« first day (2362 days earlier)      last day (299 days later) » 

4:24 AM
, nice
I did the same thing but with some unnecessary complication in the middle in place of
Why don't and ⊥⍣¯1 do the exact same thing? gives as many digits as defined in the left argument, while ⊥⍣¯1 gives as many digits as needed.
 
there is no way to determine the original length
 
5:34 AM
@rabbitgrowth Bevause Encode allows you to have mixed base
some APLs have Encode with scalar left argument be the same as inverse Decode
 
 
3 hours later…
8:47 AM
the spec subsection for Depth on the wiki seems wrong, idk how to fix tho
oh actually the one for Rank looks the same, maybe it just needs changing the glyph
> Since a depth specification greater than the depth of the argument array means to apply the function to the whole array, (⌊/⍬) or , depending on the implementation, is "depth infinity" and always specifies the whole argument array. However, in practice, any large number, for example 9, is sufficient for most uses.
9 is canonically close to infinity
 
 
4 hours later…
1:07 PM
Still trying to wrap my head around function composition / tacit stuff:

parts← (((3|⍳∘≢)⍳10)∘=)

All of parts 0, parts 1 and parts 2 work, but parts/0 1 2 does not. Why not?
 
parts/0 1 2 is equivalent to (0 parts 1) parts 2, and parts is a monad
I think what you meant is parts¨0 1 2
 
@RubenVerg yes I mean that 🙄, thanks Ruben
 
by the way, when did the transition from / being an adverb which does Reduce with verb operand and Replicate with noun operand (as A Dictionary of APL describes it) to hybrid operator/function happen? or was it already a thing? (there is a sentence that seems to imply it was a thing already and iverson just didn't like it c: )
 
 
10 hours later…
11:10 PM
@RubenVerg It was already in the A Programming Language book, and was bring up together, see page 16-17 softwarepreservation.org/projects/apl/Books/…
Actually, the concept of adverb/noun appeared at much later time
And the reason reduce and compress used the same notation is because ≢u/A ←→ +/u
 
11:26 PM
The suggestiveness of notation was preferred and there was really not much concern about consistency of parsing (or static analysis of program).
Also in the book +/a b c d is ((a + b) + c) + d
It is later in Sharp APL (and J) they start trying to fix the early design mistake that make tactic trains and forks harder to write
And trains was a relatively later time idea.
 

« first day (2362 days earlier)      last day (299 days later) »