Talk:Raku (programming language)

Perl 6 links
Hi. I maintain Perl 6 and Parrot links. Is it usefull for exernal links? --mj41 21:25, 14 January 2007 (UTC)


 * Probably not, but if you get that added to dmoz, then adding a link to dmoz here would be acceptable. Please see WP:NOT for further information. -Harmil 13:10, 11 April 2007 (UTC)

WP:NOT doesn't apply to adding a single link. Furthermore, there are links to criticism in that list of links that should be used in this article. Again I should emphasize that their correctness is irrelevant as per NPOV policy. -71.166.153.191 06:08, 13 April 2007 (UTC)

Personal comments
I just noticed the good article initiative. The article doesn't make clear the original scope of dissatisfaction with Perl 5 that justified, what, a seven-years-and-counting revision cycle. Nor is it clear reading the synopsis of language changes which of these changes were considered vital to the goals of the project, nor which of these changes most impacted the length of the project. Did the original warts run so deep it has taken seven years to remove them? The one point where that struck me as having potentially consumed a large quantity of intellectual work was the further generalization of regular expressions to regexes. Surely it wouldn't have taken seven years to accomplish sigil invariance. Perhaps the closures were also difficult, or just the interaction of so many changes. Nor does the article address the issue about why all these ambitious changes have been pursued ensemble, without a formal halfway-there release, which many projects of this scope would attempt to achieve. I'm just saying that the nature of the ambition behind the project doesn't come across in a way you can take home after reading it. It's a bit like explaining in an article about the space program that the life support and navigation systems, etc. were all challenging to achieve, without making it clear how much that was complicated by the payload restriction dictated by the Saturn V launch vehicle. In other respects, I felt this article reads quite well. MaxEnt 03:02, 8 September 2007 (UTC)


 * Try Constraints and Software Development ( chromatic, oreillynet.com blog, 16.8.2007) and others from Perl 6 and Parrot links. --mj41 20:56, 9 September 2007 (UTC)

"Considered vaporware" citation
Someone tagged the sentence "Perl 6 has been under development for over seven years, prompting some commentators to suggest that Perl 6 may be vaporware" with  despite the fact that googling [ perl-6 vaporware] reveals thousands of hits on all sorts of different online venues. Obviously a lot of people do consider Perl&#160;6 vaporware. But how does one turn that into a citation by Wikipedia? Is there even any need to provide citation for things that can be verified by anyone with half their wits in 5 seconds? —Aristotle (talk) 06:20, 17 November 2007 (UTC)


 * It's vaporware.

68.183.43.200 (talk) 22:46, 11 April 2014 (UTC)


 * That search contains references to Wikipedia and all its mirrors. A search that excludes Wikipedia and its mirrors results in only 51 hits, not thousands. Also, please don't toss around phrases like "can be verified by anyone with half their wits", especially given that your analysis hasn't been sufficiently rigorous. Mind  matrix  16:00, 17 November 2007 (UTC)


 * Interestingly, conducting the same search a few moments later yields 1230 hits (many duplicates of only a few articles, or links to those articles). My point is still valid though - most searches should be conducted to exclude Wikipedia to find relevant external resources. Mind  matrix  16:05, 17 November 2007 (UTC)


 * Conceded, but still, that’s merely an exact search for the keywords “Perl&#160;6” and “vaporware”. It does not include the many other ways to express the exact same sentiment without using that word. F.ex., one of Guido’s maxims for the Python3000 effort was “Not Perl&#160;6”, by which he meant “keep a tight grip on scope to avoid an explosion in time-till-completion.” That’s just one example of the ways in which people say “it’s taking forever and might never even finish”, AKA vaporware. Chromatic alone has probably written some two dozen substantial rants about people griping that Perl&#160;6 isn’t done yet. There is no doubt that this sentiment is widely shared, even if it’s not something for which you can point to one easy single citation. If anyone has suggestions about how to deal with that  sensibly (which might mean simply deleting it), I’m all ears. —Aristotle (talk) 21:15, 17 November 2007 (UTC)


 * Probably the best way would be pick two or three representative citations and use those. &mdash;Cryptic 21:31, 17 November 2007 (UTC)


 * The vaporware meme doesn't have much to contribute. The same was once said of Mozilla and PostgreSQL.  Many people in this profession hate slow cooking.  Not every crock pot becomes a crock.  There's certainly an important niche to fill by projects that deliver on a known schedule.  That said, why do we need to give voice to monocrop enthusiasts here?  TMTOWTDI also applies to development models.  Eventually I figured out that there were many consultants out there plying their Perl 5 expertise who started to feel the heat from consultants on rocket ships Ruby/Python.  These people would have preferred more short term effort on Perl 5.  But Perl 5 suffered from first mover syndrome.  This was both an asset and a liability.  I guess the decision was made that the asset of early mover experience had more to offer pursued as an aggressive rewrite.  The real question here is not to take a straw poll of the disgruntled, but to ask whether there's a place in the fast moving IT world for substantive, non-incremental contributions.  Good luck citing that up front with better than fluff speculation. That story might yet be written, depending on how Perl 6 turns out.  (The nay-sayers can wait for the outcome along with the rest of us.)  The negative side was "it will take too long" and the positive side was "it will be worth the wait".  Until the dust settles you can only tell one side of the story.  I'm sure there are some people out there who don't believe that any software that takes more than five years to complete is worth having.  In the energy business, there has been plenty written about self-obsolescent of excessively long procurement cycles.  I'd be the first person to read a substantive paper on the event horizon of language design.  Languages designed too quickly, I've seen a few.  &mdash; MaxEnt 18:40, 29 July 2011 (UTC)
 * Let me add another short remark: the sociology of software engineering interests me as much as any topic on earth, but eventually I figured out that the mandate of Wikipedia is not to dabble in interesting things. If we're not writing about the good stuff (for comprehensible reasons), it grieves me to see it schnuck in on straw polls.  &mdash; MaxEnt 19:01, 29 July 2011 (UTC)

Py3k
It seems unfair to the merged Py3k article that this article is allowed to be seperate from the main Perl article. I do not know the guidelines, but this discrepancy became apparent to me as the casual reader. Sorry, should I raise old issues here. --129.241.135.159 (talk) 15:24, 6 December 2007 (UTC)
 * Python 3k is a cleanup revision similar to say Perl 5.5 -> 5.8 in its scope (at most) while Perl 6 is an entirely different language from Perl 5, it's not appropriate to merge the articles. --Ævar Arnfjörð Bjarmason 23:51, 22 September 2008 (UTC)
 * One article is unfair to another? What a stupid fanboy notion. -- 98.108.223.60 (talk) 20:01, 9 July 2010 (UTC)

Image copyright problem with Image:Programming-republic-of-perl.png
The image Image:Programming-republic-of-perl.png is used in this article under a claim of fair use, but it does not have an adequate explanation for why it meets the requirements for such images when used here. In particular, for each page the image is used on, it must have an explanation linking to that page which explains why it needs to be used on that page. Please check


 * That there is a non-free use rationale on the image's description page for the use in this article.
 * That this article is linked to from the image description page.

This is an automated notice by FairuseBot. For assistance on the image use policy, see Media copyright questions. --23:54, 1 October 2008 (UTC)

Ruby influence
``Roles in Perl 6 take on the function of interfaces in Java, mixins in Ruby, and traits[18] in the Smalltalk variant Squeak.''

Just put it back in both articles (ruby and this). ktnxbye!


 * Why - that statement is not supported by a citation (one is provided for the influence from Traits, but none for Ruby mixins (or Java interfaces). Without a proper citation, this portion of the text will be removed too. (By the way, because two things are comparable does not imply that one influenced the development of the other. It is clear from Apocalypse 12 by Larry Wall that Traits was the influence for Roles.) Mind  matrix  14:43, 15 November 2008 (UTC)


 * I doubt it can be very controversial that roles «take on the function of» mixins – though if it is, I'm reasonably sure a reference can be found for it :) – but to call roles an example of Ruby «influence» on Perl 6, is taking it way too far, for the reasons you give. As for other Ruby influences, Perl is not terribly worried about where it fits on the Rubyometer.  Ruby will need to worry about where it fits on the Perlometer.  But perhaps Wikipedia also needs to worry? :) — the Sidhekin (talk) 16:14, 15 November 2008 (UTC)


 * As a Rubyist it seems a bit controversial to me, as this looks like a multiple inheritance scheme, whereas Ruby mixins are dynamic and not declared with the class. Perl doesn't have to "care" about some uncited "Rubyometer," nor does Ruby have to care about a "Perlometer." Is this more like a Ruby mixin, or the general case of multiple inheritance? It is more like the general case, and Perl can go ahead and care so little about the "Rubyometer" as to not have Ruby referenced here. If you really want to talk about Ruby here, talk about how it is different than Ruby mixins, don't do original research claiming they are related. The are both ways of dealing with interfaces, and both allow duck typing, but most languages have these concepts. 24.22.50.193 (talk) 17:20, 30 March 2011 (UTC)

Whitespace
A major idea in Perl 6 is the significance of whitespace. This really should be covered here. I don't have time to write a fresh section right now, so I'll note it here. Długosz (talk) 23:34, 20 May 2009 (UTC)

Camelia?
There's no citation that the queen bee logo... Is that even real? leaflord (talk) 14:50, 3 June 2009 (UTC)


 * Camelia is real. (Though I thought she was a buttefly, not a queen bee.  Anyway.)  I'm pretty sure I first saw it on the perl6-language mailing list.  Hang on ... — the Sidhekin (talk) 19:22, 3 June 2009 (UTC)


 * This message sums up most of what I rememeber. «Camelia, the Perl 6 bug» appears on http://www.perl6-projects.org/.  Nothing immediately says "official" (and certainly not "final"), but hey, this is Larry's work ... it's probably as good as we're going to get. :) — the Sidhekin (talk) 19:50, 3 June 2009 (UTC) Oh, and most seem to think "butterfly".  Including Larry.
 * Wow, that's cute but quite a strange logo in the sense it feels unprofessional.. Also I guess it's a bee cuz the body is yellow.leaflord (talk) 22:52, 4 June 2009 (UTC)


 * Ouch, sorry, but that logo/mascot seriously hurts the 'first-impression' credibility of this article, and probably Perl 6 as well. dr.ef.tymac (talk) 19:16, 20 February 2010 (UTC)
 * I agree with this comment. That logo makes it look amateurish, whimsical and self indulgent. No doubt some will support it because they feel it's one in the face for the big business PR way of doing things, but that attitude is short sighted. Bit like sticking a fork in your own face to protest the media's preoccupation with conventional beauty. Including this logo in this article doesn't do Perl 6 any favours (maybe that's it's purpose). Just my view. —Preceding unsigned comment added by 81.187.233.172 (talk) 09:31, 16 June 2010 (UTC)
 * Wikipedia is an encyclopedia, not a PR firm; comments here about the aesthetics of the logo are absurd and inappropriate. (And anyone who thinks its a bee rather than a butterfly is an idiot.) -- 98.108.223.60 (talk) 20:06, 9 July 2010 (UTC)
 * You're right about Wikipedia being an encyclopaedia and about it not being a PR firm. It could also be inappropriate to comment about the aesthetics of the logo, though it isn't absurd to do so, and believing the logo to represent a bee does not make one an idiot. One problem with Wikipedia is that many of the articles in it are influenced by interested parties. When that happens one faction or another sometimes promotes their particular view and the comment about the aesthetics of the logo was prompted by that thought. I got the impression that someone had put it into the article to shout out the unconventional nature of the project - but if it's the official logo then so be it. It's still awful. —Preceding unsigned comment added by 81.187.233.172 (talk) 23:11, 24 January 2011 (UTC)
 * Sounds crazy to me, it's totally consistent with the Perl attitude towards authority. If you worry that Perl isn't wearing a tie, you're not the person Perl is trying to impress. And look at Larry Wall's own "home page" if you want an example of what he feels is an appropriate nod to "professionalism." Perl cares about how well it meets its goals, how useful it is to programmers that share those goals, how effective a tool it is. It does not care what other people think.Rubypanther (talk) 17:57, 30 March 2011 (UTC)
 * Sounds crazy to me that anyone would think that there should be a Perl attitude towards authority. Perl is a programming language and doesn't have an attitude. Maybe you meant that the people writing the code have an attitude? It's irrelevant in any case. It's just an ugly logo. That's all.
 * Would this count as a logo? It does not seem official but it looks more professional. -- J.Dong820 (talk) 21:02, 19 December 2011 (UTC)

Meaning of "high-traction implementation"?
The article says

Development on Pugs, the first high-traction implementation, began in 2005 ...

I think I know what high-traction tires are, but what does "high-traction" mean here? Does anyone have a clue? I certainly don't. Neither Google nor dictionaries are any help. Please say what you mean using recognizable terms, and avoid opaque jargon, especially if it's your own private jargon. Toddcs (talk) 14:57, 2 November 2010 (UTC)
 * From traction:
 * "7. (business) the extent of adoption of a new product or service, typically measured in number of customers or level of revenue achieved"
 * Although for a FLOSS project, I guess it would be measured by the number of users. Alksentrs (talk) 15:40, 2 November 2010 (UTC)
 * In my experience it has widespread use in computers, used exactly in above linked sense, and in FLOSS I concur that it refers to users, or specifically, uptake by existing serious users. It is not nearly so strong as consensus, but it is stronger than merely being seen as an acceptable choice. It is also without the baggage and implication that word like "popular" might bring. It is a respectful way to say that many serious professionals are using something, or agree with something, without claiming it has become canonical or official. Rubypanther (talk) 18:03, 30 March 2011 (UTC)


 * The "traction" adjective floored me a bit, too. Normally such a word in a technology article is resolved by knowing the auspice of the puff artist (e.g. El Register or The Eternal Blowhard Who Must Not Be Named). &mdash; MaxEnt 18:08, 29 July 2011 (UTC)

I agree that the use of this word is goofy and unclear.--75.83.70.28 (talk) 03:33, 17 May 2012 (UTC)


 * It would have been more true if they called it "the first attempt at an implementation that got anything finished at all.94.230.147.206 (talk) 01:13, 17 December 2019 (UTC)

Quicksort pedagogy
A secondary requirement for a correct implementation of quicksort is O(lg(N)) worst case stack usage, which is normally accomplished by recursion into the shorter sub-list and iterating on the longer sublist. If the Perl 6 language has magic short-side tail recursion, that should be mentioned; if not, we're doing pedagogical damage once again. That's certainly not what Perl ever stood for. &mdash; MaxEnt 18:08, 29 July 2011 (UTC)

can not -> cannot (they don't mean the same thing)
A modification comment stands that «can not -> cannot (they don't mean the same thing)».

So the question is what is the eventual difference between «cannot» and «can not». From my understanding those are two different spellings of the same English verb in its negative form. — Preceding unsigned comment added by 86.75.160.141 (talk) 20:32, 23 October 2012 (UTC)


 * "If the three of us being there will make the group too large, we can not come."
 * It is possible to come up with situations in which there is a difference. However, they often DO mean exactly the same. TooManyFingers (talk) 04:11, 6 June 2024 (UTC)

Roles are like Traits or mixins, not Java interfaces
Of course, MIT Curl has both abstract classes and multiple inheritance with many example of mixins/traits in the class hierarchy. Indeed these are fundamental is to classes used by Curl web and GUI programmers.

see Curl (programming language)

G. Robert Shiplett 00:18, 30 July 2013 (UTC)

Unicode support
Missing notes about Full grapheme based Unicode support, including Annex #29, meaning almost unparalleled excellent Unicode support. - e.g. Unicode, Perl 6, and You, This week: Unicode normalization, many RTs, This week: the big NFG switch on, and many fixes, ... --mj41 (talk) 10:06, 28 December 2015 (UTC)

First appeared
"First appeared	25 December 2015; 4 years ago"

This is disingenuous at best, actively misleading and distasteful at worst. Sure the language rename is a thing, but that's no reason to wipe away on Wikipedia all the work that happened for the previous fifteen years! This is especially problematic as all Perl 6 named articles now redirect here. I was using Rakudo in 2010, so this is completely bogus. Just the WP article history demonstrates the lie... who edited this and why was it allowed to remain? -- MattOates (Ulti) 07:31, 4 February 2020 (UTC)
 * If anyone thinks its fine, the info boxes on programming langs are routinely used programatically to build timelines of language invention. -- MattOates (Ulti) 07:35, 4 February 2020 (UTC)

It seems that 'first appeared' has been removed, but ... RichardHainsworth (talk) 10:04, 1 March 2020 (UTC)

It is important to retain the historical sequence. The language was known as Perl 6 for the first decade and a half of its existence. Larry Wall strenuously resisted a change in name, even after requests by Damian Conway and others. Finally, he agreed to the change to Raku. In addition, referring to that language as 'Raku' when it was known as 'Perl 6' complicates verification. Larry Wall did not refer to Raku at all when he was directly involved in its early specification and development. I think that historical references should be to 'Perl 6 (later Raku)' whilst discussion of features of the language as it appears now should be to 'Raku'. RichardHainsworth (talk) 10:04, 1 March 2020 (UTC)

Consistent reference to Role/Class etc.
Various important keywords, such as 'role' and 'class', are important to Raku and should be distinguished in the text. This is not done consistently. I would edit the page for consistency, but would prefer to know first whether other editors have a preference. (I am just starting to contribute to Wikipedia, so I'm making mistakes) RichardHainsworth (talk) 10:04, 1 March 2020 (UTC)

Clarification request
What does the first line of the below even mean?!? It looks like Color is being either redefined (a la "x = x + 1", but of types), or is being defined recursively, a la the Curiously-recurring template pattern. Either way, I think the snippet on the non-Talk page should be clarified.


 * I have changed the example so as to remove the compilation error. Please take a look. It would also be nice to reference a source for this behavior, but I am not aware of any besides the official docs. — Rightfold (talk) 13:25, 24 May 2020 (UTC)

What does "Raku" mean?
As far as I can see, the article does nowhere explain the meaning of the name "Raku". I don't know Japanese, but according to Jonathan Worthington "Raku" means “comfort” or “ease” in Japanese:

Perhaps this could be added to the article? – Tea2min (talk) 08:27, 27 August 2020 (UTC)


 * I now see that there's some more info on the names "Rakudo" and "Raku" at Rakudo. – Tea2min (talk) 08:32, 27 August 2020 (UTC)

correction of quicksort example
I corrected quicksort example. It had not conserved size of the collection. Multiple same values (2,3,3,3,5,6)had been  reduced to (2,3,5,6)

I replaced "before" and "after" by "=<". (only in either one grep statement) .. I think this is at the cost of type universtality

Is there in perl something like before-or-equal ?? 193.179.3.153 (talk) 21:32, 8 February 2023 (UTC)

Replacing Perl 5
If I look for situations where Perl 5 has been adopted to replace Perl 4, I will find some. It seems clear that though Raku is not required to be 100% compatible, it still essentially does the things that Perl 5 does – i.e. it has not dropped everything and started over, and it sounds like it's supposed to do the same tasks in a better and more convenient way rather than doing something else. What is preventing it from being adopted in the way that Perl 5 has been adopted?

I expect that possible answers might include "Important parts are still not functioning" (which seems to be the case regarding lack of CPAN support), "It isn't compatible enough to make a smooth transition", "The specifications keep changing", "The way it's being developed is flawed", "The changes underway are not what most users of Perl wanted", and so on – but the article doesn't really say.

Or have answers like those become redundant because people have simply given up on it? TooManyFingers (talk) 15:20, 8 June 2024 (UTC)