Talk:Recursion

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Recursion link in "See Also"[edit]

I'd be interested in opening an RFC on the topic of putting a link to Recursion in the "See Also" section. While it is a small point, I think it reflects the much wider issue of "Can Wikipedia afford to occasionally be less than deadly serious?". I firmly believe it can, and I feel that a humorous touch like this actually creates a better encyclopedia.

Thoughts anyone? Manning (talk) 18:05, 25 June 2009 (UTC)[reply]

I agree; the link-recursion-back-to-recursion meme is so widespread on the net that it makes Wikipedia look unnecessarily po-faced if it isn't permitted here. Provided that the textual content of the article itself is genuine and relevant, there's no harm at all in referencing the joke by means of a link or "see also" item. MarkSG (talk) 21:33, 23 July 2009 (UTC)[reply]
That see also thing has been discussed to death, and it never gains consensus. The reason for that is that this is an encyclopedia, not a jokebook. It makes us look unprofessional to include jokes in our articles. Many other things are widespread on the web, such as the word 'epic' and kewl spelling; we eschew those as well. We do describe the web meme here, which is the way we should handle it. See the section above for a previous discussion about this. — Carl (CBM · talk) 21:38, 23 July 2009 (UTC)[reply]
As far as I can see, the reason it never gains consensus is because a relatively small number of contributors are very opposed to it and repeatedly make their objections known. That's why I agree with Manning that an RFC process could be useful, as it might help to get a broader spectrum of views involved. If that then does result in a consensus (or even an observable majority) against including the link, then that's fine. But, equally, if an RFC supports inclusion, then it should be included. MarkSG (talk) 07:03, 24 July 2009 (UTC)[reply]
My first reaction to finding out such a harmless spot of humor is completely taboo was actually to take Wikipedia less seriously than before (I'm a very frequent user, and a strong proponent). Wikipedia has a couple rules I disagree with (none very strenuously), but this deserves at least a vote on the topic. It doesn't have to be a disambiguation link at the top of the article, but a link to it in the "see also" is harmless. A link in the first paragraph phrased as an example might even be completely appropriate, if done properly. Prgrmr@wrk (talk) 08:22, 24 July 2009 (UTC)[reply]
Can you explain how this makes you take wikipedia less seriously? Perhaps you were looking for something other than a professional encyclopedia. — Carl (CBM · talk) 12:20, 24 July 2009 (UTC)[reply]
Jokes are inappropriate for a serious encyclopedia. Paul August 12:49, 24 July 2009 (UTC)[reply]
I think some jokes can be OK, but they should not be of the sort to annoy users who are just looking for information. I'd rather have a blanket humourlessness policy that an encyclopedia filled with the sort of joke that only friends of the (ir)responsible editor will be amused by. I think that this discussion shows that the proposed joke falls firmly in the latter category. — Charles Stewart (talk) 13:18, 24 July 2009 (UTC)[reply]
I agree with CBM, Paul August, and Charles Stewart on this. I love jokes, I make a lot of jokes, however I don't find the need to plaster them everywhere. If this joke is everywhere on the net, as MarkSG states, then how is it still funny? Now if someone came up with a really fresh joke especially suited to the wikipedia format, maybe I'd reconsider, but a stale joke like this one hardly seems worthy of inclusion. RobHar (talk) 16:10, 24 July 2009 (UTC)[reply]
I'm not even saying do it in a humorous way. Just something like "As an example, see "this link" to the recursion page. That's not a joke, and the only reason in the rules not to do it is because articles don't link to themself. I think that rule can be ignored for the one page in the entire project where a self-link makes sense. Prgrmr@wrk (talk) 04:46, 26 July 2009 (UTC)[reply]
There are many better examples of recursion that we can give, though, if the goal is to explain the topic. The main point of the self-link things seems to be to pull a practical joke on readers the first time that they encounter it. But, as RobHar says, this joke has devolved into a meme. I don't see why we need to propagate that meme further (we do discuss it already in the article). — Carl (CBM · talk) 14:12, 26 July 2009 (UTCHaylee Sharman for dayzzzzzz
OK, I started this discussion, then had to go overseas for work for a while and only just got back.
I particularly am intrigued by blanket statements such as that made by Paul August above: "Jokes are inappropriate for a serious encyclopedia." Why? Says who?
The founders of this very encyclopedia didn't agree - the original page back in 2001 actually had the recursion joke in it and I remember the discussion with both Larry Sanger and Jimbo agreeing that it should be included. The recursion joke is hardly "lame or stale" - sure it may be very familiar to anyone who has studied recursion or read Douglas Hofstadtler, but it is certainly not "well-known" beyond that.
On a MUCH deeper level, I think the refusal to include it actually reflects very badly on the entire project - are we really that dour and humourless that a single joke cannot be permitted to survive? Particularly when the joke is moderately sophisticated and also explains the topic.
In light of the project receiving so much recent media attention for alienating new users, I think there is more at stake here than a simple "no jokes" ruling. Manning (talk) 07:39, 12 September 2009 (UTC)[reply]
The problem with this joke is very simple: Jokes fail when you explain them. Since this article contains a 'Humor' section to explain the joke, any instance of the joke in the article is dead on arrival. And since the Humor section also contains two examples of the joke, its presence in the See Also section would be seen not as pointless, but as amateurish. While Wikipedia doesn't need to be deadly serious all the time, it should at least have high standards. 129.229.26.125 (talk) 01:49, 19 September 2009 (UTC)[reply]
While I do greatly love this joke, my issue is that it defeats the purpose of an encyclopedia. The whole point is to provide information, and anyone who doesn't know Wikipedia will not know how to access the actual page. Now, I'm not against jokes in an encyclopedia. Taking yourself too seriously can get you into trouble. Thus, I greatly like Manning's idea of putting a link to Recursion in the See Also section. It is clever and more accurately encapsulates what recursion is. - Austin. 23:30, 5 November 2009 (UTC) —Preceding unsigned comment added by 71.88.101.229 (talk)

I went and added a recursive joke link in the format of the other uses template. Now as to why I did this.
A. Its humorous.
B. It has a very valid purpose. See WP:GRAPE for my point. The best definition of recursion IS recursion. Therefore, not including at least some unobtrusive form of a recursive example, whether it be taken as a joke or as a serious example, seems to be required, at least in my opinion.
The Application of this joke has the capability to be both funny and educational. 67.149.81.70 (talk)

It's gone now. Seems like the point about wikipedia going down hill is valid. It's sad to me that something this small can't be added without generating the controversy it has. 67.149.81.70 (talk) —Preceding undated comment added 01:51, 29 November 2009 (UTC).[reply]
Your link was labelled other definitions uses, which it wasn't, it was the same definition. The 'joke' isn't explained, and so may confuse. Wikipedia isn't a joke book. --Escape Orbit (Talk) 01:55, 29 November 2009 (UTC)[reply]
So it was an error in presentation? Would adding a termination function, or altering the presentation make it any more appropriate? I.E., something along the lines of: For an example of recursion, see Recursion ? I know this whole point seems moot, and I do recognize that wikipedia is not a joke book ((Though I would wonder if you are against articles such as All_your_base_are_belong_to_us and Badger_Badger_Badger)). However, as the Recursive humor section indicates, there is an element of wisdom in this. Furthermore, I still reference WP:GRAPE, essay or not, as a valid and logical argument for this inclusion.67.149.81.70 (talk) 02:05, 29 November 2009 (UTC)[reply]
WP:GRAPE would support a boringly literal "For an example of recursion, clicking this link will take you to this same article.", it wouldn't support a potentially confusing "See also this link" which a naive reader might assume linked to a further article about the subject. But we already have a great big section on "recursive humor", with some good jokes on the same level (and good jokes that still work if you're reading a mirror site or a physical printout). --McGeddon (talk) 10:22, 6 December 2009 (UTC)[reply]
If someone made a worm that looks at every Wikipedia page, and it is badly programmed, it could loop forever. However, that's their problem. I say we add the link. Language Boi (talk) 18:39, 5 August 2023 (UTC)[reply]

To me, this seems sort of like if on the Rick Roll article has a citation that actually linked to a rick roll. Classless. Funny? Yea, but it doesn't go here. Nigtv (talk) 09:24, 6 December 2009 (UTC)[reply]

When I came to this page, I was upset to find that on Wikipedia, one of the best internet projects I use on a regular basis, there was no link to the recursion page on the recursion page. Wikipedia is NOT a serious business encyclopedia. Do serious business encyclopedias have a page on Lolcat? Nope! How much harm would it do to Wikipedia as a project/concept/ideal to make the word Recursion at the very beginning of the article a link to the article itself? None! And it would make me and, judging from every single person I've talked to about this, a lot of other people quite happy. Where's the harm? Why NOT? Paulcd2000 (talk) 07:53, 15 February 2010 (UTC)[reply]

Wikipedia is a serious encyclopaedia. It has articles about jokes, it quotes jokes, it explains jokes, but it doesn't make jokes. The harm in making a joke during an otherwise serious writeup is that a reader who isn't expecting a joke could be misled or confused by it - in this case, they may click the link on the assumption that it will provide further definition of the term. --McGeddon (talk) 17:26, 15 February 2010 (UTC)[reply]

I also find this joke hilarious and helpful, but perhaps inappropriate to an encyclopedia of such high caliber. I've add the joke to the Recursive humor section, where I hope it is found to be more informative as an example. ~~Andrew Keenan Richardson~~ 06:24, 24 June 2010 (UTC)[reply]

It's not really any more appropriate there. — Carl (CBM · talk) 11:28, 24 June 2010 (UTC)[reply]
A joke about recursion isn't appropriate in Recursive Humor section? It seems like that section was designed for this exact joke. ⓑⓤⓕⓐⓡ (talk) 06:24, 6 December 2010 (UTC)[reply]
The see-also-recursion joke is already explained there, at the top of the section. We add nothing to the reader's understanding by either boringly explaining "this joke would also work on Wikipedia by linking to the recursion article from the recursion article" or giggling behind our hands and saying "hey, for more information on this subject you should totally click on this link: recursion!"
Wikipedia quotes, documents and explains jokes, but it doesn't make them. --McGeddon (talk) 09:13, 6 December 2010 (UTC)[reply]


Using the principle of "Wikipedia quotes, documents and explains jokes, but it doesn't make them", as a compromise between the joke-lovers and the joke-haters, why not have a sentence along the lines of:

A hyperlink labelled "Recursion" which points to itself is regarded by some as an example of recursive humour. --Old_Wombat (talk) 11:19, 11 May 2011 (UTC)[reply]

This type of joke is already covered in the Recursive humor section. We shouldn't self-indulgently include a special Wikipedia version of the joke purely because we're on Wikipedia. --McGeddon (talk) 11:46, 4 October 2011 (UTC)[reply]
Agree. Also see WP:SELF. Paul August 11:53, 4 October 2011 (UTC)[reply]

Having just come to this article for the first time, I am disappointed that there is not a link to the article on recursion in the See Also section. I am amazed that people have actually deleted such links.81.136.210.168 (talk) 16:23, 25 May 2011 (UTC)[reply]

Let's get some quantitative rather than qualitative data - poll anyone? Casliber (talk · contribs) 11:12, 4 October 2011 (UTC)[reply]
WP:NOPOLLS. But it's already clear that plenty of editors think it'd be funny or useful to include the link. --McGeddon (talk) 11:46, 4 October 2011 (UTC)[reply]

I support having the article link to itself, and I think it can be done in a way that avoids all or most of the problems raised by those who oppose it. I agree with Prgrmr: to me, not having the link supports the image of Wikipedia as a place not to bother looking for interesting information because what you're looking for will have been deleted. While jokes are usually not included in articles, that's because they would usually be a distraction from the purpose, not because of some blanket policy against jokes; and in this case it wouldn't be a distraction, but an illustration (WP:GRAPE). To state that a joke is no longer funny is subjective -- there are those who say "funny once", and there are also many, including myself, who find some jokes funny over and over again, even sometimes funnier with each retelling. ("Oh, no. Not the Cone of Silence, Chief!") If you don't find it funny, how about just thinking of it as an unobtrusive example rather than as a joke? It can be designed so that the reader won't be confused or tricked. WP:SELF seems to me to be against saying things like "an article linking to itself, like this", but not against saying things like "a web page linking to itself", since the latter would still make perfect sense in print even if the link wouldn't be active. Coppertwig (talk) 23:25, 16 October 2011 (UTC)[reply]

WP:EGG says "make sure that the reader knows what to expect when clicking on a link" - we shouldn't have surprise wikilinks, and "linking to itself" would be expected to redirect to an article that talks about things linking to themselves. If the joke is that it's funny to click on a link without checking the URL, and then laugh when you realise you've been sent back to the recursion page, then we can only create that surprise by tricking the reader. And we should not be tricking the reader. --McGeddon (talk) 21:23, 17 October 2011 (UTC)[reply]
You're right that the reader should know what to expect when clicking; however, the point of the link is not to trick the reader. It should be worded so as to be as clear as possible. Many links on Wikipedia are ambiguous; I'm often disappointed when I click, often arriving at a page far more general than I had hoped for. We try to make them as clear as possible. If it's linked from "... or a hypertext article which links to itself" I think that's clear enough; the inclusion of the word "which" in the linking text distinguishes it from a simple link to a different article about linking to itself. (By the way, the "Recursion" article can itself reasonably be considered to be an article about things linking to themselves.) I think the point is not to create a "joke", but an illustration. You said earlier we shouldn't do it "purely because we're on Wikipedia". Here, then, are some other reasons for including it: (1) Many readers expect it. (2) Many readers want it. (3) It provides an illustrative example of recursion. (4) It enhances the article, much as an image does, creating an effect which cannot be created by text alone. (5) It follows in the footsteps of reliable sources which do the equivalent using, for example, indexes of books. WP:SELF doesn't forbid self-reference, but specifies "which types of self-references ... are acceptable". If expressed as I suggest, it would still make perfect sense when rendered on a mirror site or in print, so it meets the criteria of WP:SELF. Coppertwig (talk) 16:23, 29 October 2011 (UTC)[reply]
Many readers also do not expect this sort of thing in a professional publication. It is always a little disappointing to see in print when an author plays games by putting false items in an index (e.g. fictional people, circular "see also" entries, etc.). It would also be disappointing for us to put false items in our see also list. There is no enhancement to the article, as the link gives no more information than was already present, since the link comes back to the same article. — Carl (CBM · talk) 19:54, 29 October 2011 (UTC)[reply]
I guess it's a matter of personal preference: some perhaps find it "always a little disappointing"; others are happy to find such links and disappointed if they're not there. Coppertwig (talk) 23:20, 29 October 2011 (UTC)[reply]
Indeed. The idea of the joke occured to me yesterday and I put it in via redirect. It lasted 6 hours. I think WP would be better if it had a small number of easter eggs in it. Robert Brockway (talk) 11:36, 6 December 2011 (UTC)[reply]
I would argue that the 'joke' being discussed is not a joke, but is in fact a valid aid in understanding what recursion is. In a wikipedia page on Apple, do we not expect there to be an example of an apple? Does it not help us to understand better what an apple is. Why do we not therefore have this example of recursion? 192.68.112.171 (talk) 20:12, 23 February 2012 (UTC)[reply]
A lot of users expect exactly this sort of thing from a professional publication. I agree that in general "jokes" in encyclopedias or other reference works come off in poor taste, but it seems to be something of a tradition when it comes to "recursion." I have very rarely encountered a textbook that covers recursion in any way without its index entry for "recursion" including that page of the index. It doesn't seem to matter what the level of "seriousness" is. I personally think it would be harmless, and it would be a good show of faith that we don't have such a big chip on our shoulder about being taken seriously that we don't include the little nod that nearly every other publication does. Ltgerbilmuffin (talk) 18:52, 29 June 2012 (UTC)[reply]
So we have a bunch of people who are obsessed with keeping Wikipedia 'professional'. This isn't a business, this is Wikipedia. This is not Britannica. We are not selling Wikipedia. Complaining that it "wouldn't be professional" to have the article on recursion linking to the article on recursion is a slap in the face to countless computer science textbooks and professional works. Are you going to claim that K&R are 'unprofessional' because their book on C does it? Is Google 'unprofessional'? or do we just have an uptight view of what 'professional' means? It is also a perfect explanation of what recursion is, and therefore should not be limited to just the 'recursive humor' section. This is one of the main ways that the concept of recursion is introduced to computer science students at the undergraduate level, even at universities that are world-class research institutions. Padenton (talk) 02:51, 31 October 2012 (UTC)[reply]

FWIW, I just now came to this page for the sole purpose of seeing whether or not it had a self-reference. Numerous textbooks and Google do it, after all. TricksterWolf (talk) 02:32, 18 November 2012 (UTC)[reply]

Textbooks and Google do several things that Wikipedia articles should not do. - SudoGhost 03:24, 18 November 2012 (UTC)[reply]
You're talking about the majority of computer science textbooks, written by the most prominent computer scientists in the history of computing. Bit arrogant of you. --Padenton (talk) 21:29, 25 November 2012 (UTC)[reply]
Quoting from Wikipedia:What Wikipedia is not: Wikipedia is an encyclopedic reference, not a textbook. The purpose of Wikipedia is to present facts, not to teach subject matter. Paul August 18:59, 11 March 2013 (UTC)[reply]

I think having a recursive link to the article itself is the most visceral way of communicating the idea and is worth further discussion in this section on the talk page.108.20.48.27 (talk) 04:34, 11 March 2013 (UTC) Personally, I believe the best way to settle this is with a link in the recursive humor sectionRegicide1 (talk) 17:12, 27 October 2020 (UTC)[reply]

The issue is settled, and has been for awhile. The editorial consensus has been, and still seems to be that such a link serves no encyclopedic purpose. Paul August 21:15, 27 October 2020 (UTC)[reply]

@Paul August: I didn't see it mentioned here, but there has been a link to Recursion (disambiguation) in the See Also section of that article, which has been kept for quite a while, despite continued attempts to remove it (example). I just removed it, pointing to this discussion. In my edit summary, I suggested to discuss this again, in case the consensus has changed since 2013. Renerpho (talk) 10:47, 6 April 2022 (UTC)[reply]

@Renerpho: I see no evidence of a changed consensus. Paul August 11:15, 6 April 2022 (UTC)[reply]

Recursive redirects[edit]

Did you know that a recursive redirect looks like this ? (sadly, wikipedia fixed it for me badly) 109.201.38.128 (talk) 22:16, 26 September 2019 (UTC)[reply]

Actually, that's not a recursive redirect (in the sense of WP:Redirect and Help:Redirect), it's a "self link". The presentation of these links as bold text is deliberate. See Help:Self link. (Oh, and please see #Recursion link in "See Also" and #Recursion Recurring Link and #Suggestion to implement "Recursion: See Recursion" in a Wikipedia-friendly way and especially #Reoccurring Recursion Link on this talk page.) – Tea2min (talk) 07:08, 27 September 2019 (UTC)[reply]
It could be a recursive redirect... what is the difference? Cjjjkscratch (talk) 13:01, 26 November 2022 (UTC)[reply]
exactly... Cjjjkscratch (talk) 13:02, 26 November 2022 (UTC)[reply]

Special:Contributions/109.201.38.128 I would like to say that I tried to do the same thing in the morning, but I got a warning. Cjjjkscratch (talk) 09:55, 12 September 2021 (UTC)[reply]

Recursion in Giotto's Stefaneschi Triptych[edit]

The article currently has a {{failed verification}} tag for the following statement: Recursion has been used in paintings since Giotto's Stefaneschi Triptych, made in 1320. Its central panel contains the kneeling figure of Cardinal Stefaneschi, holding up the triptych itself as an offering. The reference clearly supports the second part of this (quote: "cardinal Stefaneschi, holding in his hands the model of the triptych"). What it does not directly support is the connection to recursion, which is not mentioned in the source (as pointed out by JayBeeEll in this edit).
However, the article does mention the Droste effect; and in that article (which has good article status), a similar claim is made in the lead section and in the paragraph about medieval art. Do we need a source that connects the Triptych explicitly to recursion, or do we even have to make changes to the Droste effect article? Renerpho (talk) 23:24, 1 April 2022 (UTC)[reply]

I too would be interested in hearing other editors' views about whether I am being overly pedantic about this. --JBL (talk) 14:17, 3 April 2022 (UTC)[reply]
@JayBeeEll: To be honest, I mostly agree with your revert. Which is to say, if you are overly pedantic then so am I. The Triptych looks like a nice example for recursion, until you realize that nobody seems to make that connection independently of the two Wikipedia articles (this one, and the Droste effect article). I don't like that, at all! Renerpho (talk) 10:03, 6 April 2022 (UTC)[reply]
Are there any reliable sources calling this an example of recursion that do not lead back to Wikipedia? I'm not saying it is (yet), but if this turns out to be a WP:REFLOOP, that would be quite funny, actually... Renerpho (talk) 10:08, 6 April 2022 (UTC)[reply]
@Renerpho and JayBeeEll: We don't necessarily need a source that connects the Triptych explicitly to recursion, a source citing the Triptych as an example of the Droste effect would seem to be sufficient, for example: Svozil p. 12 Paul August 11:10, 6 April 2022 (UTC)[reply]
Yes, I agree, that would be fine. --JBL (talk) 11:21, 6 April 2022 (UTC)[reply]
Agreed. I've replaced the {{failed verification}} tag with that new source. Renerpho (talk) 23:33, 20 April 2022 (UTC)[reply]

Semi-protected edit request on 16 July 2022[edit]

In the image byline of the Droste effect image, change the name Jan Misset to Jan Musset. (see the Droste effect page for the correct spelling) Jvandenieuwenhof (talk) 10:02, 16 July 2022 (UTC)[reply]

 Done Aaron Liu (talk) 06:02, 18 July 2022 (UTC)[reply]
@JBW: FYI. See also [1] at Droste effect. --JBL (talk) 00:48, 31 January 2023 (UTC)[reply]
Also pinging @AstroLynx: in light of [2]. --JBL (talk) 00:52, 31 January 2023 (UTC)[reply]
Yes, my earlier edit was wrong. It was "Misset", not "Musset". For a Dutch source, see "Bedenker van Droste-effect bekend", Trouw, 1 August 1994. AstroLynx (talk) 12:07, 31 January 2023 (UTC)[reply]
Lovely, thanks! --JBL (talk) 18:12, 31 January 2023 (UTC)[reply]

Semi-protected edit request on 26 October 2022[edit]

It'd be nice to put "see recursion" somewhere in the header; kinda like how Google did it. BruhMoment23578 (talk) 19:29, 26 October 2022 (UTC)[reply]

Amusing, but unencyclopedic. And also done multiple times before. --JBL (talk) 19:37, 26 October 2022 (UTC)[reply]
What if we were just to make the first usage of Recursion link to recursion? That wouldn't harm the flow at all but also still be amusing DarmaniLink (talk) 19:01, 21 February 2023 (UTC)[reply]
Yes, incredibly amusing. At least, when I was about 13 I would probably have found it amusing. Unfortunately, however, that isn't what Wikipedia is for, but the good news is that there are plenty of other web sites where children can amuse themselves by posting that kind of thing. JBW (talk) 19:44, 21 February 2023 (UTC)[reply]

Another example of recursion[edit]

Would be for a website to have a link to itself. Example file can be "m.html"

<a href="m.html">Click here to understand recursion</a> Cjjjkscratch (talk) 12:58, 26 November 2022 (UTC)[reply]

Google also does it like this. Cjjjkscratch (talk) 05:19, 27 November 2022 (UTC)[reply]

Replace image of recursion[edit]

To: Paul August. While the current image is technically correct, it is really hard to see the recursive image on the drawing of the box, and thus isn't obvious to the reader why that image represents recursion. In short, I want to add a more intuitive image. My first thought about it is the perennial recursive joke, which when applied correctly would aid comprehension for the reader. However, I'm well-aware that Wikipedia is a no joke place and WP:ASTONISHMENT applies, so I try to strike a balance and instead went with a preview for the page, with a descriptive and no-joke caption. The reason why I say WP:IAR applies is because rules can be bent slightly to aid reader's understanding, because trying to follow these rules to the letter bring catastrophic and ironic effects, as detailed in this Twitter thread. CactiStaccingCrane (talk) 10:57, 21 May 2023 (UTC)[reply]

tl;dr: Wikipedia shouldn't be unfunny in the pursuit of professionalism, and exceptions to the "no humor" rule can be made. CactiStaccingCrane (talk) 10:58, 21 May 2023 (UTC)[reply]
The iconic "Droste" image is easily recognizable as an example of recursion (the can clearly contains an image of itself—is that not obvious??) In fact, it's such a classic image of recursion that its name is used for the whole concept of recursive images, i.e. the Droste effect. There could hardly be a more representative image of recursion than this one. Paul August 11:19, 21 May 2023 (UTC)[reply]
I agree with Paul August. --JBL (talk) 21:36, 21 May 2023 (UTC)[reply]
It's not really. Most people don't found the image iconic nor the recursion apparent. The recursion image is so tiny that I doubt it's even visible in smartphones. CactiStaccingCrane (talk) 04:11, 22 May 2023 (UTC)[reply]
On my smartphone, the image of the smaller can on the larger can measures 1/4 inch tall, and so easily visible, and that's before it's enlarged when it's about an inch tall. Paul August 11:19, 22 May 2023 (UTC)[reply]
Old reply, but I've replaced the image with File:Recursion6.gif. CactiStaccingCrane (talk) 14:38, 13 October 2023 (UTC)[reply]
I've reverted your removal of the established, and "Droste" image. In my view, your proposed animation is a distraction for the reader. Please gain consensus here before replacing the current image. Paul August 17:15, 13 October 2023 (UTC)[reply]
There is a strong positive case for the Droste image (beyond the fact that it's attractive and of historical interest), made by Paul August above in May: In fact, it's such a classic image of recursion that its name is used for the whole concept of recursive images, i.e. the Droste effect. There could hardly be a more representative image of recursion than this one. Any move to replace it should involve making a similar positive case for the value of the replacement image. --JBL (talk) 21:37, 13 October 2023 (UTC)[reply]

Formal Definitions' definition is recursive ... in a bad way[edit]

The "formal" definition of "recursion" defines the base case in terms of ... "recursion". It also leaves "recursion" in "recursive step" totally unexplained.

I've recently proposed an edit, which was reverted. It replaces "recursive" with "self-referential", which breaks the unbounded recursion and in my opinion is thus helpful, turning an incomprehensible text into a proper definition.

Would appreciate feedback about reinstating this edit. Any objections to that? WillNess (talk) 01:27, 27 January 2024 (UTC)[reply]

Clearly there is an objection to your edit, as I have already undone you once, and explained on my talk page. All you have done is replaced "recursive" with a synonym. I really don't see that as an improvement. Meters (talk) 04:53, 28 January 2024 (UTC)[reply]
The current definition has been in the article since 2015, and for at least a decade before that without the mention of recursion that WillNess objects to: see https://en.wikipedia.org/w/index.php?title=Recursion&oldid=396169565#Formal_definitions_of_recursion from 2010, for example. Meters (talk) 05:03, 28 January 2024 (UTC)[reply]
The point to asking for feedback here is that it does appear here. You yourself suggested I open this issue here on the talk page. So I did. Saying "clearly there is an objection" is like saying, "yes, I know what time it is". Yes you've already replied on your talk page, but I wasn't asking for your verdict: by putting it here I'm opening a discussion.
Your objection misses the point entirely: replacing it with its "synonym" is the whole point to this: a person who does not know what "recursion" is, opens this article, skips to "formal definition", and sees it referring to "recursion". That's clearly no good, regardless of how long it was present in the article. If I the reader do not know what "recursion" is, why would I know that "self-reference" is its synonym? I wouldn't.
Thus there simply was no proper definition of "recursion" in the article, and still isn't. Of course one could object that the "synonym" is mentioned in the lede, but it just forces the reader to go through extra hoops in their mind, to understand the meaning.
It makes it unnecessarily hard to understand, and open to misinterpretations.
We should make it easier for the reader to understand the concept, not harder.
There's also no reference to the lede in the section. A person that skipped the lede to this definition is left utterly without help. -- Cheers, WillNess (talk) 18:08, 28 January 2024 (UTC)[reply]
Which edit specifically are we talking about, and what exactly is the problem it's trying to solve? --JBL (talk) 22:11, 28 January 2024 (UTC)[reply]
WillNess, yes, I suggested that you open a talk page thread, but there is no point in asking a generic "Any objections?" when you already know that there is an objection. As for your objection to the use of the term :recursion", I've already pointed out that original definition, which has been in the article since at least 2010, did not use the term recursion. I have no objection to going back to that version
JayBeeEll, the edit in question is [3], which modifies text that has been stable since 2015. Meters (talk) 08:37, 29 January 2024 (UTC)[reply]
"It's been that way for a long time" is not a good argument to revert an edit, and this bickering about who said what to whom is combative for no purpose; everyone here is obviously trying to improve the article, so let's focus on content, not on the contributor. WillNess, it would be helpful to me if you could make a concise, precise statement of what you think is wrong with the current wording, and how you propose to fix it. Thanks. --JBL (talk) 18:36, 29 January 2024 (UTC)[reply]
Please don't put words in my mouth. I did not say that I undid it because it has been that way for a long. I merely pointed out that the change was to something that had been stable. My summary was "not needed". The change was not an improvement over what we had. Meters (talk) 21:59, 29 January 2024 (UTC)[reply]
No, the change was a much needed improvement over what we had. -- WillNess (talk) 18:10, 14 March 2024 (UTC)[reply]
Thanks you for engaging. I really have already said what I had to say about that. It is all above here in this thread, including a link to my proposed edit. The definition of "recursion" uses the word "recursion"... This is unacceptable, confusing, does not explain anything. Is this really not self-evident? (I haven't read this article since then though; I'm just explaining myself as you requested me to do). All the best, -- WillNess (talk) 18:16, 14 March 2024 (UTC)[reply]
Well if you don't want to make an effort to convince anyone then I don't think you can be too upset when you find yourself in a situation where no one is coming out of the woodwork to agree with you. Your edit adds a bunch of words that don't, to me, make the text seem more correct, less confusing, or more explanatory. Your edit conflates self-reference and recursion; they are certainly related but they're not the same concept: for example, "the nth Fibonacci number" is a thing that is defined recursively but not self-referentially (each Fibonacci number is defined in terms of different Fibonacci numbers, not the number itself). And it's not clear to me why you believe that the formal definition of recursion in the article is in some way circular, nor how you think it relies on the concept of recursion itself. --JBL (talk) 18:29, 14 March 2024 (UTC)[reply]
Thank you for replying and engaging in substantive manner. Since you've raised some specific objections and misgivings I now have something to respond to. Specifically, the nth-Fibonacci procedure refers to itself. It calls itself. Its execution enters the same definition, re-traces the same block of code.
To your second remark. The formal definition in the article defines recursive procedure as such that contains a base case and a recursive case. See? The same word is used without definition, creating bottomless recursion. Why do I say that it's used without definition when it says "recursive case: one that reduces towards the base case", you ask? Because it does not explain what this "reducing towards" means. This is what my proposed edit was trying to address. Specifically, calling itself, a definition referring to self is what being recursive means. This needs to be said explicitly, in my opinion. And by the way, it might not always reduce towards base case. See the Collatz sequence for an example of that. Its definition is still recursive.
Lastly, the third and perhaps unrelated problem here is, I've read through the source which is given there, and could not find any such definition anywhere in it. So the sourcing seems to be bogus. (BTW I'm not getting notified for the responses here for some reason. Do you know perchance how can I get notified about the responses? Thanks.) Cheers, -- WillNess (talk) 01:21, 17 March 2024 (UTC)[reply]
@WillNess: (pinging because you asked for notification). I am sure your first two paragraphs make sense to you, but I have to admit that to me they seem to lie in the spectrum between "wrong" and "incomprehensible". The use of the label "recursive" in the definition of recursion to indicate that this is the part that's recursive is not even slightly circular. The Collatz conjecture is not an example of any relevant phenomenon (the place recursion is used is in defining the sequence where the number of applications of f is i, i.e., we have -- it is the index i that tracks the distance from the base case, not the value ). And merely restating your (dubious) assertion that "self-referential" and "recursive" are synonymous does not make it more true.
What I do agree with you about is that it would be nice if there were a relevant source in this section (actually large portions of the article are thinly sourced); with a few quality sources, we could follow their wording and not have to argue about this at the level of pure logic. (Of course your edits didn't help in that regard :).) I'll poke around and see what I can find (but, as I'm a mathematician, it will be biased in that point of view, so it would be good if you/others also looked for strong references maybe with other perspectives). --JBL (talk) 19:22, 17 March 2024 (UTC)[reply]
well, the use of the word "recursive" inside the definition of "recursive" certainly looks circular to a casual reader, which is supposed to be our target audience. if it's taken as just a name of the relevant clause, the whole definition reverts to being opaque (as in, not clear whatever) since that "reducing towards" is never explained, as I wrote previously.
Next, talking about a specific term of a specific Collatz sequence seems strange as it has no merit here. You describe use of recursion in counting, how is that relevant or interesting? I referred to the whole Collatz sequence for a given n, of course: collatz(1) = [1] ; collatz(n) = [n, ...collatz(n/2)] when n is even OR = [n, ...collatz(3n+1)] otherwise.
Its recursive clause of course depends on the previous value which often is not reducing but grows rapidly, so this is fully relevant.
And yes, recursion is self-reference - a recursive procedure potentially uses self to compute its result. Denying this looks very strange. Otherwise, what is it? The article doesn't seem to say, at the moment. Whether my edits helped or not is only your unsubstantiated opinion, so wording it as fact wasn't helpful. And that's a fact. :) Cheers, -- WillNess (talk) 13:33, 19 March 2024 (UTC)[reply]
Regardless whether a particular recursion is or is not self-referential, the terms "recursion" and "self-reference" are not synonyms. Paul August 14:02, 19 March 2024 (UTC)[reply]
Could you please please show me one instance of recursive definition that does not refer to self. Please. --- WillNess (talk) 15:49, 19 March 2024 (UTC)[reply]
Even if every recursion were self-referential, that does not imply the converse. Paul August 17:10, 19 March 2024 (UTC)[reply]
Yes, but this article does not deal with the converse. My sincere request stands. Thanks. -- WillNess (talk) 18:37, 19 March 2024 (UTC)[reply]
Putting a bunch of random words in italics also does not make your assertions any more valid or convincing. You seem not to have convinced any of the people you're interacting with, and you haven't brought any sources in which to ground the discussion; under these circumstances, I don't see any point in arguing with you further. --JBL (talk) 18:13, 19 March 2024 (UTC)[reply]
Thank you for the dismissive tone of voice regarding my writing style. Unfortunately, I don't think I will be able to destroy myself and rebuild in a way of which you would approve. As for the interactions, I haven't asked you any new questions, have I. Best wishes, -- WillNess (talk) 18:35, 19 March 2024 (UTC)[reply]
To clarify, I have refuted your objections to my original arguments, which remain unanswered. -- WillNess (talk) 18:38, 19 March 2024 (UTC)[reply]

Recursion in pseudo code[edit]

This is the simplest example of a recursive function in C-like pseudocode.

function recurse (int N)
{
   if (N == LOWER LIMIT) return (LOWEST_VALUE)
   else
      return some_function (recurse (N - DECREMENT) )
}


"some_function" need be or need not be recursive itself.

In the example given in the article, the Fibonacci series, the second return would simply be:

      return (N * recurse (N - 1) ) 

2001:8003:E40F:9601:F9B9:CF8C:65FD:BD5D (talk) 10:45, 12 February 2024 (UTC)[reply]

recursion in "see also" (2024)[edit]

There was a discussion about this almost 15 years ago now, and I'd be interested to see what everyone thinks now. I think putting a link to recursion in "see also" would be a really good way to demonstrate recursion per WP:GRAPE. Once again, interested in other people's thoughts. AveryMittag (talk) 11:29, 23 March 2024 (UTC)[reply]