« first day (1649 days earlier)      last day (77 days later) » 

12:04 AM
@Adám by default yes, debugging should be an option but 90%+ of my errors are probably from typos/things where the mistake is immediately obvious so closing the debug window just takes up extra unnecessary time
 
12:16 AM
@rak1507 Huh, but in such a case, I'd fix the problem and continue execution, not clear the stack.
 
@Adám doesn't work if the thing that broke was code in a previous line
 
Yes, it does, just go back to that line and continue from there.
(Unless you're mutating in place. Avoid that.)
 
@Adám doesn't seem like you can then change the line though
 
@Adám deciding where (and if) to go back to already takes more time than just re-executing the whole thing
 
also unless there's some unintuitive keyboard shortcut I don't know about you have to click on actual buttons, no thanks
 
12:20 AM
@rak1507 Shift+Enter.
 
what about going backwards
 
Ctrl+Shift+Bksp
 
that's for the repl not for the debugger
oh, never mind, when the debugger opens it doesn't focus, so you have to click in it either way
and I can't figure out if/how you can edit something too
 
@Adám I have to focus the debugger window to do that though, right? (more time loss!!!)
 
@rak1507 Ctrl+Enter.
 
12:23 AM
but then how do I even edit anything? typing doesn't seem to do anything
 
Shift+Enter.
 
doesn't work if you already have an edit window open
 
Works for me.
 
do you already have )ed f though?
 
Ah, the same function that suspends?
 
12:27 AM
the way I tend to do things is have a function open, work on it, run it in the repl, and then update it, having to close the debugger every time I misspell a variable or something greatly slows that down and if I really need to know what some value is ⎕← is pretty instant to type
@Adám yeah
 
That does fail. I guess it could simply focus on your existing editor. Can you email support with a description of that workflow?
 
my ideal setup would be: edit window open & always focused. Tracer just doesn't exist. There's a shortcut to re-execute the last REPL line. As such, I edit code, ctrl+s, ctrl+whatever-to-execute-last, and go straight back to editing.
 
^ same here
 
You could even combine saving with executing last.
 
is there a shortcut to switch between focusing the repl and the editor window?
 
12:31 AM
@rak1507 ctrl+tab apparently
 
doesn't work for me
aha, thanks
alright well at least I've learned some new things that can make it more bearable
 
@rak1507 Seems that RIDE doesn't have <JP>.
 
whatever i did to make the tracer hidden by default is close enough to good enough, but really is quite awful
 
another small quality of life thing would be if you could disable showing 'edited by <user>' just because my windows pc username is an old cringey thing from an email I made when I was 6 so it'd be nice to 'hide' that if I ever want to take screenshots and things lol
 
@rak1507 I think there's an option to force the author name to something else.
 
12:34 AM
oh cool
 
@rak1507 I can't seem to find it right now (and I shouldn't really be working at this hour) so can you ask support?
 
yeah will do tomorrow morning gonna go to bed soon now anyway
 
 
11 hours later…
11:36 AM
In J, how do I convert a string of digits (no spaces) into an array of integers?
 
@Joshua ".&>
 
how do i represent x²+y²+3x²y³ in apl ? without using parenthesis
 
Thanks @Adám
 
@PyGamer0 Who says you can do that without using multiple statements?
 
@Adám i want to know if there is a way to do it in a single line
 
11:46 AM
@PyGamer0 +.×⍨⍤,+3×,×.*2 3⍨ should work.
 
@Adám how does that work????
 
⎕←+.×⍨⍤,+3×,×.*2 3⍨
 
@Adám
    ┌─┼───┐
    ⍤ + ┌─┼───┐
   ┌┴┐  3 × ┌─┼────┐
   ⍨ ,      , .    ⍨
 ┌─┘         ┌┴┐ ┌─┘
 .           × * 2 3
┌┴┐
+ ×
 
@PyGamer0 Equivalent to (+.×⍨x,y)+3×(x,y)×.*2 3 which is equivalent to (+/×⍨x,y)+3××/(x,y)*2 3 or (+/(x,y)×x,y)+3××/(x,y)*2 3
 
12:43 PM
+.×⍨⍤,+3×⊢×××× and +.×⍨+3×,×.*2 3⍨ for shorter (infix and prefix respectively) versions
+.×⍨+3×5×/⍤⍴⊢ for prefix, taking y x
 
1:13 PM
That's some clever stuff right there.
 
1:24 PM
what is the difference between APL385 and APL386??
 
some added/updated characters
 
1:46 PM
i always see ⍝ as A
so i read comments like ⍝ binary encode as A binary encode
 
@PyGamer0 Might be a font problem, but I have tried exploiting this similarity in my quest for an APL logo:
 
 
3 hours later…
5:03 PM
Announcement: APL Campfire in one hour. Special guest: Dyalog CEO Gitte Christensen.
 
 
2 hours later…
7:05 PM
@Adám so rank wasn't invented by Whitney then? interesting
 
@rak1507 If both Gitte Christensen and Joey Tuttle say so, then I believe it. But Arthur Whitney did do the ground work of leading axis orientation, and he apparently was the first to implement it too.
 
@Adám jsoftware.com/papers/APLPersonalView.htm 'was suggested to me...by Arthur Whitney'
so it seems like he did invent it
 
@rak1507 No, "The central idea behind" it (leading axis theory), but not the operator itself.
 
'and a rank operator'
 
Yes, I also understood it like that before tonight, but it can equally be understood as "The central idea behind [both] the use of cells and [behind the use of] a rank operator"
 
7:11 PM
That would make sense too
 
Or better: "The central idea behind the use of cells [in combination with] a rank operator"
 
Do you know if Gitte misinterpreted the question on other programming languages to languages around the same time (I don't think I put it very well) or if those are the only other ones she's used other than APL even more recently?
 
@rak1507 I was pretty sure you meant more recent ones, but I'm also pretty sure she hasn't, so I didn't bother asking for more detail.
Lately, she barely ever touches APL, even. She has a company to run…
 
Fair enough
 
8:01 PM
@Adám Here's Roger Hui claiming Whitney invented it on the train to APL82. He cites personal correspondence with Whitney and Roland Pesch. What did Gitte say?
 
8:51 PM
@Adám Hm. She doesn't sound entirely certain (although I know she will say "I think" even when she is sure), and I don't see how Iverson would have written "A Personal View" that way if he'd described Rank first. I guess I'll stick with the written sources. But whether Whitney found the Rank operator or just the reason to use it isn't really important, more of a technical difference.
 
 
2 hours later…
11:06 PM
@Adám can't do tradfns in tryapl?
 
@rak1507 Can do tradfns in TryAPL: Press Alt+Enter to insert a line break without executing.
 
ah cool thanks
 

« first day (1649 days earlier)      last day (77 days later) »