 | J | 544 Members |
| 7 Servers |
| J | 544 Members |
| 7 Servers |
| Sender | Message | Time |
|---|---|---|
| 23 Sep 2023 | ||
 raulmiller | And, since the "infinite loop" form doesn't have a second inflection, the first inflection has to be about the results. | 11:22:19 |
 zard03z#0 changed their profile picture. | 17:45:17 | |
 .jtacoma joined the room. | 19:48:39 | |
| 24 Sep 2023 | ||
 anlesvavor joined the room. | 01:43:39 | |
 discodoug | I’m a little late to the discussion about this article: https://ronjeffries.com/xprog/articles/j-this-is-hard-however/, but to claim that all he needed in Ruby was a little more than some for loops seems pretty short sighted if not disingenuous. I don’t know Ruby at all and could make basic sense of his code but that’s a far cry from being able to just get ‘er done. | 06:42:08 |
| discodoug | He’s obviously more able to leverage his understanding of other languages with Ruby over J but that doesn’t make it inherently easier. | 06:44:14 |
 gluttonousporcupine | 'what part of using a complex number in this way is not an egregious hack?' he may not get it, but at least he gets it | 06:44:22 |
| discodoug | It’s not clear how deeply he’s thinking about it. Is the complaint “complex numbers should only be used for complex math dammit!” or “might there be baggage coming from using them in this way? Overhead? Unexpected behavior?” | 06:48:31 |
| discodoug | I’m not trying to be a valid booster but I find the psychology interesting. That he starts off the column with “…and I have reason to believe that I used to be quite intelligent” is pretty telling I think. | 06:52:02 |
| discodoug | I also think his style if compiler/interpreter assisted coding is somewhat at odds with J. Just typing random stuff and hoping the compiler will give you a clue just isn’t going to work in J. | 06:56:03 |
| discodoug | * I also think his style of compiler/interpreter assisted coding is somewhat at odds with J. Just typing random stuff and hoping the compiler will give you a clue just isn’t going to work in J. | 06:56:15 |
| discodoug | I’ll admit ti similar frustrations starting out but I learned to change my tack. | 06:56:44 |
| discodoug | * I’ll admit to similar frustrations starting out but I learned to change my tack. | 06:56:52 |
| discodoug | (I need thinner fingers. My typing sucks..) | 07:01:30 |
 miku.rb | ngl i did not even think to try complex nums with #, so is entirely surprising to me | 07:12:25 |
| miku.rb | ig ive never actually made it to the bottom of that vocab page, wow | 07:14:09 |
| discodoug | bobterryo posted about this recently, but it’s not really useful for solving this problem. Maybe this makes his point but he keeps talking about finding three rolls in a row but never makes it to window. | 07:21:22 |
| discodoug | Perhaps the real point is to attack it like you would learning a natural language. Pick up a few things and try to express yourself with what you know instead of being frustrated that you can’t express more complicated thoughts. | 07:23:18 |
 coltim | I read it more as surprise that that overload existed, more than anything else. probably coupled with a "how was I supposed to know that" that I guess could come from reading the docs but honestly is probably learned from another person instead | 07:47:57 |
 _elcaro_ | Chalk me up as someone who also didn't know about this with #, but I can imagine situations where it will be useful. I'm sure I've had situations where I wanted to insert something.Rather than "egregious hack" I think this comes about simply because complex numbers are in J, so it opens the door to do something interesting with them when passed to a primitive, rather than just domain error. The other one that comes to mind is i: which - given x j. y - does -x to x in y stepsI also wouldn't mind this functionality for i. too (currently, it's a domain error) | 07:48:34 |
| _elcaro_ | * Chalk me up as someone who also didn't know about this with #, but I can imagine situations where it will be useful. I'm sure I've had situations where I wanted to insert something.Rather than "egregious hack" I think this comes about simply because complex numbers exist in J, so it opens the door to do something interesting with them when passed to a primitive, rather than just domain error. The other one that comes to mind is i: which - given x j. y - does -x to x in y stepsI also wouldn't mind this functionality for i. too (currently, it's a domain error) | 07:48:49 |
| coltim | * I read it more as surprise that that overload existed, more than anything else. probably coupled with a "how was I supposed to know that" that I guess could come from reading the docs but honestly is probably learned from another person who tells you about it instead | 07:48:49 |
| _elcaro_ | * Chalk me up as someone who also didn't know about this with #, but I can imagine situations where it will be useful. I'm sure I've had situations where I wanted to insert something after each occurrence of something, and this seems a terse way to do it.Rather than "egregious hack" I think this comes about simply because complex numbers exist in J, so it opens the door to do something interesting with them when passed to a primitive, rather than just domain error. The other one that comes to mind is i: which - given x j. y - does -x to x in y stepsI also wouldn't mind this functionality for i. too (currently, it's a domain error) | 07:50:06 |
| discodoug | This post from bobterryo | 07:52:54 |
| _elcaro_ | It's not shorter than the obvious solution, but that recent Perl Weekly Challenge where you had to duplicate 0's could be expressed as (#~ 1 j. 0 = ]). If the challenge was eg. insert a 1 after each zero, it would have been nice. | 07:59:01 |
| _elcaro_ | Another use for this thing that comes to mind. I dunno if this is the best way, but no denying it's utility. | 08:00:57 |
| _elcaro_ | * Chalk me up as someone who also didn't know about this with #, but I can imagine situations where it will be useful. I'm sure I've had situations where I wanted to insert something after each occurrence of something, and this seems a terse way to do it.Rather than "egregious hack" I think this comes about simply because complex numbers exist in J, so it opens the door to do something The other one that comes to mind is i: which - given x j. y - does -x to x in y stepsI also wouldn't mind this functionality for i. too (currently, it's a domain error) | 08:05:50 |
| discodoug | Better than ((I.@]{{'-'x}y}}(#~1+]))(= toupper))'camelCase' | 08:31:48 |
| discodoug | Doesn’t even work.. | 08:33:03 |
| discodoug | I think I can fix it but not while on my phone and probably not worth it.. | 09:21:36 |