taylor@hplabsc.UUCP (07/01/86)
--------
Computers and Society Digest, Number 23
Thursday, May 29th 1986
Today's Topics:
Administration
Computer Victims...
Broadening this digest a bit...
Computers & Bad Coding and Writing (5 msgs)
[some of you may receive this issue twice. I apologize. One of the
mailers enroute crash-burned and some people got more than one copy
of the digest, while others got truncated ones, and still others didn't
receive any copies at all! It's almost enough to mistrust computers!]
----------------------------------------------------------------------
Date: Thursday, May 29, 1986 at 10:36 MST
From: Dave Taylor <taylor@hplabs>
Subject: Administrative miscellany
Just a quick note to let you know that we're still working furiously
trying to create a moderated group - the local administrators are a
bit bogged down PLUS no-one here as ever moderated a group before...
Secondly, it appears that some members of this list are using the
relay address to send messages to the group. PLEASE DO NOT DO THIS!
If you have a message of interest to the group, send it to me, as
usual, and I'll make sure it's included in the list.
Finally, I've received a lot of mail about the digest versus broken
up messages debate, and it looks like those that would like the
digest will remain on the mailing list and those that prefer individual
messages will be reading it in the news system...More soon.
Take care, everyone
-- Dave
ps: I'll be at USENIX in two weeks - I hope to bump into at least SOME of
the readers of this mailing list!
------------------------------
From: ihnp4!opus!ki4pv!tanner
Subject: Computers & Bad Coding
Date: Wed May 28 09:33:36 1986
It is not that the computer is encouraging bad coding as such.
There are certain people who will (if they write code) write bad
code. The interactive computer merely makes it possible for more
people to write code, and thus to write bad code.
It is unfortunate, but the toy (home) computer and the company
IBM-on-the-desk put the ability to write badly in the hands of more
people: thus, more bad code.
Tanner Andrews, systems
compudata south, deland
------------------------------
From: Dave Taylor <taylor@hplabs>
Date: Tue, 27 May 86 11:55:23 PDT
Subject: Computer Victims...
[this article is ripped out of last Sundays issue of the L.A. Times opinion
section. The interesting thing about it isn't so much what is being said
but rather what is being accepted: that computers are `in control' and
are `omniscient' and cannot be argued with (etc). I welcome disagreement
on my interpretation... -- Dave]
The Ultimate Computer Victim Isn't Resting In Peace
---------------------------------------------------
by FORMAN BROWN
"I first became aware of my death last May when my checks began to
bounce. Never having experiencd bouncing checks before, and knowing that I
had quite a respectable balance at the bank, I was both shocked and angry.
When I examined the returned checks and found, stamped over my signature on
each of them, in red ink, "Deceased," I was mystified. Then, when one of the
recipients of my chekcs, a utility company, demanded that I appear in person,
cash in hand plus $10 for their trouble - THEIR trouble! - I was shocked,
angry and mystified. I wondered just how they expected us deceased to
acquiesce.
"I made an immediate visit to the bank, which had, I thought, been
my friendly protector for 20 years or so. The young lady to whom I was
assigned assured me that nothing like this had ever happened before, so it
couldn't have happened now. But it had happened, and obviously the fault
was not mine since there I was, sitting efore her desk, alive, and naturally,
kicking.
"The young lady was polite and apologetic but had not the foggiest
notion of what was to be done. I suggested that she call on a higher authority,
and with evident relief she directed me to Window 1, where I could speak with
Mohammed. Mohammed? "Good Lord," I thought, "they have not only killed me,
they are sending me to the wrong heaven."
"However, I obliged, and sure enough after some delay, Mohammed, banker
personified, appeared at the window. I repeated my story and his reaction
was much the same as that of the young lady. He could in no way account for
such a "mistake," a word I felt to be singularly inept for what his institution
had done to me. he ventured that the fault must lie with the computer. (A
contrivance I have never understood or trusted) and assured me that all would
be promptly righted, and I would be restored to the land of the living.
"I went home, only slightly mollified, to rewrite the checks and a
dozen explainations, including one to a friend in Michigan who had returned
his check to the bank with quite a gratifying letter of condolence to my
"estate,", very flattering to my ego, but going little for my annoyance.
"I assumed now that I was alive and well until my June statement arrived
from the bank and I found that tthere was no record of my Social Security
payment, which for years had been automatically credited to my account on the
third of each month. Mohammed and I met again, and it now appeared that
whatever had killed me in the bank's computers had spread to those of the
Social Security system as well.
"I was advised to visit the local office of Social Security in person,
to prove that I was, beyound a doubt, alive. Since this confusion was between
the bank and Social Security, it seemed to me that a visitation by Mohammed
rather than by me was in order. Nevertheless, I went.
"After a long wait in line I was confronted by a woman who was neither
polite nor apologetic, just certain that there was absolutely nothing amiss
in HER records.
"When my July statement still showed no payment, I went off once again
to the Social Security office. This time I encountered a true gentleman who
was both polite and understanding. He wrote out in detail my sad story and
assured me to be patient and that all would be well.
"Home I went, hoping that the worst was over. The bank agreed (what
else could they do?) to credit my account each month with the amount of money
due me from Social Security until such time as the agency admitted that I was
alive and renewed my payments.
"That happened seven months later, in Dcemenber. I was relieved, and
so, I guess, was the bank. All was well.
"Or so I thought. Last week my phone rang. It was my doctors
secretary - my doctor who is both physician and friend. It appeared that
Medicare had refused to accept his bill for services rendered me because
the date of such services was six months later than the date of my decease.
So now I am lost in one more computer system.
"If I were 20 all this ado might be merely irritating, with my death
possible but remote. At 85 it can only seem probable, and much too near. Ah
well, at least after 20 years, they know me at the bank."
------------------------------
Date: Tue, 27 May 86 09:56:07 PDT
From: chuq%plaid@SUN.COM (Chuq Von Rospach)
Subject: The effects of computers on writing...
I'm just back from BayCon, the SF Bay Area Science Fiction convention.
There were a couple of panels on the subject of Computers and Word
Processing.
It is fairly obvious that the writer who doesn't use some kind of WP is
quickly nearing extinction. It is also quite clear that the benefits
of a WP aren't clear cut and differ from person to person (panelists
ran from Spider Robinson and Stephen Goldin on Mac's to James Hogan on
his TRS model 3 with a lot of IBMers in the middle.
They were all in agreement that WP has no advantage in the process of
WRITING, the first draft where the creative actually occurs. The
tendency to go back and fiddle early in the draft can get you
sidetracked and cause you to lose your train of thought. The closer
your WP can simulate a typewriter, the happier these people are.
After all of this, though, nobody except Hogan was willing to admit
even the possibility of going back to a typewriter because the WP takes
care of most of the drudge work of writing -- revisions and the
associated retyping. Authors talked about things like wholesale
renaming of characters, rearranging large chunks of text, and other
associated attrocities that they used to try to avoid become trivial,
and they spend their time creating instead of typing. It also leads
towards the infinite revision syndrome, for which someone (I think
Stephen Goldin) mentioned the famous quote:
Art requires two people: An artist to do the creation,
and another person to know when to take it away from them.
I know from my own writing, postings to various lists and groups, my
work on OtherRealms, and the serious writing I'm working on (gads, that
is a rotten sentence) that much of what they say is true.
>From a gramatical and cohesive point of view, most of my postings to
the net (and lists such as this) are sloppy at best. I know this, and
I accept it because I'm willing to take the problems and save the time
involved in doing a second draft. Frankly, my first drafts tend to read
better than most of the material on the net, so this isn't a problem to
me.
For my more serious work, though, the only real advantage that I get
from WP is the ease of revision. I write a first draft and proceed to
print it out to hardcopy, and do all my revisions to the piece of paper
-- I find it impossible to stare at a CRT screen and catch the spelling
errors, syntax and style problems, and all the other associated things
that go wrong with your words. By taking a hardcopy for each revision,
I also find that this keeps me from the infinite revision syndrome.
The thing to remember out of all of this is that WP is a tool. It
won't do the writing for you, it won't even make the writing easier.
What it DOES do is make the act of writing less onerous by helping to
fix some of the tedious parts of the craft -- revision and finishing.
At the same time that it solves some problems of writing, it creates
others -- no technology is a perfect technology.
------------------------------
Date: Tuesday, May 27, 1986 13:46:26
From: jeffries@hplabs
Subject: More on Online Systems and Bad Coding...
I hate to burst Bruce Hamilton's bubble, but I've seen exactly those
arguments (the ones in his parody ARE ONLINE SYSTEMS HELPING TO PROLIFERATE
BAD CODING?) made in all seriousness (and using virtually identical words)
by folks who propose that back in the days when people entered programs on
punched cards and had to wait a day for the results of each run, people
wrote better programs. Look at back issues of the software engineering
digest for examples of this perspective.
I think that the real answer is that any tool can be abused. My guess is
that the same people who misuse word processors, etc., also misused the
older technologies. An example that comes to mind is a colleague for whom
spelling checkers are a disaster. This man cannot spell at all. His
pre-spelling checker prose was relatively well spelled because he used an
intelligent system (aka a secretary) correct his manuscripts, but he almost
never caught the mistakes his secretary left in. Now that he uses a
spelling checker, his papers are impossible to read. The words are all
legal english words, but they are often not the words he intended. I find
such texts much harder to read -- a nonsense word is an immediate clue that
a typo has occurred, but if the error is a legitimate other word, then I am
misled to try to make sense of the sentences as it stands before looking
for the typo explanation. I, on the other hand, make extensive (and I
believe productive) use of spelling checkers. I would hate to have them
taken away from me just because they are dangerous in someone else's hands.
Every technology brings its own tradeoffs. I have yet to be convinced that,
in this area at least, the trade-offs required by computer technology are
in any way worse than those of the more traditional methodologies.
Robin
-------
------------------------------
Date: 24 May 1986 0856-PDT (Saturday)
From: Eugene miya <aurora!eugene>
Subject: Two things: Writing and Social Responsibility
On writing:
How valid a comparison is it been the existing "word processing"/writing
and the much older "hand scribe"/writing (maybe calligraphy would be better)
debate? I started typing in Fifth grade (hunt and peck) because I did not
want to bother with handwriting (unreadable any way).
I now know there is a tendency on the part of older generations to take
anything in machine printed form as WORD. I like text editors because
they free me from paper. I wish a had a better CAD type system for
drawing. (Patience.) Similarly, structure editors are nice for outlining.
Once in a machine readable form, great, spelling checking, style checking,
etc. can be applied. I'm freed from paper-(hard)-ware. Not perfect,
but better. What's really important are the ideas right? I admit the
misplaced letter can result in grave misunderstanding, but that's where
intelligence comes in (I hope), or else those pushing better writing
are just doing this mechanically.
The latter reminds me a a variation by Hamming on the purpose of computing
being insight, not numbers. The analogy on the purpose of exposition
would be ideas, not just words on a page. I do have sympathy, Dijkstra
said teaching people to program in their natural language leads to learning
that people don't know their own language.
On Social Responsibility: The original group PSR (Physicians for
Social Responsibilty) was started over a concern for nuclear war.
CPSR (Computer Professionals for SR) original was conceived for
principally the same reason, but there seem to be larger issues
facing computers. The physicians don't worry about malpractice,
they have over groups (AMA?) which worry about that. The problem
is one between keeping a focus and reaching a wider audience.
>From the Rock of Ages Home for Retired Hackers:
--eugene miya
NASA Ames Research Center
eugene@ames-aurora.ARPA
"You trust the `reply' command with all those different mailers out there?"
{hplabs,hao,dual,ihnp4,decwrl,allegra,tektronix,menlo70}!ames!aurora!eugene
------------------------------
Date: 23 May 86 21:17 EDT (Fri)
From: Bob Carter <Carter@RUTGERS>
Subject: Broadening this digest a bit...
From: Dave Taylor <taylor at hplabs>
A few propositions for you'all;
1. We're going to have this group gatewayed into a legitimate
news group finally - so I've been thinking about a possible
change to the format, namely to toss the digest format entirely
and have each message simply added to the group.
Digests are a marked convenience, especially for folks who have
(or who have hacked their own) undigestifying software.
2. Also, there has been a lot of interest in having the group tie
in somehow with the Computing and Social Responsibility groups
around the world -- I was thinking of allowing some postings
from those groups assuming that they aren't too political (as
I'd rather avoid our group falling into the inevitable jingoistic
battlefield that all emotional topics like politics would make it).
Computing and Social Responsiblity groups serve a proper function,
and that function is political advocacy, much of which is pungent and
provocative.
The problem is that advocacy is almost always shrill ("XYZ threatens
life as we know it!") and not only to drowns out gentler stuff, but
tends to shout it down as well ("How can you talk about computers and
writing style? Any message that does not deal with XYZ is immoral!")
There is plenty of pungency and plenty of shrillness on "legitimate
news groups" already. Please keep Computers and Society the way it
is. Many of us like reading something that is not apocalyptic, or
at least like it some of the time.
_B
[this sums up the comments people have been making overall, so it looks like
incorporating the CPSR stuff isn't going to happen in the near future -- Dave]
------------------------------
Date: Fri, 23 May 86 19:39:31 pdt
From: Scott McGregor <hplabs!hpccc!mcgregor>
Subject: Systems and Bad Coding
...There is a hidden risk in the new systems, in that they have enabled us
to bypass the correction and finalizing phases of coding, often
resulting in quick and efficient execution of poorly planned, sloppy,
and confusing code...
Compilers tend to be very quick to correct us on points of syntax, but
very slow to correct, or even recognize, inaccuracy or inefficiency of
algorithms in our own or others' code....
Re-read your code, and have others examine it for efficiency, absence of
kludges, and general comprehensibility, before you submit it to any
computer. Remember that online compilers are only tools, and that the
burden of effective implementation and testing still rests upon those
who use them.
-- Bruce
I agree with Bruce's comments that many people use compilers as
quick ways to check their code. I also agree that compilers are
generally good at catching syntax errors and pitifully poor at catching
semantic (or "logic") errors. And finally, I will agree that
the current result is that many people consider a compiler pass a
sufficient test for bug-free code, but actually therefore perpetrate
buggy code. To some extent I accept Bruce's dictim, IN THE SHORT TERM,
that people need to exorcise a little more restraint in racing their
code into production just because it compiles.
It is true that the advent of interactive systems and CRTs has made
the cost of another compiler pass much less expensive, and so many
people are now much less careful. But why? I would say because they find
error-seeking behavior very annoying. They probably aren't very good at
it (look at all the syntax errors that get past people that try to
desk debug). Their nonchecking behavior is very self-rewarding.
IN THE LONG TERM, attempting to force these people back into more
comprehensive checking is a mistake. What we really need is for
some intelligent individuals to apply themselves to designing
fundamentally new programming approaches. These approaches should make
the possibility of error much more difficult, (e.g. you can't misspell
if you must choose (by pointing) from existing lists!). Errors should
be reported, and be forced to be corrected as soon as possible. Lastly,
the very structure of the "program" should be such that errors in logic
become blatently obvious! (The avoidance of describing the programming
environment as a "language" is purely intentional.)
People are turning to the computers to do their checking for them, because
they want computers to serve them in that way. In the long run, we
shouldn't change the way people must act unless we can also fundamentally
alter the way the FEEL about those actions. Rather, we should mold the
machines to serve people in the way people want to be served.Of
course, such action is far from trivial--and that's a good reason not to
expect this in the short run. But in the long term that's a cop out.
Unfortunately, computer research and development is so fragemented.
There are all these operating systems, all these languages, etc.
that any particular one only gets a small number of people paying attention
to it, and base level systems are built all over the place, instead of
a few tall towers of sophisticated programming. Computing has had
its era of multiple competing technologies (ala the 2000 car building
companies of the turn of the century). Its time for the computing
revolution to shake out some of these and concentrate (like the VW bug)
on improving software engineering into an age of increased reliabilty.
I encourage every programmer to join the revolution and strike a blow
for more HUMAN programming environments!
(Lest it be assumed that the author is a naive freshman with 3 months
BASIC training somewhere, let me put this to rest now--the author has
a masters degree, 14 years programming experience, used ~2 dozen languages
on ~1 dozen operating systems.)
-----------------------------------
To have your thoughts included in this digest, or to join the mailing
list, please send electronic mail to Dave Taylor at any of the following
addresses:
taylor@HPLABS.{CSNET,ARPA -- or -- ..hplabs!taylor
This digest is published approximately bi-monthly and does not necessarily
express the views of HP nor anyone else other than the individual authors
of the messages.
***********************************
End of Computers and Society Digest
***********************************