Date:         Wed, 1 Oct 1997 19:37:39 +0100
Reply-To:     Harvey.Monder@SANDOZ.COM
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Harvey Monder <Harvey.Monder@SANDOZ.COM>
Subject:      Re: GPLOT and Symbol statement question

     Let me go back a little further.  I have worked in a number of
     disciplines with SAS.  One thing I learned is that the needs and
     'defaults' of one area may be completely different from those of
     another.  SAS may have been written with the needs of a particular
     area in mind.  It is possible that, at the time, it was more important
     that each symbol statement cycle through the color map than for each
     symbol statement to be used once in sequence for each color.  Remember
     that computer technology has advanced considerably since the 1970s.
     Hardware limitations may have been a factor.  The symbol statement has
     remained the same since it was introduced.  Now you have an issue of
     backward compatibility.  How do you change a statement without causing
     possibly millions of dollars in maintenance problems?
 
My argument on the other level is that if you use a product you have the
responsibility of knowing its peculiarities.  If you don't read the instruction
manual the result is your problem.  If SAS did not look so much like English
users would be more cautious in how they used the language.  That is the reason
for my statement.
 
I have no vested interest in keeping the language opaque.  I wish it were
possible to use 'natural language' statements.  I am just bothered by the amount
of misinformation generated by so-called programmers because they don't know
their limits or the limitations of the language.  I am concerned by the
misinformation that is deseminated by users of these languages.  To the general
public, if it appears on a computer printout it must be true.  I know that
government decisions are made on this basis with no one checking to see if the
data was validated.
 
        Harvey
 
 
 
 
______________________________ Forward Header __________________________________
Subject: Re: GPLOT and Symbol statement question
Author:  owner-sas-l@UGA.CC.UGA.EDU at INTERNET1
Date:    10/1/97 05:48 PM
 
 
On Wed, 1 Oct 1997, Harvey Monder <Harvey.Monder@SANDOZ.COM> wrote:
 
>     This addressed a pet peave that I have with so many individuals who
>     program in SAS.  SAS is a programming language.  Before you use a
>     statement learn what that statement is supposed to do.  Don't take it
>     for granted that just because you think it should do what you want
>     that it will do what you want.  Learn the rules before complaining
>     about possibly nonexistant bugs.
>     I think that the original designers of SAS may have made a mistake.
>     Instead of making it look like English they should have made it look
>     like C or APL so that users would be forced to learn the languange in
>     order to produce output.
 
Harvey. that is certainly one viewpoint, and one likely to be held by many
of those (e.g. professional programmers) who have a vested interest in a
programming language 'not becoming too easy'.  However, others may argue
that if 'the designers of SAS made a mistake', that mistake may have been
(in some cases) not to let the 'English-sounding' statements have intuitive
meanings, rather than your suggestion that they should have deliberately
made it more difficult to apply any intuition to the language.
 
As I've said before in such debates, one could produce a car in which the
gas and brake pedels were in reversed positions, the car moved to the left
when you turned the steering wheel clockwise etc. etc. - and then respond to
all the moans by telling people that this was all explained in the manual.
However, the only ones to real benefit would be those 'drivers for sale' who
had been specially trained to drive these 'back to front' cars!
 
Closer to home, many of the professional programmers' complaints and
criticisms of BASIC over the years have had much less to do with the
technical deficiencies of the language (particularly more recent
implementations) than with the fact that it was so intuitive that
12-year-olds were using it competently!
 
Regards
 
John
 
-----------------------------------------------------------
Dr John Whittington,        Voice:      +44 1296 730225
Mediscience Services        Fax:        +44 1296 738893
Twyford Manor, Twyford,     E-mail:     johnw@mag-net.co.uk
Buckingham  MK18 4EL, UK    CompuServe: 100517,3677
-----------------------------------------------------------