Wikipedia:Articles for deletion/Toki Pona (4th nomination)


 * The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review).  No further edits should be made to this page.

The result was   keep. Keeper |  76  14:31, 23 March 2013 (UTC)

Toki Pona
AfDs for this article: 
 * – ( View AfD View log  Stats )

I went over the previous deletion discussion from 2005 and didn't find strong reasons for keeping except that a lot of people supported the keeping. The article is struggling to establish notability and cites a lot of sources, but most of the sources fall into two categories: primary sources or trivial news coverage. Neither of them is good for establishing notability. Especially telling is the response from the ISO 639-3 Registration Authority (pdf): it rejected the proposed three-letter language code, explaining that it's a novelty language that produced some media interest, but expressing doubts about its continuing importance. Did anything change since then? I doubt it, but it's open for discussion.

To alleviate any doubt, I am not proposing deletion because the language has few speakers. A language may have few speakers, but be notable for other reasons. I just don't see those reasons for this language.

I am listing this under "Fiction and the arts" not to poke fun at this language as "fiction", but because I accept the idea that language construction is an art form. If anybody feels that this is not appropriate, feel free to put it in a different deletion category. Amir E. Aharoni (talk) 16:45, 15 March 2013 (UTC)
 * Update: I now noticed the other two deletion discussions. The second's deletion rationale is similar to what I wrote above, and the result was delete. The result of the third was "keep" because "as a rule of thumb, entries with circa 40 interwiki links are seldom non-notable". Well, I went over several other languages that I can read and couldn't find any sources that would establish notability. For what it's worth, it happens to be nominated for deletion in the Russian Wikipedia, too, for similar reasons: no sources to establish notability. --Amir E. Aharoni (talk) 16:55, 15 March 2013 (UTC)


 * Delete as nominator. --Amir E. Aharoni (talk) 16:55, 15 March 2013 (UTC)
 * Keep, as discussed in the 3rd nomination. Greenman (talk) 17:59, 15 March 2013 (UTC)
 * Note: This debate has been included in the list of Language-related deletion discussions. • Gene93k (talk) 18:03, 15 March 2013 (UTC)


 * Keep, this is a very important language to reference within the worldwide Esperanto community. Chuck SMITH 18:22, 15 March 2013 (UTC) — Preceding unsigned comment added by Chuck SMITH (talk • contribs)
 * Delete 12 years later and with the benefit of hindsight, it's pretty obvious that this never went anywhere. The Globe and Mail article looks like it was a good reference but 404s, and I'd bet a candy bar that whatever was in it couldn't reference most of the article which is either entirely unsourced or relies on self-references, like the now-defunct tokipona.org which shows even the creator seems to have abandoned it.  If we cut the article down to what we can reference with reliable secondary sources, what would we have left, a couple of lines, a short paragraph maybe?  If the ISO Registration Authority rejected it, and they're the experts, I have to say I agree. Andrew Lenahan -  St ar bli nd  19:16, 15 March 2013 (UTC)
 * Keep, because it has generated more interest in the Conlang community than most of the languages listed in the "Conlang" category, and there's actually a sizable number of people learning and speaking it - which sets it apart from all but a handful of conlangs. If Toki Pona is to be deleted, also delete Afrihili, Barsoomian, Blissymbols, Brithenig, Damin, Enochian, Glosa, Idiom Neutral, and so on. Also, note that Tokipona.org is NOT defunct and it has current news on the language: an upcoming book, and a query from New Yorker Magazine wanting to write about it. EDIT: I added a reference to a 5-page article on Toki Pona in the journal of the German society for Interlinguistics, and to three short films in Toki Pona, two of them professional-quality productions. Junesun (talk) 19:42, 15 March 2013 (UTC)
 * Keep. If nothing else, because Toki Pona has been used as the base language for recent research into speech recognition software and robot learning.  See, for example: Squeamish Ossifrage (talk) 20:37, 15 March 2013 (UTC)
 * Mubin, Omar, Christoph Bartneck, and Loe Feijs. "Designing an artificial robotic interaction language." Human-Computer Interaction–INTERACT 2009 (2009): 848-851.
 * Mubin, Omar, Christoph Bartneck, and Loe Feijs. "Towards the design and evaluation of ROILA: a speech recognition friendly artificial language." Advances in Natural Language Processing (2010): 250-256.
 * Saerbeck, Martin, et al. "Expressive robots in education: varying the degree of social supportive behavior of a robotic tutor." Proceedings of the 28th international conference on Human factors in computing systems. ACM, 2010.


 * Keep for the same reasons as in earlier alterations. To address a few points mentioned above:
 * Notability doesn't age. If a subject has ever been notable, it still is. If newspaper references turn up 404s now, it doesn't mean they have stopped being references. Nor does it matter whether the author has abandoned the project or not (although I don't believe it!) - the fact that the Greeks don't believe in Zeus anymore doesn't render him unnotable either. BTW, tokipona.org is not defunct, but working.
 * How many people actually use or speak Toki Pona, I cannot tell. There are a few I know of. But in a way, I am not surprised if the number is high. The learning threshold is low: all you need to know is ca. 120 words and a few rules, and *BANG*, you have a new language in your portfolio. And you can participate in an interesting "game" as well. Let's face it, the fact that Toki Pona is so unusual undoubtedly makes it attractive for some.
 * According to the Popularity stats of WP:CL the page Toki Pona has been quite popular for a long time, with more visitors than f.ex. Volapük, Solresol, Novial, Occidental, Glosa and Lingua Franca Nova. Not that I'm using this as an argument against deletion, mind, I'm merely mentioning it as a curiosity.
 * I do believe the article can do with some improvement. For example, as far as I can see the entire etymological dictionary is original research and could find itself a better home on WikiBooks. &mdash;IJzeren Jan Uszkiełtu?  20:40, 15 March 2013 (UTC)
 * Delete. Severely lacks independent sources. It was never notable and notablity is not seen to grow. A huge nuber of references given is basically original research on private webpages. Staszek Lem (talk) 21:16, 15 March 2013 (UTC)
 * Keep. The article has quality problems with a lot of dodgy factoids dates from 2002 before anyone had studied the language with any level of seriousness. But, the language, after 13 years of existence is actually notable in a way that it wasn't in earlier posts when it only had the reference to "Speed of Thought" article in Russian (in which the author said he actually didn't know anything about toki pona). Media, book and scholarly attention tends to be either a mention in a list of constructed languages. However, the big change has been the articles since 2007 is that they show toki pona being used as a topic of study because it is a small, well documented language, sort of a lab-mouse of languages e.g.


 * Mubin, O., Shahid, S., Bartneck, C., Krahmer, E., Swerts, M., & Feijs, L. (2009). Using Language Tests and Emotional Expressions to Determine the Learnability of Artificial Languages. ACM Proceedings of the 27th International Conference on Human Factors in Computing Systems (CHI2009), Boston pp. 4075-4080.
 * Logician Dr. John Clifford presented on this at Second Language Creation Conference in 2007.
 * MIT has been offering classes in toki pona from 2002 to 2009.
 * It's gotten mention in the New York Times and New Yorker (admittedly as mentions, not as the main topic as it was for the LA times article.)
 * Formal Grammar of Toki Pona, Zach Tomaszewski @ University of Hawaii http://www2.hawaii.edu/~ztomasze/ics661/ZTomaszewski-ICS661.pdf

Wikipedia routinely deletes articles written by a language inventor. In this case the inventor of the language has been doing about nothing for the last 8 or 9 years. In my opinion this makes a stronger case to keep the article because it demonstrates that what toki pona is now clearly isn't the activities of a lone inventor. Now more than ever, what people are writing about toki pona comes from people looking at people who use toki pona (i.e. corpus study), detached from any declarations or decrees from the inventor. The public corpus of texts numbers upwards 50,000 to 100,000 words written by 50 to 100 people depending on what cut off you use for negligible contributions.

There are even two people planning to teach their infants toki pona, leading the the possibility that toki pona will soon pass one of the hardest barriers for notability amongst linguists, that it have native speakers. — Preceding unsigned comment added by Matthewdeanmartin (talk • contribs) 02:04, 17 March 2013 (UTC)


 * Keep. It is the only conlang I know of that was used as a base for an independent art form called sitelen sitelen, a fascinating non-linear hieroglyphic writing system which is notable in itself. http://jonathangabel.com/projects/t47 this Toki Pona article needs revamping, not deletion. (DaBe at 08:25 UTC, Sun March 17th, 2013) —Preceding undated comment added 08:27, 17 March 2013 (UTC)


 * Keep. As mentioned above, it is a well known example of a conlanguage and often referred to throughout the conlang community. It is an excellent example of how much can be communicated with a minimal vocabulary of 120 words. The broken link to the Globe and Mail article does not mean it is not still notable. It was used as the base language for recent research into speech recognition software and robot learning as mentioned above. The article can use some updating, but it should not be deleted. DaveRaftery (talk) 21:16, 17 March 2013 (UTC)
 * Keep - I'd like to ask nominator why the plentiful Google Book hits don't count for anything? Also "Toki Pona was one of the Language and literature good articles, but it has been removed from the list." - do we have many former good articles that have nominated at AfD 4 times? cf. Lingua Franca Nova by George Boeree, Europanto by Diego Marani, Latina Nova by Henricus de Stalo, Ludlange by Cyril Brosch. In ictu oculi (talk) 04:20, 18 March 2013 (UTC)
 * Keep - I guess article may need some improvement but I think overall topic is notable enough for keeping. Quite many Google books and Google scholar hits. Also ORish fact that I actually had heard about this conlang before seeing this article on wikipedia.--Staberinde (talk) 18:50, 18 March 2013 (UTC)
 * Keep - Numerous mentions in various media (e.g., newspapers, radio, academic journals) over multiple years. Insisting that these mentions in the media are "trivial" seems short-sighted: If journalists and academics believed the topic to be "trivial," they wouldn't have reported/researched it in the first place, and a broad-brush rejection of their judgment--especially by those who have no particular academic credentials in the field(s) in question, nor any particular experience as a journalist--seems inappropriate. The allegedly poor construction of the article does not justify deletion; at best, such poor construction would merely justify improving the article, not deleting it. The ISO 639-3 rejection by SIL is neither surprising nor compelling: SIL ultimately is a Christian organization, and one of it main goals is to use language to spread Christianity by translating the Bible into many local, natural languages. Conlangs simply do not support SIL's explicitly stated mission, so discrimination toward conlangs--or at least the potential for such discrimination--is altogether unsurprising and even expected, especially for a deliberately simple conlang that would not be exceptionally suitable for translating large sections of the Bible (i.e., Toki Pona). Bryantjknight (talk) 14:52, 19 March 2013 (UTC)
 * Keep - Came to this page looking for more information on the language, after learning of its existence through the New Yorker article (http://www.newyorker.com/reporting/2012/12/24/121224fa_fact_foer?currentPage=4) and further investigation in places like Omniglot (http://www.omniglot.com/writing/tokipona.htm). Have discovered that toki pona is referenced in many places on the web both within and without a linguistic context but is described here in the most detail.  (Google Scholar yields some interesting non-trivial studies, e.g., http://onlinelibrary.wiley.com/doi/10.1002/9781405198431.wbeal0573/abstract, and "Using language tests and emotional expressions to determine the learnability of artificial languages", O. Mubin et al, CHI 2009, http://www.idemployee.id.tue.nl/o.mubin/wip_paper123.pdf)

As Toki Pona is a unique and well-developed human concept which will be (has been) similarly researched by other people interested in linguistics, scientific history and other fields, maintaining an article on the subject of this language has encyclopaedic value. Many well-reasoned arguments against deletion have already been outlined above by other users. Considering this article for deletion seems at this point ridiculous. Satkomuni (talk) 09:12, 20 March 2013 (UTC)


 * Keep - I tried to look through the criteria for deletion to see if this was mentioned as valid/invalid, but didn't see anything. So if it's mentioned and I missed it, then sorry for that, but Googling "toki pona" (include the quotes) brings up over 3.4 million hits.  The first time the article came up for deletion, the fact that it only had a few thousand hits was used as an argument against its notability.  So could we use this increase as a factor in favor of keeping?  In addition to the number of hits increasing, the relative search volume related to it can be seen in Google Trends to be holding fairly steady over the last few years.  (It has a few spikes and dips, but it's reasonably steady; about as steady as Esperanto's interest level on Google Trends.) - J. Tweed - 10:14AM, 20 March 2013  — Preceding unsigned comment added by 209.91.144.98 (talk) 14:16, 20 March 2013 (UTC)
 * Speedy keep. Both significant media mention and significant use by linguists.  – SJ +  02:47, 21 March 2013 (UTC)
 * Comment - many people on this page claim that it's notable, well-known, well-referenced, etc., but it's not supposed to be written here. It's supposed to be written in the form of reliable sources in the article itself. When I opened this AfD, the article didn't have any reliable sources at all. Now it has some references to what looks like academic articles, which may be OK, but it must be cited in the right context - the article is supposed to make statements and to back them by sources. Now it's only the story of the language's creator and some information about the language itself - there's no assertion of notability in the article. --Amir E. Aharoni (talk) 15:29, 21 March 2013 (UTC)


 * Comment - Have the criteria at Conlangs/Criteria been factored in? That seems to be a rough outline of what the conlanging community itself considers "notable".  Wikipedia does need high standards, but at the same time it is relevant to consider the wishes of the target audience; in this case, people who are interested in con langs are the ones who will be wading through this type of article, benefiting from the thinning out of vanity pages and the retaining of quality pages.  None of this to say that toki pona has necessarily met enough of those criteria - just that I think they should be considered in this case.  On an unrelated note, facebook.com lists toki pona as having 1,200 "speakers" - which of course does not mean native speakers, but "speakers" as much as it is true of any conlang. - J. Tweed  — Preceding unsigned comment added by 198.254.194.149 (talk) 23:42, 21 March 2013 (UTC)
 * Regarding Facebook, see my comment here. &mdash;IJzeren Jan Uszkiełtu?  00:02, 22 March 2013 (UTC)
 * "In any case, it is quite obvious that as a source of information [Facebook language usage statistics have] no value at all - even as a primary source." -- On the contrary, they do have value in demonstrating a spectrum of the significance of con langs relative to each other. Which is to say, it may not be possible to disambiguate all the factors that may be eating away at the absolute accuracy of the information, but one can reasonably presume – to an extent - that those factors likely affect each entry at approximately the same level.  For example, we can gather from your (excellent) examples that Laadan is roughly twice as significant as toki pona.  From there, perhaps we need a criterion that the top 10 conlangs (or whatever number is reached through consensus) are worthy of their own Wikipedia pages. - J. Tweed
 * It's an interesting thought, and I admit it has crossed my mind as well. Still, the list gives some pretty strange results. Surprising are the remarkably high scores of the relatively minor (unknown) conlangs Babm and Mirad. Surprising is also that 40.000 of Ido vs. 2.500 of Interlingua while it is commonly believed that both languages have a similar amount of users (and that Interlingua even is a tad bigger) - but would that mean you can say that Ido is 16x more notable than Interlingua? Remarkable is also the very low score of two languages that have been deemed notable enough for their own Wikipedia project, Interlingue (150) and Novial (9). So I don't know, those figures are interesting, but not reliable in any way. &mdash;IJzeren Jan Uszkiełtu?  16:59, 22 March 2013 (UTC)
 * Keep, this page has had quite a history, 4 AfDs, a DRV, one of the weakest GA assessments I've ever seen, and a failed GA reassessment, but significant coverage does exist: This LA Times piece is clearly a reliable source that fully covers the topic. Here is a copy of The Globe & Mail piece that is no longer hosted on their site, but proves excellent coverage. This The New Yorker piece is a only passing mention, but shows that coverage exists. Should it be edited down a bit, yes, but it certainly belongs here. J04n(talk page) 14:17, 23 March 2013 (UTC)
 * The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.