Wiktionary:Beer parlour/2017/February

Czech verbs ending in -ti or -t?
All my etymological Slavic sources (Derksen, ESSJa, Vasmer) consistently write Czech infinitives ending in -ti, but in Wiktionary they are entered ending in -t instead. What's going on here? Should we at least create soft redirects for the forms in -ti? Benwing2 (talk) 04:58, 1 February 2017 (UTC)
 * The -ti endings must be archaic. Most modern Czech verbs end in -t, AFAIK - brát, psát, číst, nést, být, dělat. --Anatoli T. (обсудить/вклад) 05:36, 1 February 2017 (UTC)
 * Quoting one web source: A Czech infinitive usually ends in -t (infinitives in older literary texts may end in -ti). --Anatoli T. (обсудить/вклад) 05:38, 1 February 2017 (UTC)


 * Dan has answered this for me before. Older dictionaries lemmatize with -ti, newer dictionaries lemmatize with -t. As far as I know, both forms have always existed in parallel (and are not etymologically the same form and sometimes the difference is non-trivial, e.g. péct vs. péci). I think both forms should be given in conjugation tables, with the -ti form being treated like any other non-lemma form. We should probably also give them separate names. --WikiTiki89 15:22, 1 February 2017 (UTC)
 * @Benwing2: -ti is archaic. -ti forms are used neither in modern speech nor in modern writing, as you can verify in Google Books for yourself. Nonetheless, even some 20th century dictionaries like R:PSJC and R:SSJC lemmatize on -ti. Interestingly enough, SSJC uses -ti as lemma but -t in its definitions, e.g. for "plavati". In my Wiktionary editing practice, I am ignoring the -ti forms altogether. --Dan Polansky (talk) 13:21, 4 February 2017 (UTC)

Russian-Rusyn offline dictionary (PDF)
I've got a copy of Russian-Rusyn dictionary by Igor Kercha (Игорь Керча)- 65,000 terms as two scanned PDF files, over 30 MB altogether. It seems good but I'd check translations with other sources occasionally. It's cumbersome to use (the files are not searchable) but I can share it. Please suggest how to send it to you if interested. --Anatoli T. (обсудить/вклад) 11:39, 1 February 2017 (UTC)

empty declension sections (?)
Hi, some terms show an interrogation mark in sections of their declensions, e.g. https://en.wiktionary.org/wiki/أغلب#declension. I'd like to know what it exactly means and how it could be filled. Thnks in advance. --Backinstadiums (talk) 19:33, 1 February 2017 (UTC)
 * Most elatives aren't declined any more. The ?'s indicate that the forms probably aren't attested. Benwing2 (talk) 03:46, 2 February 2017 (UTC)
 * Not even in the whole arabic wikipedia? There're some probabilistic approaches which offer really good results using such corpora, but as no possible copyright issues may arise regarding the existence of lexicographic terms, so any good dictionary, as for example the oxford arabic, or lexicon could offer so. Furthermore, OCR'd resources might be used as well. --Backinstadiums (talk) 19:08, 4 February 2017 (UTC)
 * I looked through several dictionaries to find elative forms. Benwing2 (talk) 19:23, 4 February 2017 (UTC)

Suggestions without reply in entry discussion sections
Hi everybody, I've been adding some remarks and suggestions on different 'discussion' tabs of some entries. I'd very much appreciate it if someone could check them out and discuss on them. Thank you so much in advance. --Backinstadiums (talk) 08:38, 3 February 2017 (UTC)

Seventh LexiSession: fever
Monthly suggested trend topic is . You are invited to participate in the common goal to discover what can be gathered around the word fever! It may be description of subtype of fever, creation of a Wikisaurus, add of sounds or pictures.

This is a collaborative experiment without any guide nor direction. You're free to participate as you like and to suggest next months topic. If you do something, please report it here, to let people know you are involve in a way or another. I hope there will be some people interested by this one Noé (talk) 10:50, 3 February 2017 (UTC)

Standardizing homophone qualifiers
I've noticed we don't really have a standard for designating in which accents a word is a homophone. What I see most commonly is after the IPA, assumed to be for all accents (unless indented under one) unless there is a qualifier template with "in some accents". There must be a way to do these consistently though, such as directly in the homophones template ("ex1|ex2|q1=some|q2=caught-cot" or similar?). As it stands, if there are 3 homophones and the last 2 are "in some accents", there is no clear solution. It's also tricky with automated translations, such as in ca-IPA where Central Catalan has many, many homophones compared to Valencian, and I think it is worthwhile to specify where two words are homophones, and it is clearest under the line in question instead of after them all. Ultimateria (talk) 22:18, 4 February 2017 (UTC)

User:BrunoMed
This contributor has been warned about, and blocked for, questionable entries several times. First they were creating entries of protologisms, and finally unblocked when they agreed not to do that any more. Since July, they've been mass-creating entries from lists, all with exactly the same content even when it doesn't make sense. In September, it was country-name prefixes, all with exactly the same etymology except for the source country-name and the entry name- an etymology that was correct for only a handful of the entries. Today they created entries for about 154 given names (there may be a few in there without the English section), each with an English and a Serbo-Croatian section, without any evidence that most of these Serbo-Croatian names have ever used in English. Many of them, such as "Ognjemir", would be unpronounceable to the vast majority of English speakers. They also created 48 Latin proper-noun entries based on Serbo-Croatian names. After seeing how much work it took to clean up their September run, I simply mass-deleted the the whole batch from today (and also reverted an edit where they changed the declension of a Latin noun ending in -a from first to second declension).

At the same time I blocked them for two weeks. Given that they continued to create bogus entries using both their own account and IPs long after being warned and blocked (they blanked their talk page, so most of the warnings aren't visible), and that they were blocked and warned again in September for mass-creating entries, but came back and did it again, I would like to extend that block much longer, to make it indefinite- but I would prefer to have the consensus of the community before doing so.

They've been wasting a lot of other people's time and effort with their actions over the past couple of years. The specific types of bad edits have changed, but the pattern of reckless disregard for accuracy has been quite consistent. Even if they stop their current practices, I see no reason to expect that they won't come up with something equally irresponsible later on.

What does everyone else think? Chuck Entz (talk) 03:51, 5 February 2017 (UTC)


 * When someone consistently adds bad stuff I think it becomes reasonable to demand a discussion and block them if they refuse to engage/reply. So far it looks as though all he has ever done with his talk page is to blank it. Equinox ◑ 08:58, 5 February 2017 (UTC)


 * Don't think I'm vandalizing again. I blanked my talk page because I have right to do so. I clean every WhatsApp message when I read it, I did the same with my talk page. I didn't realize that my mass creation could be a problem, because there are thousands of entries much worse than mine (and I think mine are good).


 * Today they created entries for about 154 given names (there may be a few in there without the English section), each with an English and a Serbo-Croatian section, without any evidence that most of these Serbo-Croatian names have ever used in English. Many of them, such as "Ognjemir", would be unpronounceable to the vast majority of English speakers.


 * This is the proof that the name is used in English. English is a universal language, thus every Serbo-Croatian name without diacritics is also English name and it belongs to English language. That's why I love English. You cant just delete all my entries without asking what I'm doing. If you told me, I would make everything right. And without people like me, on Wiktionary would be no more than a hundred entries. And there are much more unpronounceable words than "Ognjemir" like Hradec Králové, hryvnia, hsianghualite, ZWNBSP, Þrymskviða, Łukasiewiczian, ǂKxʼauǁʼein, Świętochłowice, Ṛgvedic... is that pronouncable to vast majority of English speakers???


 * The problem is with Latin entries. Please recreate those entries because they have nothing to do with Serbo-Croatian names in English, they are completely fine and correct:
 * Književnik: Časopis za jezik i poviest hrvatsku i srbsku i pirodne ..., Opseg 1
 * Vojnomir Slaven | Hrvatska enciklopedija
 * Jezik: - Opseg 9 - Stranica 34


 * I think he is now . SemperBlotto (talk) 15:20, 7 February 2017 (UTC)
 * You don't think correct. --Bjelun (talk) 15:22, 7 February 2017 (UTC)

Reconstructed Latin terms are not linked
Hi, the etymology of meddle states "from Late Latin misculare", yet misculare is in red, without specifying it's a reconstructed term despite there being an entry for it. I think I have seen reconstructed terms indicate so with an asterisk. Thanks in advance. --Backinstadiums (talk) 19:35, 5 February 2017 (UTC)
 * Mhm, so you add the asterisk when you see something like this. I've fixed it now. — Kleio (t · c) 19:44, 5 February 2017 (UTC)
 * Okay, I'm going to give you some basic information on linking. You can fix this stuff yourself. The way to link to the Reconstruction namespace is to place an asterisk before the term in the linking template (or etymology template in this case): thus, or . The asterisk and language code tell Module:links to add the text   to the wikilink. Also, you can link to entries using   or   (or with the linking template ) rather than copying out the whole URL. — Eru·tuon 06:05, 6 February 2017 (UTC)
 * Also, etymologies in English-language dictionaries tend to use the infinitive, but we use the first-person present indicative for our lemmas, so it's a good idea to change to that form when you see it in our etymologies. Otherwise users click on the link, only to get a "form of" entry referring them to the lemma. Chuck Entz (talk) 13:14, 6 February 2017 (UTC)

Glyph origin vote
I created Votes/2017-02/Glyph origin.

It was based on the discussion Beer parlour/2017/January. --Daniel Carrero (talk) 18:47, 6 February 2017 (UTC)
 * Thank you so much. I do support it. --Backinstadiums (talk) 14:11, 7 February 2017 (UTC)

"only present in non-English entries"
WT:EL contains this line:


 * Inflection, or Conjugation for verbs, or Declension for nouns and adjectives, or Mutation, only present in non-English entries

I'd like to remove the part "only present in non-English entries", and I guess we don't a need a vote for that.

Rationale for the edit: the entry be has a conjugation section, and I believe that's OK. --Daniel Carrero (talk) 14:24, 7 February 2017 (UTC)
 * Mutation is entirely separate from Inflection, both co-occur in entries. So it should be changed so that Mutation doesn't appear to be a variation on Inflection. —CodeCat 14:27, 7 February 2017 (UTC)
 * Maybe we can just list them like this:
 * Inflection
 * Declension
 * Conjugation
 * Mutation
 * This would be consistent with Synonyms, Antonyms, Hyponyms, etc. that are all in separate lines. --Daniel Carrero (talk) 14:49, 7 February 2017 (UTC)
 * But Inflection, Declension and Conjugation are mutually exclusive, whereas Synonyms and Antonyms are not. —CodeCat 16:17, 7 February 2017 (UTC)
 * OK. Then it probably should be:
 * Inflection, Declension or Conjugation
 * Mutation
 * --Daniel Carrero (talk) 16:19, 7 February 2017 (UTC)
 * ✅ --Daniel Carrero (talk) 21:50, 16 February 2017 (UTC)

Vulgar Latin hypothetical verb forms and Ibero-Romance
I was wondering if there should be a distinction noted for the descendants of Vulgar Latin verbs in Ibero-Romance (not including Catalan). Specifically, these languages merge the Latin second and third conjugation types (though the infinitives derive from the second type, with the long e in -ēre). For example,, , , , etc. I understand there's no need to make a distinction for a main Latin entry (of an existing word), but if we're making the Vulgar Latin entries more detailed for those who wish to study or learn about it more, including differentiating various types of conjugation for the types of Romance, should there be a distinction made for these languages (for example in or )? Or should it just be assumed that Portuguese, Spanish, Asturian, Galician, etc. always have the stress on the final syllable of infinitives, even if the Vulgar Latin form they're listed under has stress on a different vowel (which would apply for the rest of Romance). Does anyone know if this was a later phenomenon in that region or if they presumably used a separate form of Vulgar Latin in this regard? Since the first records of these languages comes much later, it's hard to come to a conclusion on this. In other words, would Spanish have come from a, as a variant or later evolution of , or simply just changed the stress for all its verbal infinitives later, analogically? Word dewd544 (talk) 22:39, 7 February 2017 (UTC)

Petitioning to change Turkic Khalaj (klj) to Khalaj
Anyone who's heard of Khalaj knows Khalaj is a Turkic language, Iranic Khalaj appears to be a ghost language and it would be a good idea to stop perpetuating confusion about this. Crom daba (talk) 02:23, 8 February 2017 (UTC)
 * I support the change. My materials always call this language simply "Khalaj". --Vahag (talk) 08:57, 12 February 2017 (UTC)


 * ✅. I have renamed the  to "Khalaj" and removed the ghost code  . - -sche (discuss) 22:37, 25 May 2017 (UTC)

finding an unknown word
I would like to be able to find all possible words with particular first and last letters by typing in at least the first letter of a word, and a certain number asterisks (or ?), representing subsequent letters and then the last known letter. Is this something I can do in Wiktionary? Or perhaps someone knows another site that can help me do this. Clduncan (talk) 16:57, 8 February 2017 (UTC)
 * This would be extremely helpful, but I'm pretty sure you can't do it on Wiktionary. Andrew Sheedy (talk) 17:24, 8 February 2017 (UTC)
 * Try the simple way using '*' for a wildcard. You may want to restrict yourself to English lemmas using 'incategory:"English_lemmas"'. It works for me. DCDuring TALK 18:39, 8 February 2017 (UTC)
 * Well, look at that. I could have sworn I'd tried it before without success. Andrew Sheedy (talk) 18:45, 8 February 2017 (UTC)
 * I hadn't realized it until I went to the Mediawiki documentation (CirrusSearch) today. I thought it would take some regex search, which I didn't know how to limit to entry titles. DCDuring TALK 19:54, 8 February 2017 (UTC)

Saxon
Since the Saxons didn't name themselves after the seax, where does the word come from? ÞunoresWrǣþþe (talk) 19:42, 8 February 2017 (UTC)

Display order of Template:cite-book
Right now, it displays the author(s), then the year, then the entry, then the title of the book. This doesn't really make sense to me. A more sensible order would be to have the entry first, then the title, then author(s), then year. Can this be changed? —CodeCat 19:02, 9 February 2017 (UTC)
 * IMO, the year needs to come first. This helps with ordering the citations in chronological order. --Daniel Carrero (talk) 19:05, 9 February 2017 (UTC)
 * Yes, and Quotations calls for putting the year first. —Aɴɢʀ (talk) 10:37, 10 February 2017 (UTC)
 * These are citations though, not quotations. —CodeCat 13:32, 10 February 2017 (UTC)
 * Aren't citations just quotations found in the "Citations:" namespace? Is there any reason to format them differently at all? (except you add a "#" when the quotation is directly below a sense in an entry) --Daniel Carrero (talk) 14:07, 10 February 2017 (UTC)
 * See the documentation of . —CodeCat 14:39, 10 February 2017 (UTC)
 * Can this be implemented please? contains a link to a web page with the entry, but this link is hidden in the middle of the text rather than being placed first. —CodeCat 15:33, 8 March 2017 (UTC)
 * I understand how sorting by date is important for quotations, as you want to know when a word first came into use, but for references, I think the author is much more important than the date. I think that's also the reasoning behind it being the standard in published works. I'm opposed to switching them around. --Victar (talk) 05:23, 27 March 2017 (UTC)

Rhymes:Telugu/త
Huh? Why is this not in IPA? —CodeCat 14:01, 10 February 2017 (UTC)
 * -త- Wyang (talk) 10:50, 13 February 2017 (UTC)
 * So can it be moved please? —CodeCat 17:09, 13 February 2017 (UTC)
 * Sorry I don't know Telugu. This only looks like a concerned face to me. You may wish to tag the creator Dr. Rajasekhar for his opinion. Wyang (talk) 09:41, 14 February 2017 (UTC)
 * I have explained this previously around here somewhere. I think User:Renard Migrant asked about it, probably on the Beer Parlour. Be that as it may, Dr. Rajasekhar and I discussed this matter at User talk:Stephen G. Brown, where we said as follows:
 * STEPHEN: In general, you could create pages such as Rhymes:Telugu/ప్పా [...] Or you could show the rhymes in IPA as we do in English: Rhymes:Telugu/pːaː. Choose the method that you prefer.
 * RAJASEKHAR: [...] Since many Telugu words does not have IPA ; it looks easier to use the first method Rhymes:Telugu/ప్పా. But none of the language rhymes in English wiktionary have used this method.
 * STEPHEN: Almost all of the language rhymes on English Wiktionary are languages that use the Latin alphabet. I think only Hindi and Russian are using IPA instead of their standard alphabet. The Greek rhymes are using the Greek alphabet for its rhymes. See Rhymes:Greek/έος for example. —Stephen (Talk) 04:40, 21 April 2017 (UTC)

Actualités: Monthly news of French Wiktionary in English
Hi all, Here comes the 22th issue of Actualités, the periodical news about Wiktionary, translated in English specially for you! Well, it's January edition with a small delay (only 11 days!).

This is a huge issue. Massive. You're gonna love it. It's wiktionarilly tremendous. There is three amazing articles: a focus on thesauri in French Wiktionary, a description of a published dictionary of insults (more than 9.300 insults in French!) and an article about the names for the snow in Inuit. Plus: a Wiktionarian coded a small game based on Wiktionary database and it was improved collaboratively. As usual, we also mention some briefs and provide metrics.

This time, it was quite hard to translate, so I am sorry but it is barely better than a draft and you may have to guess the meaning for some part, or to help to finalized the translation if you can. It is still made without any funding for anyone, craft by Wiktionarian hands. I hope it will be inspiring for those who will read it and I'll be delight to answer to any questions or to listen to comments on our work! Cheers, Noé (talk) 20:16, 11 February 2017 (UTC)
 * I wish I had time to help with the translations this time, but I haven't even found time to read it yet! Andrew Sheedy (talk) 03:54, 12 February 2017 (UTC)

Hunnic Lemmas
There were previously no Hunnic Lemmas on the site. I've added 18 of them, with 2 of those being reconstructions. ÞunoresWrǣþþe (talk) 12:47, 12 February 2017 (UTC)
 * They should all be reconstructions IMO since we don't have any primary sources. Not to mention that Pritsak's etymologies are speculative and that alternative theories exist. Crom daba (talk) 14:12, 12 February 2017 (UTC)
 * The Greek forms count as attested ones. ÞunoresWrǣþþe (talk) 15:58, 12 February 2017 (UTC)
 * We usually count those as belonging to the language they're attested in, compare 🇨🇬. Crom daba (talk) 16:34, 12 February 2017 (UTC)
 * This is different. ÞunoresWrǣþþe (talk) 16:40, 12 February 2017 (UTC)
 * Yeah, to my eye, these should all be reconstructed, as there is not direct attestation of the words in the Hunnic language. We have sometimes been inconsistent (Gaulish names in Latin are sometimes put in Gaulish), but in the absence of a proper written Hunnic corpus, all lemmata should be reconstructed, especially those words that are not directly attested (,, , etc.). — JohnC5 18:01, 12 February 2017 (UTC)
 * You moved all of those words to Reconstruction, but they need to be in Reconstruction:Hunnic/. Currently, they are not legal entries. — JohnC5 19:10, 13 February 2017 (UTC)
 * Would like to note that quite a few scholars (e.g. D.H. Green) still consider Attila to likely be a Gothic name (or a Gothicized version of a Hunnic name), his Hunnic name may very well have been different. — Kleio (t · c) 17:31, 12 February 2017 (UTC)
 * Those historians are a minority. ÞunoresWrǣþþe (talk) 18:52, 12 February 2017 (UTC)
 * What makes you say so? — Kleio (t · c) 19:31, 12 February 2017 (UTC)
 * Because it's a fact. ÞunoresWrǣþþe (talk) 21:08, 12 February 2017 (UTC)
 * Then back it up. Korn &#91;kʰũːɘ̃n&#93; (talk) 21:32, 12 February 2017 (UTC)
 * No thanks. I'm done wasting my time in pointless arguments. ÞunoresWrǣþþe (talk) 14:51, 13 February 2017 (UTC)
 * Love how you live in some pathetic little bubble where basic facts have to be sourced. There's no "source" to the fact a minority of historians believe it's Gothic and a majority believe it's Turkic-Altaic. Want a list of every publication that's ever leaned in favour of one or the other? Good luck. It's impossible. And since you with your superior opinion are clearly correct... Please, source the opposite claim: that there are enough historians that believe Attila is a Gothic name to even warrant the inclusion of that fringe theory in a dictionary? By all means, try, but I expect you only "contributed" to this discussion to be contrarian. Oh, don't forget to source every word in your response, otherwise we can't be sure you're speaking English or that your words mean what they do. ÞunoresWrǣþþe (talk) 15:04, 13 February 2017 (UTC)

Sorry to break it to you but some of the sources you have provided do not meet Wiktionary's standards (e.g. Reconstruction:Hunnic/strava, Reconstruction:Hunnic/medos). In addition to this, you haven't followed appropriate naming regulations either (e.g. Reconstruction:Ōybárs, Reconstruction:ōy, Reconstruction:čérkün, Reconstruction:tōn). I believe that the concerns raised by other users are well-founded – the subject at hand is not well documented and therefore prone to scrutiny, and rightfully so. I also advise you not to refer to words in languages you don't master as you did here. And last but not least, will you do us the courtesy of abstaining from personal attacks, they won't do you any favours around here. --Robbie SWE (talk) 20:43, 13 February 2017 (UTC)
 * Also, where do the guidelines for orthography in About Hunnic come from? Why are some of the vowels starred (*ȫ, *oː, *aː) and others are not, especially when they are all reconstructed? Also, why is there a mix of IPA representations (*oː) and orthographic ones (*ȫ)? — JohnC5 22:18, 13 February 2017 (UTC)
 * In that case, Robbie, never edit a Hunnic page again. You don't master the language. You clearly have some racial bias here, I can see from your user page you're Romanian. You're the last person I'd trust on this, given your source contradicts you, and was posted in a foreign language with no translation to confuse or mislead people. ÞunoresWrǣþþe (talk) 23:18, 13 February 2017 (UTC)
 * Cucura > Koúkouron > Cucură or Cucura > Cucură. I must say it's very "hard" to decide. Reminds me of my fringe theory where the word boat doesn't come from Old English bāt, but instead bāt > qóng > boat, came to English through Mandarin. ÞunoresWrǣþþe (talk) 00:12, 14 February 2017 (UTC)
 * Wow, do we have another Uther on our hands? —CodeCat 23:51, 13 February 2017 (UTC)
 * Interesting question, let's look at the evidence. This user created the page About Hunnic by copying WT:ACEL-BRY and how the  template works. The user has edited primarily Celtic and Anglo-Saxon pages. The user does not understand how namespaces work. The user has taken on a language for which (s)he clearly lacks a comprehensive or consistent understanding of the literature. The user has a name associated with the British Isles' mythology (Old English for Thunder's Wrath). And the user is quite rude, resorting to incomprehensible personal assaults when challenged for evidence. All in all, I'd give this one a strong 8/10 on the “Uther-o-meter”™. — JohnC5 00:14, 14 February 2017 (UTC)
 * I am not that person, though you may of course believe what you want. I will say, however, that you are a 10/10 on the failed abortion-o-meter. A grade-A cunt, not those made in China cunts but a bon and bred American cunt. ÞunoresWrǣþþe (talk) 00:30, 14 February 2017 (UTC)
 * And pssst, Thor's Wrath, you bescittende cunte, and fukkende horninsunu. Chupama y chingate, cago en la leche de th madre. ÞunoresWrǣþþe (talk) 00:32, 14 February 2017 (UTC)
 * Oh sorry, that is my bad as it is completely ambiguous. I'm gonna' upgrade you to a 9/10 on the Uther-o-meter. Does this mean we block now? ? — JohnC5 01:23, 14 February 2017 (UTC)
 * After that, I support a block. —CodeCat 01:33, 14 February 2017 (UTC)
 * I don't know what the official policy for declaring someone a sockpuppet. Do I need a CheckUser ? Or can we just block Uther? Regardless, I'll block the account for a few days for offensive behavior. Anyone wanna' nuke some Hunnic entries? — JohnC5 01:50, 14 February 2017 (UTC)
 * I think what has happened so far is bad enough, without them being confirmed as a sockpuppet. So block away I say. —CodeCat 01:51, 14 February 2017 (UTC)
 * I always have trouble judging these things. I gave them a week block. If someone wants to upgrade that, feel free. — JohnC5 01:54, 14 February 2017 (UTC)

Wow, nothing like a good ol' cup of bigotry in the morning. Not trying to defend myself here since I don't believe that anyone buys what you're selling, I just want to point out that I did not do any substantial changes to Hunnic reconstructions because I'm admittedly not well versed in reconstructed languages. However, unlike you, I speak Romanian and I know Romanian history and orthographic norms, so I'm not going to abstain from correcting obvious and elementary errors. I'm somewhat taken aback by your claim that I'm racially biased – it says much more about you than it does about me. Not that it matters since you don't know me, you apparently have no respect for anyone else around here and you do remind me an awful lot of Uther who was impossible to discuss with, but for the record I'm a Swede. Mind you, it doesn't make me more or less objective but it exposes once again your flawed logic and antagonistic behaviour. I would've blocked you indefinitely if it were up to me, but thankfully it's not. I have respect for my peers, hence I trust their judgements regarding the duration of you block and your future participation in this project. --Robbie SWE (talk) 10:34, 14 February 2017 (UTC)
 * I have upgraded the block; regardless of whether or not he is Uther, this is grossly inappropriate. I do think that a CU investigation would be handy all the same. The real question, which I am linguistically ignorant of, is whether we should nuke all these Hunnic lemmas, reconstructed and otherwise. —Μετάknowledge discuss/deeds 03:21, 14 February 2017 (UTC)
 * Thanks for the upgrade. As to the Hunnic lemmata, I possess not the knowledge to evaluate whether the entries are salvageable nor know-how to fix them if they be so. Regardless, they cannot stay as they are. — JohnC5 03:30, 14 February 2017 (UTC)
 * I've moved Reconstruction:Hunnic/adám to an acceptable name and tidied it up in accordance with the one cited source. I have neither the time nor the inclination to do the same for all the remaining Hunnic entries, but this one at least has a scholarly background. —Aɴɢʀ (talk) 09:05, 14 February 2017 (UTC)


 * For what it is worth, User:ÞunoresWrǣþþe is almost certainly User:UtherPendrogn per check user investigation. - TheDaveRoss 12:28, 14 February 2017 (UTC)
 * I've upped the block to to indefinite on that information. —Aɴɢʀ (talk) 13:25, 14 February 2017 (UTC)

We should probably delete all his Hunnic reconstructions, if only because they do not list descendants. --Vahag (talk) 07:20, 17 February 2017 (UTC)
 * Would you do the honors? — JohnC5 07:27, 17 February 2017 (UTC)
 * ✅. All that was salvageable has been moved to Reconstruction:Proto-Mongolic/köküür --Vahag (talk) 09:52, 17 February 2017 (UTC)

De-Recognition of Wikimedia Hong Kong
''This is an update from the Wikimedia Affiliations Committee. Translations are available.''

Recognition as a Wikimedia movement affiliate — a chapter, thematic organization, or user group — is a privilege that allows an independent group to officially use the Wikimedia trademarks to further the Wikimedia mission.

The principal Wikimedia movement affiliate in the Hong Kong region is Wikimedia Hong Kong, a Wikimedia chapter recognized in 2008. As a result of Wikimedia Hong Kong’s long-standing non-compliance with reporting requirements, the Wikimedia Foundation and the Affiliations Committee have determined that Wikimedia Hong Kong’s status as a Wikimedia chapter will not be renewed after February 1, 2017.

If you have questions about what this means for the community members in your region or language areas, we have put together a basic FAQ. We also invite you to visit the main Wikimedia movement affiliates page for more information on currently active movement affiliates and more information on the Wikimedia movement affiliates system.

Posted by MediaWiki message delivery on behalf of the Affiliations Committee, 16:25, 13 February 2017 (UTC) • m:Special:MyLanguage/Affiliations Committee/MassMessages/De-Recognition of Wikimedia Hong Kong - February 2017 • Get help


 * I think this is a big mistake. ÞunoresWrǣþþe (talk) 18:06, 13 February 2017 (UTC)
 * Have you tried contacting anyone at WM-HK? —Justin ( koavf ) ❤T☮C☺M☯ 19:57, 13 February 2017 (UTC)

Proposal: Implementing Wikidata access
I suggest implementing Wikidata access on Wiktionary. I believe this requires a vote and then if it passes we'll have to fill a request on Phabricator.

Apparently we can't access Wikidata using wikitext like or Module:wikidata right now.

I'd like to use Wikidata to contain all character information (including character names and image links) that is currently on subpages of Module:Unicode data, for use by other Wiktionaries as discussed here: Grease pit/2017/February.

There are probably other things we can do with Wikidata.

I wonder if we need all the 19 extensions from the groups "Wikibase" and "DataValues" that Wikipedia is currently using. Some of these extensions seem optional like a geographical parser and a JavaScript API, but seem nice anyway.

See these pages for the lists of extensions:
 * w:Special:Version (in Wikipedia)
 * Special:Version (in Wiktionary)

--Daniel Carrero (talk) 18:59, 13 February 2017 (UTC)
 * I absolutely believe that we need to integrate with Wikidata and the rest of the WMF projects. How that will look exactly and the extent to which Wiktionary will become more like an OmegaWiki project that is primarily a database are important discussions to have but there's no doubt that there are substantial advantages to opening up to Wikidata. —Justin ( koavf ) ❤T☮C☺M☯ 20:00, 13 February 2017 (UTC)
 * Strongly support. If we want to raise the quality of all language Wiktionaries we need ways to cooperate, instead of duplicate loads of data on every site. –dMoberg 21:42, 13 February 2017 (UTC)
 * Absolutely. - TheDaveRoss 21:58, 13 February 2017 (UTC)
 * (naturally) and also comment: If we are going to use Wikidata to contain any information in all languages (for example: the periodic table in all languages or something), then I'd also suggest eventually changing all our exceptional language/family/dialect codes to be ISO-compliant. (which can be discussed and/or done later, it does not have to be right now) Reason: it appears that any information we put there will be accessible by other people rather than only by Wiktionary, so it might make sense to use a universal standard. For example, maybe it would be nice to change our "gem-pro" to "gem-x-pro", because I believe that the former basically means "Germanic-Provençal" instead of "Proto-Germanic". (here's a previous discussion about this: Beer parlour/2013/February) --Daniel Carrero (talk) 06:06, 15 February 2017 (UTC)
 * Basically, this is a proposal that would involve more typing for zero benefit. It would be one thing if the ISO assigned a meaningful code for Proto-Germanic (they won't), but if we are to make our own codes, they should be consistent with the ISO codes and not too much trouble — and both of those things are currently true. —Μετάknowledge discuss/deeds 07:11, 15 February 2017 (UTC)
 * Re "they should be consistent with the ISO codes": I think I know what you mean, because our codes don't conflict with actual ISO codes. Still, we are not using strictly ISO-compliant codes. Probably the use of made-up language codes is an issue to be discussed at Wikidata too. We might want to ask them: "Are you cool with keeping some information using langcodes that Wiktionary made up like gem-pro meaning Proto-Germanic and roa-opt meaning Old Portuguese?" I guess they have a right to want ISO-compliant codes but I didn't check that yet.
 * If in the future we decide to change all our codes to be ISO-compliant for whatever reason, for "gem-pro", instead of "gem-x-pro" we might use "qge-pro" and keep within the established 7-character limit. (we may even shorten it somehow) Admittedly, it would take work to change all the codes. --Daniel Carrero (talk) 15:23, 15 February 2017 (UTC)

There is an ongoing project in Wikidata to include Wiktionary data and I invite you to participate with your proposal. In my opinion, it is mainly developed by people looking at benefices for Wikidata and not so much people looking for how it can be useful for Wiktionary. I tried to make them clarify some aspects of the development, but it is still quite hard for me to understand how it can not create a fork. Anyway, I am very happy to see your discussion about Wikidata Noé (talk) 09:11, 15 February 2017 (UTC)
 * - also volunteering to help with the integration. to clarify, this proposal is a bit different and complementary to the one you linked. We need a 2-way exchange of data, this proposal deals with the integration of already existing data (such as language codes) back into Wiktionary. – Jberkel (talk) 09:28, 15 February 2017 (UTC)
 * I am still not sure to understand which data do you want to use from Wikidata. Module where mentioned, but it means to host them in Wikidata and I don't know if it is a good repository for that kind of file. I will mention this discussion in French Wiktionary to invoke more comment on your proposal, and see if we may not do a common ticket on Phabricator Noé (talk) 09:48, 15 February 2017 (UTC)
 * The obvious first modules to migrate are the ones which are faux databases already, those supporting language codes/names/families and those supporting categorization spring to mind. This is data which cannot currently be well supported (those modules are just a flat-file database which is extremely limiting).  Beyond that, the sky is really the limit concerning what could be migrated, the limiting factor will probably be UI development more than anything. - TheDaveRoss 12:26, 15 February 2017 (UTC)
 * I suppose you are talking about e.g. Module:languages/data2? The only thing I'm worried about is performances: those lists are basically loaded once per page, which allows a page full of translations to be rendered quite quickly. I'm not sure it will be as fast if we have to query the database for every language (unless we get all the languages at once with a single query). — Dakdada 14:11, 15 February 2017 (UTC)
 * Also, those faux databases are the subject of much contention. Working out Serbian vs Serbo-Croatian and all the other headaches between different Wiktionaries would make the databases a lot more complex, unless you expect the individual Wiktionaries to choose to lose their autonomy. —Μετάknowledge discuss/deeds 17:48, 15 February 2017 (UTC)
 * About moving language codes/names/families to Wikidata... Would we use Wikidata to expand "pt" into "Portuguese", "ojp" into "Old Japanese", etc.? We might have a problem if someone wants to change language names in a way that we don't approve, like if someone decides that "ang" should be "Anglo-Saxon" instead of "Old English". Someone might decide that "en" looks better as "Modern English" rather than "English" for some purpose outside of Wiktionary, which could then affect Wiktionary.
 * That said, Wikidata could have a complete list of languages, including the fact that "sr" means "Serbian", but we could have an internal module blocking the use of "Serbian" in some places like translation tables. --Daniel Carrero (talk) 09:46, 16 February 2017 (UTC)
 * The reason this could be a problem is simple: the language information used on Wiktionaries are not data but metadata, and this metadata needs to be controlled by the respective Wiktionaries (as in controlled vocabulary). In other words, Wikidata is not the right tool for that, and it should be reserved for actual content. — Dakdada 09:55, 16 February 2017 (UTC)
 * The majority of data could probably come directly from Wikidata. For the remaining edge cases (the mentioned custom language codes, Proto-Germanic, contentious denominations like Serbian etc) we should try to see if there's a way to get this data into Wikidata, if that's not possible we could have “overrides” on the Wiktionary side(s) which would take precedence. – Jberkel (talk) 10:46, 16 February 2017 (UTC)

Another comment: apparently Wikidata already contains information (author, publication date, etc.) about some books. The page Q43361 is about Harry Potter and the Philosopher's Stone. Maybe we can use it for quotation purposes. We could type something like and have all the information filled automatically when possible. (the template does not have to be called "auto quote", it was just an example). --Daniel Carrero (talk) 13:13, 16 February 2017 (UTC)

Past February 2017:
 * I need at least one way to link Wikidata items with 1. translations 2. senses 16:21, 20 April 2017 (UTC)


 * Has anything happened on this? - [The] DaveRoss  20:41, 25 May 2017 (UTC)
 * Votes/2017-05/Installing Wikidata Vote started. - [The] DaveRoss  23:09, 25 May 2017 (UTC)

A new Labs Tool to visually explore etymological relationships extracted from the English Wiktionary
Hi all! I have developed a tool to visualize etymologies. Please check it out at tools.wmflabs.org/etytree. My work is funded by an IEG grant. Please leave your feedback here. It will help improve it. As a first release, it's is impressive how well automatic extraction of data works (with some bugs of course...). This is because Etymology Sections are written using well defined standars. I would like to get some feedback about some difficulties I have encountered while extracting data and some ideas I have about new templates. I wrote some notes here. Please add your comment there if you have any. Some additional notes follow:
 * 1) I could not use trees as in the nicer demo because there are loops between words that cannot be fit in trees (in trees branches don't merge). Loops are conflicting etymologies. Many are due to simple inconsistencies that users can easily fix, others are real conflicting etymologies and should be represented using multiple trees. I will work on this in a future release.
 * 2) Etymology Sections rarely link to words and their sense/pos, generally only link to the lemma. This is a problem for homographs, cause they generally have different etymological trees which get mixed up in this current implementation. See for example the discussion on the Etymology Scriptorium. It would be nice to have more precise links in etymology sections that link to the correct word.
 * 3) I am not plotting all derived words as of now to clean up a bit the visualization.

Looking forward to your comments! Epantaleo (talk) 18:22, 15 February 2017 (UTC)

The use of also
It seems PapiDimmi and I disagree on how to use also. Please advise on this reversion:. Relevant policy should be here: Entry_layout. Equinox ◑ 19:31, 15 February 2017 (UTC)


 * I wouldn't say "identical letters" but rather "similar letters" as for example oe, œ, ö and ø, are different letters but can have an like övrig an øvrig. However, after reading WT:EL I wouldn't think that chauffeur and choffer belong on Chauffer. chaufer, chaffer, caûffer and maybe many more entries are also "similar" to Chauffer. And then there's the matter of interpreting "similar". Similar in spelling, in pronunciation? As wiktionary covers many languages, fish and Fisch would be similar too, both by spelling and pronunciation. And wouldn't fis, visch, ish, gish, tish and wish be similar to fish too? With all those "similar" entries the also would become too long and the "similar" would be too vague, too open for interpretation. -Slœtel (talk) 10:32, 16 February 2017 (UTC)


 * I think you were right in removing those links. is for typographical similarity. For language-dependent similarity, there is the See also section of entries. — Ungoliant (falai) 11:44, 16 February 2017 (UTC)


 * That template is supposed to show words that are identical except for capitalization; also for words that are identical except for diacritics. Originally there were two reasons why was needed. First, if you typed, for example, mike, and then later typed Mike, you would go to mike and would not be able to access Mike. I think this problem has since been fixed. Second, since most people with an English keyboard have trouble typing diacritics, we also put forms with diacritics (cafe, café).  is definitely not for variant spellings, synonyms, translations, or the like. However, in the case of other alphabets and scripts, it might be useful to include words spelled with a letter that many English-speakers confuse with another letter (B, ß). —Stephen (Talk) 12:14, 16 February 2017 (UTC)
 * I agree with Equinox, Ungoliant, and Stephen that also is for word that differ only in capitalization (mike/Mike), diacritics (sake/saké), punctuation (wont/won't), spacing (everyday/every day), and the like. I'm undecided on visually similar characters from different scripts, though (e.g. to/το; hug/հաց). —Aɴɢʀ (talk) 12:39, 16 February 2017 (UTC)
 * Agreed with all the above. Another instance in which might be useful is for confusion between the long s and lowercase f (e.g.  could have a link to, if the latter is attested as ſilly). Andrew Sheedy (talk) 12:44, 16 February 2017 (UTC)
 * In these cases it might be useful to actually use the long s in the template, so readers can recognize it. Andrew Sheedy (talk) 12:46, 16 February 2017 (UTC)
 * I have long thought that the basic use of above the first L2 was most useful to help someone get to an entry with different diacritics or capitalization in a different language. That is the reason for its placement above the first L2. IOW, having terms that are merely alternative forms in the same language would be not just redundant but wrong IMO. Whether this would be something worth the effort of correction is a separate matter. DCDuring TALK  13:56, 16 February 2017 (UTC)

Review of initial updates on Wikimedia movement strategy process
''Note: Apologies for cross-posting and sending in English. Message is available for translation on Meta-Wiki.''

The Wikimedia movement is beginning a movement-wide strategy discussion, a process which will run throughout 2017. For 15 years, Wikimedians have worked together to build the largest free knowledge resource in human history. During this time, we've grown from a small group of editors to a diverse network of editors, developers, affiliates, readers, donors, and partners. Today, we are more than a group of websites. We are a movement rooted in values and a powerful vision: all knowledge for all people. As a movement, we have an opportunity to decide where we go from here.

This movement strategy discussion will focus on the future of our movement: where we want to go together, and what we want to achieve. We hope to design an inclusive process that makes space for everyone: editors, community leaders, affiliates, developers, readers, donors, technology platforms, institutional partners, and people we have yet to reach. There will be multiple ways to participate including on-wiki, in private spaces, and in-person meetings. You are warmly invited to join and make your voice heard.

The immediate goal is to have a strategic direction by Wikimania 2017 to help frame a discussion on how we work together toward that strategic direction.

Regular updates are being sent to the Wikimedia-l mailing list, and posted on Meta-Wiki. Beginning with this message, monthly reviews of these updates will be sent to this page as well. Sign up to receive future announcements and monthly highlights of strategy updates on your user talk page.

Here is a review of the updates that have been sent so far:


 * Update 1 on Wikimedia movement strategy process (15 December 2016)
 * Introduction to process and information about budget spending resolution to support it
 * Update 2 on Wikimedia movement strategy process (23 December 2016)
 * Start of search for Lead Architect for movement strategy process
 * Update 3 on Wikimedia movement strategy process (8 January 2017)
 * Plans for strategy sessions at upcoming Wikimedia Conference 2017
 * Update 4 on Wikimedia movement strategy process (11 January 2017)
 * Introduction of williamsworks
 * Update 5 on Wikimedia movement strategy process (2 February 2017)
 * The core movement strategy team, team tracks being developed, introduction of the Community Process Steering Committee, discussions at WikiIndaba conference 2017 and the Wikimedia movement affiliates executive directors gathering in Switzerland
 * Update 6 on Wikimedia movement strategy process (10 February 2017)
 * Tracks A & B process prototypes and providing feedback, updates on development of all four Tracks

More information about the movement strategy is available on the Meta-Wiki 2017 Wikimedia movement strategy portal.

Posted by MediaWiki message delivery on behalf of the Wikimedia Foundation, 20:30, 15 February 2017 (UTC) • m:Special:MyLanguage/Strategy/Wikimedia movement/2017/Updates/Initial announcements review • Get help

Titles of pronunciation modules
I recently created the Czech pronunciation module Module:cs-pronunciation. said on the talk page that it should be moved to Module:cs-IPA, and I asked why, because -IPA is currently not the most commonly used form for the module names in Category:Pronunciation modules. (23 modules use -pron, 11 use -IPA, 6 use -pronunciation, and 4 use -pronunc.) said that it was decided in a discussion that -IPA should be used, but did not link to the discussion. I couldn't find the discussion by searching. Does anyone know where that discussion was? — Eru·tuon 20:21, 17 February 2017 (UTC)
 * FWIW, I would prefer them all under  as well. — JohnC5 20:30, 17 February 2017 (UTC)
 * Aside from what name all modules should have, if the template's name is, it only makes sense to use the same name for the module. —CodeCat 20:32, 17 February 2017 (UTC)
 * I don't remember where the discussion was but I also don't see why the module can't be renamed. --Anatoli T. (обсудить/вклад) 22:48, 17 February 2017 (UTC)
 * Not all pronunciation modules are pure IPA modules. Wyang (talk) 02:32, 18 February 2017 (UTC)
 * I think they should be standardized under '-pron' because that's the most common case currently, it's easier to type, and for Wyang's reason. Benwing2 (talk) 02:34, 18 February 2017 (UTC)
 * Which pronunciation modules are not pure IPA? I have not encountered that yet. — Eru·tuon 20:09, 18 February 2017 (UTC)
 * Module:my-pron, for example, also generates a phonetic respelling and romanizations in addition to IPA. —Aɴɢʀ (talk) 20:26, 18 February 2017 (UTC)

I guess I personally prefer -pronunciation, because it does not require editors to know what IPA stands for, and because is an ambiguous abbreviation that can mean either  or. — Eru·tuon 20:09, 18 February 2017 (UTC)
 * I recall a discussion in which it was decided that templates that automatically produce pronunciation info should be named "xx-IPA"; I don't think there was ever a discussion about what the corresponding modules should be called. Since ordinary editors very rarely have to type the names of modules (as they frequently do the names of templates), succinctness is less important for module names than for template names. I agree that "xx-pron" should be avoided as templates named "xx-pron" are usually headword-line templates for pronouns, and it would be confusing for "pron" to stand for something else in modules. Other than that, I don't really mind what the module is called as long as the template is called "xx-IPA" as expected, since the module name only has to be written once, namely in the template that invokes the module. —Aɴɢʀ (talk) 20:24, 18 February 2017 (UTC)

Tsonga Dictionary
Via OTRS someone has released a digitized Tsonga-English/English-Tsonga dictionary under CC-BY-SA. I currently have the Tsonga-English portion as a 438 page Word document, and I would like to figure out what the best thing to do with the information is. I am happy to do some work in terms of reformatting and uploading the content, however I do not speak Tsonga nor am I familiar with Bantu or any South-African languages. I gather that the language is a bit politically complicated as well. We have very little coverage of this language as it stands.

Are there any contributors who would feel comfortable helping to review the dictionary and help with getting the data ported into Wiktionary? One possible starting point is that I can reformat and upload all of the content to a collection of subpages, which could then be moved into the main namespace by contributors who feel comfortable verifying. I could also provide the Word document to interested people who could advise on the content and whether it is trustworthy enough to just upload in its entirety. - TheDaveRoss 21:00, 17 February 2017 (UTC)
 * How many words is it? What kind of information is present- definitions, pronunciations, inflections? DTLHS (talk) 21:01, 17 February 2017 (UTC)
 * I am not sure how many defined words or definitions there are exactly, there are 220,564 words in the document so I would guess that the number of defined terms is on the order of 20,000. (edit: there are 27,451 line breaks, so nearly that many terms.)
 * Here is a sample of the information provided:
 * abuxeni, (salutation) see avuxeni.
 * abvana 5, see bvana, maabvana.
 * accelerando 9, accelerando (music).
 * adajiyo 9, adagio (music).
 * adirese 9, (Eng.) address of a letter, postal address.
 * -adiresela, address a letter.
 * afirikati 9, (Eng.) affricate.
 * Afrika 1, the continent of Africa; -- -Dzonga, South Africa; -- wa Dzonga, Southern Africa.
 * agenda 9, (Eng.) agenda.
 * Agoste, (Eng.) August.
 * ahanti, ahati, interj. to express uncertainty, dubiousness, indecision, perhaps, weak refusal: May be! I don't know I am sure.
 * ahe, ahee, interj. as an answer to a greeting, also to express thanks, approval, assent; interj. to children and boys among themselves.
 * ahee, ahehe, interj. used as a warning against danger: mind, look out, beware; here you are, there it is.
 * -ahlama, be wide open; open one's mouth, gape.
 * -ahlamela, stay with open mouth. (Idiom) -- munhu, to pick a quarrel with a person.
 * -ahlamisa, cause mouth to open, make gape; hold open, as a sack. (Idiom) -- nomu, to gasp, to speak; -- tinhlaya, to feel drowsy.
 * -ahlamula, yawn, open the mouth.
 * -ahluka, detach, separate itself from.
 * -ahlukana, become divided, separated, as husband and wife by divorce.
 * -ahlukanya, detach, separate.
 * -ahlula, judge, adjudicate, try a case.
 * ajenda 9, (Eng.) agenda.
 * -aka, 1 build, construct, edify, elevate, erect, as a hut; -- tinghalava, shipbuild-ing. 2  dwell, inhabit. 3 be ingrained, e.g. mukhuva wu ta aka, the habit is ingrained. (Idiom) -- vuxaka, to foster friendship; -- mhaka, to prepare a court case; -- tiko, to develop a country; -- muti, to stay peacefully together, as husband and wife.
 * -akana, help each other to attain a high status; mould each other's character.
 * -akademiki, (Eng., adj.) academic.
 * The portion of the document that I have does not indicate what all of the annotation means, so there may be more information there (such as the 1s and 9s) than I can interpret. - TheDaveRoss 21:10, 17 February 2017 (UTC)


 * I'd guess the numbers are noun classes. Equinox ◑ 21:13, 17 February 2017 (UTC)
 * I'd recommend looking at Zulu and Swahili entries. Zulu and Swahili are the best developed Bantu languages on Wiktionary. In Zulu, verbs and adjectives are placed at the form without the hyphen but the headword is displayed with the hyphen, as are links. See for an example. As for the noun classes, comparing it to the list at Tsonga language, it appears that your list does not actually give forms that include the noun prefixes. Perhaps there are different allomorphs of the prefixes, so that for example class 9 nouns have no prefix if the noun begins with a-, but I don't know. I may look into it more. Feel free to ask me for any help as I'm already experienced in handling Zulu. —CodeCat 21:23, 17 February 2017 (UTC)
 * I've created a basic entry at, using which I just created. You can use this as a basic template at least for the verb entries. We probably want a reference template for this dictionary, but I don't know what it's called, who made it or where it's located. —CodeCat 21:31, 17 February 2017 (UTC)


 * It will have to be done partially by hand, and that will require a lot of effort. Right now, the key is to make this publicly available by posting it on Wikisource. —Μετάknowledge discuss/deeds 21:32, 17 February 2017 (UTC)
 * I'll try and get it up on Wikisource once I have the remainder (especially since there may be more meta information available in the other part. - TheDaveRoss 21:50, 17 February 2017 (UTC)
 * For what it's worth, AWB would probably help in creating them as well. Ping me if you'd like me to assist. —Justin ( koavf ) ❤T☮C☺M☯ 03:06, 18 February 2017 (UTC)

Rollback
Is rollback available to user who are non-administrators? Pkbwcgs (talk) 19:03, 18 February 2017 (UTC)
 * It is possible to create a rollbackers group but that doesn't exist here. If you want, w:WP:TWINKLE has a rollback feature. —Justin ( koavf ) ❤T☮C☺M☯ 19:13, 18 February 2017 (UTC)
 * I would like to use Twinkle but it is not available on Wiktionary. I tried to create the gadget on my own but it didn't work. See User:Pkbwcgs/Twinkle.js and User:Pkbwcgs/Twinkle warn.js. Pkbwcgs (talk) 19:17, 18 February 2017 (UTC)
 * Looks like we both learned something today (me twice!). At Special:ListGroupRights, you can see that we do in fact have a rollbacker group. —Justin ( koavf ) ❤T☮C☺M☯ 03:03, 19 February 2017 (UTC)
 * Not having ever had the ability to use all the sysop buttons, I've often wondered: what the difference is between using rollback and simply undoing an edit? Andrew Sheedy (talk) 03:13, 19 February 2017 (UTC)
 * Rollback is instantaneous. No opportunity to give an edit summary, just undo the edit as fast as possible. —suzukaze (t・c) 03:14, 19 February 2017 (UTC)
 * Ah, OK, thanks. Andrew Sheedy (talk) 03:17, 19 February 2017 (UTC)
 * Rollback also marks the reverted edit as patrolled, so those of us who go through Recent changes with patrolled edits filtered out won't see them anymore. That's the main reason to be careful about who gets rollback: we lose the ability to correct mistaken reverts or to look for larger problems/patterns that the rollbacker might have missed in fixing the obvious immediate problem. Chuck Entz (talk) 04:43, 19 February 2017 (UTC)

Placement of synonyms
In, synonyms were moved away from their Synonyms section, to definitions. I don't like this and it does not match our long-standing practice. Can someone please point me to a relevant discussion? User:CodeCat may like to comment. --Dan Polansky (talk) 19:31, 18 February 2017 (UTC)
 * I think putting synonyms and antonyms under their respective senses is a good idea, especially for words with lots of meanings, like head. Keeping them in separate Synonyms and Antonyms sections makes them harder to find, makes it harder to find which sense they belong to, and makes it easier for them to be assigned to senses that are not actually listed. Separate sections are fine for words that have only one or two meanings, but for words like head with a very large number of senses and subsenses and subsubsenses, I think the new arrangement is an improvement. —Aɴɢʀ (talk) 19:57, 18 February 2017 (UTC)
 * I have noticed these edits too. I think they are structurally an improvement, but they do suck up vertical space in a way that is awkward for somebody who just wants to read the list of definitions. I suppose I would like a way to hide *nyms until explicitly choosing to view them. Equinox ◑ 20:00, 18 February 2017 (UTC)
 * Especially in long entries, it is important to be able to skim definitions, which these changes make harder to do. Incidentally, the head entry is now a horrible mess of inept subsensing; there is e.g. "Mental or emotional aptitude or skill" as a subsense of "The part of the body of an animal or human which contains the brain, mouth and main sense organs". --Dan Polansky (talk) 20:02, 18 February 2017 (UTC)
 * I'm not fond of subsensing either, but that's an entirely separate issue that has nothing whatever to do with the placement of synonyms. —Aɴɢʀ (talk) 20:27, 18 February 2017 (UTC)
 * What happened to the idea that was proposed last time we discussed this, to put them in a collapsible form, identical in structure to quotations? I oppose placing them under definitions, unless they display in that format, in which case I fully support having them under definitions. Maybe we should put it to a vote so it gets more attention? Most people seemed to be in favour of putting them directly under definition lines, IIRC. Andrew Sheedy (talk) 03:16, 19 February 2017 (UTC)
 * FWIW Simple English Wiktionary has collapsible functionality: simple:language. —suzukaze (t・c) 03:59, 19 February 2017 (UTC)
 * Interesting. That's exactly what I'd like to see here. Andrew Sheedy (talk) 04:01, 19 February 2017 (UTC)
 * I'm very fond of the format of the Simple English Wiktionary as well. Wyang (talk) 05:12, 19 February 2017 (UTC)

onelook.com problems
While onelook.com is a useful site, obvious errors never seem to be fixed, and the people operating it seem entirely unresponsive to any reports submitted through their "Contact us" page. There has for a long time been a glaring problem with Wiktionary links whereby most words incorrectly link to a capitalised form that usually does not exist. Does anyone have any inside track for contacting them about this? Or perhaps someone else will have more luck with their "Contact us" than I ever have. Mihia (talk) 21:49, 20 February 2017 (UTC)
 * I think the problem should be apparent from the lack of advertising. I haven't been able to get them to correct glaring problems like deadlinks to dictionaries (no just entries in dictionaries). We could produce our own equivalent I suppose. We might even be able to include some sources that they don't/can't. DCDuring TALK 00:45, 21 February 2017 (UTC)


 * I don't understand what you're saying about capitalised links. A lot of useless spam-scum sites either steal material from Wiktionary, or link to us, in an attempt to get credibility, hits, or whatever those sick fucks want. I think they are just best ignored. Equinox ◑ 00:46, 21 February 2017 (UTC)
 * I think calling onelook spam scum is unwarranted. DTLHS (talk) 01:56, 21 February 2017 (UTC)
 * Quite disgraceful. Mihia (talk) 03:06, 21 February 2017 (UTC)


 * I reached out to OneLook initially back in 2006 to suggest they include Wiktionary, and Connel and I worked on generating the list of English terms which they originally used to link back to Wiktionary. Those lists are long since defunct, and I am not sure what they are now using.  I still have the email address of the creator of the site, and the guy who was maintaining it, I can try and reach out to them.  A lot happens in 10 years, so they may not even be involved any longer.  Re motives, at least at that point they were all about making data more open, so I wouldn't put them in the same class of folks as those who mirror content to try and generate ad revenue. - TheDaveRoss 13:49, 21 February 2017 (UTC)
 * I would expect that it would be better maintained if they did generate more ad revenue. I use the site often to see and show what the lemmings are doing, often adding it as an External link to our entry. DCDuring TALK 15:49, 21 February 2017 (UTC)
 * While OneLook indeed curiously links to Rubber from its rubber, it is generally a great service that I use when I want to compare the content of multiple dictionaries, which I do quite often. OneLook's Alexa rank (~ 24,000) is worse than Wiktionary's (626), so they don't really appear to steal readers from Wiktionary. I am glad the service exists. --Dan Polansky (talk) 11:28, 25 February 2017 (UTC)

Category:Antonomasias by language
I see we have Category:Genericized trademarks by language, which are basically a special case of antonomasia. Could we have the more general category, to treat cases like and, which come from  and  respectively? --Barytonesis (talk) 08:39, 21 February 2017 (UTC)
 * That seems like a very unknown jargony term. I'd rather stick with something that people do understand. —CodeCat 14:23, 22 February 2017 (UTC)

New game
Hey. Anyone fancy another game of Wiktionary Scrabble? The last one was pretty epic, with a pretty table from Kenny. Granted, there was shambolic organisation, spelling, rulekeeping and general gamesmastership from WF. But it's been a while since we played a game, and Equinox is getting bored. I started a page at Random Competition 2017. --Quadcont (talk) 18:45, 21 February 2017 (UTC)
 * I've been hoping for a game! Now I just need to figure out which language I can start learning that exploits these rules most effectively... —Μετάknowledge discuss/deeds 00:54, 22 February 2017 (UTC)
 * I don't think there's been a game since I started editing here! Hopefully my weekend isn't too busy for me to play. (Oh good, this isn't scheduled to start till the end of the month) Andrew Sheedy (talk) 03:35, 22 February 2017 (UTC)
 * We start the game on Feb 24th, with the seven letters EWERSOA. --Quadcont (talk) 18:26, 22 February 2017 (UTC)
 * Does the first word have to go through the center square? DTLHS (talk) 18:31, 22 February 2017 (UTC)
 * Yes, it does. --Quadcont (talk) 18:43, 22 February 2017 (UTC)


 * I hope Scrabble doesn't sue us. --Quadcont (talk) 18:44, 22 February 2017 (UTC)
 * I hope they do, because then they might merge Wiktionary with the Scrabble dictionary and it wouldn't be missing so many common words. - TheDaveRoss 19:07, 22 February 2017 (UTC)
 * We need more competitors! People who like this kind of thing:, please join in! —Μετάknowledge discuss/deeds 19:34, 25 February 2017 (UTC)
 * Already in (and thinking about next Christmas). SemperBlotto (talk) 20:00, 25 February 2017 (UTC)
 * ("sasowere" seems to be a Japanesy-type word (transliteration) but I can't quite get my head around it.) SemperBlotto (talk) 20:02, 25 February 2017 (UTC)
 * You won't get anywhere with that, as /we/ is a disallowed syllable in modern Japanese. —Μετάknowledge discuss/deeds 23:58, 25 February 2017 (UTC)
 * I'll join in! I was away, so I couldn't join in till today. Andrew Sheedy (talk) 17:22, 26 February 2017 (UTC)

Another UtherPendrogn clone?
. Can someone check? —CodeCat 19:25, 26 February 2017 (UTC)

THE ONLY VANDAL IS YOU! STOP! Llacheu (talk) 19:36, 26 February 2017 (UTC)
 * Thank you for the confirmation... :) As long as you act like Uther, you can't fool anyone. You only lasted as long as you did with the last sock because you managed to behave- up until you finally blew it. I pretty much knew it was you from the beginning, but decided to let you have a go at turning over a new leaf- which you almost did. I, for one, appreciate your trying. Chuck Entz (talk) 22:57, 26 February 2017 (UTC)


 * Late to the party, but confirmed. - TheDaveRoss 15:49, 27 February 2017 (UTC)

Syntax of Template:label/example
You left a change message saying "odd syntax, but makes the code much shorter". I don't think making for shorter Lua code is necessarily a good reason for using an odd syntax. I'm not opposed to using en: to specify a language code instead of a separate parameter, because it may be easier to read, but having the examples separated by a semicolon instead of as separate params seems a bit dubious. In reality I don't think it would actually make the Lua code longer to iterate over separate arguments instead of splitting on semicolon. Benwing2 (talk) 19:57, 26 February 2017 (UTC)
 * I used the odd syntax because I am not sure how else to make the template work when there are multiple examples in a row. How would the module determine where one example ends and another begins? It's not shorter Lua code; rather, shorter template code. — Eru·tuon 20:03, 26 February 2017 (UTC)
 * Can you explain further? It seems to me you could do various things:

etc.

Using a blank argument to separate sets of template arguments is fairly standard. In this case you'd probably have to strip whitespace off of the arguments, since this isn't done by default for unnamed args. Benwing2 (talk) 20:14, 26 February 2017 (UTC)
 * Well, okay, I guess there are a number of ways to do it. Actually, I came up with the method of indicating the parameters with punctuation at some point, and simply never changed my ways. I would change my method if there is a more generalizable method, one that can be made into a general template examples module. — Eru·tuon 20:40, 26 February 2017 (UTC)
 * Question: Why not keep the old template-based syntax? Adding an entire new function and worrying about its syntax seems like overkill for a static demonstration of the template. —suzukaze (t・c) 00:36, 27 February 2017 (UTC)
 * Because I was adding a few more examples and I don't like typing out and then  with the same parameters, and adding the arrow between. It seems simpler and more reliable to have a module do it (though I admit the function takes quite a while to write). — Eru·tuon 01:10, 27 February 2017 (UTC)

User:Cl adding incorrect Dutch pronunciations
This user has been butting heads with me and User:Lingo Bingo Dingo over their edits to Dutch pronunciations. They've introduced nonphonemic features to phonemic transcriptions, as in, , , ,. When pointed out, they removed the phonemic transcription altogether, and replaced it with a phonetic one without specifying which varieties of Dutch it applies to:, , , (these pronunciations are not universal). They also removed one of the two possible pronunciations in. When reverted, they start an edit war. Can this please be sorted? —CodeCat 19:13, 27 February 2017 (UTC)
 * I refer to reliable sources when making changes, yours are none. Cl (talk) 19:15, 27 February 2017 (UTC)
 * A dictionary is not a valid source, I've already pointed this out to you. —CodeCat 19:16, 27 February 2017 (UTC)
 * A dictionary can't but be a reliable source. Whatever you say isn't (unless supported by a reliable source). Cl (talk) 19:23, 27 February 2017 (UTC)
 * What is "reliable"? And no, a dictionary is not a valid source for Wiktionary. Have you read WT:WFW yet? —CodeCat 19:24, 27 February 2017 (UTC)
 * WHere exactly is stipulated dictionaries aren't a valid source for Wiktionary? This text, e.g., implies they are: []. It'd be highly illogical if they shouldn't be accepted as reliable sources. Cl (talk) 19:30, 27 February 2017 (UTC)
 * Those are references, not sources. There's a big difference. Also, it's common sense: a dictionary can't use another dictionary as a source. And again, what is a "reliable" source, and what makes your dictionary one? —CodeCat 19:33, 27 February 2017 (UTC)
 * Wiktionary isn't a dictionary by itself, but a secondary source. And of course existing dictionaries should be checked as a reliable source, should some question arise. How come can't a dictionary be a reliable source {of a phonetic transcription in this case) if Wiktionary itself demonstrates they are? Cl (talk) 19:45, 27 February 2017 (UTC)
 * Wiktionary is a dictionary and says so in its logo. Professionally published dictionaries are not the work of gods and we do not trust them unquestioningly. Equinox ◑ 19:51, 27 February 2017 (UTC)
 * You should have the official Wikipedia reliable source policy changed, if you adhere to a different point of view. I don't think you're going to succeed, though, because that'd result in chaos. Cl (talk) 19:54, 27 February 2017 (UTC)
 * Wiktionary does not follow Wikipedia policies, so their definition of "reliable source" is completely irrelevant to us and always has been. —CodeCat 19:56, 27 February 2017 (UTC)
 * One still needs some kind of verification, and yours has been null so far. Cl (talk) 19:58, 27 February 2017 (UTC)
 * Well then, what's the procedure for verifying IPA? —CodeCat 20:03, 27 February 2017 (UTC)
 * Pronunciation dictionaries and works on the Dutch phonology, in my opinion. Should you have a better idea, I'm ready to listen. Cl (talk) 20:07, 27 February 2017 (UTC)
 * It was a question asked to everyone else too. I genuinely don't know, so some other experienced editors should clarify hopefully. —CodeCat 20:10, 27 February 2017 (UTC)

When it comes to pronunciation info, clearly our usual criteria of use in a permanently archived source by multiple independent authors spanning the course of more than a year cannot apply. When I add pronunciation info for English, German, and Burmese, I always rely on information from other dictionaries; when I add it for Irish, I rely on published phonetic descriptions of specific dialects combined with my own knowledge of the phonology of the language. For Welsh and Lower Sorbian, I rely on the spelling-to-pronunciation rules described in dictionaries as well as as phonological descriptions of the language. And for English and German, still I often come into conflict with other users who have different opinions on how best to transcribe these languages, even if we don't actually disagree on how a given word is pronounced. In general, phonemic transcriptions are more important here than narrow phonetic ones, but there's no reason not to include both. Is there any objection to ? —Aɴɢʀ (talk) 20:40, 27 February 2017 (UTC)
 * @Cl: With a language that has so much variation, it requires some judgement to apply the information in references to the language in general. A phonological study is a single data point which may not represent larger patterns, unless it's specifically designed to do so. Conventions for representing a particular phoneme may be different in different sources. You have to know your "tools", and use the correct one for a particular task. Simply copying transcriptions wholesale- even from excellent references- without considering such factors would be like using a scalpel to clear brush and a chainsaw to perform brain surgery- they may be the best of their kind available, but your results won't be very good. Chuck Entz (talk) 03:55, 28 February 2017 (UTC)
 * Pardon, but when I'm presented with two possible transcriptions and the argument for one is we should use mine because I consider it correct and here's some works that back me up and the other is we should use mine because I say so, I don't care what we define as reliable source, one side clearly has more merit than the other. I'm sure CodeCat's transcriptions can be sourced too – but then now would be the time for CodeCat to do so, preferably with a source on top which shows why the other system should not be applied by us. I consider it a bit rude to imply that Cl has a lack of judgement and blindly copies things he doesn't understand because...really, there's no reason to. We have to refer to some source when making judgements about languages and phonological works are the obvious choice. Korn &#91;kʰũːɘ̃n&#93; (talk) 12:37, 28 February 2017 (UTC)
 * "when I'm presented with two possible transcriptions and the argument for" "the other is we should use mine because I say so"
 * Can you cite where that argument is made? The contention is in fact mostly about replacing phonemic transcriptions with phonetic transcriptions, e.g., . Lingo Bingo Dingo (talk) 14:04, 28 February 2017 (UTC)
 * That, and adding nonphonemic detail to phonemic transcriptions, as I noted in the first set of diffs I provided. There are no, , phonemes in Dutch.  is one possible realisation of the combination , but this is entirely allophonic and applies across word boundaries as well. The glides , ,  are automatically inserted between high vowels and a following vowel to break up a hiatus. —CodeCat 14:21, 28 February 2017 (UTC)
 * That argument is made by the action of reverting cited pronunciations to uncited ones and then demanding the change to cited pronunciations stop, effectively demanding to take good faith over professional publications. That's simply not good process and we shouldn't let such lapses creep in. At least my understanding of the proper Wiki process is: "Good faith until challenged, then sources." Consensus doesn't cut it, with consensus you can introduce and maintain any nonsensical junk if you have a big enough clique behind you. (Which, to be clear, nobody is trying to do here.) What if one day a bunch of POV‐pushing speakers of a remote language come and nobody can check on them and the only sane person gets shot down because his sources count nothing against their claims of phonetic irrelevancy? I simply don't want this to be a precedent for a loophole for wrong data. Korn &#91;kʰũːɘ̃n&#93; (talk) 18:02, 28 February 2017 (UTC)
 * I'm definitely okay with it if Cl adds phonetic transcriptions (or even indicates variants in phonemic transcriptions using parentheses or multiple transcriptions), while preferably also indicating what lect is being described. As for a reference for a voiced pronunciation of  in veertig, see the footnote here. Lingo Bingo Dingo (talk) 14:04, 28 February 2017 (UTC)
 * In case that link goes dead:
 * Another example is the present-day tendency in standard Dutch to voice initial fricatives /f/ and /s/ in the numerals veertig 'fourty', vijftig 'fifty', zestig 'sixty' and zeventig 'seventy'.
 * From Auer, P. & Hinskens, F., "The convergence and divergence of dialects in Europe. New and not so new developments in an old area", 1996, page 9, also see page 10. Lingo Bingo Dingo (talk) 14:16, 28 February 2017 (UTC)
 * I feel that both phonemic and phonetic presentation not only can but should be included, both are valuable information (even if the second should theoretically be derivable from the first). Crom daba (talk) 23:18, 28 February 2017 (UTC)

Changing the text of Template:bor from "borrowing" to "borrowed"
This change has come up before, and I'd like to start implementing it. However, the template is sometimes used as part of a larger sentence, such as "a ..." and if we change the phrasing here, it will break the sentence. Is there a way we could fix these cases easily? —CodeCat 14:31, 28 February 2017 (UTC)
 * As we discussed in WT:RFM and User talk:Daniel Carrero/2016, I'd prefer doing this: I agree with your RFM comment when you said "We could get rid of the text from the template altogether (emphasis mine), and add it manually to the entries instead. Then it would work like and, which don't include text either." I consider it an annoying inconsistency that  currently returns "borrowing from" while  and  do not return "inherited from" and "derived from", respectively. In fact, I've seen many entries, with code like "  and  ", which apparently means "borrowed from English and French" but the second template was  (or ), likely to avoid showing up "borrowing from" twice or the need to use "notext=1".
 * The table below is what I'd expect to do. Also, I don't think that all instances of "borrowing" or "borrowed" need to link to the glossary. That's why I removed the glossary link from the "proposed" row.


 * {| class="wikitable"

! ! Syntax ! Result ! Current ! Proposed
 * Borrowing from Italian pizza.
 * Borrowing from Italian pizza.
 * Borrowed from.
 * Borrowed from Italian pizza.
 * }
 * --Daniel Carrero (talk) 14:48, 28 February 2017 (UTC)
 * That works too. Perhaps a poll is in order. Also, I think if we decide to change it, we could rename the old template to or similar, and gradually replace the old one with the new. —CodeCat 14:53, 28 February 2017 (UTC)

Option 1: Leave as is

 * 1)  and change the other templates to be more descriptive. DTLHS (talk) 23:49, 1 March 2017 (UTC)
 * Are you suggesting that is more descriptive than other, similar templates? If so, in what way? Andrew Sheedy (talk) 00:24, 9 March 2017 (UTC)

Option 3: Remove text altogether

 * 1) [[Image:Symbol support vote.svg|20px|link=]] Support per my reasons above. --Daniel Carrero (talk) 15:00, 28 February 2017 (UTC)
 * 2) [[Image:Symbol support vote.svg|20px|link=]] Support. Let's remove this inconsistency. --Barytonesis (talk) 15:00, 1 March 2017 (UTC)
 * By the way, it would be nice if the same were done for as well. --Barytonesis (talk) 15:03, 1 March 2017 (UTC)
 * 1) . The inconsistency has always bugged me, and there are times I've wanted to put it in the middle of a sentence, which doesn't work because "borrowing" is currently capitalized by the template. Andrew Sheedy (talk) 23:30, 1 March 2017 (UTC)
 * Just as an aside, many such templates (including ) have a nocap parameter to force the first letter to lower-case. ‑‑ Eiríkr Útlendi │Tala við mig 00:23, 2 March 2017 (UTC)
 * Good to know, thanks. Andrew Sheedy (talk) 00:21, 9 March 2017 (UTC)
 * , but this is not particularly simple., how do you expect to be able to find and fix all the places where editors have structured a sentence in the etymology section based on the text it currently outputs? We may end up with a lot of ungrammatical etymologies if they are not checked. —Μετάknowledge discuss/deeds 07:36, 9 March 2017 (UTC)
 * Alright. Well, we already have a lot of ungrammatical etymologies. If the whole etymology consists of the template, then it currently is "Borrowing from ." when it could have been either "A borrowing from  ." or "Borrowed from  ."
 * I created these template tracking categories: Category:bor without notext, Category:bor with notext, Category:bor with nocap.
 * I suggest fixing all entries and adding "notext" in all uses of where the entries are fixed... Some may be done automatically, including the hundreds or thousands of etymologies that consist simply of that template and nothing else.
 * The category "bor with nocap" is an interesting case, since it might contain "A ", meaning "A borrowing".
 * Many etymologies in, say, Ido and Esperanto contain multiple uses of bor with "notext". It goes without saying that all uses of bor with "notext" are already good and don't need to be changed.
 * When all entries are fixed and have "notext", then we could remove the "Borrowing" text from the template by default and then remove the "notext" from all entries. I realize it's a two-step process. If someone comes up with a one-step plan, I'm interested and open to hear it. Gotta give credit where it's due: this two-step plan was proposed by CodeCat in User talk:Daniel Carrero/2016. --Daniel Carrero (talk) 11:24, 9 March 2017 (UTC)
 * To provide the same text, perhaps invocations of without 1 should be replaced with  or  . — Eru·tuon 20:40, 9 March 2017 (UTC)
 * : 9/10 times I'm adding adding notext anyway. --Victar (talk) 04:23, 27 March 2017 (UTC)

Letter definitions and phonemes
I noticed that does not include any parameters to specify what phoneme a letter typically represents. Is this by design? If so, why? Also, should we have entries for digraphs and trigraphs and the phonemes they represent? —CodeCat 19:31, 28 February 2017 (UTC)