The APL Orchard

« first day (1869 days earlier) ← previous day next day → last day (802 days later) »

9:06 AM

@RikedyP did my blog clarifications make sense?

10:03 AM

Hi all, first of all thanks for responses to my previous stuff, it's highly appreciated.
Also, I would like to thank for the Array Cast as educating, down to earth content, with inspiring people involved.
I feel like abusing this thread with again a noob question regarding iterations in apl, so excuse me please for annoyance:
Regarding combinations w/o repetitions, I know there is cmat dfns and it's nicely documented. But in general, I'm struggling to bend my mind in reflecting simple python iterations like this one

Adám

@MiljanNC Try using outer products.

Or even ⍳ with a vector argument.

Oh, wait, I didn't understand that Python code correctly.

rak1507

10:20 AM

@MiljanNC there's not really a nice way other than lots of ¨s

⋄ {⍵⌿⍨∧/2</⍵}↑,⍳5 5 5 or something inefficient like this

TryAPL

Fawn Locke

There's an obvious index pattern of ⍳3 4 5. But yeah, just kinda an adhoc problem

MiljanNC

thanks!

Sixtyfive

10:38 AM

@xpqz began reading your book (videos are not my thing for learning) and right at the start it has those links to J, K and Q ... which in the case of K, presents you with these ... interesting ... claims: why: hundred times faster than biqquery redshift databricks snowflake .. and

compare vs biggie shifty sparky flaky .. / taxi: 1.1Billion  10-100 times faster/QP$ / taq: 1.1Trillion 100-INF times faster/QP$

... any clue what all that is in English?

Adám

@Sixtyfive The first is a list of "well-known" databases. The second are some standard database queries on standard data sets.

rak1507

not sure what QP$ means, queries per dollar?

Sixtyfive

I see., thanks Adám :-)

xpqz

As a database, KDB+ is pretty cool. Expensive, though undeniably fast. As with all benchmarks, the comparisons in this case are a smidge apples-vs-oranges.

Adám

Relevant reading: Wild claims about K performance

Sixtyfive

10:43 AM

Was all just double Dutch to me and wanted to knwo what it's about in general.

This is such a different world from what I'm used to. Even the comments. In J you seem to start a comment with NB. .

rak1507

yeah, pretty weird

PyGamer0

NB. Yes you do. I think NB stands for NoteBook

Adám

Nota Bene.

xpqz

NB stands for Nota Bene. Latin for "note"

Adám

Note well, I think.

PyGamer0

10:49 AM

@Adám oh

Sixtyfive

As someone who has been known to annoy his coworkers by using Emoticons as variable names, I do like ⍝ to start a comment :-D

Adám

What does it look like to you?

Sixtyfive

(Yes, note well, if it's indeed nota bene)

@Adám APL, you mean?

xpqz

Using three symbols for a comment is grating, I must admit

Adám

@Sixtyfive ⍝

@xpqz Lots of languages use two.

Sixtyfive

10:51 AM

@Adám Looks like a raised index finger ... or perhaps one of these little coloured markers you can buy to put on pages of dead tree to find them again later.

PyGamer0

@Adám i think intercal uses REM (or was it BASIC)

Basic, for sure

REM is from BASIC.

For "remark".

so intercal has PLEASE NOTE

Sixtyfive

10:52 AM

Had completely forgotten about that ... so many years have passed.

Adám

@xpqz I wonder what they didn't just make it n. — that even looks a little like ⍝

Sixtyfive

Aha, so the Unicode spec calls it "APL Functional Symbol Up Shoe Jot."

xpqz

I'm with @Sixtyfive on lamp looking like a raised index finger.

Sixtyfive

(Whatever an "up shoe" is ... but then, I'm not a native speaker of English)

Adám

A horse shoe when the horse is walking up ↑: ∩

PyGamer0

10:54 AM

and a jot: ∘

i wonder why it is called jot

Sixtyfive

Ooooooooooh

PyGamer0

yeah both are combined to form ⍝

many characters are like that

Adám

@PyGamer0 Because it is small. Something small is traditionally associated with the letter i/⍳/י, known as jodh=jot.

c.f. "jotting down".

PyGamer0

⍤

Adám

Officially, ⍝ is a lamp. The ring and the shoe are supposed to overlap:

Sixtyfive

10:58 AM

Ah, with that overlap I can see the lamp in it.

Adám

Although I think it looks even more like an LED.

Sixtyfive

One of those traditional 5mm THT LEDs perhaps. But one of the legs would have to be a tad longer and there's a flat side missing :-)

/me wonders why APL has persisted for so long while at the same time having so many offshot languages ... J, K, oK, ...

Oh. No /me here. Bummer.

PyGamer0

@Sixtyfive sand the flat side and cut the legs to equal length....

Adám

@Sixtyfive You can put * before and after your message to put it in italics.

Sixtyfive

Ah, but that's still different from the IRC way, where it would have become something like * sixtyfive wonders why ..., as though one was allowing the others a glance into their thoughts.

PyGamer0

11:04 AM

> Using APL can "spoil" a programmer, in the sense that they get frustrated by the lack expressiveness in other programming languages.

Adám

PyGamer0

i see the upshoe jot

nice

Adám

@PyGamer0 … but also make them hunger for even more expressiveness, leading to a new APL offshoot.

PyGamer0

although its too long

@Adám yeah true

Sixtyfive

Hmm. It would then seem that there is also such a thing as too much expressiveness, hence allowing the original APL to retain its popularity. That'd make sense I guess...

Sixtyfive

11:22 AM

"The programs, if you can call them that, resemble the results of a prolonged fistfight with a keyboard."

xpqz

@Sixtyfive Where did you find that quote?

Sixtyfive

beyondloom.com/blog/denial.html ... linked to by you

xpqz

Ah yes, thought I'd heard it before. John should blog more.

3 hours later…

Brian BED

2:44 PM

agreed. when i program in python now it just irritates me, so i always stick to APL. Actually something i want to do now is program a discord bot in APL but seems really annoying as i still can't get py'n'apl to work. what i'm doing now is trying to make an apl console app which evaluates input so i can just throw apl code in apl and listen in with python.

Oh that reminds me. how do i make a bound executable? i'm stuck on it needing the find "dyalog180_64rt_unicode.dll" but i can't find it myself

Adám

@BrianBED Are you sure it doesn't work straight from the File menu?

PyGamer0

i just write little apl snippets but im now irritated by python's lambda syntax, lambda is a long word to type...

RikedyP

3:38 PM

@xpqz Yeah I think so; I believe the same link has the latest draft (red text removed). The only thing for me is maybe the bit before Rodrigo's equation should have some connecting text? Maybe " ... explicit version... which can be derived from this simplified mathematical expression, suggested by Rodrigo"?

Brian BED

@Adám what file menu? File explorer?

RikedyP

@BrianBED In Windows IDE, you can File → Export

Also re: pynapl - let me know if you want to email some details (rpark at the domain dyalog), see if we can help

it's also going to fall over as soon as 18.2 goes out, so I need to look at getting that fixed too

xpqz

@RikedyP Sure, that works.

RikedyP

@xpqz if there's nothing else, we'll publish next week, or maybe in tandem with the launch of this year's contest

which should be soon(ish)

1 hour later…

Brian BED

4:53 PM

@RikedyP oh yea where else would i make a bound exe? i used that and made a standalone exe but it didnt work with eval ofcorse which i should have seen coming so i expect bound executable to work but it needs this dll file that it cant find

3 hours later…

Adám

8:02 PM

> didnt work with eval ofcorse

@BrianBED Care to explain ^ a bit more?

Brian BED

8:57 PM

wait first, diffrent issue.

game2←{s←' '~⍨⍕?⍨6⋄W←0⋄_←{⋄⍞←'Guess: '⋄g←' '~⍨7↓⍞⋄s≡g:W ⎕←1 'WIN!!!'⋄⎕←+/s=g⋄⍬}⍣{W}⍬}

why does this function give error on tryapl? spesifically illigal token? on RIDE it works fine

@Adám yes ok so, i know how to run an exe and get the output with python so i wanted to take advantage of that so i can run apl with python. i decided to use eval and ⍞ to get apl code input from console/python and output the answer, and that should make my python be able to run apl no problem. but bound executable i cant get to work. let me see if i can get the error to show

« first day (1869 days earlier) ← previous day next day → last day (802 days later) »

all rooms

Transcript for

Mar8

Mar '229

Mar10

apl.chat ― Learn, teach, ask, code, golf, & discuss usage. See ...

apl array-manipulation j tips

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC