|
00:07
stanrifkin left
01:28
sibl joined
01:30
librasteve_ left
|
|||
| disbot6 | <jubilatious1_98524> @rcmlz @antononcube @librasteve I know this is gonna sound kinda selfish, but cracking 4K, 5K, or 10K points on StackExchange would go a long way towards getting people interested in Raku for DevOps: unix.stackexchange.com/users/22773...bilatious1 | 01:54 | |
| <jubilatious1_98524> Right now: 622/629 of my answers on Unix & Linux StackExchange mention "Raku". That's 98.89% | 01:56 | ||
| <antononcube> Hmm... I do not know... My answers on StackExchange do not fair well. Except, if they are posted on MathematicaStackExchange. | 02:18 | ||
|
03:05
arkiuat left
03:19
arkiuat joined
03:33
arkiuat left
03:49
arkiuat joined
03:56
arkiuat left
04:17
arkiuat joined
|
|||
| disbot6 | <rcmlz> Or we ask PewDiePie for a video „I switched to Raku (and so should you)“ youtu.be/pVI_smLgTY0 | 06:56 | |
|
07:00
sivoais_ left
07:02
sivoais joined
08:06
arkiuat left
08:18
arkiuat joined
|
|||
| disbot6 | <simon_sibl> good idea ! | 08:22 | |
|
08:22
arkiuat left
08:27
arkiuat joined
08:33
arkiuat left
09:03
arkiuat joined
09:08
arkiuat left
09:12
dakkar joined
09:38
arkiuat joined
09:43
arkiuat left
09:53
sibl left
10:03
arkiuat joined
10:08
sibl joined,
arkiuat left
10:14
sibl left
10:32
arkiuat joined
10:37
arkiuat left
11:06
arkiuat joined
11:11
arkiuat left
11:30
arkiuat joined
11:35
arkiuat left
11:42
hudo joined,
hudo__ left
12:05
arkiuat joined
12:09
arkiuat left
|
|||
| disbot6 | <jubilatious1_98524> @rcmlz Well...if I've posted 622 Raku one-liners and a substantial proportion (the majority?) have ZERO upvotes, that says to U&L StackExchange that the community isn't supportive. | 12:23 | |
| <jubilatious1_98524> And if the community isn't supportive, why should anyone else adopt the language? | |||
| <antononcube> Exactly that! I can say the same for my Raku packages. Granted, that is easily explained. | 12:35 | ||
| <antononcube> Most people do not reach for Raku in their everyday tasks, and/or, treat Raku programming as hobby. I often read statements proclaiming the latter in some way. | 12:38 | ||
|
12:38
arkiuat joined
|
|||
| disbot6 | <antononcube> These, BTW, are understandable.I personally, use Raku in professional settings only for the LLM workflows (very often) and for text processing or grammar-based parsing (ocassionally). | 12:41 | |
| <antononcube> And, yes, I have/had plans to use Raku for sysadmin tasks. (Very rarely I do.) | 12:42 | ||
|
12:43
arkiuat left
12:50
arkiuat joined
|
|||
| disbot6 | <jubilatious1_98524> weekly: unix.stackexchange.com/a/803066/227738 | 12:50 | |
| <jubilatious1_98524> @antononcube I'd happily upvote you packages if I knew where to do that! | 12:52 | ||
| <antononcube> Let me be more constructive here. I have several related suggestions. | 12:53 | ||
|
12:54
arkiuat left
|
|||
| disbot6 | <antononcube> 1. Make a Raku script -- or package -- that scrapes all answers of StackExchange user in a given community 2. Write and share)a JSON file with DSL examples for doing UNIX tasks with Raku (one liners) | 12:55 | |
| <antononcube> I assume given the package in 1, generating the JSON in 2 is via LLMs is possible and that hard. | 12:56 | ||
| <antononcube> That JSON file with UNIX-by-Raku DSL examples is going to be included into "DSL::Examples", and, hence, be easily accessible by LLMs. We can run an all-Raku server that gives those relevant Raku one-liners (or multi-liners) upon request. | 12:58 | ||
| <antononcube> Instead of doing 1 or 2, one can just: 3. Try to see are UNIX-by-Raku one-liners easily retrievable -- and correctly adjusted -- using a dedicated prompt with general LLMs. | 13:01 | ||
| <antononcube> This might require significant prompt engineering and lots of experiments. | |||
| <antononcube> I hope it is clear about the reason for this suggestions -- if LLMs easily and readily can produce functional Raku code for everyday tasks, then Raku's mindshare would increase. | 13:03 | ||
|
13:07
arkiuat joined
13:12
arkiuat left
13:18
arkiuat joined
13:24
arkiuat left
13:35
arkiuat joined
13:45
arkiuat left
13:50
arkiuat joined
13:59
arkiuat left
14:27
arkiuat joined
14:32
arkiuat left
14:41
librasteve_ joined
|
|||
| disbot6 | <librasteve> note that weekly bot is not working here (unless you get an ack) | 14:44 | |
| <jubilatious1_98524> @librasteve I got an ack... | 14:50 | ||
|
14:51
arkiuat joined
|
|||
| disbot6 | <librasteve> @jubilatious1_98524 - I note that your posts start with "Using Raku (formerly known as Perl_6)" - since Perl is generally considered to be a dead language, I wonder if this is part of the problem - one major benefit of the rename is to cut the immediate link from raku to the perl legacy ... btw I would not judge upvotes as a negative on either raku or your (very nice) one-liners ... more a symptom of (i) wtf is raku | 14:51 | |
| (dont care), (ii) I do not have raku installed (pain), (iii) looks nice in raku, but I already prefer bash/perl/awk. | |||
| <librasteve> <notable6> @librasteve_ Noted! (weekly) <== this is what you are looking for ... I am pretty sure these acks are visible to all | 14:53 | ||
| <librasteve> I think you could write a book! Even better would be a prompt setup where you put in what you want to do and get the raku one-liner back ;-) | 14:56 | ||
| <jubilatious1_98524> @antononcube That solves the problem of adoption but not the problem of advertising. Upvotes are advertising. Sed, AWK, Perl, Ruby have tons of upvotes. So do csvkit, mlr, etc. A naive reader of the StackExchange forum would conclude that mlr is a better choice based upon upvotes--which of course it isn't. | |||
|
14:56
arkiuat left
|
|||
| disbot6 | <antononcube> I was not clear - I want to be able to scrape answers of a particular user (say, you) and automatically make DSL examples. | 14:58 | |
| <librasteve> I 100% agree with that analysis - imo problem is that you (we) are trying to "boil the ocean" ... read this soloway.pbworks.com/f/Crossing-The...LEBOOK.pdf | |||
| <jubilatious1_98524> Oops, YOU got an ack... . | |||
| <librasteve> ;-) | 14:59 | ||
| <librasteve> the idea is to pick a small number of valuable "bridgehead" applications that are sustainably differentiated | 15:00 | ||
| <antononcube> For the record, I am trying to speed up the coming of the next AI winter, so if @jubilatious1_98524 is trying to boil the ocean, that is very counterproductive. | 15:01 | ||
| <librasteve> www.reddit.com/r/rakulang/comments...exemplars/ is my "starter" to have us as a community work out what these bridgeheads could be ... please do vote (btw Anton, I voted for "LLM Prompt Engineering") did you vote for it? | 15:03 | ||
| <antononcube> @librasteve I have to add the LLM opinions about Raku-for-LLMing in the GitHub page you created. I will do it later today. | 15:04 | ||
| <antononcube> (Forgot to do it yesterday.) | |||
| <librasteve> oh and the bridgehead apps should target small niche applications that are early on in their evolution and do not already have well established incumbents (which is where I think we are with unix one-liners, where AWK/SED/Perl/Ruby already have 99% share) | 15:05 | ||
| <librasteve> @antononcube well I will make it anyway as soon as I can find tuits | |||
| <librasteve> I am just surprised that it is not winning the poll | 15:06 | ||
| <jubilatious1_98524> Not trying to boil the ocean, trying to become the "Jon Skeet" of Raku! www.reddit.com/r/dataisbeautiful/c...tation_in/ | 15:07 | ||
| <antononcube> This discussion makes think my 4th Raku advent post should be about Number theory visualizations, not code highlighting. | 15:11 | ||
| <antononcube> The latter is a "solved problem" for many... | 15:12 | ||
| <antononcube> ... And nobody get Number theory much... | |||
| <librasteve> lol - btw Jon Skeet is a great illustration of what I mean, he entered the race early (2008) when there was little competition and was able to establish as the market leader in the short time in the early evolution when there is a chance to break in - after that, the market shares get kinda of frozen (due to the reinforcement effect of upvotes you mention) and the leaders hold their share broadly (bit of a horserace, | 15:13 | ||
| sure ) throughout the rest of the market evolution | |||
| <librasteve> as an extreme counter example, if you made a new SO group called rakuonelines.stackoverflow.com, then you would be the #1 (of nothing) | 15:14 | ||
| <antononcube> I was thinking about that -- what does it take to make a new Raku StackExchange group. | 15:16 | ||
| <librasteve> hmmm - would need to be a non-raku topic (that raku can be the leader of) - I have a yen for DSLLMs (Domain Specific Langiages for LLMs) | 15:18 | ||
|
15:24
arkiuat joined
15:29
arkiuat left
15:42
arkiuat joined
15:46
arkiuat left
16:11
arkiuat joined
16:16
arkiuat left
16:17
arkiuat joined
17:46
dakkar left
17:50
arkiuat left
18:02
arkiuat joined
18:34
stanrifkin joined
18:46
stanrifkin left
|
|||
| disbot6 | <jubilatious1_98524> www.forbes.com/sites/benjaminkomlo...the-ocean/ | 18:58 | |
|
19:34
stanrifkin joined
19:58
arkiuat left
20:00
arkiuat joined
20:09
arkiuat left
20:11
arkiuat joined
20:15
arkiuat left
20:21
librasteve_ left,
arkiuat joined
20:29
arkiuat left
20:31
arkiuat joined
20:36
arkiuat left
|
|||