NetHistory

VM/COM, July 1985

1
-
OO OO OOO OOO // OOOOOOO OOOOOOO OOO OOO
OO OO OO OOO OO // OO OO OO OO OO OOO OO
OO OO OO O OO // OO OO OO OO O OO
oo oo oo oo // oo oo oo oo oo
oooo oo oo // oo oo oo oo oo oo
oo oo oo // ooooooo ooooooo oo oo
0 ------------------------------------------------------------
July 1985 edition Volume 2 Number 2
- CsNews Network Newsletter
-
-
Staff:
0 Michele Robinson CSMICH at MAINE Editor
Andrew T. Robinson ANDY at MAINE CsNews Director
Prof. G. Markowsky MARKOV at MAINE Faculty Advisor
0 (Special thanks to Dave Eckhardt, DAE@PSUVAX1, for error,
spelling, and grammar checking)
-
0 Ôçççççççççççççççççççççççççççççççççççççççççççççççççççççççä
³ Newsletter article contribution Userid: CSNEWS@MAINE ³
³ ³
³ Contributions from readers welcomed and encouraged! ³
¨ççççççççççççççççççççççççççççççççççççççççççççççççççççççç]
1
0 Vm-Com Issue 2.2
-
0 Table of Contents
+ _____ __ ________
- Introduction to Vm/Com 2.2 . . . . . . . . . . . . . . . 1
CSNEWS Notes . . . . . . . . . . . . . . . . . . . . . . 2
Designing a DataBase . . . . . . . . . . . . . . . . . . 3
IBM/370 I/O Architecture . . . . . . . . . . . . . . . . 6
A Look at ISPF . . . . . . . . . . . . . . . . . . . . . 10
NetCon -- Spring '85: A Summary . . . . . . . . . . . . . 12
A Poem . . . . . . . . . . . . . . . . . . . . . . . . . 14
OpCodes . . . . . . . . . . . . . . . . . . . . . . . . . 17
-
-
-
-
-
-
-
-
-
-
-
-
-
- 2
1
0 Vm-Com Issue 2.2
0 Introduction to Vm/Com 2.2
+ Introduction to Vm/Com 2.2
-
Welcome to issue 2.2!!! In this issue you will see
articles on IBM/370 I/O programming, an introduction to ISPF
(Interactive System Productivity Facility), a guide on
designing databases, a report on the recent NETCON in New
York, and a humor section consisting of a poem about LISP
and a collection of rather unusual OPCODES.
0 If any of you are interested in contributing articles
for future editions drop me a note. We really could use help
from all our readers!!! Also if you have anything to say
about Vm/Com or even something about CsNews drop us a
mailfile... If it's ok with you I'll publish it in a future
Vm/Com issue.
0 Hope you all enjoy it!! and I'll be seeing you again
next issue...
0 Michele Robinson
Editor
-
-
-
-
-
-
-
-
-
-
- 1
1
0 Vm-Com Issue 2.2
0 CSNEWS Notes
+ CSNEWS Notes
0 Andrew T. Robinson (ANDY@MAINE)
- As I'm sure most of you can see, the Vm/Com production
effort is (finally) in gear. We had a bit of a problem
between issue 1.1 and issue 2.1, but hopefully you will see
a new Vm/Com every month or so (god forbid, maybe bi-
weekly!).
0 While I am here, I may as well tell you what is going
on with CSNEWS this summer. At the current time, funding
has been allocated for the development of a locally-run
(MAINE users only) user-friendly interface system. This
interface will initially consist of three major components:
a mailer, a conferencing system (local use only), and an
academic database system.
0 And while all this is going on, I am attempting to re-
write portions of CSNEWS (specifically, the command
dispatcher and lower-level routines) in IBM/370 Assembler to
decrease CSNEWS' response time, as well as its CPU usage.
As one example of our current usage, between the months of
August 1984 and May 1985, CSNEWS used approximately $40000
worth of CPU time (assuming the current departmental 'funny
money' rate of $3.00 per Computing Resource Unit). That's a
lot of CPU time! I'm glad Maine *doesn't* charge for system
usage...
0 One thing I have noticed is that the CSBBoard system on
CSNEWS doesn't seem to be getting much use. Our other
forum, FLAME, is getting more use that I'd ever imagined,
but CSBB doesn't get updated more than once every couple of
weeks. All I can say is, C'mon people! CSBB is one of the
simplest and easy to use BBoards there is. We will be
implementing a much needed SEARCH command this summer, and
the system is already set up to allow easy location of
desired information... So use it!
0 Speaking of FLAME... It's been getting a LOT of use, and
there are a lot of interesting conversations going on in it.
Originally, FLAME was intended to be a Vm/Com column, but as
it turns out, I'm glad we made it a separate service! But
just one word of caution (a pleading one)... Don't get too
carried away in your FLAMEs. Remember that CSNEWS is still
subject to the long arm of administrative rule, and I would
prefer not having to delete FLAMEs because someone in the
hierarchy is complaining about the content of the FLAME
file. Keep it clean, if at all possible.
0 I guess that's about it for now. Thanks for reading (if
you've gone this far) and I'll see you next issue!
- 2
1
0 Vm-Com Issue 2.2
0 Designing a DataBase
+ Designing a DataBase
0 Howie Ducat (Howie@Bklyn).
-
The idea of having a database is that it provides a
centralized control for a user's/company's data. This in
itself will reduce redundancy of data, hold inconsistencies
in data to a minimum (insure integrity), allow data to be
shared by more than one user if necessary, and make a data
security system somewhat easier to implement. The following
is an outline of considerations when designing a database.
0 Requirements:
+ Requirements:
0 Before we can begin, we must first look at the
requirements of the company. We must see the type of data
that the company feels will be relevant to their daily
processing. At the same time we are viewing the data
specifications, we should also be defining the data
elements. The two previous steps will familiarize us with
the general types of data, how we might store the data, and
define what exactly is needed and not needed in the
database.
0 Another good idea at this point is to find out from the
company how they will use the database, when they will use
it, and why they believe they need one. Will they require
immediate updating of records or will end-of-day batch
processing serve their purpose? What type of response time
will they need (not want, as in the real world, it is plain
to see that everyone wants everything 2 minutes ago)? We
should try to ascertain the company's rate of growth to
allow for extra space/room to expand if necessary, as
easily and with as little downtime as possible.
0 After the requirements have been outlined we feel that
a few things should be done. First, we must consider
personnel planning, cost comparisons, cost/benefit analysis,
and a project budget. We will need to know how to divide
the total project budget between all needs in order to
complete the project. We must hire two or three programmer
analysts (depending on the urgency of the project), and one
or two systems personnel to take care of the operating
system's needs (as well as the possible formatting of
external storage devices, and channeling I/O flow for
maximum speed and efficiency). Also needed is a database
manager, who will make key decisions on insertion and
deletion of key data elements, expansion of storage
capacity, and purchases of additional hardware and software
after the database is in production. After this, a clear
schedule of installation and development of the database
must be made, involving matters such as how long each step
0 3
1
0 Vm-Com Issue 2.2
0 would take, how much time would be spent on aspects, and who
will take care of each aspect. We should allow for one or
two small setbacks per step so as to bring the scheduling
closer to a real world situation.
0 ACCESS METHODS:
+ ACCESS METHODS:
0 We are applying the following rules to our choice of
access methods:
0 1) We are not considering purely sequential media, such as
magnetic tape, since it imposes too many restrictions to be
useful to our database EXCEPT as a medium for journaling
activity.
0 2) We make no attempt to predict what future DASD or other
external storage classes will be like, but if a new large
capacity DASD were to be produced, we will be able to take
immediate advantage of it, provided we set up data
independently from disk addressing keys (random/direct
accessing).
0 We must take into account the volume of data and how it
will be used in order to choose the access method. If more
than say, 70% of all records will be posted each day,
sequential files might be best. However, if you post few
records but do a large number of inquiries, ISAM, random, or
direct access methods might be a good idea (no updating
though!). If you will be doing little to moderate updating,
and comparing many data elements, an inverted file structure
is one possibility. A hierarchical file structure might also
be used, which seems an even better choice, because when a
user does a data comparison he will see "one record" when
there are really multiple records, existing in different
files, possibly with different LRECL's and BLKSIZ's, thus
providing more flexibility.
0 SECURITY & PRIVACY:
+ SECURITY & PRIVACY:
0 With the ever increasing number of "Hackers", security
is becoming more and more important. First, we will briefly
discuss data security. A simple yet effective precaution
against internal (within Data Center) sabotage is a magnetic
field detector. The detector alerts management that someone
is attempting to enter the Computer Center with a magnetic
device. A magnetic device can do serious damage to Direct
Access Storage Devices, in-CPU program instructions, and
Magnetic Tapes (which transaction logs are usually kept on)
by simply erasing, or by reversing the information's
meaning. Another type of Data Center security are Magnetic
Card Readers with a teletype, (to produce a hardcopy of who
enters which areas of the Center) and a door Lock/Unlock
system. This will (almost) ensure that no one gets into an
area where they do not belong, therefore lowering the
0 4
1
0 Vm-Com Issue 2.2
0 probability of internal sabotage.
0 Access security should be considered at this time as
well. If data is highly confidential, different access
levels can be set up so that only certain users or
administrators can view it.
0 Examples:
0 A) Database Administrator: Unrestrained access
to
to entire relation for ALL types of
operations.
B) Payroll Department: May see ONLY Payroll
related
data. ie: Name, Address, SS#, Salary, # of
dependents
C) Employee: Can only VIEW his/her record. NO
UPDATING is permitted.
0 The above can be achieved by what is called Identification
and Authentication, which is comprised of user names and
passwords along with link passwords to specific devices
where certain data lie.
0 MANAGEMENT CONSIDERATIONS:
+ MANAGEMENT CONSIDERATIONS:
0 Management must consider the following when the
database is in the designing stages: user's needs, what
types of reports will be required by different department
personnel, what maximum response time should be, and how
"User Friendly" should the database be.
0 Management should also try to provide maximum
processing time (or minimum downtime), design the database
within the budget of the company, and try to make program
functions as modular as possible, thus facilitating easy
changes/additions to program code in as short a time as
possible.
- These are just the very basic considerations for even
beginning to design a database. Once specifications have
been received, and work has begun, there are many other
things that must be considered.
0 Special thanks to Robert Dong (KAMIKAZE@BKLYN) whose
presence in my database class prevented my premature
insanity. -H.D.
-
-
5
1
0 Vm-Com Issue 2.2
0 IBM/370 I/O Architecture
+ IBM/370 I/O Architecture
0 Andrew T. Robinson (ANDY@MAINE)
-
Computers in general would be pretty useless if they did
not have some method for interacting with the outside world.
The intent of this article is to introduce the structure of
the IBM 370 I/O system, and I/O programming in general.
0 IBM s/370 computers use a powerful interrupt-driven I/O
system. Doing I/O does not tie up, or INTERLOCK, the CPU.
Programs can continue executing while I/O is being completed
on a device. This is accomplished by CHANNELS. A channel is
basically a small computer that handles I/O operations.
Every device attached to the CPU has an ADDRESS. An example
of this address is the line-number that is returned from a
QUERY NAMES command next to the userid. This address is
often referred to as 'cuu' where 'c' is the channel address,
and 'uu' is the unit address. This means that up to 16
channels (0-F) may be attached to a 370 CPU. Each channel
may have attached to it either 256 devices, or 16
subchannels with 16 devices each.
0 Channels come in two basic varieties, SELECTOR, which
handle high-speed I/O devices such as tapes or disks, and
MULTIPLEXOR, which handle low speed devices such as
terminals. Selector channels are dedicated to one device at
a time. If the tape drives are on channel 5, only one tape
drive at a time is serviced. Multiplexor channels handle
several devices simultaneously, such as terminals. This is
necessary to provide the real-time response needed from
interactive terminals.
0 The IBM/370 machine language actually possesses very few
instructions dealing with I/O. There are instructions to
start, test, and halt I/O operations on a specific device,
but the actual operations are performed by a CHANNEL PROGRAM
kept in main storage. A channel program consists of a
series of 8-byte words called CHANNEL COMMANDS WORDS (CCWs),
which specify operations to be performed, such as a read or
a write to a device, etc. In general, the channel commands
are device specific, which is to say, a WRITE command for a
terminal will not usually work on a disk drive. The channel
command also contains such things as the address of data to
be written, or into which data is to be read, and how many
bytes are to be transferred. There is also a 'transfer in
channel' CCW (channel command word) which allows loops in
channel programs. A channel program may be arbitrarily
long, and perform many operations in a single invokation.
-
0 6
1
0 Vm-Com Issue 2.2
0 To initiate I/O operations the programmer must first
place the address of his channel program in a predefined
area of storage known as the CHANNEL ADDRESS WORD (CAW). The
CAW is located at the fixed storage address x'48' (48
hexadecimal). Once the address of the channel program is
stored in the CAW, you initiate the I/O operation with a
Start I/O (SIO) instruction. SIO causes the channel to begin
executing your channel program.
0 Now the user program may continue to execute. However,
it is generally recommended that the I/O operations are
allowed to complete before execution continues. This is
accomplished by one of two ways: a Test I/O (TIO) loop, or
a PROGRAM WAIT. In this article, we will discuss the TIO
loop method for determining I/O completion.
0 Generally you set up the TIO loop immediately after the
SIO instruction. It is simply a TIO instruction for the
specific device, followed by a series of BRANCH instructions
for specific condition codes. For a TIO, the following
condition codes are defined:
0 CC = 0 -- channel and device available
CC = 1 -- channel status word stored
CC = 2 -- device is busy (executing channel program)
CC = 3 -- device is not operational
0 If the condition code is 3 you should leave the loop
immediately. In more sophisticated applications, you would
test the Channel Status Word stored when you get a CC of 1,
but in this example, we will wait for a CC of 0 and then
exit the loop. The Channel Status Word is stored after a
channel interrupt, and contains detailed information
regarding the progress of the I/O operations. The CSW will
be discussed in a later article.
0 The following assembler program illustrates the
principles discussed above. Instructions for assembling and
executing it follow the program. The program simply writes
a line of text to your terminal, which is (usually) at
device address X'009' (that is, channel 0, device 09):
-
-
-
-
- 7
1
0 Vm-Com Issue 2.2
0 WRITERM CSECT
STM R14,R12,12(R13) save registers passed by CMS
LR R12,R15 load our entrypoint into R12
USING WRITERM,R12 address this program
*
* In order to store in location x'48' we must have a nucleus
* storage key,which we accomplish with the CMS DMSKEY macro.
*
DMSKEY NUCLEUS establish a zero storage key
LA R4,CHANPROG get the address of our channel
* program
ST R4,X'48' store into the CAW
DMSKEY RESET restore the last storage key
*
* now having stored the ADDRESS of our channel program, we
* initiate I/O on the CMS console (device x'009')
*
SIO X'009' X'009' is the device on which
* I/O is done
*
* now do a TIO loop to make sure I/O is complete. We simply
* wait for a condition code of zero, and then drop out.
*
TESTLOOP TIO X'009' test device
BO OUTOFIT BO is for CC=3,
* device not operational
BNZ TESTLOOP if CC Not Zero, do loop again
*
* Now we are through doing I/O, go back to CMS.
*
OUTOFIT LM R14,R12,12(R13) get registers back
XR R15,R15 make a zero return code
BR R14 return to CMS
*
* the following area is the channel program, consisting of 1
* channel command word, which writes the message at MSG to
* your terminal. The X'01' is the command code for a WRITE,
* X'20' is the CCW flag which indicates no more CCWs in
* chain. If the channel program is more than 1 CCW long, the
* X'20' would be changed to an X'60', in all the CCWs except
* the last which would be kept X'20'.
*
CHANPROG DS 0D align on double word boundary
CCW X'01',MSG,X'20',MSGLEN
*
* The message area to be written
*
MSG DC C'This is a line output to the terminal.'
MSGLEN EQU *-MSG length of message
*
* following are required macro definitions
*
REGEQU
END WRITERM
0 8
1
0 Vm-Com Issue 2.2
0 This program actually works!! To assemble and execute
it, issue the following commands to CMS:
0 GLOBAL MACLIB CMSLIB DMSSP
ASSEMBLE WRITERM
LOAD WRITERM
GENMOD
WRITERM
0 This program illustrates the principles discussed above.
It is relatively simple but should give you an idea of what
is described in the article. My next article will discuss
in detail some of the basic ideas presented above, including
the use of interrupts and checking the CSW for I/O
information.
-
-
-
-
-
-
-
-
-
-
-
-
- 9
1
0 Vm-Com Issue 2.2
0 A Look at ISPF
+ A Look at ISPF
0 Code B. Shive (33000003@UNFVM)
-
The following is part one of a series of articles
devoted to the use of an interactive dialog system called
ISPF. Please keep in mind that this is a project I am
working on, and I am writing this as I learn. I have no
sources of reference other than the IBM manuals furnished
me. For those of you familiar with the system I am
describing, I am interested in talking. However, since I
feel that this is something fairly new (as far as student
work goes), I will continue with a description of ISPF.
Perhaps in later articles I can discuss setting up and
actually running a program with ISPF.
- Most installations employ full-screen terminals for
both students and faculty. There is a definite advantage in
using an entire screen since the screen is designed to suit
the particular application whether it is editing a file or
entering data.
0 There are various systems across the nation that use
full-screen applications. There are systems like CMS and TSO
that are used to edit files. There are also systems devoted
solely to on-line applications like flight bookings on
airlines. These systems are different from those of CMS or
TSO, but are similar in that they are used to communicate a
screen of information at a time.
0 A popular on-line system is CICS (Customer Information
Control System). With CICS, the application program runs in
a special region that is set up especially for CICS to
operate. The next time you see a data-entry screen, it
probably is from an application that is operating under
CICS.
0 I have recently discovered (though it was not lost) a
new system designed for creating screens and manipulating
them in dialogs. Unlike CICS and other on-line systems,
ISPF (Interactive System Productivity Facility) offers an
easier approach to programming in an interactive environment
without the expense incurred from running CICS. ISPF is
usually packaged with PDF (Program Development Facility).
With PDF, you can edit new screens for ISPF and generate
code for the various libraries ISPF requires you to have.
Using PDF is not necessary, but it is a nice addition.
-
- 10
1
0 Vm-Com Issue 2.2
0 Both of these systems run under their respective virtual
machine that runs disconnected. To see if you have the
facility available, query the names on your system and look
for ISPVM, ISP, or PVM. Chances are that if you have ISPVM,
you have PVM. If you don't see any of the above, you are
probably out of luck. If you do, type in the keyword:
0 ISPSTART
+ ISPSTART
0 and hit the enter key.
0 If ISPF is installed correctly, you are presented with a
screen in the form of a menu. Select the tutorial, and page
through the screens. Unfortunately, this tutorial is not
designed for a beginner who is trying to create the screens,
but it is a nice demonstration of how the finished screens
will look.
0 To end the tutorial, use the PF15 key to return you to
the first menu screen. Then, select the option to end the
session.
- Later, I will discuss how to design a screen and what is
necessary for running programs and functions under ISPF. If
you are interested in more information, I am available.
-
-
-
-
-
-
-
-
-
0 11
1
0 Vm-Com Issue 2.2
0 NetCon -- Spring '85: A Summary
+ NetCon -- Spring '85: A Summary
0 Marc Kriguer (VSS6751@UCLAVM)
- This 'summary' is intended more towards those of you
who did not go to the last Bitnet convention, both to let
you know what you missed, and to give you some reasons as to
why you should go to the next one (or at least give it some
serious thought!). With that in mind, here is a very quick
summary of the Memorial Day event.
0 Due to circumstances beyond my control, I was not able
to make it to the party on Friday night, where most of the
Bitnauts first met each other. But that didn't really
matter too much, as most of them were there the other days
as well, and I got to meet them then.
0 On Saturday morning, (my roommate and) I arrived in New
York, and were met by Miriam, Mr. T., and Ferret (Marvin).
We took a taxi to the Essex house (where the convention was
held), joined a large group about to have lunch (at a nearby
delicatessen), and got to meet a lot of the people there.
0 After lunch, the group started to split up into about 4
subgroups, each of which went touring a different part of
New York. (For example, my group, led by Lynn Snyder, went
to the World Trade Center.) After a few hours, all the
groups got together and redivided again for dinner. Our
group went to Chinatown, another went into Greenwich
Village, etc. After dinner, some people in each group got
together again and went to see the Rocky Horror Picture
Show, while others went dancing, others went bar hopping,
and still others did (or tried to do) all three. It was
basically Do What You Want To Do Night.
0 The next morning: Was anyone awake? I doubt it!
0 The next afternoon: Breakfast. Although SOME people
woke up before 3:00 PM, I was not among them... so my
(relatively) small group of 4 went on a quest through
Central Park -- In Search Of Ray's (Original) Pizza. At
5:30, we gave up and went to The Village and found one
there. We made it back to the hotel at around 7:00, just in
time to join the groups again. (But after spending a great
time in the rain with Steve Goldsmith (SEGXU) and Teri Blake
(Blonde Bomb) in Central Park).
0 That night, the plans were to meet at the Hard Rock
Cafe, but there was an hour's wait, and Bitnauts (at least,
on the average) don't like waiting that long, so we went to
Danceteria, found too high a cover charge, and decided to
have our own party in the hotel. (See -- Bitnauts know how
to find fun, or at least make their own).
0 12
1
0 Vm-Com Issue 2.2
- I won't go into too much detail about the party... but
I will say that a BITNET party HAS to be experienced. :-)
0 What I will say is this -- speaking seriously for a
moment (but only for a moment!) -- I had never been at a
drinking party before... and now I can add that (as well as
many other "firsts") to my "list of experiences."
0 The party lasted till 5:30 in the morning, even though
only four people actually made it that far. Most of the
people could make it out on their own, but we had to help a
couple back to their rooms.
0 Speaking of rooms -- the rooms at the Essex were really
nice, and the key mechanisms (somewhat resembling punch
cards) allowed every person in a room (maybe an average of
4) to go in and out at will without disturbing the others.
The management was very helpful in replacing bent/lost keys,
and changing locks.
0 Finally, Monday came. (Or for those few of us still
awake after Sunday night's party, breakfast time came). We
went out in little groups to various dellies and cafes, and
got back to the hotel lobby for numerous goodbyes.
0 Most likely, the majority of the Bitnauts attending the
Spring NetCon will be there at the upcoming one in
Washington in October, but there will be quite a lot of
"first timers," too.
0 Even if you only know a couple of people there, or
hardly ever chat, that's not a real reason for not going.
NetCon is an ideal place to meet your old friends, and make
several new ones.
- Well... that about covers it... (Spring) NetCon '85 in
a 85-line glance. I could go on for pages and pages with
details about New York, but I won't, mainly for three
reasons: I have a deadline on my hands, this next NetCon
will be in Washington, and so the physical surroundings will
all be different, and, finally, because there is no way to
do justice to 3 days in less than that many pages... (not
to mention it would probably bore a few of you!) but you'll
just have to go to the next NetCon and see for yourself!
- Note: Although this is not REALLY meant as an advertisement
for the upcoming BITNET 'Convention,' it will almost have to
be taken in that vein. -M.K.
-
0 13
1
0 Vm-Com Issue 2.2
0 A Poem
+ A Poem
0 Written by Guy L. Steele Jr.
Submitted by Mike Dow (CSNM065@MAINE)
-
I think that I shall never see
A matrix lovely as a tree.
Trees are fifty times as fun
As structures a la PL/I
(Which Dijkstra claims are too baroque).
And SNOBOL's strings just can't compare
With all the leaves a tree may bear,
And COMIT strings are just a joke.
Vectors, tuples too, are nice,
But haven't the impressive flair
Of trees to which a LISP is heir.
A LISPer's life is paradise!
0 Many people think that JOSS
And others too, are strictly boss;
And there are many BASIC fans
Who think their favorite language spans
All that would a user please.
Compared to LISP they're all a loss,
For none of them gives all the ease
With which a LISP builds moby trees.
0 RPG is just a nurd
(As you no doubt have often heard);
The record layouts are absurd,
And numbers packed in decimal form
Will never fit a base-two word
Without a veritable storm
Of gross conversions fro and to
With them arithmetic to yield
And decimal places represent
Truncation loss to circumvent:
Thus RPG is second-rate.
In LISP one needn't allocate
(That boon alone is heaven-sent!)
The scheme is sheer simplicity:
A number's just another tree.
When numbers threaten overflow
LISP makes the number tree to grow,
Extending its significance
With classic tree-like elegance.
A LISP can generate reports,
Create a file, do chains and sorts;
But one thing you will never see
Is moby trees in RPG.
-
14
1
0 Vm-Com Issue 2.2
0 One thing the average language lacks
Is programmed use of push-down stacks.
But LISP provides this feature free:
A stack -you guessed it- is a tree.
An empty stack is simply NIL.
In order, then, the stack to fill
A CONS will push things on the top;
To empty it, a CDR will
Behave exactly like a pop.
A simple CAR will get you back
The last thing you pushed on the stack;
An empty stack's detectable
By testing with the function NULL.
Thus even should a LISPer lose
With PROGs and GOs, RETURNs and DOs,
He need his mind not overtax
To implement recursive hacks:
He'll utilize this clever ruse
Of using trees as moby stacks.
Some claim this method slow
Because it uses CONS so much
And thus requires the GC touch;
It has one big advantage, though:
You needn't fear for overflow.
Since LISP allows its trees to grow,
Stacks can to any limits go.
0 COBOL input is a shame:
The implementors play a game
That no two versions are the same.
And rocky is the FORTRAN road
One's alpha input to decode:
The FORMAT statement is to blame
But on the user falls the load.
And FOCAL input's just a farce;
But all LISP input comes pre-parsed!
(The input reader gets its fame
By getting storage for each node
From lists of free words scattered sparse.
It parses all the input strings
With aid of mystic mutterings;
From dots and strange parentheses,
From zeros, sevens, A's and Z's,
Constructs, with magic reckonings,
The pointers needed for its trees
It builds the trees with complex code
With rubout processing bestowed;
When typing errors do forebode
The rubout makes recovery tame,
And losers then will oft exclaim
Their sanity to LISP is owed -
To help these losers is Lisp's aim.)
-
15
1
0 Vm-Com Issue 2.2
0 The flow-control of APL
And OS data sets as well
Are best described as tortured hell.
For LISPers everything's a breeze;
They neatly output all their trees
With format-free parentheses
And see their program logic best
By how their lovely parens nest.
While others are by GOs possessed,
And WHILE-DO, CASE, and all the rest,
The LISPing hackers will prefer
With COND their programs to invest
And let their functions all recur
When searching trees in maddened quest.
0 Expanding records of fixed size
Will quickly programs paralyze.
Though ISAM claims to be so wise
In allocating overflow,
Its data handling is too slow
And finding it takes many tries.
But any fool can plainly see
Inherent flexiblity
In data structured as a tree.
0 When all their efforts have gone sour
To swell fixed records, losers glower.
But list reclaimers hour by hour
By setting all the garbage free
Yield CONSequent capacity:
Thus trees indefinitely flower.
(And trees run on atomic power!)
0 To men of sensibility
The lesson here is plain to see:
Arrays are used by clods like me,
But only LISP can make a tree.
0 -The Great Quux (with apologies to Joyce Kilmer)
((c) 1973 Guy L. Steele Jr. -@MIT-)
-
-
-
-
-
16
1
0 Vm-Com Issue 2.2
0 OpCodes
+ OpCodes
0 Various and Assorted Creative Minds...
-
Here, back by popular demand... OPCODES!!! This is the
second collection of humorous machine language opcodes,
written by a number of clever thinkers of unknown origin.
- GBB Go to Back of Bus
JRAN Jump RANdom Õnot to be confused
with IRAN - Idiots RANdomþ
AAR Alter At Random
LAP Laugh At Program(mer)
BAW Bells And Whistles
PAS Print And Smear
AAD Alter All Data
AAC Alter All Commands
SOAWP SOlve All the World's Problems
CAIL Crash After I Leave
ABR Add Beyond Range
DW Destroy World
DW Destroy Work
DTC Destroy This Command
DTVFL Destroy Third Variable From Left
DSPK Destroy Storage Protect Key
DSH Destroy Sector Header
DPR Destroy Program
DMPK Destroy Memory Protect Key
DK Destroy Klingons
DD Destroy Disk
DPK Destroy storage Protect Key
RM Ruin My files ÕUNIXþ
WAD Walk Away in Disgust
DKP Disavow Knowledge of Programmer
JOM Jump Over Moon
TOG Time Out, Graduate
COWHU Come Out With your Hands Up
SWOS SWap out Operating System
OML Obey Murphy's Law
EBRS Emit Burnt Resistor Smell
MSIP Make Sure Plugged In
ICB Interrupt, crash and burn
IOP Interrupt processor, Order Pizza
CSN Call Supervisor Names
LSBL Lose Super BLock ÕUNIX onlyþ
MBTOL Move Bugs to Operator's Lunch
FB Find Bugs
CRASH Continue Running after Stop or Halt
WCR Write to Card Reader
ROPF Read Other People's Files
GEWJO Go to the End of the World, Jump Off
0 17
1
0 Vm-Com Issue 2.2
0 UAI Use Alternate Instruction set
GENT GENerate Thesis
ITML Initiate Termites into Macro Library
SDPO Spin DASD Platters Opposite
DSR Detonate Status register
GSI Generate Spurious Interrupts
GRMC Generate Rubber Machine Check
IRPF Infinite Recursive Page Fault
ZRB Zero Random Block (of storage)
GREM Generate Random Error Message
EM Evacuate Memory
GMC Generate Machine Check
GMCC Generate Machine Check and Cash
GLC Generate Lewd Comment
GIE Generate Irreversible Error
SHON Simulate HONeywell CPU Õpermanent NO-OPþ
BOMB Branch Out and Make Bugs
IDC Initiate Destruct Command
ICMD Initiate Core Melt-Down
ELEF Erase Last-Edited File
RLC Relocate and Lose Core
BOMR Branch Out of Machine Room
UP Understand Program(mer)
STD Stop, take drugs
CBE Crash and Burn on Error
RIC Reverse Instruction Counter
-
-
-
-
-
-
-
-
-
- 18