffy new keysets with
the light grey NEXT key?

---------- response 5
  02/25/76     21.40     cole       langlab

    You may also be having problems with heavy
autobreaking. If you type very fast sometimes
the terminal doesn't accept the keys since
there is a limit on how many keys/timeunit it
will accept.

D. Cole

 ---------- response 6
  02/26/76     01.45     d williams       unidel

   It can even be dangerous! Sometimes I type "IS2", and
it only gets "S2", which destroys the saved lines. The
precautions to take are obvious ("Look before you NEXT"?),
but....
   I, too, assume it happens to everyone; it DOESN'T just
happen on these newer keysets, does it?

   It couldn't be caused by hitting keys too quickly, since
It's usually the first keys that are ignored. (Or could
the second interfere with the first?) My superstitious
impression has always been that I hit the key too "abruptly"
(hitting instead of pressing), but don't comment on that,
because I don't believe it. ¬↑ˇ-- Dan

 ---------- response 7
  02/26/76     09.38     rader       s

Most of these editor control keys are going in at a
pause, and it is quite possible to lose one (or more)--





your best protection at present is to wait until you see
what you typed before pressing NEXT!

If you miss keys while inserting or replacing text in the
editor, you must be either typing very fast or have some
other problem (keyset, line errors,....).

---------- response 8
  02/26/76     10.40     dulle       shs

We use an old style keyset.

Dave Kehoe

---------- note 341 econ
  02/25/76     10.30     hanoka       ed

      Does anyone out there have a drawing of the
      ecological poster? ( the one with the big
      green e on it?)

      mike

---------- response 1
  02/25/76     15.42     pierce       cs196

      A green e on a plasma panel?

---------- note 342 datafile!
  02/25/76     11.03     fay       o

Thank you, Ruth, for the spiffy (and quite useful) new
data sorting options in the datafile machinery.

I hope also that systems staff is considering (soon) some
sort of facilities to accomplish the following:
1) Sorting of data by student prior to printout.
2) The analysis of summary stats (similar to the PEER
   package) as mentioned in "projects".
3) Providing some means of getting more legible numeric
   info into the datafile. The current scheme of octal
   and a VERY short field integer ¬+ floating nos. (w/out-
   putl) is, ah, frustrating and frequently practically
   useless.

Nonetheless, this most recent addition is welcome indeed.

---------- response 1
  02/25/76     11.08     darlene       pso

I agree with Tim. The new data sorting feature is going
to be useful to many. But don't stop there. Head copy from
the datafile is impossible to use now but the ability to
have it sorted and printed would be a great aid. It is





almost impossible to do all data review on line and the hard
copy available doesn't aid in reviewing the data as it now
stands.
The other points in Tim's note should also be worked on in
making the student data collection on PLATO truely useful.

Let's see the student data area of PLATO grow with the rest
of the system.

---------- response 2
  02/25/76     11.25     dvm       ee

i agree.
i often have this problem:
i want all responses to an (one) arrow in one place
AND the datafile must be flushed for subsequent S use.
if i could sort the data (by unit, say) and print it,
my life would be happier, and my lessons better.

---------- response 3
  02/25/76     12.57     chabay       s

I don't understand Tim's point ¬$3. Do you mean that
the display shown by the datafile editor isn't adequate
for outputl's?

---------- response 4
  02/25/76     13.02     friedman       csa

A related request: Please make it possible to specify
-output-s and -outputl-s separately when listing what to
include/exclude while viewing a datafile.

 ---------- response 5
  02/25/76     14.40     fay       o

To clarify my point ¬$ 3, Ruth: yes, the display generated
by the datafile editor is inadequate for showing the
numeric information generated by outputl's. Specifically,
it is only capable of showing relatively small integer
and floating values. In some cases I can see where the
current format (octal, integer, floating) is useful; in
others it is desireable only to see an integer or floating
or octal value. Perhaps some new variations on the
-outputl- command are in order:
-outputl- to dump vars to datafile w/current format;
-outputo- to dump only octal
-outputi- to dump only in integer
-outputf- to dump only in floating
-outputa- for alph
At any rate, the important thing to me in this area is
being able to see a floating or integer value carried to
a larger number of places.

 ---------- response 6
  02/25/76     15.15     mcneil a       uicc





While outputo/a/n is being considered the folowing would be
be useful:

1) Allow us to set what format to show outputl's in like
the common editor does (any combination of a,n,o,v ).

2) -reado- to read output (not outputl) from the file.
So far we almost have access to all the information in
the file but (unless I missed a new system feature buried
in replies to notes) so far no access to output.

---------- response 7
  02/25/76     16.23     fay       o

I especially like McNeil's first point. In fact, it would
probably be a lot more sensible, convenient to use, etc,
than implementing the host of different command I
mentioned.

---------- response 8
  02/26/76     08.06     avner       s

The PEER package is now being expanded and consolidated
(primarily by Tamar Weaver) to allow use of datasets for
storage, editing, and analysis of selected area and
output data (in a relativley restricted format--but at
very low cost to te user in terms of setting up data
handling procedures). This package is, of course, intended
mainly for early analysis of lesson data for initial
revisions. It is expected that each user will want to
subsequently design a tailored package to suit their own
needs (which would make use of some of the other suggestions
made earlier here). Al Avner

---------- response 9
  02/26/76     09.06     fay       o

Al: the PEER package is indeed a powerful one, but it is
very expensive (in terms of disk space and ECS) relative to
a widely used system-level lesson (like the savings involved
in using "mrouter"). The poor guy running a one part course
with a 4 part datafile and perhaps 5 two part lessons is
probably not that enthusiastic about going thru the trouble
of adding a dataset to his collection and a couple of
"dummy" lesson spaces (to contain the -use- commands).
While the PEER package will continue to be most useful to
many people, I feel very strongly that the system should
still expand and improve its own handling of student per-
formance data.

---------- response 10
  02/26/76     11.12     t weaver       peer

The PEER package no longer needs the use command or commons.
One lesson will serve all at the same time. Datasets are
the only space needed, The ECS needed to run is lower
than 3000 words, and each user will have a storage of about





1300 in the most used parts of the package. An option for
printout sumaries in reasonable looking tables is available
too. Will probably be released in April.
(More information will be available at that time.)

---------- response 11
  02/26/76     11.38     avner       s

In addition, the main purpose of the datasets is to store
data from the datafiles (in squeezed form) for periods
of several weeks. There is thus no alternative to them
given that you wish to save data from several datafiles
(in order to make comparisons over time, etc.). As Tamar
points out, the only overhead will be an ECS load during
processing. This would be present even with systems
level lessons (even if hidden from view by fictional 1500
word lesson lengths). Al

---------- note 344 red cross
  02/25/76     11.13     badnarik       iu

                        NOTICE!
The RED CROSS will be having another BLOODMOBILE next
week at INDIANA UNIVERSITY!
        Date:   March 2,3,4
        Time:   10:00am till 4:00pm
        Place:  IMU solarium
        Call:   332-7292 for appointments
Thank You,

---------- response 1
  02/25/76     11.24     b sherwood       s

I can't resist pointing out that this note will be of
use to an extremely small number of PLATO users, given
the geographical location. Can't people use a little
more judgement in filling up this forum?

---------- response 2
  02/25/76     11.44     buz       phar

actually, not much less useful than the seminar notices...
which are usually very geographically limited, also.

---------- response 3
  02/25/76     12.31     mike b       cornell

       There are a lot more PLATO users in Urbana than
there are at the University of Idiana, I'm sure.

M¬ ike≤≤≤≤ˇB

---------- response 4
  02/25/76     18.15     rowlen       mflu





       Not to mention the fact that PLATO seminars are
       directly connected with PLATO where as the
       connection with a red cross blood drive escapes
       me.

       But then again, I guess I'm fairly new on the
       system and wouldn't know.

Jim

---------- response 5
  02/25/76     20.29     curly       iu

  Please!....Its Indiana University, not
University of Idiana.....but I do agree with
the point being made....This particular note
should have been in 'iukidnotes' if anywhere.

curly≤≤≤≤≤-↑ˇ↑ˆ↑ˆ↑ˆcurly

---------- response 6
  02/26/76     22.40     wong       cs317

You guys got something against giving blood?

 ---------- response 7
  02/27/76     11.14     broadus       css

wong, we dont mind giving blood

Bob Broadus

I thought having just one type of general notes was supposed
to stop all the deciding in which file a note should go!

 ---------- note 345 I.D. aid
  02/25/76     11.14     dvm       ee

If CERL were to create an "aids"-type lesson
regarding instructional design (as the
present lesson "aids" regards the TUTOR
language), what kind of things would be





the most helpful to you?

When you answer this question, please remember
to indicate the approximate length of time you
have been programming PLATO lessons.
Thank You,
Dave Meller

---------- response 1
  02/25/76     13.05     olson       ced

An instructional design type lesson would be very
useful. I've been programming a little over 1 year.

Dennis

---------- response 2
  02/25/76     14.37     avner       s

There is an instructional design section in AIDS now, I
would be curious to know how many people
a) are aware of it
b) have used it
c) feel that it meets this need.

---------- response 3
  02/25/76     17.17     dvm       ee

I would also like to have answers to al's question.

However, please do not confuse my original question
with what is only currently available.
These are two different questions.

Feel free to answer either or both of them.

---------- response 4
  02/25/76     22.55     clegg       amesrad

        I've been programming for a year and a half, have
discovered the section on instructional design in "aids",
and also think that a lesson on ID is a good idea, altho
I've used this section in "aids" and it probably does
meet this need fairly well.
        First. It would be a big help if all the ID lessons
were indexed together. "Aids" has a list of additional
lessons on ID, eg. "style and "test".
        I'd like to see a lesson that could be used in a
CAI seminar to teach proper instructional strategies for
CAI, especially simulations and student-control of the
subject matter.
        Also, how about something for teachers who will
not necessarly be programming, but will be teaching
alongside of PLATO?

Janet Clegg

---------- note 350 ECS msg





  02/25/76     11.23     michael       english

Is there a system page reached by INSTRUCTORS who have
executed a -jumpout- with an -inhibit jumpchk- to a large
lesson when site ECS was over allotment that says something
like:

  "There has been an error -- you may choose
  another lesson or press NEXT to try again."

It may not be worded exactly like this, but is there such
an ECS message for instructors?

---------- response 1
  02/25/76     11.52     koning       csstaff

I have seen that page several times after having an
exec error in a lesson but I don't know about the
message for ECS shortage. Would suspect it is the same
though...

---------- response 2
  02/25/76     13.01     chabay       s

Right -- instructors should get a message saying:

        An error has occurred
 
        You may wish to try the lesson again or
        to choose another lesson

There's no special message for ECS problems -- perhaps
a line of explanation should be added in this case.

---------- response 3
  02/25/76     17.48     michael       english

PLEASE and the ECS message.....one of our site directors
insists that it must be a program error since it doesn't
say ECS.....

---------- response 4
  02/25/76     22.49     chabay       s

OK, instructors will now get a message stating the type
of error (execution error, not enough ECS, etc.)
(After the next system reload, that is.)

---------- note 351 ansv idea
  02/25/76     11.24     p cohen       med

Dear s(ystems):
        A very frequent use of the -ansv- command is to
see if an input lies within a certain range rather than
if it equals a number within a certain error limit.
        If an author wants to see if an input is between





¬a and ¬b, his only recourse is to use to customary:

ansv    (¬b+¬a)2,(¬b-¬a)/2

which is rather obscure for so simple an idea.
        How about another command such as:

rangev  ¬a,¬b

to cover this case. Is it possible? Any sentiments?
--paul

---------- response 1
  02/25/76     11.26     warner       iu

        Good idea, It is far more useful to define a range
of numbers thn a tolerance in most cases...
but instead of a new command, how about a different ansv
tag, say
ansv    3,5,2.5 becomes equal to
ansv    1;6 or
ansv    1,6,range

---------- response 2
  02/25/76     11.28     b sherwood       s

I certainly favor it. Your proposed name sounds ok. except
what would be the name of the complementary command which
would do a "judge wrong"? How about -ansr-/-wrongr-??

A related suggestion from users has been for a judging
command which would be matched if the tag was -1:

arrow   1215
store   x
test    x>3  $and$ ntries<5
write   ......

(Your range case is so common that it seems worthwhile
just by itself, of course.)

---------- response 3
  02/25/76     11.28     b sherwood       s

Warner's suggestion sounds good, too, for syntax.

---------- response 4
  02/25/76     11.15     p cohen       med

I also agree with warner about using -ansv-.
Er...um...mathematical usage would suggest:

ansv (¬a,¬b)

for all inputs strictly between ¬a and ¬b and:

ansv (¬a,¬b)





to include the endpoints as well.
--paul

---------- response 5
  02/25/76     11.55     koning       csstaff

I like that last notation... and the other commands
suggested look very useful; I have often found myself
doing a -store-, -ok- and -judge- just to make the
idea of a range obvious in the code.

---------- response 6
  02/25/76     13.10     maggs       law

        I have been using something along this line, though
I haven't really thought it through to see if it is fool-
proof.

define ifit
       it=v1
       if(u)=(it=0)≠(u=0)-it≠u
unit   testit
arrow  2020
store  it
ansv   if(it¬>3 $and$ it¬<8)
write  You've got it!

---------- response 7
  02/27/76     11.08     shirer       vu

   An -ansv- range is clearly an innovation which could 
benifit many people. Would the difference between
ansv 10,2
and
ansv 10;12
be easily apparent to most authors? That seems simplest
to implement.

---------- response 8
  02/27/76     11.11     ahasic       med

 I like Paul's idea too and I have some further suggestions.
It would be nice if you could specify two or more allowable
ranges in a ansv. Being able to use ­²­ and -­²­ as endpoints
of the ranges would also be usefull. Then an author wanting
the student to input values for the equation ¬[ln x/(x-7)¬]
would only need:

ansv (0,7),(7,­²­)

---------- note 359 ecs parity
  02/25/76     11.45     tamar       cerlcc

 While using a student record to transfer/crunch data from
datafile to a dataset, a system message apeared:
 System failure





ecs parity
The dataset which was involved in this, and was reserved,
kept giving zreturn as it was reserved long after it wasn't.
  When same record tried to get the dataset, the sys
crushed emidietly. (by chance?)

---------- response 1
  02/25/76     14.40     rader       s

Thank you for the report, Tamar. The first ecs error was
in part of your records, and so you had to sign in again.
The dataset was still reserved, due to the abrupt nature
of your being thrown off the system when the first error
occurred. The fact that we crashed just as you tried
again is merely chance.

---------- note 364 crashmess
  02/25/76     12.36     kehoe       shs

...
Well, we still have the problem of no visual, coherent,
and otherwise pleasing display during crashes.

We are still only receiving a flashing-to-constant red
error light. When the button is depressed, we receive
random lines and a mess of char displays.

We wrote about this before, and we understood that
something was being done. Well, apparently, no
such luck.

Dave Kehoe
Springfield High School.

---------- response  1
  02/25/76     14.20     hinton       ssu

Today (2/25/76) the down tapes were not plotting well
here at ssu either...it was the old problem of messed-up
words ("our b estimat is tht Pl will return..."), and
occasional random line plotting....

---------- response 2
  02/25/76     14.42     rader       s

Thanks for the reports. We will try and get some of the
tapes re-made, and the heads cleaned, and see if that
helps.

---------- response 3
  02/25/76     15.42     frankel       p

These things take time...

---------- response 4
  02/25/76     19.44     hody       med





and how about finally changing things so that during
reload and other bobbles the screen doesn't read
the ever welcome "press NEXT to begin" when it
doesn't mean it!?

---------- response 5
  02/25/76     22.13     warner       iu

        An observation from here at Indiana University:
the 15-minute tape is _much_ cleaner than the few-minute
tape. Simple inspection suggests that the few-minute
tape has been getting the most usage lately.
        So why don't you make two or three different
tapes with the times most commonly used...and three
or four of the few-minute?

---------- response 6
  02/26/76     02.00     d williams       unidel

   When the U of Delaware switched from "network" to
site 4, the tape reception changed-from what was
described here, to perfect legibility.

---------- response 7
  02/26/76     09.37     parrello       uimatha

And FLB is STILL a nightmare, except on physical site 13,
which comes in clearly for the most part.
As it is now, the down tapes on physicals 12 and 19 simply
exercise the slide projectors. Virtually NO line or
character output ever reaches the screen.

---------- response 8
  02/26/76     10.19     frankel       p

So, you see, there are problems in many areas...we are
working on it...

---------- note 365 Notes-err
  02/25/76     12.49     kehoe       shs

...
There is an error here in NOTES.
Message number 364 shows on the screen when this
section is first entered, but doesn't exist when
either 1) it is selected or 2) you advance forward
to it.
...
Dave Kehoe
Springfield High School

---------- response 1
  02/25/76     12.50     kehoe       shs

Kindly disregard this note.





Apparently, the lesson was only temporarily malfunctioning.

---------- response 2
  02/25/76     14.23     t little       research

Someone probably had stored a note and then decided to
edit it. During editing the note no longer exist for
anyone to see.

Todd

---------- response 3
  02/25/76     14.26     putch       arizona

Then why should its number be shown at all?

---------- response 4
  02/25/76     17.03     woolley       p

The re-edit probably began between the time the note
was displayed on the directory page and the time that
he requested to see the note.

---------- response 5
  02/26/76     10.39     dulle       shs

There was no re-editing involved....I had written the
note, but couldn't take a second look at it.

But after writing this note, the problem cleared up.

Dave Kehoe

---------- note 367 biz mgmt
  02/25/76     12.51     john m      catid

Principles of Business Management: Does anyone know of a
PLATO course which deals with the general overview of the 
subject of business management. namely the Planning,
Organizing, Directing and Controlling of the resources
of the corporation to achieve corporate objectives?
Also, is there any interest in developing such a course
on PLATO? Would appreciate any comments.

John Maloney, Consultant.

---------- note 370 conference
  02/25/76     12.56     a1       iututor





   You are invited to attend the ''Second Biennial
Conference on Computing in Indiana'' sponsored by the IU
Student/Professional Chapter of the ACM which will be held
at IU on March 13, 1976.
        The major areas of interest we plat to discuss are
Computer, Science Curricula, Security, Hardware-Software
Integration, CAI, Minis and Macros, and Artificial Intellig-
ence. This will be done with panel discussions, presenta-
tions, exhibits, etc. We will also have student paper
work sessions where students will have a chance to present;
their work. This should provide an opportunity for industry
an education to meet and exchange ideas, plans, and concepts.
        In conjunction with the conference, we also plan to 
have a Job registry and will have companies, mainly from
(continued in responses) press Lab

-------------response 1
  02/25/76     13.03     a1       iututor

the Midwest and Eastern states, participating in this.
  We are most fortunate to have as our keynote
speaker Dr. Ken Knowlton of Bell Telephone Laboratories,
Murray Hill, New Jersey. The topic of the keynote address
will be ''Computer Art and Animation''.
   If you are interested in attending the conference
please leave your name and address in pnotes of Brown/iucs.

Sincerly,

AL Ciccone
Indiana University
Bloomington, IN

-------------response 2
  02/25/76     14.17     campanini       mfl

Humble query: Is there a registration fee required in 
order to attend the conference and/or to participate in
the job registry activities?

------------response 3
  02/25/76     18.12     carter       comm

The provision to allow student speakers at this conference
is an excellent idea, When I get some more time, I hope
to encourage the Computer Science department here to hold
an annual program review consisting totally of student
presentations preceded by brief faculty introductions.
Such activities provide valuable graduate training.

However, you guys at Indiana really need to improve the
preliminary program you mailed out. At minimum, it should
contain the titles of the talks and the names of the 
speakers. Just citing broad subject areas gives the
reader the impression this event is a low-priority deal,
not worth driving over to see.

------------response 4
  02/27/76     11.13     shirer       vu





   You should also put the Valparaiso Computer Aided
Learning Committee (Valpo U 46383) on your mailing list
for such affairs. We were doing Computer aided learning
when youse guys were in rompers. Any mailing to our
"computer" lab goes to the payroll people and they
dump it in the wastebasket!

Donald L. Shirer address c/o above.

---------- response 5
  02/27/76     14.08     brown       iucs

Please read Note ¬$32 (CS Conf) in the

GROUP notes file labeled 'iuppf' for

complete info. regarding the Second

Biennial Conference on Computing in

Indiana to be held at IU Bloomington

on Saturday, March 13, 1976.
 
Thank you.

---------- note 378 lesnam
  02/25/76     13.14     errol       kka

I have a lesson in which students make comments.
The lesson shows from which lesson the student
came (uses: ¬0a,lesnam¬1). In almost every case
I get a reasonable lesson name, but a couple times
I got )

Why did I get that strange mark instead of a lesson
name?

---------- response 1
  02/25/76     14.15     warner       iu 

         First of all, there is no such systems variable as
"lesnam". Therefore I presume you are defining it as a
defined variable, and setting it when the student leaves
his lesson for the comments section.
        This variable must be getting clobbered somewhere,
that is, being altered after you enter the comments section.
Look for places. where the variable is redefined as another
name (remember that n** and v** refer to the same number.)
Especially look for a place where a variable is set to 50,
since an integer variable containing 50 will show up as 'j'.

---------- response 2
  02/25/76     16.43     friedman       csa





In lesson cscomments, I store a lesson name obtained via
the old form of -from- (e.g., -from n1-). Works fine.
I have also used successfully a lesson name obtained from
a leslist number and -lname- command (don't forget to
reference the 2nd word returned by the -lname-, or for
safety, reference all 3 words).

---------- note 409 condense q
  02/25/76     22.19     stewart       arizona

        While I was sitting in the condense queue at 56
out of 60, I began to realize how useful it would be to
be able to check the length of the queue without actually
entering it. Maybe shift-q (which now goes to aids) could
be changed to show how many people are in the condense
queue. How about it system people??

Bruce Stewart

---------- response 1
  02/25/76     22.59     curly       iu

  I ask you..How hard is it to shift-stop out of
queue?...Much less entering it?...sheesh.

curly≤≤≤≤≤¬↑ˇ↑ˆ↑ˆ↑ˆcurly 

---------- response 2
  02/26/76     08.59     fay       o

Curly: I guess the answer to that depends on whether or
not STOP1-ing out of the queue updates your condense pri-
ority or not. If not then I guess its 6 of one and half
dozen of the other. If it does update your condense pri-
ority, then I too would like to be able to see how long
the queue was before I tried it!

---------- response 3
  02/26/76     10.10     stewart       arizona

        That was my reason for asking. I had heard that
your condense priority gets updated when you enter the
queue. If this is wrong, shift-q is not needed.

---------- response 4
  02/26/76     10.28     curly       iu

  As I understand it, the updating of the condense
priority is dependant on the size of the lesson you
are trying to condense. If you never reach the
condensor, the size of the lesson is unknown, and
updating prioritys is difficult. Therefore, I assume
stop1ing out of the queue is harmless. Someone correct
me if my logic's wrong.





curly≤≤≤≤≤¬↑ˇ↑ˆ↑ˆ↑ˆcurly

---------- response 5
  02/26/76     11.49     al       mfl

I suggest that "Q" be left alone and go to AIDS as always.
Why not use "q", since EDIT doing a jumpout q won't do
any good.

---------- response 6
  02/26/76     15.10     golden       s

Your condense priority depends only on the amount of
condensing you have done today compared to the time you
have been on and the same facts for the others who wish to
condense. You lose nothing but a key-press by entering the
queue and then leaving.

---------- response 7
  02/26/76     15.18     stewart       arizona

Thanks, 
That is all I wanted to know.

---------- response 8
  02/26/76     19.45     dave fuller      uimc

        But how much set-up time is wasted with a stop1
condense request? And how many times will you get an
uneeded condense when all you want to do is check the
queue length?

        Could someone upstairs kindly answer my babbling?

        Also, how many of you out there could benefit
from such an option?

---------- response 9
  02/27/76     12.34     battenberg       css

I'd like it! There are times when I am editing that I would
like to know how long the queue is WITHOUT actully
condensing a lesson. There are times when I get into the
queue and find that I am 1 out of x and try to shift stop
out because all i wanted to do was check the queue. Often
I am not quick enough and I end up condensing a lesson when
I am not ready to.-----Thanks------rex

---------- response 10
  02/27/76     13.25     kent       unl

   It's been my (admittedly relatively limited)
experience that the only time the condense queue has any





substantial length is immediately after a system crash
when EVERYTHNG  has to be recondensed--all at the same
time. Other than that, I can never recall finding it
longer than 2 or 3; which I regard as insignificant.

---------- response 11
  02/27/76     15.11     mailman       ustaf

         You've been lucky. I've been backed up as far as
15 without benefit of a system crash.

---------- response 12
  02/27/76     18.11     tricorn       hebrew

Why would someone want to know the length of the queue if
they weren't ready to condense???
              Δ

---------- note 417 nooknook
  02/26/76     03.50     blomme       s

    On the latest version of the system, the problem
mentioned regarding the erroneous erase of ok/no
judgments in sized writing when specs nookno is in
effect has been fixed. If you don't know what on
earth this is all about, that's ok, because this condition
had (apparently) never been noted by anyone for the
many moons the problem has existed.
   Thanks to A.W. Eyl for pointing this out.

---------- response 1
  02/26/76     10.44     judy       pso

Good deal, Rick! thanks....

---------- response 2
  02/26/76     11.19     eyl       mtc

    Good Grief! Does no one else use sized writing at an
arrow with a specs nookno?????Strange!!

---------- note 423 tapecontes
  02/26/76     10.00     fay       0

     *****DOWN TAPES CONTEST*****

Yesirree, folks! It is time to remake some of those old,
worn out and all too familiar PLATO DOWN TAPES!!! Before
replacing some of the worn out tape with fresh tape contain-
ing the same old stuff, we thought it would be a good idea
to sponsor another contest! So, for your entertainment and
convenience; we have resurrected lesson "tapentry".





Enter "tapentry" in the author mode and read all about it.
(Incidentally. the original contest winners are going to be
kept around: we're just going to supplement them wiht a
little fresh blood.)

Have fun!

---------- response 1
  02/26/76     11.15     meers       wright

         Will reentries of losers in the firstcontest
be considered? I thought some of them were very good,
but only 5 or so made it to the tapes.

---------- response 2
  02/26/76     20.24     frye       mfl

When does the contest end?

---------- response 3
  02/26/76     21.10     warner       iu

  Are linesets allowed?

---------- response 4
  02/26/76     22.35     koning       csstaff

Could someone put the entries of the old contest
back somewhere? I had one that would need some work
to conform to the new rules and I would like not to
have to type the whole thing back in. Thanks...

---------- response 5
  02/26/76     23.46     kent       unl

         ..So I try to enter "tapentry" and it bounces me
back to the author mode page! Gee, Thanks!!!!...

---------- response 6
  02/27/76     09.52     parrello       uimatha

Press NEXT after the name, not DATA.

---------- response 7
  02/27/76     13.27     kent       unl

         (...talk about consistancy on this system! *grouse*
*crab*)

---------- response 8
  02/27/76     13.46     fay       0

The contest will run until Friday, March 5, at 5 PM.





Sorry, the older, non-winning entries from the first tape
contest were not saved.

Until the system allows the conversion of -linesets- into
a series of commands (like -char- commands), we will not be
able to allow the use of linesets in the contest. The
entries must be completely self-contained and we cannot
afford to have them dependent on other lessons (for char-
sets or linesets) nor do _we_ want to carry around a large
library of charsets and linesets.

---------- response 9
  02/27/76     17.41     obrien       uicc

Really?

I would have thought that the tapes were cut by recording
the audio going to a terminal. Just how DO you make
them. anyway?

---------- note 426 jumpoutkey
  02/26/76     10.19     bob y       reading

what does a jumpout do to the value of -key-?

*
pause
jumpout key=next;q;x
at      810
showo   key
*

if you are an author, this executes as i would expect --
showing the true value of key, and jumping out if i press
next. but as a student in a router, i get 0302 every time,
even if i press next.

is that the way it's supposed to work...?

---------- response 1
  02/26/76     11.00     white       p

Incidently, it only does that if you are a routed
student, in your own router.

Since you don't expect that to happen, it is not what
should be happening. However, when I first saw it
four months ago, I was hesitant to change something
that was very deliberately coded that way, and I knew
nothing about.

---------- note 437 printreq
  02/26/76     12.53     luke       cornell

         Question:

         If you request a print of a file, what version
of the lesson gets printed? The version that exists





at the time the request is made, or at the time the
request is actually carried out/
       For instance, if one is pressed for space and
wants an archive print of his code before he deletes
it, can he make his request and delete the code, or
must he wait until the print is actually made?

        Luke Kaven

---------- response 1
  02/26/76     13.02     golden       s

You put in a request. The actual printing happens several
hours later. The file will be printed as it is then. In
general, I wouldn't delete anything I valued until I had the
print in my own hands.

---------- note 439 defines
  02/26/76     13.10     al9000       german

How smart is the condensor?
If I define the following:
*
define test
       start =12
       array(i)=n(i+2+start)
*
will the condensor change +2+start to +14?

---------- response 1
  02/26/76     13.17     midden       s

Yes.

---------- response 2
  02/26/76     15.23     stewart       arizona

         Is this a new change?? I had always heard that the
defines were a straight replacement. So that 2+start would
be replaced by 2+12 and not 14. The reason I ask is that I
tried to use a similar define in an array, and the condensor
didn't like it.

---------- response 3
  02/26/76     15.32     midden       s

Well, it likes my !arrays" (don shirer's matrix arrays).
I don't know where you had problems, but using defined
constants works.

Yes, condensing is just text substitution, but later when
the text is compiled into machine code it optimizes
constants like 2+12.
                M.Midden

---------- response 4
  02/26/76     22.28     koning       csstaff





If condensing is just text substitution, why the requi-
rement that the substituted string must make sense when
enclosed in parentheses, which means that
        define plus=+
is illegal? This is not a useful example, but I have
had cases where the ability to do this kind of thing would
have been quite useful...

---------- response 5
  02/27/76     17.45     white       p

Paul, because it is just text substitution, if we didn't
insert the parenthesis, the following might occur:

define index=base+offset
...
calc   n1‎index*row

which would be: n1‎base+offset*row
or, after operator priority: n1‎base+(offset*row),
which is clearly not wanted in most cases.

Thus, we insert the parenthesis around our internal
representation of the text before substitution.

---------- note 441 help readr
  02/26/76     13.42     maggs       law

Does someone have a routine I could adapt for my own use
which reads student records, saves soem of the variables,
computes overall averages, then formats the whole thing
in common for printout (or a routine that does some of
the above?).

---------- response 1
  02/26/76     15.59     galcher       uimatha

I don't know about for common, but if you wish to store
the data in a dataset, Silas Warner wrote lesson "reader",
which formats up individual and average course statistics
and writes then out to a dataset.

    Bill

---------- response 2
  02/26/76     16.09     campanini       mfl

Contact hart of mfl by pnote re: -lsort- (may be just
what you're looking for?)

---------- note 442 term-thanx
  02/26/76     14.02     marty smith       mtc

i can't remember seeing the thank you for this anywhere,





and since it does deserve a heap of thanks, (in my humble
opinion, of course), THANKS for the way the -term- feature
remembers where you were while editing!! in other words,
if you*re in the middle of a line of code, and someone
pages you, the line doesn't get wiped out when you exit
the term machinery. BRAVO.

---------- response 1
  02/26/76     16.28     hinton       ssu

I hadn't had a chance to notice this, or I would have
thanked people too. Now I don't h≤ ave to refuse a call
just because I am editing...only for other reasons.
                ndh

---------- response 2
  02/26/76     21.56     blomme       s

    Todd Little and I are glad that there are people who
find (at least some of) the changes we made to the term
handling an improvement. We'll probably make a few
related changes in the near future.

---------- response 3
  02/27/76     09.30     jmk       pso

I gather that this change will assuage the typical
user of "consult" who, after laboriously creating a line
of code, rashly presses TERM for the consultant's evaluation
of his efforts (while still in "insert" or "replace" mode),
and discovers that it was all for naught.

The consultants thank you, and I'm sure the consutees do
too.

---------- note 452 buff alt
  02/26/76     17.21     bowery       comm

There is presently no way to tell if a -put- type command
has caused any change to the judge buffer short of -storea-
ing before and after the execution of the -put-.

Could some flag or sys var be set to indicate a change to
the judge buffer has occured?

---------- response 1
  02/26/76     17.34     rick       mfl

You could save "jccunt", as below:

arrow
...





no
calc   before‎jcount
judge  continue
putd   +.+ dot+
no
calc   after‎jcount
...

---------- response 2
  02/26/76     17.53     elston       rhrc

Of course, jcount does not tell you a change has been made
if the replacing character(s) is the same length as the
original object.

         2c≤/

---------- response 3
  02/26/76     21.07     fritz       amesrad

Which was exactly what bowery was doing...

---------- response 4
  02/27/76     11.24     tenczar       s

yes. there seems to be some real need here...will put the
job on my list...but again. don't hold your breath

---------- response 5
  02/27/76     19.14     bowery       comm

thanx

---------- note 455 score err
  02/26/76     18.21     yamada       mxc

under what circumstances is score an illegal command?
will a computation involving division by zero cause an
execution error (listing illegal score command)?

---------- response 1
  02/26/76     19.51     frye       mfl

Yes, a division by zero will cause an execution error in
-score-. In fact. any number greater than 100 will do the
same thing -- on this machine, a divide by zero produces an
infinitely large number in some cases (i.e. 1/0).

G. David Frye.

---------- note 456 tape talk
  02/26/76     19.13     j thiher       amesrad





Well the person whom I depended on to tape Paul Tenczar's
talk didn't get the job done so ...

Is there anyone out there who taped his talk? We here
at Iowa State would like to hear it and would have a
courier pick it up at your locale (or pay mailing charges)
Help out anyone?
Reply via p-notes.
       jim

---------- response 1
  02/26/76     23.47     b sherwood       s

My guess is that Paul's lively and highly graphics-oriented
talk won't come through very well without the slides....

---------- response 2
  02/27/76     08.56     berger       mfl

Al Groupe, al - mfl has a tape of the seminar. Perhaps you
can convince him to make you a copy. He likes Garcia's
pizza...

---------- response 3
  02/27/76     13.16     al       mfl

I like Papa Dels' pizza better.

Actually, the tape didn't come out very well at all. We
made a trial copy here to send to Eisenberg and the copy
was almost inaubible over the tape noise. Rick Simkin
(rick/mfl) is planning to transcribe the tape and put the
text into a tutor lesson, at least temporarily. With
this medium, we could also duplicate the slides on the
panel and try to reenact the speech as best we can.

---------- response 4
  02/27/76     14.46     j thiher       amesrad

All things considered, that sounds like the best solution
to the "problem". Thanks and keep us posted.

In light of all this furor perhaps Paul should consider
a speaking tour?? (chuckle)

  jim

---------- response 5
  02/27/76     17.45     obrien       uicc

Somehow I can see a revival of the Chautauqua.

---------- note 466 indent?
  02/26/76     22.28     j wilson       unidel





Is indenting still an unannounced feature of TUTOR?
If so, what more needs to be decided/done? Would
it be reasonable for the C-search to find the "command
field" wherever it may be?

---------- response 1
  02/26/76     23.49     b sherwood       s

I THINK it is still "unannounced" -- Paul?

And no one has treated Judy's point that a slip by a
student from micro-o (¬o) to micro-i (¬D ) makes it
very hard for the student to erase the error. Add my strong
vote for putting that thing on some other key!

---------- response 2
  02/27/76     08.31     judy       pso

How about micro-‎? by analogy to micro-/ turning the

I assume that the command field search is fast because it
only examines b letters. I would rather use the X-search
for those times I was looking for indented fields, and
keep the command field search efficient.
/ around to ¬/.

---------- response 3
  02/27/76     09.35     jmk       pso

micro-tab seems to me the most logical place to put the
indent symbol. It also seems the least likely to conflict
with anyone's existing micro definitions.

---------- response 4
  02/27/76     10.00     k mast       p

Actually, I think the command field search would be just
as fast as it presently is. It uses a -find- command, thus
it can only find commands whose first character is the
first character in a word. This is the case in the
indenting.

---------- response 5
  02/27/76     11.34     tenczar       s

indenting has such huge ramifications that we are going
very slowly and trying to find the "best" way of doing it.
the current "micro-i" method is just an attempt to see
what one sort of implementation would look like/invlove.
it is incomplete in that it needs further programming in
the editor with respect to many of the search options and
needs a lot of work in our system "change" programs that
change all TUTOR source code to a different form once we
decide upon changes to TUTOR (e.g, like the soon to come
removal of the x-y-coordinates from the tag of the -circle-





command).

while the current method has some benefits, it also has
drawbacks...like all the damn typing of "¬D "!
another method being studied would require some special
marker on the continued lines of those TUTOR commands that
can have the "special" continued lines...like -write-
cont.

---------- response 6
  02/27/76     11.41     tenczar       s

e.g.,
            write   Hello there, Sue. Please
                    write me a letter!

the second line above is very "strange" in that it is an
"internal" continuation of the previous -write- command
and interpreted just as a "CR".
If "free-form" indenting was allowed...then the second line
would be ambiguous...should it be interpreted in the
"CR" mode, or does it represent another -write- statement:
        write me a letter!

another form of indenting being studied would force one
to put a marker on continued lines of commands with
internal continuations:
        write   Hello there Sue.
        ¬W     write me a letter!
then, there would be no ambiguity to free-form indenting
of the first word on a line.

---------- response 7
  02/27/76     11.49     tenczar       s

so, in summary, we are currently investigating indenting
of the following two forms:
1.
arrow   1010
answer  George Washington
write   Yep, Quite a great man!
        By the way, are you watching the
        Adam's chronicles on educational TV?

2.
arrow   1010
        answer  George Washington
                write   Yep....
                ¬W     By the...
                ¬W     Adam's...

Currently, the problems of NOT using some kind of mark and
fixed-distance between command and tag, are beyond our
thoughts...

---------- response 8
  02/27/76     11.54     tenczar       s





by the way, indenting seems to be a pre-requisite to
putting IF-THEN-OR-ELSE structures into TUTOR.

---------- response 9
  02/27/76     13.11     judy       pso

As for micro-TAB, that is "reserved" for backwards
writing (when it comes). I think that instead of TAB
you would press micro-TAB, and lo-and-behold, you'll
be going leftwards from the right margin.

---------- response 10
  02/27/76     13.51     tenczar       s

well, if we get indenting...then maybe micro-tab should
be indenting and micro-B should be backwards...surely
backwards will be infrequently invoked in author mode by
most authors and automatically invoked (ala force backward)
in student mode

---------- response 11
  02/27/76     16.06     j wilson       unidel

Thanks much for the explanation. I'm happy with
either suggested appearance of indenting, and would
like the C-search to find indented commands if it's not
too expensive.

To me, MICRO-TAB sounds good for indenting,
and MICRO-‎ good for right to left writing.

---------- response 12
  02/27/76     16.42     david       hebrew

Why does indenting seem to be a prerequisite for
putting IF-THEN-ELSE into TUTOR?

It appears to be the same as saying "indenting is a
prerequisite for putting regular commands which are
contingent upon judging commands into TUTOR".

To be sure,

arrow  505
answer indent
write  Good
wrong  micro i
write  Use -indent-
endarrow

is more readable than without indenting, but it
makes no difference in terms of the utility of the
abstract concept of answer-contingent writing.

(continued)

---------- response 13
  02/27/76     16.53     david       hebrew





Perhaps I would much rather read

this than this:

if    trials<3 then ¬I if trials<3 then
if    score=7 then ¬I if score=7 then
write Good ¬I write Good
else  ¬I else
write Fair ¬I write Fair
endif ¬I endif
else  ¬I else
write Crummy ¬I write Crummy
endif ¬I endif

But I would rather spend the extra time reading the
unindented version than having to use some entirely less
readable control structure to do what I want!

---------- response 14
  02/27/76     16.58     benveniste       css

   If I am not mistaken, micro ‎ already is being used
to cause a carriage return without margin. Would we lose
this feature?

---------- note 468 recondense
  02/26/76     22.28     benveniste       css

  I don't know if this has been brought up before or not
but if, when editing a lesson, you Press -SHIFT LAB- and
the last binary is unavailible, would it be possible to
write a message with an option not to recondense
instead of just going directly into the condense queue?

Mike Benveniste

P.S. The same thing holds true off the Error Page, where
it would be especially helpful.

---------- response 1
  02/26/76     22.58     mike b       cornell

This is a good idea.

M¬ ike≤≤≤≤ˇB

---------- response 2
  02/27/76     08.58     berger       mfl

Why would one indicate that he wants to see the student
version of a lesson and change his mind in the time of
a single keypress?

---------- response 3





  02/27/76     09.02     michael       cornell

The thing is "I want to reenter my lesson if the binary is
still available, but if it is not, I don't really want to
bother to recondense." I like it too.

M↑ˇi↑ˇk↑ˇe O↑ˆl↑ˆt↑ˆz

---------- response 4
  02/27/76     10.41     michael       english

If there is no line in the queue, it would take almost as
long to recondense as to read the message. If there is,
you can STOP1¬¬ out of the line.

---------- response 5
  02/27/76     11.29     artman       uimatha

But your number of condenses goes ¬W,

and your condense priority goes ¬X.

eric artman

---------- response 6
  02/27/76     12.07     baillie       cerlcc

As response 6 to note 409 states, you do NOT ruin your
condense priority by entering the queue and then leaving
before your condense actually takes place.

---------- response 7
  02/27/76     12.57     bargen       conn

Ah, but one DOES perform unneeded condenses (and downgrade
one's priority) if the queue is empty when one enters
it. I for one have often thought I would like such a
feature.

What about presenting a message such as:

Binary unavailable----entering condense queue

for 2 - 3 seconds. If the author pressed -stop1- while
on that display, he would be returned to his code;
otherwise he would enter the condense queue.

Would that brief delay be objectionable to others?
Would reducing condensing by a small amount be helpful?

Mark Bargen

---------- response 8
  02/27/76     13.19     judy       pso

A binary is not made for STOP1 condenses, so if you are





"editting", you will likely never have a binary available.

---------- note 471 CompCon
  02/26/76     22.37     walter winchell       infe1

Announcing:

The Chicago Area Computer Hobby Convention

This Sunday, Feb. 29 at 1:30
at Northwestern University, Evanston
in Lecture Room 2 (first floor)
in Tech Building
at Sheridan and Nyers Roads.
Many Altairs and other fun beasties.
Computer Lib/Dream Machines recommends this one.

See You There!

Your roving reporter

---------- response 1
  02/27/76     11.13     travers       mxc

That's Sheridan ¬+ _Noyes_.

---------- response 2
  02/27/76     17.20     lieber       altair

Anyone know how late it might last?

---------- response 3
  03/01/76     10.11     mcneil a       uicc

It lasted 'til 5.

---------- note 479 languagel
  02/27/76     03.36     osborne       iucs

The name of the lesson is 'languagel' now and is undergoing
major rewrites. I am the author so if you have any
questions concerning the lesson/language please contact me.
I wasn't aware that anyone out there in Platoland was
using/interested in it except the students here at IU in
C251. Glad to hear it!
OZ

---------- note 489 note gone
  02/27/76     10.35     t obrien       politics

Something just happened that I think should not have.
I had just responded to note ¬$488, and pressed -next-
from the 'Responce stored' page, when I found myself
looking at note ¬$0, and I soon found that note ¬$488 had
disappeared! Was it deleted, or just being edited
when I got finished? Either way sounds strange, since





it couldn't be edited if I had responded (I think), and
I don't know of any means of deleting a note. Is it
a feature I hadn't heard of?

        Terry O'Brien

---------- response 1
  02/27/76     10.40     woolley       p

A note is not marked as having received a response
until the response is stored. Thus, the note could have
been re-edited or deleted while you were writing your
response.

By the way, pressing SHIFT-EDIT gives you the option of
either re-editing or deleting the note.

---------- response 2
  02/27/76     12.31     t obrien       politics

Ok. Just thought it might have been one of those
things.

       Terry O'Brien

---------- note 504 discuss2
  02/27/76     12.14     nate       iumusic

    anybody know who wrote the now-nonexistant -discuss2-
or know what the lesson got changed to? (or not to)

        Thanks
        Nate

---------- response 1
  02/27/76     12.37     carter       comm

I did. There is a copy of the lesson in "discuss". The
"discuss2" lesson space became "commnotes". "discuss" is
primarily of historical interest now.

---------- response 2
  02/27/76     13.22     carter       comm

In fact, this happens to most old plato lessons. When
people ask why I didn't use student records (name and
course) I simply explain that those features were not
available then.

---------- note 519 IU Conf
  02/27/76     14.19     brown       iucs

         Please read Note ¬$32 (CS Conf) in the





GROUP notes file labeled 'iuppf' for

complete info. regarding the Second

Biennial Conference on Computing in

Indiana to be held at IU Bloomington

on Saturday, March 13, 1976.

Thank you.

---------- note 524 cstop
  02/27/76     16.23     travers       mxc

A suggestion (that I think I made before):

How about allowing one-argument tags for the -cstop- and
-cstart- commands? The argument would have to be a
numerical expression that evaluates to a constant. If it
evaluates to -1, the command has its usual effect. Any
other value causes the command to be ignored.
        This would be of value mainly to writers of -use-d
code, effectively allowing them to parameterize their
routines. This can be done now only by putting seperate
components in differently-named blocks, but this can
be clumsy and wasteful in many situations.
        So how about it? (If it's impossible for some
reason or other, apologies in advance).
                Mike Travers

---------- response 1
  02/27/76     17.50     white       p

Sounds good to me, but I expect it will be a while
before it is implemented.

---------- response 2
  02/27/76     19.15     al       mfl

While we're at it, is there any chance of a reserved word
telling the name of the lesson you are currently in? It
would also be useful for lessons using other lessons.

---------- response 3
  02/28/76     13.11     friedman       csa

This was proposed before, while the systems staff was
considering moving to multi-word lesson names, and so
instead of a reserved word with the lesson name, we got
one with the leslist number. But this is less than 100;
satisfactory: what about a block of -use-d code intended
for the use of anybody on the system? Obviously, a leslist
of all the -use-ing lessons would be impossible. So how
about a reconsideration, please, systems?

---------- note 531 -find- ¬=





  02/27/76     17.32     michael       cornell

I have often wished that the -find- command were a bit
more versatile. For example, I want to look for the first
word that is NOT equal to, say, zero. Wouldn't it be
possible to allow a format like this:

find   object,start,length,return,incr,mask,¬a

where  ¬a is a member of ¬[=,¬=,<,>¬], meaning "find the
first value that stands in the given relationship to the
object". The default would of course be =, so only those
authors using the extra argument would be affected.

I have a feeling that the ¬= would be easier to add than
the others. Do I hear a second?

M↑ˇi↑ˇk↑ˇe O↑ˆl↑ˆt↑ˆz

---------- response 1
  02/27/76     18.06     guerra       uimc

I would definately have many uses for this, if it could
be implemented.

---------- response 2
  02/27/76     18.39     mike b       cornell

         This is an excellent idea which would banish a lot
of kludgy code from the system. I'm all for it.

M¬ ike≤≤≤≤ˇB

PS: And it's not because he's got the same name.

---------- response 3
  02/27/76     19.38     hinton       ssu

s e c o n d !

ndh

---------- response 4
  02/27/76     20.26     clark       lawyer

agreed...it would make compressing code in common
much easier among other things...

---------- response 5
  02/27/76     22.17     k mast       p

Someday....you're format is very interesting, however.

---------- response 6
  02/27/76     23.29     kent       unl





        Now if you include ¬< and ¬> in your ¬a set, you will
have the -index- function I suggested in note no. 328.
With this addition, I agree wholeheartedly! Hear, Hear!

---------- note 542 ee lessons
  02/27/76     20.21     pogue       uimatha

Does anyone know where I can access some Electrical
Engineering lessons? I heard that there was a router
set up somewhere that has EE260 and other EE lessons
in it....
Thanx,
mike

---------- response 1
  02/27/76     21.37     fries       ee

         Try 'uieeindex' for a large listing and I think
a short explanation (it is at least subdivided into areas
of interested). It can be refrenced as an author to see
if its what you want (in student mode).

EˇE

---------- response 2
  02/27/76     22.08     pogue       uimatha

Thanx! That is what I was looking for....

mike

---------- note 544 help(parm)
  02/27/76     21.24     copland       csa

         What about the possibility of implimenting a
feature allowig commands such as 'lab', 'data', and 'help'
to pass a parameter to the specified unit. It seems that
this could be used to great advantage in long help
sequences, etc. Also it would be an aid toward the
development of better structured lessons.

            Opinions?

---------- response 1
  02/27/76     21.24     osborne       iucs

How about making it _all_ key branch commands so one could
pass parameters while changing main units, i.e.,

next someunit(34,yourmom,iflag)

  I could have used this last week but got around it with
some less elegant/structured code.





---------- response 2
  02/27/76     22.37     b sherwood       s

Nope. The branching command is executed at the time it is
encountered, and merely places a unit name in the associated
marker word. Then, later, when the student presses HELP
or DATA, we look at the help- or data-associated marker
word to decide where to go. There really is no way there-
fore in the present structure to allow for an argument
list for each control key. And would the arguments be
evaluated when the command was executed, or when the
function key was pressed?

---------- note 551 russ chars
  02/28/76     10.28     b sherwood       s

After much consultation, character set "charsetx,russian"
has been cleaned up to contain only the basic letters
and punctuation marks necessary for typing Russian
sentences. This set is essentially compatible with
that used in the UI Russian lessons. One minor change
was made: the soft sign was moved from the minus key
to the 6 key, where it is next to the hard sign (on the 5),
it looks like a 6, and it is not taken by TUTOR to be
a punctuation mark.

---------- note 552 comlengths
  02/28/76     10.54     a appel       uni

How about this: Changing the length of a permanent common.

Blocks of common could be added at the end of the existing
common blocks, and blocks (or chunks of words) could be
taken off of the beginning or end of existing commons.

This is possible to do now, but it requires extra space
(with copy a block etc). And just now, lacking the space
in the lesson the common was in, I had to use a different
lesson. With problems involving different security codes,
different common codes. forgotten common codes, and moving
between lessons, the whole operation took ten minutes
instead of the 30 seconds it should have.
(nice sob story, isn't it!≤?)

Is there any chance of such a feature?

---------- response 1
  02/28/76     13.59     wagle       iu

   Did you try converting the common to source, and
 manipulating the resultant blocks?
   This sounds like what you want...

---------- response 2
  02/28/76     14.46     white       p

We expect to provide (soon?) a feature to extend and





shorten common blocks without the trouble of converting
them to source and adding and deleting blocks. However,
Wagle is correct, changing the common to source is
currently the easiest wat to do it.

---------- note 553 branch
  02/28/76     13.10     k gorey       research

    I am now getting the following warning in code that
never gave me a warning before:

*warning*:system variable *branch* redefined

    I suppose this is permanent, but not in the prime
time version. If so, what does the branch variable tell
one?

---------- response 1
  02/28/76     13.46     white       p

It is not actually a variable, but the special symbol
with special meaning in calc statements...

calc  n1‎0
0     n1‎n1+1
      branch n1<10,0

Other special symbols in the same list are
"doto" "skip" "¬p" and "¬0".

---------- response 2
  02/28/76     14.37     k gorey       research

     The question though is, why was it not flagged as
an warning in the past, and will it continue to be flagged
as a warning in the future?

---------- response 3
  02/28/76     19.11     blomme       s

    It was a possible source of difficulty that was
overlooked for "flagging" in the past, but will now
continue to be marked in the future.

---------- response 4
  03/01/76     11.01     michael       english

Actually, I'm glad it's now done. I once spent all day
trying to figure out why a -draw- would't work right.
I had inadvertently defined a variable called "skip".
thus overriding the system definition, but had forgotten
about it since there was no system warning. This discus-
sion does, however, point to the desireablity of allowing
programs to turn off condense WARNINGS -- not ERRORS --;





it is frustrating to have to change a variable name in an
old working lesson in order to avoid the condense error
page. I know students do not see the page, but authors do,
and such unneeded warnings are annoying.

---------- note 557 topics!
  02/28/76     14.06     clark       lawyer

Suggestion for lesson "topics" (which is now a very
valuable tool! thanks!!). It might be useful to put
an asterick or some other type of marker next to lessons
which can be reached through sample. That way, you would
know which lessons you could already use without having
to ask for author consent.

---------- response 1
  02/29/76     21.24     tebby       pso

Thank you for the suggestion for "topics". I will think
about your idea. However, eventually "sample" is to dis-
appear because "finished! lessons which are considered pub-
lishable will be in a systems catalog which will access each
lesson if desired. "sample" is just supposed to be a
sampling of "this and that" and does not in any way contain
all the available finished lessons which are mentioned in
"topics", now is it intended to do so.
                Tebby Lyman

---------- response 2
  02/29/76     21.30     clark       lawyer

ok...thanks

---------- response 3
  03/01/76     15.23     jackson       cs196

 Sounds great, Tebby. When can we expect to see this
systems catalog? Thanx. G.J.

---------- note 560 b board
  02/28/76     15.16     friedman       csa

ANNOUNCEMENT:↓
Ever feel the need for a PLATOized "bulletin board" for
your classes? A way to leave notes for the whole class
to read, such as a description of the next exam assignment,
or a change to a printed handout? Well, here's your
chance to set one up with a minimum of effort.↓
Lesson "bulletin" is now available for -use- by other
lessons. It provides a bulletin board editor (highly
graphical, an elaboration of principles pioneered by
lesson "service" and the SS and SD options of the system 
lesson editor). It also provides a small lesson for the
students to read the bulletin board. Messages may be
created. edited, and destroyed as required by the
instructor, even while the students are reading them.↓
For more details, including how to set it up for yourself,





execute lesson "bulletin".
Comments/complaints should be addressed to me by pnote.

---------- note 561 indent
  02/28/76     15.20     obrien       uicc

I have been having thoughts as regards structured pro-
gramming in TUTOR, mainly because I've been earning my
bread working with a structured language. As Paul has been
pointing out, TUTOR always has been a structured language.
Well and good. The "indent" feature is designed to make
the code more readable to humans, thereby allowing one who
does not live in symbiosis with TUTOR to have a reasonable
chance at reading TUTOR code. Also well and good. And,
should TUTOR ever get if-then-else-while structures, it
will be almost mandatory to have an indented format in
order to make TUTOR readable by anyone, even the author.
All true.

        However, the current method of indenting will
make it all but impossible to use for all but "showcase"
examples as long as disk allocations are fixed in length.

                *cont.*

---------- response 1
  02/28/76     15.26     obrien       uicc

The main reason for this claim is that the current indent
feature eats up almost one word per indentation, whereas
in almost all other systems, each indent eats up only
one character (the TAB character, which PLATO lacks).
Therefore, if we consider the average structured program
which uses indenting, we will find that it can eat up to
two to two and a half times the "white space" (black space,
on one of these brown boxes) as it would if it were not
indented. On PLATO, each such space is a character, so
lessons which use this feature as freely as it should be
used will eat up quite large amounts of lesson space.
Therefore, either the structuring itself will not be used
(best case), or "hot-shots" will use it to write unreadable
code (worst case).

        At a minimum, therefore, I ask that consideration
be given to the implementation of a TAB character on
PLATO.

---------- response 2
  02/29/76     12.10     b sherwood       s

This is indeed under consideration, for these and other
reasons.

---------- note 563 errtype=2
  02/28/76     16.47     friedman       csa

I recently asked for help trying to figure out what could
be causing errtype=2 returns to my router's error unit.





Errtype=2 is supposed to mean "fatal condense error",
but not "no ECS". Well, it just happened to me.↓
I was signed on with a student record, and tried to
execute a certain lesson. After waiting for it to
condense (seemed like a rather long wait after the
"now preparing lesson" message stopped flashing), my
router reported errtype=2! I retried it several times;
each time I got a very quick response with same message.
As soon as I could, I got in with an author signon and
condensed the lesson; no problem at all" (There is
absolutely NO possibility that anyone else could have
edited the lesson.)↓
I know it's non-prime time, but this doesn't seem to
depend on which version of Plato is running.↓
Can anyone suggest a cause?

---------- response 1
  02/28/76     17.57     clark       lawyer

errtype = 2 when you try to jumpout to another router
with a student sign-on also...perhaps this is the problem

---------- response 2
  02/28/76     18.49     parrello       uimatha

This also happened to me with lesson "phon1". Student
got errtype=2 but author condensed perfectly on the
first shot. "phon1" is NOT a router, and you can check
if you don't believe me.

---------- response 3
  02/29/76     12.34     weible       german

Check to make sure there isn't an error in the leslist.
An attempt to jumpout to a lesson whose _correct_ name
is not on the necessary leslist does result in an
errtype=2.

---------- response 4
  03/01/76     09.56     parrello       uimatha

I did a - jumpout resume-.

=rstartl= was 'phon1'
=rstartu= was 'done'

Yes, "phon1" has a unit *done*, since the -jumpout resume-
worked after the lesson was condensed.

---------- response 5
  03/01/76     10.23     mcneil a       uicc

I'm in favour of a few more errtype values. Having several
different errors all return 2 is a pain.

---------- response 6
  03/01/76     10.44     friedman       csa





The jumpout was absolutely NOT to another router, and
was a one-tag jumpout.↓
SYSTEMS, HELP, PLEASE!

---------- note 564 inspect
  02/28/76     17.01     hecht       ed

Frequently new authors enter their lesson in "inspect only"
and don't know how to enter their security code. When I
first started authoring, this happened to me, and I'm
sure it happens to many others because i've been asked
several times this week "What's wrong with this lesson?"

Would it be possible to add one line to the "inspect only"
message which appears on the lesson contents page telling
authors to press SHIFT-DATA to enter the code. This
minor modification could end a lot of frustration for
new authors who aren't used to the system.

---------- response 1
  02/28/76     17.59     j wilson       unidel

This would be very handy.

---------- response 2
  02/28/76     18.05     koning       csstaff

If is on the help pages... but a more conspicuous spot
would be useful.

---------- response 3
  02/28/76     19.15     d zweig       iu

No, lets not put it on the BLOCK LISTING Page.

After having to look for it on the HELP pages once or
twice, you'll remember!

Let not cluter up the Listing page any more than
necessary.

I, personally, don't want to have to look at it plot
everytime.

---------- response 4
  02/28/76     19.30     fritz       amesrad

Yes, it's on the HELP pages, if can decipher it...
It could be reworded mush more clearly.

---------- note 572 currentu
  02/28/76     19.46     bowery       comm

A couple of suggestions that would go hand-in-hand:





icurr <unit name>
* unit <unit name> would be "appended" at the beginning of 
* each new current unit
*
* ¬+
*
showa curu
* name of current unit in -curu-
* -curu- ¬= name of -icurr- unit if it is refered to in the
* -icurr- unit by the -icurr- logic

   This would be very helpful in debugging things. 

(better names for these things?)

---------- response 1
  02/29/76     12.32     al       mfl

Unless I'm very confused, what is the difference between
icurr and imain?

---------- response 2
  02/29/76     12.51     woolley       p

The current unit may not be a main unit (e.g., -join-ed
units.) I'd like to see something like this, too.

---------- response 3
  03/01/76     09.07     fay       o

Ah, many have been the times when I have wished to do some-
thing based on the _current_ unit! Having the reserved
word (curru, or whatever) _and_ the -icurr- command would
be a real treat.

---------- response 4
  03/01/76     17.35     bowery       comm

Well how about it there sysprogs? Any action planned?

---------- note 573 pure+easy
  02/29/76     00.27     ostuni       iu

There once was a note pure and easy
Playing so free like a breath rippling by...

---------- response 1
  02/29/76     06.45     j thiher       amesrad

huh?

---------- response 2





  02/29/76     13.32     nate       iumusic

   comes off of Pete Townsend's solo album entitled
"The Who Came First"....a little bit of prose or
what-have-you.

Nate

---------- response 3
  03/01/76     11.39     michael       matha

Yeah, right.

Who's going to be the first to transcribe "The Rhyme of the
Ancient Mariner" into NOTES/

---------- note 580 cpa exam
  02/29/76     13.07     leviton       ed317

for all those accounting students, there is a new
program out caleed cpa review. It is a series of
questions taken off past cpa exams, put in the form
of the cpa test.
it is currently just being started, but can be used.
to get to it, the course is ed317l.

---------- response 1
  02/29/76     17.18     al       mfl

"all those accounting students" can't read public notes
"cpa review" does not exist
course "ed317l" also doesn't exist

---------- response 2
  02/29/76     18.03     temmerman       com

for the record, ed317l just worked for me. If you're
worried about accessibility, I think that it could be
put under one of the accounting multiples quite easily
why don't you send a note to either mckeown of com or
petersen of com. I think the topic is a good idea

---------- response 3
  02/29/76     20.45     anderer       ee

  Perhaps there is a _lesson_ ed317l, but Al is right in
observing there is no _course_ ed317l.
dg

---------- response 4
  03/01/76     10.13     maggs       law

After tring the cpa exam, you can take the bar exam in
lessons bar1, bar2, bar3,(bar4 and bar5 not quite finished)





---------- response 5
  03/01/76     10.25     judy       pso

The cpa exam is interesting, but all that sized writing
is a pain! I can understand "you're doing great" just
as well when it is in size 0.

---------- note 584 survey
  02/29/76     19.32     david       hebrew

If you have a few minutes to spare, I would appreciate
it greatly if you would help me by going into lesson
"driver" which will, for the next two weeks (or until
it fills up), contain a survey of indentation style.

You will be shown sections of tutor code indented
four different ways, and the point of the survey is for
you to rank order them (in terms of preference).

The reason for this survey is, basically, pure curiosity
as to whether my hypotheses about indenting are correct.

(An explanation of the hypotheses will be forthcoming
soon, i.e. when I get my head together on it.)

Again, thank you for your help.

P.S. No, you need not remind me that PLATO isn't a
democracy.

---------- response 1
  03/01/76     11.44     tenczar       s

first, all your samples have lines that are of "fixed
format" in that there is an 8 character space from start
of indent to command (or next indent)...it would be nice
to see what people thought of "free form" indenting...
(by the way, the current unannounced indenting method "¬D"
would have been "." if I had realized that ". backspace ,!
used 3 characters as well as "¬D"...since I wanted the whole
indenting "package" to use 10 characters = 1 word, and hence
be trivial to implement and execute)

secondly, what is to keep the same person from "voting"
n-times in your current survey and hence "stack the vote?"

---------- note 585 guano gap
  02/29/76     20.09     roper       siu

Would anyone else who wants a print of the Great Guano
Gap please let me know soon, as the lesson will be
undergoing changes shortly.
Thanks,
Quetzal

---------- response 1





  03/01/76     03.30     kent       unl

        What sort of changes? I think it's great the way
it is--it should remain in all its pristine glory for
the edification of future (and perhaps as yet unborn?)
PLATO authors.....

Aeolus

---------- response 2
  03/01/76     08.19     mike b       cornell

Well, let's not lose our heads.

M¬ ike≤≤≤≤ˇB

---------- response 3
  03/01/76     13.36     warner       iu

        He means that the Scrooge who hands out lesson
spaces at SIU is pressing for more productive use of
the Great Guano Gap.
        (I stick with the term "Scrooge,, by the way...)
        If you are going to print it soon, I would like
a copy. Maybe someday I can get it back in proper form
but in any case it should be saved in as many places
as possible!

---------- note 589 autosign
  02/29/76     23.46     berger       mflu

On the non prime-time version. autosigned terminals
are allowing only the autosigned records to be used
at those terminals - ie: no other records can sign
on.

I thought that autosigning was for convenience, rather
than restriction.

Is this change intentional?

---------- response 1
  03/01/76     00.14     andersen       s

Fixed now.

*** end of notes ***