I like to tell my younger colleagues that I am disappointed when my code works correctly the first time. No bugs to hunt down? Where’s the fun in that? 😀
On Sun, Aug 22, 2021 at 14:49 Bill Johnson < 00000047540adefe-dmarc-requ...@listserv.ua.edu> wrote: > You claim to know of a 1 line APL super complex program but when asked to > prove it can’t. I get out of bed on the same side every day. > > > Sent from Yahoo Mail for iPhone > > > On Sunday, August 22, 2021, 2:21 PM, Jeremy Nicoll < > jn.ls.mfrm...@letterboxes.org> wrote: > > On Sun, 22 Aug 2021, at 18:43, Bill Johnson wrote: > > Anyone who writes a compiler or assembler is quite complex. And very > > likely thousands of lines of code that took years to develop. > > More than just a few thousand, I'd expect, unless it's very-table-driven. > > > > More in line with the COBOL programs I was referencing. Not some 40 > > line REXX program that took a day or two. > > You seem to have a bee in your bonnet about 40-line REXX programs. > Why? What makes you think that REXX programs can't also be much > much longer? > > > In College, I wrote an ATM machine. It > > took the entire semester and was my class project. Way more complex > > than a 40 line REXX/CLIST > > There you go again. Did you get out of bed on the wrong side today? > > > or the APL mirage you mention. Show me the 1 > > line complex APL program. Remembering, I’m a math major. > > I haven't programmed in APL since the early 1980s, so I cannot produce > an example of my own. But you could look at the "Game of Life" and > "HTML tags removal" code examples at > > https://en.wikipedia.org/wiki/APL_(programming_language) > > to get an idea of how unreadable APL can be. > > Note that it has a right to left execution order, and a huge range of > operators which almost all do different things depending on whether > they are used in a monadic or dyadic way - like the difference between > the way "-" in arithmetic is either used for negation or subtraction. > People who like deliberately writing impenetrable APL expressions > tend to write heavily nested expressions. Of course such things are > unmaintainable - it's a bit like the impossibility of understanding a > complex regular expression at a glance. > > It's definitely a programming language where people learn idiomatic > ways of achieving something then don't look again at the nitty-gritty of > what a series of operators actually do. If someone strings a sequence > of idioms together in one statement it's hard to see where each one > starts and ends. And if someone subtlely changes something inside > what looks like an idiom (but isn't) it'd be hard to spot that too. > > Although APL is good at handling multi-dimensional matrices of numbers > it can be used for non-mathematical things too; eg I wrote a simple full- > screen text editor, in APL, for editing APL functions. At the time the > supported (built-in) way of editing APL functions was much more like > line-mode editing of BASIC code. > > -- > Jeremy Nicoll - my opinions are my own. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > > > > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
