[Date Index] [Thread Index] [Author Index]

Re: A Question About Directive [off-topic]

• To: mathgroup at smc.vnet.net
• Subject: [mg111292] Re: A Question About Directive [off-topic]
• From: Murray Eisenberg <murray at math.umass.edu>
• Date: Mon, 26 Jul 2010 06:39:14 -0400 (EDT)

No, not really!

As with nearly any programming language, it's possible to write nearly 
inscrutable code -- even with Mathematica.

In APL, lots of folks did, and may still do, try to write "one-liners" 
and use cryptic names for variables and functions. But lots of APL code, 
even though very, very terse, could be, and is, very readable -- if you 
know the language!

In fact, one reason for APL's popularity, and continued use, is that the 
developer can throw together a prototype quickly. And modify the 
prototype until the results meet the client's ever-changing need. 
"Read-never" could hardly be the case with such code that one needs to 
modify and otherwise maintain frequently.

The "write-once, read-never" designation is thus to some extent a 
slander perpetrated by people who did not bother to learn the (very 
descriptive) symbols, syntax, or semantics, or who were unable or 
unwilling to do the higher-level thinking involved in manipulating 
entire arrays at once.

On 7/25/2010 2:00 AM, AES wrote:
> In article<i2eadd$q1l$1 at smc.vnet.net>,
>   Murray Eisenberg<murray at math.umass.edu>  wrote:
>
>> Fewer keystrokes is not the sole metric for simplicity! Code readability
>> is another.
>
> Agreed!
>
> Wasn't APL the classic example of a language with near-minimum
> keystrokes, near-maximum unreadability?  -- such that APL programs were
> sometimes characterized as "write once, read never".
>

-- 
Murray Eisenberg                     murray at math.umass.edu
Mathematics & Statistics Dept.
Lederle Graduate Research Tower      phone 413 549-1020 (H)
University of Massachusetts                413 545-2859 (W)
710 North Pleasant Street            fax   413 545-1801
Amherst, MA 01003-9305