What is the correct result?

[carey]

(03-21-2024 07:08 PM)John Keith Wrote:  

(03-21-2024 04:44 AM)DGM Wrote:  ... I also got into J (jsoftware.com), which Kenneth Iverson and Roger Hui came up with as “APL without the funny characters”, using regular symbols like “,” and “;” and “<“ and “?” to do pretty much the same thing: have a simple way to write terse code.

J seems even less readable than APL to me due to the seemingly random use of ASCII characters for mathematical operations. And people think RPL is unreadable. Harrumph!

Yes, while APL’s eventual character set (not fully formed in Iverson’s 1962 book “A Programming Language”) was constrained to use characters (glyphs) that could be included in an IBM selectric typewriter font ball, at least APL glyphs evoke some visual connection to their name or function (e.g., grade up ⍋, grade down ⍒), little visual connection between ASCII symbols and name is found in J, whose appearance has justifiably been described as “line noise.” While a pain to use, APL glyphs are an endearing feature of APL, and are iconic (pun intended).

It may be of interest that ELI is a largely ISO APL compatible APL variant (with additional features e.g., complex #’s, tables, etc.) that also uses an ASCII character set, but unlike J, ELI’s ASCII character set bears at least some visual connection with APL glyphs, names or functions (e.g., compare iota ⍳ (iota in APL) to ! (iota in ELI). Having such visual cues to APL glyphs while using an ASCII font is a nice feature of ELI.