User talk:Adam78/archive1

Szia, csak kérdezném, hogy szerinted biztos egybe kell vonni ezeknek a fordítási táblázatát? Merthogy az egyik tudományos megnevezés, a másik meg köznyelvi, más nyelvekben is létezhet rá két külön szó. Pl. a sodium chloride se irányít át a table salt fordításához. De ha már összevonjuk, akkor a dizygotic fordítása miért nem mutat a fraternal szócikkre? Nem kritizálni akarom, csak egy vélemény.--84.236.117.96 17:42, 19 January 2019 (UTC)Reply

Szia, itt válaszolok, mert nem jelentkeztél be, az IP-címedhez tartozó vitalap meg nem feltétlen mindig hozzád fog tartozni. (Javaslom, hogy jelentkezz be, akár vadonatúj felhasználónévvel, ha kvázi anonim akarsz maradni.) Meg amúgy is egyszerűbb így talán a válaszadás.

A fraternal kapcsán teljesen igazad van, ez elkerülte a figyelmem. A tárgyban említett két megnevezés viszont egyaránt tudományos alapvetően, és kétlem, hogy érdemes lenne róluk két külön szócikket alkotni. Az egyik nyelvben tudományosabb szóval illetik, a másikban köznapibb szóval, de attól még a referencia azonos, az extenzió azonos; legfeljebb az intenzió, a nézőpont különbözik, de az is csak némelyik nyelvben (és sok esetben nem is válik ketté, vagy nem következetesen). Adam78 (talk) 19:17, 19 January 2019 (UTC)Reply

Nem egészen így értettem. A fordítási táblázat az, ami következetlenül van elhelyezve: identical / dizygotic. Az identical twins amúgy nem tudományosabb kifejezés angolul, mint magyarul az egypetéjű ikrek, vagy angolul a table salt, egyszerűen így mondják hétköznapi beszédben, azok is, akiknek fogalmuk sincs a reproduktív biológiáról és azt hiszik, hogy a babát a gólya hozza. De ha neked szívügyed, akkor maradjon így, csak fölvetettem megfontolásra. A dinamikus IP engem nem zavar, szerintem téged se zavarjon.--84.236.117.96 21:01, 19 January 2019 (UTC)Reply

It might be misleading to add a declension table to the inflected pronouns. It is fine to add it to the lemma entry (véle) but the form entries themselves should contain only a See also section. Otherwise, the user might interpret it as the declension of véled, etc. Panda10 (talk) 21:23, 27 January 2019 (UTC)Reply

All right, I don't mind. In fact, it wouldn't even make sense to copy the same content to many places, rather than inserting a custom template. Adam78 (talk) 22:12, 27 January 2019 (UTC)Reply

The Hungarian Wiktionary could benefit from using the template {{zh-pron}}. --Lo Ximiendo (talk) 12:02, 29 January 2019 (UTC)Reply

I'm having second thoughts about creating the new prefix entries ugyan-, se-, mind-, minden-, akár-, and bár-. Is this really the correct way? The Hungarian dictionaries don't treat these words as prefixes. All words formed with them are compound words. Would you mind explaining your reasoning? I'd like to make sure we are not making a mistake here. Thanks. Panda10 (talk) 16:23, 30 January 2019 (UTC)Reply

It all depends on the definition of the elements. Among its meanings listed, ugyan did not have anything similar (or related) to 'the same'. Does this word exist in itself with this meaning? Ugyan! :) If there is no independent word with this meaning, then it must be something else. As it only occurs in this meaning in compounds, as a prefix, then it has to be a prefix, I suppose. After all, it wouldn't make sense to link the independent word ugyan from ugyanaz etc. if ugyan does not (and will not) mention this meaning. I don't think this meaning of ’the same’ belongs there, so the link should point to another lexical entry, which can hardly be anything but a prefix. It does have a distinct meaning, doesn't it? Hungarian dictionaries rarely ever list prefixes or affixes anyway, despite the fact that compounding and derivation are quite common properties of Hungarian, and these prefixes and suffixes do have their own meaning, just like verb prefixes like el-, át- or oda- or noun suffixes like -nak and -ből. As far as I'm concerned, I'd need much stronger arguments for ugyan- as a prefix to be treated at the same place as ugyan as an adverb, a conjunction or an interjection. There seems to be no overlap in meaning whatsoever. But if you want, we can look up other Hungarian monolingual dictionaries or grammar books, or even consult MTA Nytud. Int. for confirmation. Adam78 (talk) 16:59, 30 January 2019 (UTC)Reply

It makes sense the way you explained it. But if you wouldn't mind spending time with getting confirmation from any other resources you find appropriate, that would be helpful. I feel that we are breaking new ground here. Not just the fact that we are explaining Hungarian in English (is there another dictionary that does that?) but also the way we are presenting suffixes, inflected words as entries, inflection tables, etymology, etc.

I have two more questions:

1) Should the prefixed pronouns and adverbs be listed in their own category or should they be presented mixed in the current Hungarian words prefixed with ugyan-? For suffixes, the different POS are separated, but for prefixes they are not always.

2) Regarding se-, mind-, minden-, akár-, and bár-: I assume you have the same reasoning as for ugyan-. Right? Would you help me with their meaning? I can handle the rest of the work. Panda10 (talk) 17:24, 30 January 2019 (UTC)Reply

For (1), do you mean separating the pronouns from the adverbs into two different categories? I'm not sure it's necessary.

For (2), yes, I have the same reasoning, except that it would be theoretically possible to link mind and minden from the relevant compounds, since these words do have the same meaning as the prefix (’all’). However, I think it's much clearer to treat them separately, and it's more consistent with the system too. For their meanings, this chart should explain it, I hope: Table of correlatives in Esperanto or this explanation. Adam78 (talk) 17:38, 30 January 2019 (UTC)Reply

I normally use "a lexicalized form of" for the Hungarian "megszilárdult ragos alakulat". Is grammaticalized the same? Panda10 (talk) 21:13, 5 February 2019 (UTC)Reply

Originally I was about to write lexicalization, but when I checked the term in Wikipedia, I realized it was just the opposite (that is, a lexical word "sinking" to the level of mere function words, i.e. becoming grammaticalized, such as turning into a postposition). Now I've become a bit unsure, and I also have some doubts about the other words you categorized previously. What if you created a category for words that you labelled as "as lexicalized form of" something, probably within Hungarian terms by etymology, so that we could review the words in question as well as other words concerned, and check their categorization together? Adam78 (talk) 21:30, 5 February 2019 (UTC)Reply

Enter "a lexicalized form of" with quotes into the search field and you will get the list we need. Panda10 (talk) 21:33, 5 February 2019 (UTC)Reply

Better: "lexicalized form of" without a. This will include an "inflected lexicalized form of". Panda10 (talk) 21:38, 5 February 2019 (UTC)Reply

OK, I'll need to do a little more research on this. That's all I've found up till now:

…A kérdés az, hogy lexikalizálva van-e az őszibarack, ill. sárgabarack a magyarban, vagy ezek csak körülírások? Hogyan lehetne ezt eldönteni? Ha ugyanis úgy érvelek, hogy igen, ezek lexikalizált magyar elemek, mert rendszeresen együtt használjuk őket egyetlen dolog kifejezésére, akkor a méhenkívüli terhesség, valamint a természetvédelmi terület alakulatokat is szónak kell neveznem, hiszen rendszeresen együtt használjuk őket egyetlen dolog kifejezésére. No de akkor nincs megállás, mert az engedély nélküli építkezés és a jegybanki alapkamat is szótári egység, azaz szó lesz. Oké, de akkor ezzel kimondtuk, hogy a magyarban van arra szó, hogy „jegybanki alapkamat”, méghozzá így hangzik: jegybanki alapkamat. (Vigyázat: az nem érv, hogy mi van egybeírva, hiszen azt az MTA szakemberei döntik el, s bármilyen bölcsen döntenek, tudatos emberi döntések nem tartoznak a nyelvészet vizsgálati területébe.)… A lexikalizáció (by Ádám Nádasdy)

Adam78 (talk) 21:44, 5 February 2019 (UTC)Reply

The way I understand it these were originally inflected forms that became independent words, real dictionary entries. I am working on a list based on Zaicz: User:Panda10/Words#Lexicalized_inflected_forms_(megszilárdult_ragos_alakulatok_per_Zaicz). Panda10 (talk) 22:34, 5 February 2019 (UTC)Reply

Thank you very much, it's really useful.

Going back to kedvéért, see grammaticalization and lexicalization. I tend to think it may be both at the same time. Lexicalization, because a new, stable, fixed lexical element has come into existence, and grammaticalization, because it has "sunk" from the level of lexical words to the level of function words. If it's true, the question is which feature is more relevant for categorization and/or description. Adam78 (talk) 23:01, 5 February 2019 (UTC)Reply

Thanks for the articles, they were very useful. Yes, kedvéért could be a grammaticalization, although it wasn't a dictionary entry before it became a postposition. But what about the suffixed words becoming a dictionary entry? For these, our approach could be not using any of these terms and just show the base word + suffix, as in végül. Here are a few more articles I've found:

1. Dér Csilla Ilona: Grammatikalizáció
2. Csilla Ilona Dér: Why and what to teach about grammaticalization in Hungarian?
3. Forgács Tamás: Lexikalizálódási és grammatikalizálódási folyamatok frazeológiai egységekben. Panda10 (talk) 18:09, 6 February 2019 (UTC)Reply

Maybe we could bypass (and postpone) the problem by creating a template for it, and its specific content could be changed later, too -- supposing these cases belong to the same type. I don't mind leaving out the term altogether for the moment, but it would be convenient to let people change the content (description or the lack thereof) together any time in the future. Would it be feasible? Adam78 (talk) 23:40, 6 February 2019 (UTC)Reply

The Hungarian dictionaries treat these words as the same type: megszilárdult ragos alakulat. They all started out as humble inflected forms that later became stabilized and acquired a new part of speech deserving a dictionary entry. It's just that the available English terms do not exactly have the same meaning. That's all I wanted to clarify with my original question. I also would like to use the same term throughout. Up to now, I wasn't aware that I used at least three different variations of lexicalized form. :)

To create and apply a template will require some initial decisions and a lot of work:

1. Template name: hu-grammaticalization? hu-lexicalization? something else?
2. Template content: It should contain some text, cannot be empty. E.g.: Grammaticalization of
3. Template function: Display the given text and place the word in the given category.
4. Category name: Category:Hungarian grammaticalizations? Category:Hungarian lexicalizations?
5. Parent category: Since this is a non-standard category, it cannot use the {auto-cat} feature and the parent directory has to be hard-coded. Should the parent be Category:Hungarian terms by etymology?
6. This template will have to be added to each existing entry (about 300).
7. And the final decision: What is the value added? Is it worth the effort? Or would this time be better spent with creating new entries to reduce the number of red links?

Another option would be to create an appendix (Hungarian grammaticalizations?) and just copy all the words from my subpage to the appendix. Perhaps merged and correctly sorted. They could also be grouped by parts of speech to make it more interesting. Even if we don't do anything, I would still have to go through those entries that contain a variant of lexicalized form and either remove it or standardize it. Panda10 (talk) 18:04, 7 February 2019 (UTC)Reply

I need a bit more time to think it over. Adam78 (talk) 18:39, 11 February 2019 (UTC)Reply

I went ahead and created {{hu-lex}} and the corresponding category and added különb. Panda10 (talk) 17:12, 18 February 2019 (UTC)Reply

Thanks a lot! I'm very glad. It seems like a very good solution; something that can be modified in the future, in case it seems necessary. I'm sorry for not going back to the topic earlier; I felt tired for big decisions. :) Adam78 (talk) 14:59, 19 February 2019 (UTC)Reply

Adam, if possible, please use {{af}} instead of hu-suffix. Due to a naming issue, we are trying to remove this template. It still has a couple of functions that {{af}} doesn't have, but the vast majority can be replaced. Panda10 (talk) 18:52, 19 February 2019 (UTC)Reply

All right, I'll try to keep it in mind. Adam78 (talk) 23:39, 21 February 2019 (UTC)Reply

For the formatting example.

I was wondering about including valaki and its many inflected forms in headwords, such as szabad kezet ad valakinek. Wouldn't szabad kezet ad be sufficient? The entry itself would explain -nek/-nek. Since this is an English language dictionary, how should this be presented? I know I created vacsorát ad valaki tiszteletére where valaki is in the middle. I couldn't decide at the time how to resolve this. I'd like to leave them out if possible, but is it possible? Let me know your thoughts. Panda10 (talk) 17:44, 23 February 2019 (UTC)Reply

The phrase you mentioned could become simply vacsorát ad a tiszteletére, I guess, and I'm happy to omit valakinek in the other. I also don't like having pronouns as parts of the headwords but I seem to recall (??) that occasionally this someone or something is used in English headwords as well. Maybe we could set out and find some policy regarding this matter, find examples or counterexamples, or discuss it with other editors. Adam78 (talk) 19:28, 23 February 2019 (UTC)Reply

I've found some examples: float someone's boat, tickle someone's fancy, turn someone's crank, up someone's alley. As a matter of fact, there are 885 hits for someone in the search field, restricting the search to page titles and the main namespace, and 126 hits for something. However, these already include redirects as well, where editors apparently wanted to eliminate these words from the page title, e.g. beat to the punch redirected from beat someone to the punch. Adam78 (talk) 20:22, 23 February 2019 (UTC)Reply

And there is another but related question. What to do with phrases like tesz róla, gondja van rá? Shall we include róla and rá in their headwords? Adam78 (talk) 20:50, 23 February 2019 (UTC)Reply

I was aware that there are English headwords with someone or something in them. To me, in an English language dictionary, this is fine, while foreign language entries using their own corresponding pronouns is different. It seems to me that there is a conflict. Hungarian dictionaries constantly use vki, vmi, vkihez, vmihez, valahol, valahová, etc. and everyone understands what that means. I'm not sure if I can assume the same when I describe Hungarian in English to users who may know very little Hungarian. I'd rather explain it in a gloss next to a sense that to someone requires the suffix -hoz/-hez/-höz and add examples to demonstrate it. This is not a customary notation in Hungarian dictionaries, but here we constantly have to think about the English language user who has a different mindset than a native Hungarian. On the other hand, I have no problem with tesz róla and gondja van rá as headwords. By the way, I deleted vacsorát ad valaki tiszteletére and added it as an example to ad. Panda10 (talk) 21:31, 23 February 2019 (UTC)Reply

What is this? Per utramque cavernam 16:06, 1 March 2019 (UTC)Reply

A translation hub. Adam78 (talk) 16:07, 1 March 2019 (UTC)Reply

Thanks for moving them under see also, it didn't occur to me at the time. You're doing a great job by the way. --84.236.127.70 11:59, 9 March 2019 (UTC)Reply

When -ként is added to foreign words ending in -o (casco, Leonardo), will it lengthen? Cascoként or cascóként? Thanks. Panda10 (talk) 16:30, 11 March 2019 (UTC)Reply

Very good question. Since other vowels like -a and -e do not lengthen before -ként (e.g. grafikaként, meseként), it seems that -o and -ö will not lengthen either: cascoként, Leonardoként. (Word-ending -i, -u, -ü never lengthen anyway.) This criterion is given at Section 216 in A magyar helyesírás szabályai, 12. kiadás (’The Rules of Hungarian Orthography, 12th edition’). Budapest: Akadémiai Kiadó, 2015. →ISBN. The suffixes -kor, -i, and -szerű do not lengthen the word-ending vowels, either. Adam78 (talk) 16:44, 11 March 2019 (UTC)Reply

Thanks. I corrected Leonardo. Panda10 (talk) 16:57, 11 March 2019 (UTC)Reply

I'm sorry I disagree with you. A non-Anglophone looking up αναισθητοποιώ would need to know that THERE IS an alternative. How else would they know if they didn't follow the link? — Saltmarsh. 07:17, 17 March 2019 (UTC)Reply

I've added the correct template. So should I delete all the redirects? Which one contains the hidden character? The non-lemma forms should be created with the accelerated method, just by clicking on the green links in the declension table. If there are no green links, please set it in Preferences -> Gadgets -> Add accelerated creation links for common inflections of some words. I've also created bonszai. Panda10 (talk) 18:07, 30 April 2019 (UTC)Reply

Thank you very much for fixing the inflexion chart for Thai and creating it for bonszai. (The comment said it was not possible so I was looking for a way to fix it somehow. I tried to outsmart the system and then you outsmarted me.) Yes, please delete all the redirects. I think all I created as redirects contain the hidden character. – The gadget you mentioned is switched on for me too, but the green links sometimes appear, sometimes they don't, and even if they do, I think they usually don't supply the pronunciation and hyphenation block, nor the etymology section, so I might still be better off with copypasting from a similar entry and then making the necessary changes by hand. But if it does appear, I'll consider using it.

Another thing, do you think the compound words listed under köz should be treated as prefixed words? (That is, treated as {{af|hu|köz-|…}}). Anyway, there are plenty of compound nouns in the category Hungarian compound words, do you think they could be moved to their own category? I noticed that it doesn't exist when I used pos=noun in this template and the result was a red link for this category. Of course, subcategories for other parts of speech could be created, as well. Adam78 (talk) 12:25, 2 May 2019 (UTC)Reply

Thai: No problem. The functionality was there, but it wasn't documented. In the future, if the declension table is not working correctly for a specific entry, the best solution is to add code or to create a custom table. Hidden characters should be avoided, they can be very confusing, although you added good comments to clarify, but still.

Accelerated method: The green links are set up only for the declension table. They are not available for the possessive and conjugation tables. The accelerated form contains the bare minimum. The rest of the sections all require manual edits, so they can't be automated. There are non-Hungarian editors who like to create these entries and leave it as is. Also, not everyone agrees with adding additional sections to a non-lemma form, since it's extra work, but I think they are useful for learners. I keep a text file on the side for copy-paste purposes. You can use any method, just make sure the syntax is up-to-date. For the grammatical case abbreviations are used instead of the full word. The parameter lang=hu is now being moved to the second position (inflection of|hu|...).

Köz: "A magyar helyesírás szabályai 12. kiadás" lists both a prefix köz- and a suffix -köz. It's probably a good idea to do the same here.

Compound categories: Hungarian seems to be the only language (or one of the very few) that uses the part of speech categories, except for prefixes and compound words. The other languages collect their entries under words. It would be a lot of manual work to move the 5000+ compound words to a new category. It might make sense to create a compound verb category since there are not too many compound verbs. For compound nouns, we will have to keep the current category. Panda10 (talk) 18:08, 2 May 2019 (UTC)Reply

Curious what you meant by the gloss you added here? Did you mean that spoken idősb is only used to express the sense of Sr. in people's names? (No snark intended at all, just confused by the on-screen presentation.) ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:52, 14 June 2019 (UTC)Reply

Yes, exactly. At least I've never heard it in any other sense. I'm sorry if my wording was misleading. If you can phrase it better, please do change it. Adam78 (talk) 00:04, 15 June 2019 (UTC)Reply

Thank you! No time at the moment, but I may have a go at tweaking the wording later today. Cheers! ‑‑ Eiríkr Útlendi │^{Tala við mig} 19:54, 17 June 2019 (UTC)Reply

RMaung (WMF) 14:34, 9 September 2019 (UTC)Reply

RMaung (WMF) 19:14, 20 September 2019 (UTC)Reply

RMaung (WMF) 17:04, 4 October 2019 (UTC)Reply

I see you added sorting to categories at the end of entries. Have you noticed something incorrect? It should be automatic. Although, words starting with ö/ő and ü/ű are usually out of order, at the end of categories. Panda10 (talk) 17:50, 6 October 2019 (UTC)Reply

Yes, I noticed mistakes with all á's and é's, for example, which went to the end of the alphabet, rather than being sorted like a and e. I only sorted words where accented vowels occur after the first letter. Adam78 (talk) 18:21, 6 October 2019 (UTC)Reply

@Panda10 I checked out the sorting of German lemmas in this (English-language) Wiktionary and I saw that words beginning with ö are collated there together with words beginning with o, see e.g. Öblarn, Öbstler, öd, öde, Öde, Ödem (like we collate long ó together with o, although separately from ö and ő). No sorting key is visible in the article; it just seems to work automatically. The same can be seen among French language lemmas, words with é placed peacefully among words beginning with e, or words with ç among words with c, so ca is followed by ça. You can notice the same for Spanish, ébano listed before ebanuz. Based on this all, I decided to collate manually those approx. 20-30 Hungarian three-letter words beginning with accented vowels. If possible, this should be achieved automatically with Hungarian lemmas as well, with respect to our standard, sorting á the same as a, é the same as e, and so on. Update: I've just noticed that it does work in the category for Hungarian lemmas, so the problem seems to affect a narrower range of words than I thought. For example, I fixed ásó but írógép is still at the very end of the alphabet among hu:Tools and bukó is followed by bácsi among Hungarian uncomparable adjectives. It should be fixed somehow. Adam78 (talk) 11:49, 10 October 2019 (UTC)Reply

Sort keys are coded in Module:languages/data2 (search for Hungarian). French has an extra standardChars parameter. Some languages match capital letters, as well. In Category:Hungarian lemmas, varga-Varga, Kovács-kovács - inconsistent. It should be Varga-varga. The examples you mention (bukó and bácsi) are correctly sorted in Category:Hungarian lemmas. We shouldn't add sort keys to category links in entries. I can ask in Grease pit. Panda10 (talk) 17:35, 10 October 2019 (UTC)Reply

Hi. You've been very active in adding Hungarian translations to our translation tables lately. Do you keep track of Category:Requests for translations into Hungarian? I've added requests for translation into Finnish on a number of entries (generally idiom entries; here are some examples: diff, diff, diff, diff), because I know Surjection regularly checks Category:Requests for translations into Finnish; if that interests you, I could do the same for Hungarian. Kindly let me know. Canonicalization (talk) 19:03, 26 October 2019 (UTC)Reply

These forms are not in the inflection table. Originally, they were in the case table (not the possessive table), but were removed. They were labelled incorrectly as "genitive case" which doesn't exist in Hungarian. I couldn't decide for a logical place for it. Any thoughts? Panda10 (talk) 18:47, 28 October 2019 (UTC)Reply

I made Google searches for the four combinations of predicative vs. non-attributive and possessive vs. genitive (between quotation marks) with "Hungarian" added, and I found that "non-attributive possessive" is the most widely used term, so I'll change it back. Thank you for noticing and raising this point! A Practical Hungarian Grammar (1995) uses "the non-attributive genitive" (p. 75) but you're right that genitive doesn't make much sense in Hungarian. On the other hand, Carol H. Rounds uses "non-attributive possession" in his book Hungarian: An Essential Grammar (page 152). Thanks again! (I'll also reply in connection with the verbal prefixes above, I just need a bit more time to think it over.) Adam78 (talk) 19:06, 28 October 2019 (UTC)Reply

Yes, that's where I took it from originally (Carol H. Rounds). So we will call them non-attributive possessive suffix. I assume it should be added to the main case table, at the end maybe? Panda10 (talk) 19:18, 28 October 2019 (UTC)Reply

First I thought it should be added to the end of the table of possessive endings, as it also deals with the possessive and it may be followed by case suffixes like other possessive endings (e.g. barátoméhoz). However, when I came to think of the fact that it can follow a word with a possessive suffix (e.g. "apám"), and the declension chart of this latter type of word doesn't have the possessive endings, I realized that it should still be added to the end of the chart of case suffixes. Adam78 (talk) 19:40, 28 October 2019 (UTC)Reply

Here is a sample table: User:Panda10/Declension. Please check out the labels. Example values (not included yet): row 1 column 1: házé, row 1 column 2: házaké, row 2 column 1: házéi, row 2 column 2: házakéi. Panda10 (talk) 22:09, 28 October 2019 (UTC)Reply

I think it's fine, except it's not a possession that one will find in the given cell of the table but a "possessive form", in accordance with the other labels such as "nominative (case/form)" etc. Thank you (as always!). Adam78 (talk) 22:24, 28 October 2019 (UTC)Reply

I'd like to delete the entire series. These are artificially derived words with no meaning, just for demonstration. They won't pass CFI (Wiktionary:Criteria for inclusion). Do you agree? Panda10 (talk) 14:36, 8 November 2019 (UTC)Reply

Honestly, I don't agree! I'd like to keep them, only the meaning should be clarified, with some note about the demonstrative nature of the word. Where else are people supposed to look for it if not in Wiktionary? Even if the meaning can be debated, the word as such and its elements certainly exist. I think it can be described nicely, based on what we know and what we can assume on this and this basis. Adam78 (talk) 14:44, 8 November 2019 (UTC)Reply

I've just noticed that you opened a new request in Wiktionary:Requests for verification/Non-English and I added my comment there before I saw your reply here. We can continue the discussion there. Panda10 (talk) 15:02, 8 November 2019 (UTC)Reply

When creating a new Hungarian template, please follow the established naming conventions. All our templates start with hu-. In this case, it would be hu-seealso. It should be placed in Category:Hungarian templates. Thanks. Panda10 (talk) 14:42, 8 November 2019 (UTC)Reply

OK, thank you, sorry. Better late than never. Adam78 (talk) 14:44, 8 November 2019 (UTC)Reply

Done. Adam78 (talk) 15:00, 8 November 2019 (UTC)Reply

I checked the new width in Firefox, Chrome and Safari. In every case, there is a large white area on the right side and the blue colored cells remain on the left, always the same width. Not sure what causes the difference, maybe the default text size? I'll leave it as is. Panda10 (talk) 22:16, 9 November 2019 (UTC)Reply

I wonder if I should rework this appendix. Possible changes:

1. Rename Appendix:Hungarian pronunciation pairs to Appendix:Hungarian minimal pairs.
2. Update the structure. The page is getting very long and the number of templates mean slower loading. Create subpages for each pair type, or maybe one subpage for all vowels, another for all consonants.
3. Improve navigation. Add the TOC after each pair type, so users don't have to go back to the top each time.
4. Change {{audio-list}} to {{audio}}. The latter will take more space but it plays the word without switching into another page (on desktop computers) or trying to download the sound file (on mobile devices).

Let me know your thoughts. Panda10 (talk) 18:44, 12 November 2019 (UTC)Reply

@Panda10 I'm sorry, I forgot about your message. If you still feel like doing this: I agree, except that if the title mentions minimal pairs, then "halas - hálás" and similar pairs (with more than one difference) need to be moved into a separate subsection (IMHO). Also, I'd much prefer having en-dashes or em-dashes or long quotes between members of the pairs, instead of the current hyphen. (By the way, I'd like to expand it some day with R–L and A–O minimal pairs.) Thank you in advance if you decide to implement these changes. Adam78 (talk) 21:58, 7 December 2019 (UTC)Reply

Would you give me a couple of examples for R–L and A–O minimal pairs? Panda10 (talk) 17:13, 8 December 2019 (UTC)Reply

@Panda10 Here you are:

• bél – bér, tél – tér, fél – fér, vél – vér, tol – tor, fúl – fúr, lop – rop, túl – túr, él – ér, élt – ért, ló – ró, vált – várt, fal – far, szál – szár, vall – varr, leng – reng, lejt – rejt, kelt – kert, tölt – tört, velem – verem, lohad – rohad, válok – várok, malom – marom etc.
• van – von, lap – lop, alt – olt, fagy – fogy, hagy – hogy, hal – hol, kar – kor, kas – kos, pap – pop, tar – tor, tart – tort, sara – sora, zárak – zárok, várak – várok, vágyak – vágyok, tárak – tárok, alkat – alkot, tartam – tartom, tava – tova, hava – hova, nyomat – nyomot etc. (and many adjectives used as nouns as well, e.g. pirosak–pirosok, objektívak – objektívok)

Adam78 (talk) 17:41, 8 December 2019 (UTC)Reply

I started working on the changes. So far I created {{hu-toc-min-pairs}} and two subpages: Appendix:Hungarian minimal pairs/g-gy and Appendix:Hungarian minimal pairs/t-ty. The old audio files sound bad, I might have to re-record them. If you have any thoughts about the new format, please let me know. Panda10 (talk) 20:24, 3 April 2022 (UTC)Reply

@Panda10 As a rule, splitting up the page is a very good idea and it's definitely useful, especially when there are many audio recordings. I agree with this layout.

On the other hand, I'm not sure each and every example needs an audio file (although of course there's nothing wrong with audio files, just it seems to be an overkill). To my mind, the number of minimal pairs demonstrate the importance of a phonetic distinction, e.g. for language learners who hesitate whether to invest more (or all necessary) energy on that little accent or whatever. For this pedagogical reason, I think the more the better, and learners can look them up to check that they're completely different in meaning. However, sound recordings will usually be in delay (you can't always provide them instantly, obviously) and they may also slow down the loading time of pages.

Then again, audio examples for different phonetic combinations are always good, though their selection doesn't necessarily be linked to minimal pairs. Adam78 (talk) 22:14, 3 April 2022 (UTC)Reply

OK, thanks. I'm still working on improving the table structure. Take a look at Appendix:Hungarian_minimal_pairs/t-tt. Instead of using separate tables for the different syllables, I've combined them into one. The identical table width provides a more organized look. Maybe I could add one blank row after each table to make it clearer. It might be a little confusing that the IPA is after the English translation. I think it would make more sense to display Hungarian (English). Panda10 (talk) 18:34, 4 April 2022 (UTC)Reply

@Panda10 I think I'd suggest a simpler layout. The current layout may suggest at first glance that the right column is just a continuation of the left column; only after closer inspection can one discover that these items belong together, despite their distance. I think they should be closer. Also, I'd say the visible table structure could be avoided; maybe invisible cell borders would make a breezier impression. In addition, the meaning and the pronunciation might have their own distinct cells. It would also allow for easy editing, having to insert only "||" between them. Adam78 (talk) 18:58, 4 April 2022 (UTC)Reply

I don't know how to resolve the distance problem. Adding the longer audio template, the IPA and the English translation created a distance between the minimal pairs. I will look into the other suggestions. Panda10 (talk) 20:31, 4 April 2022 (UTC)Reply

Please take a look at it again. I rearranged the entire table. Thanks. Panda10 (talk) 21:33, 4 April 2022 (UTC)Reply

@Panda10 Much better! Though I still think that the meaning and the pronunciation could have their own columns. It's not so easy to enter the right dash and it would also make its structure clearer, I think. Also, I'd prefer not to type "(no audio)" whenever the audio of a newly entered pair is not available yet. What if this column is left empty when the audio is not available? Adam78 (talk) 21:42, 4 April 2022 (UTC)Reply

I've made the changes. It does look cleaner. :) Panda10 (talk) 21:48, 4 April 2022 (UTC)Reply

@Panda10 I think pairs can be in the same row if there are no internal border lines. I've made this change myself. You can revert it if you really want to. Adam78 (talk) 08:03, 5 April 2022 (UTC)Reply

OK, thanks. I will use this layout for the rest of the pages. Panda10 (talk) 17:06, 5 April 2022 (UTC)Reply

@Panda10 It looks good now, in comparison to the amount of information that is intended to be packed in. However, I still have some doubt if all this information is necessary; whether it would be better to make it available at the links to their own entries. Most of the pronunciation is redundant, since the only difference is the length or quality of one sound, all the rest follows from other rules of pronunciation. The sound file can also be linked. Honestly I'd prefer something simpler, at least additionally, a list that can be expanded easily and that collects only the essential pieces of information, i.e. the words themselves plus maybe the meaning. I have some reservation about it being a bit too heavy due to partly redundant information. Being "leaner" and more focused might benefit it more. Let's think it over a bit more before you start implementing the changes everywhere else... Adam78 (talk) 17:21, 5 April 2022 (UTC)Reply

Yes, it's fine to think more about it. There was a time when I added a comment after the hu-IPA template (Not to be confused with...) but the community did not like the location of the comment, so instead of deleting it, I moved it to the Usage notes section because I thought it was important to emphasize the difference. See intézett. If we keep the minimal pairs Appendix on a single page, the audio should be removed and only the list should be kept. It may still load slowly because each word is linked with {{l}}. Panda10 (talk) 17:31, 5 April 2022 (UTC)Reply

@Panda10 I wonder if a bridging solution made sense, like creating a Hungarian-specific template that invokes {{l}} but with two superscript links, one for the IPA and another for playing the audio, e.g. {{hu-sound|lap|sheet||Hu-lap.ogg}} instead of displaying all these bits of information, with a result like lap ^{pr audio_link} before a similar link to láp, so as to reduce the need for formatting. This way if someone just wants to add a minimal pair without any more details (simply with "{{l}}") it would still be possible, without disfiguring an existing table structure.

However, if you're fine with the most recent version you just created on Appendix:Hungarian minimal pairs, it's also great with me. Honestly, I think I prefer the latter. The appendix may still be split up. Maybe one section could be longer, e.g. one section for vowel length differences (including a–á, e–é), another for consonant length differences, and a third for consonant quality differences (voicing, palatalization etc.). Adam78 (talk) 18:35, 5 April 2022 (UTC)Reply

I prefer the latest solution at the current a-á section of minimal pairs. Let me make the simplifications on the entire page, adding back the sections I deleted, then we can see if it's need to be split up.

I'm done with the changes. I also added sections L-R and A-O. What should I do with the subpages? I can delete them or keep them and link to them from under the corresponding section for a detailed list. On the other hand, I realize it's a lot of work to create the subpages, maybe it's not worth it. Panda10 (talk) 21:09, 8 April 2022 (UTC)Reply

@Panda10 As a matter of fact, I'm not convinced (either) that it's worth keeping the subpages, unless as a starting point in terms of design for any future development. I'm sorry for being ambiguous in my earlier communication, which may have cost you extra time and work invested.

The only bit of information that might be worth keeping in my opinion is meaning, but I wouldn't like to duplicate it, what's more, cluttering the page may not be ideal. A long-term solution (which could be used in compounds as well) could be something like the one used in Chinese-language entries for the primary meaning of individual characters when they occur as part of compounds, see e.g. 購買 / 购买 (gòumǎi, “to buy/purchase”), which links to {{zh-forms}}, invoking {{module:zh-forms}}, invoking {{module:zh}}, invoking {{module:zh/data}}, especially {{module:zh/data/glosses}} for displaying the same short glosses in all entries of compound words. However, obviously, it's by far not as justified for Hungarian as it is for Chinese.

Thank you for restructuring the existing appendix! Adam78 (talk) 21:38, 8 April 2022 (UTC)Reply

No problem at all. I've found your comments very helpful. Directions often change when charting a new territory. When a better option comes along, the old one is discarded. I'll leave the subpages for now. As you said, the Chinese system may not be the best for Hungarian. In compound words even the same element can have different meanings, see for example mű-. Panda10 (talk) 20:37, 10 April 2022 (UTC)Reply

I thought the superessive singular of the noun terem was termen. Panda10 (talk) 21:21, 7 December 2019 (UTC)Reply

@Panda10 I'm sorry, my mistake. I've just corrected it. Adam78 (talk) 21:25, 7 December 2019 (UTC)Reply

In E-szókincs it's iva and iván, not íva and íván. I corrected the conjugation template but I thought I'd ask before changing the entry. Did you have a specific source for the long í? Panda10 (talk) 21:09, 4 February 2020 (UTC)Reply

@Panda10 Yes, I did: Osiris Helyesírás and Osiris Diákszótár – Helyesírási szótár, both authored by Krisztina Laczkó and Attila Mártonfi. It's always good to have these at hand! As you probably know, the former is more comprehensive but the latter is more recent (and therefore perhaps less likely to err). – By the way, I think we'd better put égy in the second place and in parentheses in the conjugation table (if we do want to mention it), because it's so rare that it's almost non-existent. I barely managed to find one or two instances for this form. Adam78 (talk) 21:24, 4 February 2020 (UTC) PS: We could try to find out whether (el)hivés or (el)hívés is the correct form for the relevant table. Adam78 (talk) 21:29, 4 February 2020 (UTC)Reply

Okay, thanks. Corrected them. I do have Osiris Helyesírás, not the other one yet, but I forgot the check it. I trust E-Szókincs too much, even though I know well there are errors in it. In Osiris Helyesírás: hivés. Panda10 (talk) 21:39, 4 February 2020 (UTC)Reply

Indeed, thank you (I was looking for it under hisz). Adam78 (talk) 22:21, 4 February 2020 (UTC)Reply

I don't understand why the verbal noun használás was removed from the conjugation table of használ. It's a valid word. Panda10 (talk) 19:01, 6 February 2020 (UTC)Reply

@Panda10 It's extremely rare and uncommon. Magyar értelmező kéziszótár has no entry for it, even though this dictionary comprises some 70,000 Hungarian words. A magyar nyelv értelmező szótára only refers to it but it doesn't define it either, and it only has three instances for it among all the definitions or examples.

If you still want to keep it, it could be inserted after használat, definitely in parentheses, so as not to mislead non-Hungarian speakers. By the way, we could also consider adding -ván/-vén as alternatives to the adverbial particle. (And perhaps we could consider adding the historical past forms to the inflection table: olvasék, olvasál, olvasa etc.) Adam78 (talk) 19:37, 6 February 2020 (UTC)Reply

Google returns 39,800 hits for használás. Most of them are archaic or part of a verbal prefix phrase such as fel nem használás, but it is still there. I thought we stick to -ás/-és verbal nouns and add the others to Derived terms. If you don't want to add használás to the conjugation table, then add it to the Derived terms section, but it should be mentioned somewhere. We already include archaic terms so why not this?

• -ván/-vén: I agree, these variants should be included in the conjugation table.
• Archaic past tense (elbeszélő múlt): I really don't know how to include this without cluttering the currently nice and compact layout. I understand it might be interesting to linguists or even useful for someone reading older literature. Adding the archaic past tense would mean two additional rows (indefinite and definite): olvasék, olvasál, olvasa, olvasánk, olvasátok, olvasának and olvasám, olvasád, olvasá, olvasánk, olvasátok, olvasák. Panda10 (talk) 22:12, 6 February 2020 (UTC)Reply

@Panda10 Fel nem használás is the negation of felhasználás, so I think we rightfully exclude these hits from the conjugation table of használ, but anyway, we might mention it in parentheses in the table, just like other verb forms that normally occur with a certain prefix. I just prefer not to clutter Wiktionary with odd-sounding hypothetical word forms if they are unheard of by native speakers and they can mislead language learners.

These form can actually occur as mistakes in performance, even in writing (and nowadays there are more and more casual content to be found on the Internet). Also, Google parsing is deficient since many of the hits are actually fel- használás in two parts. Some results are machine translations; some are repetitions. And if you scroll a couple of pages among the results, you'll see that it's not even that many hits, because it's an estimate; leafing through the pages will come to an end after some 30 pages. For me, the total number of hits was 316 (searching for "használás" with quotation marks), including all the additional, hidden (repetitive) results.

As far as the form is concerned, we can't stick rigidly at all costs to the -ás/-és form, because if one form comes into existence in a particular sense (like használat in this case), the other competing potential forms (like használás) will be prevented in turn by the generation processes. (I seem to recall this principle vaguely from my erstwhile linguistics studies.) As for the rest, I'm glad that you're open to these ideas. I'll think about how to implement them in a handy way. Adam78 (talk) 22:45, 6 February 2020 (UTC)Reply

@Panda10 If you have the time, please open the template hu-conj-gyek for editing, insert {{hu-conj-experimental in the first line (instead of hu-conj) and check out "Show preview". If you scroll down a bit (without saving the changes, of course), the second table you'll see is a possible layout for the verb vesz with the historic forms included. Do you have any suggestions? Adam78 (talk) 21:53, 9 February 2020 (UTC)Reply

Can you make the label simpler? "Archaic past" would be easier to understand than "Preterite (historic past)". The table is getting a little crowded, so it would be nice to make the archaic past forms seem a little less important than the modern forms. I wonder if a smaller font would help. I don't want a different color because that would actually highlight them. Panda10 (talk) 22:03, 9 February 2020 (UTC)Reply

@Panda10 I've made the label simpler, although we should keep in mind that there are 3 archaic past tenses (megyek vala and mentem vala being examples for the other two) so this term is a bit loose, but anyway, we don't need to be so precise here. As far as the location of this section is concerned: What if we placed it at the very bottom of the entire table, under the "Other nonfinite verb forms"? (Small letters may easily interfere with reading these uncommon forms.) I updated it, you can check it out, but I doubt it's the best way possible. We can brainstorm some more. Adam78 (talk) 22:44, 9 February 2020 (UTC)Reply

I don't see the changes, not even in the previous format. Yes, we can think about it some more. Panda10 (talk) 22:49, 9 February 2020 (UTC)Reply

I have a question about the etymology of second-person singular subjunctive present definite verb forms where the lemma ends in a consonant + t (kelt, gyújt) or a long vowel + t (nagyít) or a short vowel + t (mutat, fut). The -j- that becomes an -s- should be shown somewhere. The ety which I created, currently is kelt +‎ -d, but kelt + -j + -d might be better. What do you think? Panda10 (talk) 20:38, 9 March 2020 (UTC)Reply

@Panda10 I don't really agree, because if there were a modified -j- there, then this -j- should also be present in forms like várd. I think the regular subjunctive endings (as those found after vár or kér) trigger peculiar subjunctive stem variants in the other verbs you mentioned. So I think the suffix doesn't change but it's the stem whose alternative forms are seen in kelts-, gyújts-, nagyíts-, and fus-, just like bokr- or madar- appear as stem variants before certain suffixes (and not before others). Adam78 (talk) 20:51, 9 March 2020 (UTC)Reply

The verbs vár and kér are not in the above mentioned category since they are not ending it -t. I'm still not comfortable with the current ety. I have now Hegedűs Rita's Magyar nyelvtan. On p. 126, -t + -j = ss, so the way I interpret it: keltsd = kelt + -j + -d. Panda10 (talk) 21:29, 9 March 2020 (UTC)Reply

@Panda10 I'm very glad you got this book. Hegedűs Rita's main concern is not linguistic history but language pedagogy; that's where she excels. First of all, I can't imagine how várd and keltsd could have two different etymologies, one without -j- and the other with -j-, since they take the very same place in the verb paradigm. I suppose várd and keltsd might have been shortened from várjad and keltsed, and therefore the -d itself must be assumed to incorporate this -j in whatever form. So we if we manage to support this hypothesis with facts, then we could mention -j in the etymology section of the -d suffix. If it's a subjunctive ending, this -j- must be hiding there somewhere after all, both in várd and in keltsd. Adam78 (talk) 21:47, 9 March 2020 (UTC)Reply

I've always imagined the Hungarian section of this wiki as a learner's dictionary with more emphasis on language pedagogy than linguistic history. The suffix -d is the second-person singular suffix, so I don't see how the subjunctive suffix -j can hide in there. I guess I need some more time to think about this since I just can't get rid of the discomfort I feel with the current etymology but at the same time I can't grasp and explain it scientifically. I want to make sure that our etymologies are supported by reliable sources and they are not our personal interpretation of grammar rules. :) Panda10 (talk) 23:31, 9 March 2020 (UTC)Reply

Here is what I've found in Magyar Grammatika, p. 117: A felszólító mód jelének alternánsai: -j, -gy (egyed), -ggy (higgyed), -s (ránt-s-uk), -Ø. (Note: Other sources also list -jj as in jö-jj.) "E/2-ben árnyalatnyi használatbeli különbséggel két alak is lehetséges: egy hosszabb, például vár-j-ad és egy rövidebb, például vár-Ø-d. E/2 rövid alakjában a felszólítómód-jel -Ø. A fordít-s-ad : fordít-s-d típusban a felszólítómód-jel a rövid változatban is megtalálható. Az sz/v-s tőtípusba tartozó igék E/2. rövid alakjában a személyrag megnyúlik, például: e-dd, hi-dd." Based on this, I think keltsd is kelt-s-d. The question is, should the alternate suffixes have their own entry? What do you think? Panda10 (talk) 18:34, 10 March 2020 (UTC)Reply

@Panda10 Thank you! It's really useful. I also checked out Magyar nyelvtörténet and Etimológiai szótár (the latter at -d and -j-) but I didn't find in them anything about this short subjunctive form.

It would be interesting to have an entry for the null morpheme in Hungarian, listing its various functions like:

• an alternative form of the -j- subjunctive suffix
• the suffix of the third-person singular indicative present in most verbs (except for nincs-en, sincs-en, tegy-en etc.)
• the suffix of the nominative singular,
• etc.

I'm mentioning it because if we supply the etymology for keltsd as kelt + -j + -d, then we should also supply the etymology for várd as vár + -∅ + -d. (On a side note, it's indicated with an empty set sign, ∅, not the Scandinavian slashed O, i.e. Ø, to my best knowledge. It's available in the collection below the editing field, at the "Miscellaneous" drop-down item of symbols.)

My suggestion is that we could link both -s and this -∅ to -j, because that's where it is defined, and there we could mention all its alternate forms (allomorphs), a bit like linking -vé from the form széppé.

In fact, I'm wondering if we should merge harmonic variants like -hoz/-hez/-höz into the same entry (named "-HOZ" or "-hOz" or "-hoz/-hez/-höz" or any other combined form that is advisable here), because most suffixes have several different meanings (e.g. four senses are already listed at -a/-e/-ja/-je) and they should be listed at the same place, rather than repeating the definitions across different entries and trying to keep them synchronized. For example, the various article senses of a are not listed again at the form an. Adam78 (talk) 19:51, 10 March 2020 (UTC)Reply

I added the entry for the null morpheme but I don't think we should use it in etymologies. It would be too confusing to use it everywhere (Wikipédia Zéró morféma lists 11 cases) and we either use it in all types or none of them. I made changes in -hoz/-hez/-höz, I don't like the idea of -hOz, think about the verb-forming suffixes: each variant can have its own conjugation. We need to keep suffixes consistent as much as possible. About -j: I'm wondering why Magyar Grammatika never mentioned verbs ending in -s, -sik, -z, -zik, -sz, -szik, -dz, -dzik. The suffix -j assimilates in these endings, so besides the allomorph -s, we should add -z, -sz and -dz and in the glosses we should mention the -ik verbs. Would you agree? Panda10 (talk) 18:33, 11 March 2020 (UTC)Reply

@Panda10

• I think we can consider it a general rule (although not without exception) that the least marked cases tend to have the null morpheme, so I agree we don't need to insert it everywhere. However, there are particular cases when we cannot ignore the lack of something (one of the best examples is a plural genitive in Russian, like рабо́т being this genitive form of рабо́та). As far as Hungarian is concerned, I'd say we need to be consistent at least in this particular field: as long as we want to indicate it in keltsd, we should also indicate it in várd, since (as I wrote) they take the same place in the verb paradigm and they have the same semantic elements – they just can't have different etymologies. If we claim that the suffix -d doesn't have such a sense that implies the subjunctive (and it seems so, based on your citation of Magyar grammatika), then we are bound to assume some form of -j also present in these verb forms, overt or covert.
• I agree about mentioning the other assimilated forms of -j; it's a good point indeed.
• I see that you added the Hungarian section to the "slashed O" character. I noticed Wiktionary claims that this character is used for null morpheme, although the English-language Wikipedia says the opposite (like in w:Zero (linguistics), w:Null morpheme, w:Null allomorph, w:Null sign, w:Silent letter). Anyway, this entry can still be moved to the empty set page later providing we find enough evidence for that being the right character.
• As far as -hoz is concerned, I absolutely agree with your recent improvement but it doesn't really address the issue that I raised. Maybe I wasn't clear enough. Here's an example from w:hu:A magyar nyelv eseteinek listája:

A sublativus (-ra/-re) például kifejezhet
  • a hely mellett (postára)
  • időt (szerdára, két percre),
  • módot (kedvünkre),
  • mennyiséget, mértéket (kőhajításnyira),
  • okot (felébred a kopogtatásra),
  • célt (hivatalos látogatásra érkezik),
  • eredményt, következményt (zöldre fest, 30 fokra emelkedik),
  • ezenkívül vonzata lehet a nézve névutónak
  • és több tucat igének, melléknévnek (biztat vmire, gondol vmire, kíváncsi vmire stb.).

I don't think we'd like to list these many senses separately in the entry of -ra as well as in the entry of -re and keeping them synchronized, when we're describing the meanings of the same suffix. There are two ways I could currently imagine it. One is if we reuse the existing template of the form variants (allophones) for the purpose of listing the senses (in which case the template would include not only the Usage notes but also the preceding section with the meanings, right after the closing brackets of head). The other is that we make an "alternative form of" type of redirect from the -hez and -höz forms, and we don't even start giving details about their meanings in the entry of -hez and -höz, only at one (primary) form, in this case, -hoz. The former would be more user-friendly, although technically a bit unconventional perhaps (I don't know). What do you think? Adam78 (talk) 20:14, 11 March 2020 (UTC)Reply

@Panda10, please do check out the short chapter on the subjunctive allomorphs in Attila Mártonfi's theses and dissertation (pp. 141--149 in the latter, although the former might give you an easier overview). It reflects on multiple different earlier studies. We should do our best to avoid the "spelling trap": one must focus on the pronunciation rather than the spelling. (It's not the letter s that is to be accounted for but the modified pronunciation.) Unfortunately, this result apparently still won't save us from mentioning the subjunctive -j in whatever form, since that's what induces the alternative stem. Still, this paper might help us tackle the problem in a more convenient way. Adam78 (talk) 01:58, 13 March 2020 (UTC)Reply

I'd like to add the IPA symbol to our appendix since it is used. Can you help me explain it? Is it used elsewhere besides euró? Thanks. Panda10 (talk) 16:54, 29 April 2020 (UTC)Reply

@Panda10 It's a good idea. I used it to indicate that it is part of a diphthong, that is, the word is pronounced with one less syllable than the number of vowel letters as one of the two neighboring vowels becomes a glide or semivowel. It mostly occurs in loanwords; autó being another example (along with its compounds like autóbusz) as well as augusztus, though optionally in some cases. If you ask native Hungarian speakers how many syllables autóbusz has, they're likely to answer three. Here's a more detailed discussion on other ways of how these phones can be shown to be diphthongs: A kérdésben a válasz.

For the notation, see Semivowel:

"In the International Phonetic Alphabet, the diacritic attached to non-syllabic vowel letters is an inverted breve placed below the symbol representing the vowel: U+032F ̯ COMBINING INVERTED BREVE BELOW."

In a similar vein, the pronunciation of óriás and óriási could be given with , in accordance with this article:

"Vannak azonban olyan magánhangzók, amelyek mássalhangzóként is tudnak viselkedni. Ilyen például az , amit mássalhangzói előfordulásában -vel szoktunk jelölni. Valójában pontosabb volna a jel (ahol a betű alá írt kis félkör azt jelenti: ebben az esetben nem a szótag központi szereplője az )."

Other less common phones in Hungarian that I know of are found in arra and erre (ɒː, ɛː) and also in the word haló as pronounced when answering the phone, contrasted both with haló (dying) and háló (net or bedroom). Adam78 (talk) 17:26, 29 April 2020 (UTC)Reply

Thank you! Panda10 (talk) 18:04, 30 April 2020 (UTC)Reply

It would be fine to add if you want to. Things to think about: This will need a new parameter or two (with or without -j) and will have to be added manually to each noun. The headline will be much longer. For example:

asztal (plural asztalok, third-person singular possessive asztala)

makroobjektív (plural makroobjektívek, third-person singular possessive makroobjektíve or makroobjektívje)

Are you planning to add the accusative, as well? Panda10 (talk) 16:25, 5 May 2020 (UTC)Reply

@Panda10 I think the description should be abbreviated, like pl instead of "plural" (several other declension types are abbreviated for other languages), with the dotted underline, so that anyone interested can look up the full form. Otherwise, it's the kind of job that could and should be aided by a bot. If we add a parameter to put most types of the possessive form into specific subcategories (one of them already exists), then a bot could be made to do the required change on every element of the given category. This way we could review and check the scope of its operation in advance and prevent any mistakes. We might as well keep these categories, or at least those with a finite number of exceptional cases (e.g. hídja, dereka).

Yes, we could do the same for the accusative, too (at the same time as the possessive subcategorization or separately, depending on which looks more convenient). The accusative is simpler and can be algorithmized almost completely, based on the existing plural parameter and the final consonant. I think the first form after the dictionary form should be the accusative, then the plural, and finally the possessive form(s), cf. Hegedűs, p. 91, and also because the possessive might be more complicated. The only other form that could be considered is the superessive, as it may be regular as opposed to all the others, e.g. hetet, hetek, hete, but héten. Also: földet, földek, földje, but földön. If we include it, then the head could look like this (I left this form to the very end because it's different from all the others in cases like hét):

hét (acc hetet, pl hetek, pos hete, spe héten)

makroobjektív (acc makroobjektívet, pl makroobjektívek, pos makroobjektíve/-je, spe makroobjektíven)

All these could be arranged by adding the nouns into subcategories first. What do you think? Adam78 (talk) 17:21, 5 May 2020 (UTC)Reply

I'm not sure about the abbreviated grammatical terms. In the past, the community's opinion was that this is an online dictionary and no need to worry about saving printed pages by using abbreviations, so it's best not to use them to make things more readable for end users. Who will know what spe is? You mentioned that "several other declension types are abbreviated for other languages". I clicked all the translations under table and have not found any abbreviated case names in the headlines of other languages. Some even list the diminutive. See Tisch, kotak. As for the number of cases, the superessive might be a little too much. But again, this is just my opinion. I was aware that other languages and our own Hungarian dictionaries list more cases than just the plural, but I always thought that we have the declension tables, so anyone interested in more cases, they can just look it up. It's up to you, though. If you see it as useful information to learners, by all means, go ahead. Panda10 (talk) 18:59, 5 May 2020 (UTC)Reply

@Panda10

• "making things readable for end users": I agree, and it's for this very purpose that I suggest using abbreviations. Anyone who starts browsing Hungarian entries will find out the structure after a couple of pages. We could even use something like this: ◆ between them, because the endings should suffice by themselves. I doubt that mixing relevant and irrelevant information could be useful for anyone; all we need is not to leave anything unclear for outsiders.
• "several other declension types": see Module:gender and number. You don't need to know what a class 1 means unless you deal with a language that uses such classes. Similarly, you don't need to know what spe (perhaps we should use sup instead) or even superessive means unless you deal with Hungarian (or another language that employs it). At the page I linked, you can find an for animate and in for inanimate. On the other hand, I don't think the method used at kotak is the way to go…
• Nevertheless, you raise important points. I think the main goal would be to list the forms that are indispensable for creating the other forms, such as ferō 3 tulī lātum or ferrum, -ī n, which are traditionally supplied in Latin dictionaries. All the other declension or conjugation below should be superfluous, the exposition of the above core information, which should be sufficient and necessary for a language learner. Adam78 (talk) 19:29, 5 May 2020 (UTC)Reply

It would be great to have a complete table. Thoughts:

1. The length of the table: Your new layout has a different look and feel than the current declension template, so even if both tables would be completely closed, there would be one short and one long. I wish we could keep the two short tables under each other to maintain a balanced look in the entries. I think there might be a way to keep the table short when closed and resize it to full size when open.
2. Closed by default: I've just noticed that the current tables remain closed in mobile view - this is great. There must have been some kind of change in the past few years. So the code for the full table could use the same code as the current one to ensure that it is closed in mobile view. This is important because the full table is very long and if it's open all the time, it will push down all the other important information and the user will have to scroll down to see them.
3. Parameters: Lots of new parameters, I did use them in one of the examples in {{hu-infl-pos-table-comparison}}, such as 1sg_sg_acc, 2sg_sg_acc, etc.
4. Functionality: The possessive templates do not have identical functionality and parameters.
5. Template name: I have already used the {{hu-infl-pos-table-full}} name, so I wonder if the new template should be called {{hu-infl-pos-table-full-horizontal}}? This template is used only internally by {{hu-pos-atok}} etc.

That's all I can think of for now. Panda10 (talk) 16:56, 5 May 2020 (UTC)Reply

@Panda10 I think first of all we should agree on the look, then we could figure it out how to implement it. My replies:

1. I think it's useful if the six persons are arranged in columns (next to each other), like for verb forms, and the cases are listed in rows (under each other), like in the existing table with case suffixes.
2. I agree that we should save space so as not to push down the remaining information too far at the bottom, but I think the means to this end is to have wide tables, just like the one I created. I think it's short enough with its mere four lines. (Sorry, originally it was six but I've just reduced it by two lines, if not in the most elegant way.) I'm not sure if the "" or "" link could be inserted to the right side of the nominative row; if so, this could be a way to save two more lines. (If you want to save even more space, then maybe the table of contents should be hidden for terms where no other language entry is given, only one in Hungarian.)
3. I don't think we need to worry about parameters because I think the existing possessive template should be replaced with this new template, so it would get all the values that are needed to generate the forms.
4. I don't understand this point of yours. (Or maybe I replied to this above?)
5. If we replace the existing template (after careful testing, of course), we don't need to worry about the new name. Adam78 (talk) 18:06, 5 May 2020 (UTC)Reply

About the look: In point 1, I was talking about the horizontal length (probably should have used width) of the table. The current narrow declension table with the new wide possessive table will look a little weird. I think the new possessive table should be completely closed without showing any values. Panda10 (talk) 19:05, 5 May 2020 (UTC)Reply

@Panda10 I have an idea about the look. We could put 10 rows on the left side and 10 rows on the right side. The underlying reason is not merely symmetry and better space saving (and consistency with the planned new possessive table) but the fact that there are cases that (originally) express location (semantic meaning, cf. hu:w:Eset) and there are other cases with a more abstract, mostly syntactic meaning. What's more, those in the first group have similar morphological properties, e.g. almost always employing the dictionary stem (bokorban vs bokrot, nyárról vs. nyarat, with some exceptions at a/e final words: órában but óraként). Those in the first group behave the same way so their listing is actually less informative than the listing of the other group, so possibly the other group might be given more emphasis. The first group comprises the following: terminative, inessive, superessive, adessive, illative, sublative, allative, elative, delative, ablative. Do you think we could move these rows to a new column, thus creating two blocks with the existing three columns (altogether six columns, plus a thin separator)? Adam78 (talk) 14:16, 6 May 2020 (UTC)Reply

I think we should leave the declension table as is. I believe the current code allows a narrow closed appearance, but when it's open, it jumps to full width. This would work for the new possessive table. Panda10 (talk) 17:54, 6 May 2020 (UTC)Reply

@Panda10 I suppose the nominative forms should be made bold (and/or perhaps some headers), but here you can check out an experimental version in terms of functionality:

I'm afraid the additional -j forms cannot be linked automatically because they'd cause problems, so I had to remove the brackets, but the following table still doesn't work, so it needs more work (think twice before opening it ;):

The good thing is that the entries don't need to be updated with new parameters, only the tok/tek/tök/otok/etek/ötök/atok possessive templates need some extra parameters. Adam78 (talk) 22:17, 8 May 2020 (UTC)Reply

One option to fix it is if the j-optional tables invoked themselves twice, once with the j, and once without. But actually it might make this section far too lengthy; they should be compressed a little bit. Maybe by making links like sivatag(j)aik, indicating and linking both versions but still within the same cell. Or do you think two independent tables would be better under each other? They're closed by default anyway and it would provide a neater view, I think. Adam78 (talk) 10:40, 9 May 2020 (UTC)Reply

The table looks really good to me. I modified {{hu-infl-pos-table-test}} to make the colors and formatting the same as the current table. I didn't finish all rows, just the top. The sivatag(j)aik solution would save a lot of space, but would the user be able to click on the small j to see sivatagjaik? And click the rest to see sivatagaik? If this is not feasible then a second table has to be used. Are you planning to add the non-attributive possessive forms: sivatagomé, sivatagoméi? I noticed that you did not use new parameters for the cases (e.g. 1sg_sg_acc) and this is fine, but I wonder if acceleration could be applied eventually to this structure. The current one doesn't have acceleration. About the future new name: what about {{hu-infl-pos-table-wide}}? I'd like to leave the current {{hu-infl-pos-table}} unchanged, just in case. Panda10 (talk) 16:41, 9 May 2020 (UTC)Reply

@Panda10

• The user wouldn't have to click the small j but the (j) along with the parentheses. (And yes, the rest of the word to see sivatagaik.) Like this: ]]], that is, sivatag(j)át. However, links that include a (j) or even the stem sivatagá simply don't want to work (for me). I'm not saying it's not feasible but I haven't been able to manage it. And I'm not even sure if the other way (with two distinct tables) would be easier to implement.
• Yes, the non-possessive forms should be added too. I just wanted to make sure first that the most difficult parts work properly. For the moment, they don't. :(
• Yes, these new parameters and acceleration could be used but I don't know how to do it. I'm not familiar at all with Lua and this type of WikiMedia syntax! I wouldn't mind it if someone else did it but I don't see people scrambling to do it. :(
• First of all, I wanted to make sure that the new table is completely functional and compatible with the existing parameters, in order that we don't have to replace its link in each and every entry (thousands of them).
• I don't understand your question about the name. If we want it to be functional from all the existing articles, then we have got to keep its name the same as the current name. Of course, we should keep a copy of the current template, probably by moving it to a new name (something like "……-archived") so that the existing name can be used with the new content. Adam78 (talk) 18:33, 9 May 2020 (UTC)Reply

1. Acceleration: If you look at {{hu-decl-table}}, you'll see lines with accel-form. This is the only Hungarian template with acceleration function and this is what the possessive forms used (not the nominative, though). I tried to figure it out when I added the non-attributive possessive forms, but it is not perfect. It does create the correct lines, but the non-attributive possessive is not displayed. Instead, the parameter name shows up which has to be edited manually. There is a Module:accel, Module:accel/hu and MediaWiki:Gadget-AcceleratedFormCreation.js. As you said, this can be added later.
2. It was never the goal to edit thousands of possessive entries for the new wide layout. The planned change was always internal and everything else (template names and parameters the user would see and use) would remain the same. I didn't even know that new parameters would have to be added to the -tok/-tek/... templates (such as the v parameter you added). I figured the {{hu-infl-pos-table}} will be replaced with a new one to provide the new layout and the -tok/-tek/... templates would contain the actual values as today, except more of them. I think your solution is different. Unfortunately, I'm not that familiar with the wiki script to determine which direction is more beneficial. Panda10 (talk) 21:03, 9 May 2020 (UTC)Reply
3. What do you mean by "I just wanted to make sure first that the most difficult parts work properly. For the moment, they don't." What is not working? The j part? I wonder if we tried the other solution (where the table template provides the layout and the -tok/-tek... templates the values) would resolve this?
4. The template name: It's fine if you want to keep the current name and archive the content. Panda10 (talk) 21:03, 9 May 2020 (UTC)Reply

@Panda10

• I don't like the idea of having two tables under each other with the only difference of the j being present or absent. If we dump the reader with so much information, which turns out to be mostly irrelevant (because it's predictible after the first couple of occurrences), they will have all the reason to dislike this entry as a time-waster. When there is so much information involved, the editors need to be prudent about their handling.
• What doesn't work properly at the moment is the linking. If you check out a form like sivatag(j)aimat, you'll see that only a part of this form (sivatagaim) is a link, not the entire form that is displayed.
• I've been thinking about another solution, something like sivatagaitok (+j) in every cell involved, where the second form is linked to the correct j-version. If you agree, I might try to implement this. Adam78 (talk) 19:36, 10 May 2020 (UTC)Reply

Okay, let's go with the one table solution. You're right, even that is a little intimidating just to look at. I'm not sure about the (+j) solution, not totally intuitive. The linking problem could be asked in Grease pit. I need to ask you again about acceleration. Do we want acceleration in the wide table at all? If no: We could live without it because acceleration works in the declension template which is used in the possessive form entries. This assumes we will still add the declension table to each possessive form. The wide table would be still useful because search will point the users to the main noun even if the possessive form doesn't exist. If yes: I need to find out whether each form needs a unique variable name for acceleration to work in the wide template. If a unique name is needed, the table will have to be reworked. I brought up in my previous comment another possible solution where the v variables are not used in the -tok/-tek... templates but the actual variable names are added. Is there a reason you didn't like this solution? Panda10 (talk) 20:58, 10 May 2020 (UTC)Reply

@Panda10 It's not that I don't like that solution (yes, of course it would be nice to have acceleration in that table) but I'm simply not motivated to deal with it until the most difficult part of the problem is solved. I say, let's make sure that it works perfectly and then (afterwards, later) we can refine, improve, and possibly even rework it. One thing after the other, please, please. ;) Adam78 (talk) 23:22, 11 May 2020 (UTC)Reply

What do you think about the idea of a nested table layout? It is described in Wiktionary:Grease_pit#Double_linking_a_single_entry_in_declension_templates. Panda10 (talk) 23:14, 13 May 2020 (UTC)Reply

Would you mind making this table closed by default? It's too long and runs into other language sections. Thanks. Panda10 (talk) 15:56, 22 May 2020 (UTC)Reply

@Panda10 I've made the less essential parts hidden, so it's about half height now. I don't think we should make it entirely closed by default because it's important to point out the connection between azon and rajta, as well as the others. It's very important for language learners. Is that OK with you?

I'm sorry I stopped reacting about the table. I got frustrated that we were stuck and the discussion turned to relatively peripheral directions. You suggested moving the forms into the other table to keep the look and feel separate, which I can understand but I suppose you know that it would involve creating some 200 parameters and forwarding them all to the other template; I don't know how it could be worthwhile. Also, I'm not very fond of the idea of nested boxes (because "double hiding" makes sense to me when something is as obsolete as historic past tense forms in Hungarian), although I could accept that point nevertheless. Adam78 (talk) 16:42, 22 May 2020 (UTC)Reply

Yes, the height is fine now. Thanks. I just didn't want it to run into other languages. About the nested table: I couldn't make it work, anyway, so go with your original plans. Panda10 (talk) 16:47, 22 May 2020 (UTC)Reply

@Panda10 What do you think about this similar template for postpositions and related terms? Do you have any suggestion before I start inserting it into the articles? {{hu-suff-postp}} Adam78 (talk) 00:09, 23 May 2020 (UTC)Reply

It's good to have a summary list. Here are my thoughts:

1. Location: Right now the table is on the right side. Anything on the right side (images, wikipedia boxes, etc.) will show up in mobile view on the top, before anything else, so users will have to scroll down quite a bit to see the actual entry. On desktop, it stays on the right, but it squashes the main content a little too much and it becomes less viewable. I think the table should be on the bottom left side, this way you can create a full size table if needed.
2. Him/her (it): I don't understand the parentheses. I would change it to him/her/it. Example: Hol volt a szék? Az asztal mellett vagy alatta (under it)?
3. Postpositions: Should it be noted somewhere that not all items are postpositions?
4. English translations: Do we need it? At least for the first column? Just a thought.
5. Description: I had to study the table a bit to figure out what it wants to say. A short description might help.
6. Template category: This template might be a usage template, not sure, though. In any case, can you please add the documentation and category?
7. Template name: If you decide it's a usage template, would U:hu:postposition-comp be better? I assume comp = comparison. Usage template naming convention is U:hu:part of speech-unique identifier. There is one already: {{U:hu:postpositional-adjective}}.
8. Ez alatt vs. ezalatt: It would be helpful to point out the difference between the two types using example sentences at ezalatt (and for the rest, too). Obviously, this cannot be done in the table. I am still struggling with making this table a little clearer.
9. Last row: Remove közben. It's a duplicate. Add comma to separate the postpositions to add clarity. But it's still not clear why they are listed there. A short explanation would help.

Panda10 (talk) 16:29, 23 May 2020 (UTC)Reply

@Panda10

1. Thank you! I had {{number box}} in mind when creating these two templates. This number-box template is used by dozens of languages: English, German, French, Spanish, Dutch, Russian, Czech, Danish, Latin, Arabic, Ancient Greek, Old English, Esperanto, Ido, Swahili, Georgian, Sanskrit, Arabic, Finnish, Veps, Egyptian, Mongolian, Tibetan, Dzongkha, Hindi, Assami, Marathi, Gujarati and even more, aside from Hungarian. You can't say as a rule that the right-hand location is wrong! I noticed that some lines are broken on the left but actually very few lines are affected and even those few broken lines look all right to me. We can discuss the content, the arrangement, the layout and whatever else, but I think the right-hand location is the best place for this kind of navigational templates. They can add a lot more usability to Wiktionary, especially in cases when seemingly different things belong together in terms of grammar or meaning (like kettő and második or azon and rajta).
2. Him/her and it. I think there is a primary interpretation of these suffixed postpositions. If you want to express I won't leave without him/her/it in Hungarian with Nem megyek el nélküle or Nélküle nem megyek el, isn't the listener likely to assume that you insist on a person (or at least a living, sentient being for sure), rather than an object? Also, if you say szerinte, it cannot refer to a research paper, only to a living person. Please do give it some thought and let me know what you think. (Of course, if there is a particular context with an object, it may override this interpretation, like in your example.)
3. "not all items are postpositions" What do you mean? It's only the header of the first column, not the entire table. Aren't all items in the first column actually postpositions?
4. Yes, maybe some brief English translation would be good; I'll think it over.
5. Description: the header is meant to make things clear; what else do you think should be added?
6. Well, I can add this category, although I view it as a navigational template (that's why it's placed in the top right corner of the entry). To my mind, a usage template is something with full sentences, which belongs in the Usage notes section.
7. As above, I'd prefer not to rename it to make it look like a usage template. On the other hand, I could just remove the part "comp" and leave "postp" or "postpos" (or maybe "postposition"), although we could indicate somehow that it includes other derivations of postpositions as well. Do you have an idea how it could be phrased appropriately in the template name?
8. As a matter of fact, this point may well be vague for the majority of native Hungarians. See Section 121 in A magyar helyesírás szabályai, 12. kiadás (’The Rules of Hungarian Orthography, 12th edition’). Budapest: Akadémiai Kiadó, 2015. →ISBN and Osiris Helyesírás pp. 121–122. Yes, we could and should add examples for these terms in their own entries. You can also look up their entries in ÉrtSz. or Nagyszótár for example sentences.
9. OK, I'll remove közben. I made this row (1) to indicate that the above list is not comprehensive, offering a selection of the most important remaining postpositions, (2) because some existing categories could not be linked from the above list, and I didn't want to omit their links, and (3) because I didn't know where to insert the link to this Appendix and it seemed a waste of space to add an extra row solely for the link "All".

Adam78 (talk) 18:33, 23 May 2020 (UTC)Reply

1. I never said that "as a rule the right-hand location is wrong". I just said in mobile view it shows up on the top before anything else.
2. I don't have other arguments and it's OK with me if you leave the parentheses. Nem megyek el nélküle: this can be a book, a DVD, or any object. The understanding will depend on the context.
3. The first column contains postpositions but the rest are not. That's all I meant.
4. OK.
5. Compound postpositions? Or comparing compound postpositions to xxx? Some kind of a title above the first row. What was the reason you created this table? There must have been something that you thought this was important.
6. Navigational template is fine. Just add the Hungarian template category + some documentation.
7. I'm fine with the current name if it is a navigational template.
8. Thanks.
9. OK.

Panda10 (talk) 18:57, 23 May 2020 (UTC)Reply

@Panda10

• Indeed, I forgot about one thing, the syntactic role of this word. If you say Ülj le mellé, it cannot really refer to a box. But if you say, Leteszem mellé, it may imply a box too. If you say Ellene nem véd, it will probably imply a human, but if you say Ez sem véd meg ellene, it can imply a force or substance as well. If you say Rajta nevetünk, it'll imply a human, but if you say Nevetünk rajta, it can be a human as well as a thing. Maybe this feature should be mentioned concisely at the bottom.
• The reason for this table: because this group of words is so unintuitive for non-Hungarians and it even gives a headache to Hungarian professional grammarians. (You surely remember the problems and inconsistencies we encountered earlier about these words in dictionaries.) Maybe Pronominal adverbs from postpositions? Based on Category:German pronominal adverbs. I think "postpositions" need to be specified here because this group also includes adverbs deriving from case suffixes in Hungarian, comparable to English thereon, thereof, thereto etc., so it's just a subtype, and they're closely related to tőle, nála, hozzá etc. In fact, I think these two tables should link to each other somehow.

Adam78 (talk) 19:41, 23 May 2020 (UTC)Reply

I copied the letter names from Magyar ábécé (Hungarian Wikipedia). I also looked at Balázs Géza's article in the Külső hivatkozások section. So, there is now a discrepancy. I will change back elipszilon to a full entry because there are more hits in Google than with ellipszilon, especially in declined forms. I can use superseded spelling and give a reference. Panda10 (talk) 15:41, 31 May 2020 (UTC)Reply

@Panda10 Even Balázs Géza writes the following in the referenced article:

…. Talán szívesebben látná így: ellipszilon? (Nem, ezt is aláhúzza, bár a Nyelvművelő kézikönyv ezt a formát ajánlja.)

Thank you. Well, Wikipedia is not an ultimate source, and certainly less authoritative than the dictionaries published by the Academy. Anyway, this point will need to be corrected in Wikipedia. I think you should use "misspelling" instead of "superseded spelling", because A magyar nyelv értelmező szótára (published 1959–1962) consistently mentions these letter names in the long form, including ellipszilon, and the previous edition of the regulations, effective between 1984 and 2016, also has it with double l. OH. also has ellipszilon (and ejj) on page 1017, as well as Magyar helyesírási szólista, Hagyomány és újítás, and A szakmai nyelvművelés alapjai. This source only mentions ejj as its name but it doubles the consonant in every letter name in question. Adam78 (talk) 16:45, 31 May 2020 (UTC)Reply

Thanks for the corrections and the useful links. I will try to add the sources to the entries where appropriate. Panda10 (talk) 17:50, 31 May 2020 (UTC)Reply

According to Gábor Zaicz, there are four different etymologies:

1. From Proto-Uralic *-tt:
   • Noun-forming suffix (hit, lét)
   • Verb-forming suffix (instantaneous: ért, tilt; causative: kelt, teremt)
   • Past participle suffix
2. Of debated origin: Accusative suffix
3. From the past participle: The past tense suffix
4. Not given or not clear to me: Locative suffix

Do you have other sources besides Zaicz's etymology dictionary? Panda10 (talk) 20:24, 2 June 2020 (UTC)Reply

@Panda10 I don't have other sources. I think etymologies that produced different parts of speech (like the noun-forming and verb-forming descendants of Proto-Uralic *-tt) may well be treated separately for the sake of convenience, while of course it's good to provide references from one to the other ("from the same stem as…", like it is done by Zaicz). We should definitely treat them separately if they have different harmonic or other variants. Another reason for their separation is the fact that e.g. the accusative has several different senses in itself (length of time, e.g. két órát maradunk ott etc.), which will have to be listed sooner or later, although I suppose it's beyond doubt that the accusative is a separate matter. If you feel like filling in the etymologies based on Zaicz, I'll appreciate it. However, you don't need to repeat the same explanation for -t/-at/-ot/-et/-öt, you can just mention that the latter four are composed of a harmonic linking vowel plus the -t, and linking the latter. Adam78 (talk) 20:47, 2 June 2020 (UTC)Reply

Sure, I can add them. Do you have a preference for the order? I think we should keep the same order at each entry (main -t and linking vowel+t). E.g. first the accusative, then noun-forming, etc. Panda10 (talk) 20:59, 2 June 2020 (UTC)Reply

@Panda10 Originally I thought that the more common senses of each suffix should be presented first, and while it may be the accusative for -t, the case of -at is somewhat different as -at is the exception there for nouns. However, you may be right that we should keep the same order and -at is still pretty important in several other word forms. The accusative could be the first at each, and the rest could stay as it is now, leaving archaic, limited-use, or in other ways less productive, less typical, or less characteristic meanings to the end (e.g. Kaposvárt, hit/lét, teremt). Adam78 (talk) 21:09, 2 June 2020 (UTC)Reply

Mártonfi Attila: A magyar toldalékmorfémák rendszere lists these as compound suffixes. We could do the same and remove it from -at/-et. You have already mentioned it at -at (which I'm editing now) in this comment: "it's probably two distinct etymologies, but their category, linked below, is lumped together". Panda10 (talk) 18:13, 4 June 2020 (UTC)Reply

@Panda10 A very good idea, great, thank you, I support it. Adam78 (talk) 18:16, 4 June 2020 (UTC)Reply

I still don't know if the -at is the same as the noun-forming suffix. Panda10 (talk) 18:41, 4 June 2020 (UTC)Reply

@Panda10 I think we can treat it like that one, as it does form a noun, even if not from a verb but from another noun, and I don't see any semantic restrictions forbidding it from being that suffix. Adam78 (talk) 19:03, 4 June 2020 (UTC)Reply

I'm not sure about the following nouns: fodrászat, gyógyászat, horgászat, kohászat, orvvadászat, csempészet, művészet, régészet. Are they really suffixed with -ászat/-észet? It sounds more like -at/-et. Other than these, this is done. Panda10 (talk) 20:55, 4 June 2020 (UTC)Reply

@Panda10 I'd say yes, e.g. if a computer creates some kind of art, the computer or the software won't become an artist (művész), while the result may be a work of art (művészet). Also, healing might happen by means of objects, materials, or devices alone, which may be gyógyászat, without the involvement of a gyógyász. Or, that American student who stole a propaganda poster from North Korea might not be called a csempész, just a reckless tourist, but he was probably accused of csempészet. Will kohászat always necessarily involve a kohász? In short, the involvement of a human agent appears to be forced in several cases. For sure, it needs to be mentioned that it's a compound suffix, but I think this indication is enough. Adam78 (talk) 21:17, 4 June 2020 (UTC)Reply

Hey, many thanks for your work on Hungarian rhymes, it looks great! I have a question because I want to avoid making a mistake when adding rhyme information to entries. I see you have created pages for a number of two-syllable rhymes (Rhymes:Hungarian/ɒmlik, Rhymes:Hungarian/ɛksik etc.). Are these only used for -ik verbs? If so, is there any phonetic law that causes them to be treated differently than other words (második vs álmodik)? – Einstein2 (talk) 10:06, 9 June 2020 (UTC)Reply

@Einstein2 A very good question. Yes, I only intended them for -ik verbs, although I failed to indicate it (an omission for sure). The reason why I wanted to keep them separate from other (noun, numeral, pronoun etc.) forms is that with the exception of this form, they behave identically to verbs that have the ik-less stem (nyomkod, csapkod etc.), except for first-person singular present-tense forms like álmodom vs. csapkodok. So basically I intended them (álmodik etc.) as sub-pages of the latter pages (nyomkod etc.). As you can see, the page that has második mostly has nouns, adjectives, and numerals, rather than verbs, and thus a meaningful comparison becomes possible across inflections as opposed to conjugations. (Similarly to the distinction between e.g. verbs and numerals suffixed with -ad, rather than lumping them all in the same category.) In this way the rhyme pages become more expedient. At least this is my idea. It should be clarified indeed. @Panda10, any idea? Adam78 (talk) 10:27, 9 June 2020 (UTC)Reply

I understand. It is indeed a good idea to differentiate the -ik verbs. Maybe there could be a short explanation and a list of all the relevant rhyme lists on Rhymes:Hungarian/ik, as well as links to the -ik pages on the respective -ikless pages. – Einstein2 (talk) 11:32, 9 June 2020 (UTC)Reply

@Einstein2 Great idea, I've done some part of it already. Adam78 (talk) 13:22, 9 June 2020 (UTC)Reply

Looks good to me. Panda10 (talk) 15:43, 9 June 2020 (UTC)Reply

I've noticed that you made some changes in the conjugation templates. There is an extra space below the conjugation table which is usually caused by an extra space in the wiki code. I removed a couple of them, but it is still there. See szemerkél. Would you take a look at your edits and remove the extra space? Thanks. Panda10 (talk) 19:24, 9 July 2020 (UTC)Reply

Also csepp. I did remove a space from {{hu-infl-nom}} but there is still something there. Panda10 (talk) 19:32, 9 July 2020 (UTC)Reply

@Panda10 Thank you; I've fixed both. You should be able to see the difference, in the worst case by appending ?action=purge to the requested URL. Adam78 (talk) 19:38, 9 July 2020 (UTC)Reply

In this case ön would mean not self but you as a formal address. He went to the costume party as you? (dressed as you, pretending that it was you). Panda10 (talk) 14:30, 13 July 2020 (UTC)Reply

@Panda10 I find it rather hard to imagine (I'd say önnek öltözve) and contrived to suppose, partly due to the strongly present other meaning of the word, which would have to be avoided. But if there is some actual source for this usage, I'd say yes, let's include it. Adam78 (talk) 14:40, 13 July 2020 (UTC)Reply

E.g.: veszett fejsze nyele, the template was changed from {{head|hu|phrase}} to {{head|hu|genitive noun phrase}}. A change like that will remove the phrase from the lemma category. Are you planning to remove phrases from Hungarian lemmas? Panda10 (talk) 14:12, 12 August 2020 (UTC)Reply

@Panda10, no I didn't mean that, sorry. In this case I think we'll have to add them to lemma categories manually, one by one. I wouldn't like to leave them in the catch-all "…… phrases" category because it was really a mess. On the other hand, I'm open to your ideas to handle this problem. Adam78 (talk) 14:16, 12 August 2020 (UTC)Reply

I'd leave the current head template as is and would add {{cln|hu|genitive noun phrases}} at the bottom. This will keep the entry in the lemma category, in the general phrase category and in the new genitive noun phrase category. We do have other cases where entries belong to both a general category and a more specific subcategory. Another solution is to request an update to the Module:headword/data to recognize the new parameters as lemmas. Panda10 (talk) 14:28, 12 August 2020 (UTC)Reply

@Panda10 I admit that there are things in common between various different items in this category, which may call for a category of its own, but I doubt there is really something common between ALL of them, e.g. "veszett fejsze nyele" and "jó napot kívánok" are phrases in rather different senses of the word. Maybe if we could remove elements of the "phrasebook" category (NB, only these have inflected verb forms) out of "phrases", then the rest might be something coherent and meaningful. (I'm not sure yet, but it looks like a sine qua non condition.) What do you think? – For example, van is rather different in a kocka el van vetve and ma rossz idő van. Putting these in the same category looks kind of unreasonable to me, to say the least. On the other hand, I see that elements of the category "phrasebook" are automatically placed in "phrases" so if we want to keep separate things separate, then all we can do is remove the non-phrasebook items from phrases, which is actually what I did. I'm clueless… @Einstein2, your suggestions would be welcome as well. Adam78 (talk) 14:57, 12 August 2020 (UTC)Reply

We could use {{head|hu|phrasebook}} as a head template for phrasebook items. This is what I used in the past but MewBot went through all of them and changed it to {{head|hu|phrase}}. I have no idea why. Phrasebook is a valid parameter for {{head}}. This could be a Beer parlour question because you don't want to put all the work in just to see a bot change it back again. Panda10 (talk) 15:10, 12 August 2020 (UTC)Reply

@Panda10 OK, it sounds like a good idea. Would you mind raising the issue there? After all, you know first-hand what happened. Or would you prefer me to do so? Adam78 (talk) 15:37, 12 August 2020 (UTC)Reply

I will submit a question. Panda10 (talk) 15:40, 12 August 2020 (UTC)Reply

One more question: I tested the phrasebook parameter. It doesn't put the phrase into the lemma category. It will be only in the phrasebook category. Is that ok? Panda10 (talk) 15:52, 12 August 2020 (UTC)Reply

@Panda10 Yes, I think it's OK. If something is a lemma on its own right, we can still add it to "phrases" or one of its subcategories, but I don't see why e.g. "nem beszélek angolul" should be considered a lemma. Adam78 (talk) 16:02, 12 August 2020 (UTC)Reply

I thought the correct English translation of ikerszók is coordinated pairs and not dvanda compounds. We should have a category for ikerszók, the dvanda category contains too many other types (like mellérendelő összetétel such as búbánat). It's fine to build new categories but is it really necessary to empty out the existing ones? Panda10 (talk) 18:22, 13 August 2020 (UTC)Reply

Actually, I think that ikerszók is better translated as reduplications (Category:Hungarian reduplications) and valódi mellérendelő szóösszetételek as coordinated pairs, szóismétlések as reduplicated coordinated pairs. Would you agree? Panda10 (talk) 18:53, 13 August 2020 (UTC)Reply

(edit conflict) @Panda10

• Please look at the description of "coordinated pairs": "Terms in Hungarian consisting of a pair of terms joined by a coordinating conjunction." Hébe-hóba has no conjunction, while dvandva compounds do. This is one reason.
• Please also look at Category:English coordinated pairs: almost all of them contain a conjunction.
• The other reason why I wanted to remove terms from coordinate pairs is exactly the reason you brought up: it already contains different types dumped together.
• Please look at the etymology of hébe-hóba (maybe you wrote it yourself), it is practically hőben-hóban, just like the other dvandva compounds that are already included. Please also look at this source for more examples in Hungarian and distinctions from similar types (search for the term "dvandva").
• I don't know which dictionary you use; I happen to use (and basically trust) Angol egyetemes nagyszótár created by two professors at ELTE, and it supplies "doublet; reduplicated form" as a translation for "ikerszó", not "coordinated pair". "Doublet" seems a bit too loose; however, "reduplication" is quite close, compare w:reduplication.
• If we do want to exclude historical linguistics and prefer to focus on synchronic linguistics exclusively, then hébe-hóba may look like an ikerszó (similar to tyúkom-búkom or mendemonda), but I still don't very much like this categorization of this particular term because if you look up ikerszó (for example in A magyar nyelv könyve by A. Jászó, Anna , pp. 301–302), you'll see that it mostly comprises onomatopoeic words (see also its Hungarian Wikipedia article).
• Emptying out existing categories: If you mean the category "phrases", you already suggested two ways how it could be solved, and I'm open to both, just please let's wait until we get a reply at Beer Parlour. (By the way, you may have noticed that I already put about a dozen pages back to "phrases".)
• I'm sorry I don't usually provide detailed explanations in advance, which you may find annoying.

Your later suggestions look good at first glance; we'll see how they fit the particular cases. Adam78 (talk) 19:04, 13 August 2020 (UTC)Reply

PS: we can still split up dvandva compounds into subcategories if we do find it necessary. Adam78 (talk) 19:09, 13 August 2020 (UTC)Reply

Isn't app a clipping instead of an apocopic form? Panda10 (talk) 18:03, 8 September 2020 (UTC)Reply

@Panda10 According to the relevant Wikipedia articles, clipping seemed to be a broader term, implying omission from the beginning (like plane for airplane or bus for omnibus), both ends (like flu for influenza), the middle (like o’er for over), or the end, and this latter subtype is called apocope, see w:Apocope (especially the section "Informal speech"), w:List of English apocopations, and hu:w:Apokopé, especially the examples taken from Szathmári's book. On the other hand, I noticed that most (almost all) similar cases in English are classified under clippings. Also, Nyelvi fogalmak kisszótára lists these cases under Szórövidülés (although it doesn't mention apokopé, but then again, it's a fairly small dictionary). I don't have anything against recategorizing these Hungarian terms under clippings (and bizalom from bizodalom, győzelem from győzedelem could still be separated via the subcategory of syncope) but in this case we should move all such terms there, so the category of apocopic forms could be deleted. Do you agree? Do you think we should do that? – I don't mind either way. Adam78 (talk) 19:11, 8 September 2020 (UTC)Reply

No, we can leave it as is. I didn't realize that clipping was a broader term. The glossary could be better. Thanks for the clarification and the detailed information. Panda10 (talk) 19:24, 8 September 2020 (UTC)Reply

Am I doing this correctly when I add a rhyme that starts with a long consonant such as in ablakká? Panda10 (talk) 20:33, 10 September 2020 (UTC)Reply

@Panda10 Actually, I don't think we should add inflected forms, unless they are single-syllable (or perhaps disyllabic) words and/or they are the only cases when a particular rhyme occurs (the reason why I added forms like döntsd). Also, if such words need to be added, I think we'd better categorize them by their last syllable, which can only start with one single short consonant (unless it's a monosyllabic word, of course).

In fact, it turned out to me after I started working on rhyme pages that the intended rhyme categorization should always start with a vowel, even if the whole word ends in a vowel, so e.g. róka should be placed under a rhyme like oːkɒ. However, it didn't seem convenient or practical to me (as essentially similar words could not be accessed from each other) and I think I found at least one counter-example among rhyme pages of various languages, so I suppose we can keep the existing consonant + vowel rhymes like kɒ as they are.

In short, I don't think syllable-initial long consonants should be useful or practical (but if you give me a good reason what unique information they can convey, I can be convinced), especially if it's not a lemma and not a single-syllable word (within which latter type I'd like to aim at comprehensiveness). Adam78 (talk) 20:49, 10 September 2020 (UTC)Reply

I see that at többé the rhyme is . So would words like többé and karalábé really rhyme? As for not adding rhymes to inflected forms: Most nouns/verbs in poems would be inflected. I'm not adding them to the rhyme page, I just add the {{rhymes}} template to the form entry. About starting the rhyme with a vowel: You can still do it. One word could be added to multiple rhyme pages. English has several multi-syllable rhyme pages. Panda10 (talk) 21:23, 10 September 2020 (UTC)Reply

@Panda10 Sorry, I think you're right, adding the rhymes template to a form entry may be all right. Also, adding one word to multiple rhyme pages; why not. – On the other hand, you need to consider that English multi-syllable rhyme pages are based on the stressed vowel, which may be the very last or the second or the third from the back, varying with each word (unlike in Hungarian), so we cannot follow this principle anyway, we can make our own guidelines about how many syllables and initial consonants to include in a rhyme.

Concerning többé and karalábé – well, I can't really tell what's the right thing to do. I just thought rhymes are closely linked to syllabification (independently of hyphenation), but if you do advise so, we can disregard syllabification in this case. Adam78 (talk) 21:44, 10 September 2020 (UTC)Reply

I don't think I know enough about the rules of Hungarian rhyme. Looking into poems, it is more complicated. In Tóth Árpád: Egy lány a villamosban: Utasok - olvasott, könyvét - könnyét, tisztelt - minisztert, hadak - csak, lélek - élet. There are examples for the többé / karalábé type of rhyme: "Várj csak, hogy is kezdjem, hogy magyarázzam? / Te ismered a házam" (Kosztolányi: Hajnali részegség). So you were right, the correct rhyme in ablakká is . Thanks. Panda10 (talk) 16:12, 11 September 2020 (UTC)Reply

I still can't see how holta is the same derivation as járta and tudta which are in Category:Hungarian nouns suffixed with -ta. While there is a noun holt which can have a possessive derivation, the nouns *járt and *tudt do not exist, so instead the verb is suffixed with -ta. Have you had a chance to look at the entry holt in Tótfalusi's etymology dictionary? To me it clearly states that holta is from the noun holt. Panda10 (talk) 21:08, 17 September 2020 (UTC)Reply

@Panda10 It is pretty rare if a noun suffixed from a verb with -t (like hit) can be used without a possessive suffix; the existing other examples are rather scarce (ét, hit). I looked into Tótfalusi but it didn't give me much news and I don't see how it should disprove what I'm saying. Tótfalusi doesn't give examples, neither historical, nor contemporary ones. On the other hand, if you look up Bárczi for the definition of "holt" as a noun in the sense of "death", you'll see it hardly exists, except in the two fixed forms holtig and holtra (and of course all the possessive forms). Have you encountered forms like holtot, holtnak, holttal, holttól, holthoz, holtban etc. in the sense of "(to/from/with/in/etc.) death"? You can also look up Czuczor, who describes: "2) Halál v. meghalás. Ez esetben a ragok o-val járulnak hozzá holtom, holtod; tárgyesete volna: holtot, de mint rendszerént az ily fajta ige-neveknél nincs szokásban." "Holt" in this sense is a barely existing, transitional, intermediate form, practically only inferred as a construct based on two peripheral forms and of course the widespread possessive forms. The analogy with similar words (tudta etc.) is so clear, with the only exception of o vs. a in the stem, which happens over the course of several centuries. Why do you insist on forcing such a form into being despite all the available historical evidence and going against all those numerous analogies that would let it neatly fit into an extensive system? Adam78 (talk) 21:52, 17 September 2020 (UTC)Reply

@Panda10 I came across another example where Tótfalusi's wording might be misleading: if you look up menten, you might get the impression that ment is an existing noun in the sense of "going". However, it appears to be more of an intermediate, potential form, so the dictionary form should be mente instead.

I've made a compilation of such nouns (only those with a deverbal sense, some only with inflections or in compounds, although the list is far from exhaustive): áldozta​/​feláldoztával, állta​/​beállta, aludtában, bukta​/​buktában, dőlte, (épülte) megépülte, este/eleste, (fejezte) befejezte, fekte, félte/féltében, fogyta/fogytán, folytán​/​egyfolytában, (fordulta) postafordultával, futtában​/​kutyafuttában, hallata/hallatán, hasadta, (hunyta) elhunyta, ijedtében, járta​/​lejárta​/​jártányi​/​ottjártakor, jötte​/​feljötte​/​közbejötte​/​létrejötte, kellete, kelte​/​felkelte​/​járta-kelte, készülte​/​elkészülte, következtében, közeledte/közeledtével, (küldte) visszaküldtével, látta​/​láttán, lépte​/​fellépte, mente​/​lemente, múlta​/​elmúlta, nevettében, nyugta​/​nyugtával, repte/röpte/reptében, siettében, (szakadta) inaszakadtáig​/​torkaszakadtából​/​körmeszakadtáig​/​körömszakadtáig, szállta/szálltában, szánta​/​jószántából​/​önszántából, szűnte/megszűnte, telte​/​tölte​/​eltelte​/​letelte​/​(hold)tölte, (térte) visszatérte, (történte) megtörténte, tudta​/​tudtával, (tűnte) eltűnte/letűnte, ülte​/​elülte​/​ültében, végezte​/​végeztével, veszte​/​vesztébe, virradta, volta​/​jóvoltából, zárta​/​lezárta​/​lapzárta, (zárulta) lezárulta. It would be good to treat this group consistently (e.g. nouns rather than noun forms, with possessive forms on their own), and holta could fit here, both semantically and morphologically. On the other hand, there are barely ten examples for those that can exist without a possessive suffix. Of course, we could sacrifice this system in this case and add this term to the latter group if this word is attested in this form (without possessive suffixes) and in this sense (as a deverbal noun). But all we have is holtig (and possibly holtra, if it derives from the same sense), which look more like isolated forms. Adam78 (talk) 20:31, 18 September 2020 (UTC)Reply

I see you updated the entry. So let's look at the consistent representation of verbal participles (igei igenév) and the verbal nouns. Maybe this is something that should be part of Wiktionary:About Hungarian. The page is not viewed too often, though.

1. For verbal participles: Use {{hu-participle}} under the ===Verb=== header. See járta as an example. The template will place the entry in Category:Hungarian verbal participles.
2. For verbal nouns: Use {{verbal noun of|hu}} under the ===Noun=== header. See járta as an example. The template will place the entry in Category:Hungarian verbal nouns.

   Comment: I find it somewhat confusing that we call both járás and járta a verbal noun and they are collected in the same category. Also, to me it's a little odd that the template is the first element of the definition but I know the English entries use it the same way.

3. Etymology for verbal participles and verbal nouns: At járta, the current etymology for verbal participle is the same as for the third-person singular indicative past definite. I'm not sure this is correct. Zaicz says that -a/-e is "egyes szám 3. személyű igei személyrag. Az -a/-e, -ja/-je birtokos személyjellel azonos eredetű toldalék." So in the case of the past definite, -t + -a are past-tense suffix + personal suffix. In the case of the verbal participle, -t + -a are past-participle suffix + possessive suffix. According to Magyar Grammatika:

   p.235 Igei igenevet csupán tárgyas cselekvő igéből hozhatunk létre.

   p.236: Az igei igenevek a -t/-tt képzős melléknévi igenevek személyragos formáinak tekinthetők.

   p. 247 Az úgynevezett igei igenevet a -t/-tt melléknévi igenévképzővel hozzuk létre, és a melléknévi igenév sajátos alcsoportjának tekintjük.

   p.249 Ha az igei igenév alanya a magam, magad, maga stb. személyes névmás, az igenév személyragozásában nincs személybeli egyeztetés: a magam sütötte kalács, a magad sütötte kalács, a maga sütötte kalács.

4. Question: Is the -ta/-te suffix really a "noun-forming suffix"? Or the noun sense of járta was derived from the verbal participle?
5. Comment: The -t is currently indicated as noun-forming suffix in tudta. Based on the above, this may not be correct.

Panda10 (talk) 18:02, 25 September 2020 (UTC)Reply

@Panda10 Thank you for the systematic approach. If you don't mind, I'll start at the end.

• On 5: I think we'd best consider this -ta as a compound suffix (you may have seen that I added it to their list earlier). By this I mean that I agree, we shouldn't separate the t and the a/e in the etymology section of words. Sometimes this analysis is possible (e.g. járta) and sometimes it is not (e.g. fekte, röpte, nyugta). We shouldn't give different etymologies to words that have the same morphological makeup and the same kind of semantic derivation, only because one group can have an additional analysis and the other group cannot. In short, I think we should provide its etymology as tud + -ta and of course the entry of -ta will give more information to those interested.
• On 4: I think we have to consider -ta/-te as a noun-forming suffix no matter what it happens to be derived from, on the grounds that such derivations behave as nouns (which is how linguistics tends to work, at least since Bloomfield). What's more, it's fairly productive, so it's not something ad hoc. For example, I randomly coined a word not included in the list above, vizsgáltakor, and I got several results for it in Google (even if some of them might be misspellings of vizsgálatakor).
• On 3: Based on your references, the right solutions seems to be that we separate past-tense forms from verbal participles, since the latter group doesn't have a past-tense element in them (provided that we are positive that the functions of the two -t suffixes is different, if I understand your quote from Magyar grammatika correctly). Apparently, we should treat the verbal participle and the verbal noun under the same (second) etymology. In other words, Etymology 1 should contain only the verb etymology while Etymology 2 could comprise both the verbal participle and the verbal noun. What do you think about this?
  • We should keep in mind that the above statement (no past-tense element in the participle-forming suffix) also applies to indefinite past-tense forms, so e.g. látott should also have two different etymologies on the same basis. This split might be supported by the fact that sometimes the past-tense form has a shorter form than the past-participle form (e.g. font vs. fonott, tört vs. törött, varrt vs. varrott, even if their overlap is considerably more common).
  • Also, this could be a way for us to place past participles (like látott) together with the adjectives derived from them, since both can have a noun-like declension but the past-tense verb form cannot. In fact, it's often difficult to tell whether an actual adjective have come into being or it's still a participle only. (Sometimes it takes syntactic criteria, e.g. in a tőle elvont javak is a participle but in elvont jelentés it's an adjective: only the former can take the arguments of the verb and only the latter can have comparison.) – Do you think we should make a similar split in giving their etymologies?
• For verbal nouns, I think it's good to keep in mind that it's basically a semantic category, so it doesn't always align with morphology. On the one hand, it means that it can have more than one forms: not only -ás/-és and -ta/-te but also certain cases of -tal/-tel, like jövetel, hozatal, tétel, vitel, but obviously not étel). On the other hand, it also means that it will apply to one sense of a term and not to the others, like in the case of írás (or tojás in ne zavard a tyúkot tojás közben), in which both cases the result of the action might be named deverbal nouns (a piece of writing and an egg). As we saw in the case of holta, being synonymous with meghalása, elhunyása, and as we know how vevés and vétel etc. can be used as synonyms in most cases (kézhez ~, semmibe ~, alapul ~, bérbe ~, birtokba ~, használatba ~ etc.), I think we can safely claim that all these three suffixes create actual verbal nouns (as opposed to épület or festmény, which are deverbal nouns, as is mostly the noun döntés, unless you want to say something like döntés közben mindig ráncolja a homlokát, if it's possible, which would be a verbal noun sense again). In short, the verbal noun can't be used in parallel to participles, which latter type has (I think) a fixed form and a fixed semantic relation with the original verb.
• In accordance with the above conclusions (if you agree with them, though you don't have to, of course), apparently we should put the participle and the verbal noun under the same etymology, separate from the verb etymology. It also means that participles don't need to be categorized together with verbs anymore, or even placed under the ===Verb=== header. In fact, if you look up Entry layout, you'll see that "Participle" as such is listed among the allowed parts of speech. I noticed this is the case with German in many cases (cf. getan, gebaut, gegessen, gegangen, gesagt, gemacht, geglaubt, though not consistently), different from the case with the corresponding English and French forms. (The reason why they're often placed under "Verb" may be that many other meanings do belong to the verb sense, like in the case of untersucht or English done. Nevertheless, Category:German past participles is still a subcategory of German verb forms.) We might consider using this header, too, if we need to separate it from the verb sense anyway. Of course, the noun form (if it exists) would be placed under a different header.

I'm looking forward to your comments. Adam78 (talk) 20:39, 25 September 2020 (UTC)Reply

The more I read about this, the more confusing it becomes. Linguists do not agree about the structure of -ta/-te. There are studies about the verbal participle, but nothing about the corresponding verbal noun. If we can't accurately identify the elements of the ending, it will be just our personal guess without supporting documentation. I really don't want to do that. Just a few examples:

• Jakab István: Személyjel alakú képzők a magyar nyelvben, Magyar Nyelvőr 111, 276–278, 1987

  Minél jobban belemélyedtem az ún. igenévi-igei alakulások személyragozásának vizsgálgatásába, annál erősebbé vált az a meggyőződésem, hogy — a rokon nyelvi példák bármit sugallnak is — a magyarban nem névszói személyjelezésről van szó, hanem a tárgyas igeragozású alakok melléknévi használatáról.

• A.JÁSZÓ ANNA: Megjegyzések az „igenevek”-hez

  Az isten adta gyermek szerkezet egyféleképpen magyarázható: ’isten adása gyermek’, tehát az igenév nomen actionis jelentésével. A szerkezetben személyragos -t képzős igenév van. Ezért nem fogadtam el az ige-igenév terminust, használatát következetesen kerültem. Külön kategóriát semmiképpen sem alkotnék belőle egy szófaji rendszerezésben (vö. Magyar leíró nyelvtani segéd-könyv 44), hiszen — mint láttuk — elég sok ide-oda billegő igenév van (a holta esetében beszélhetnénk főnév-igenévről, a jövet esetében határozószó-igenévről, s ha elfogadjuk VELCSOVNÉ megállapítását a -ván/-vén alkalmankénti külön életéről, akkor ez is ige-igenév volna).

• Nádasdi Péter: Egy vonatkozó szerkezet a nyelvemlékek tükrében

  Ugyanakkor az is empirikus tény, hogy a vizsgált szerkezetünkben a -ja/-je sohasem fordulhatott elő. Míg a birtokos szerkezetben mind a -j-s, mind pedig a -j- nélküli személyjel megjelenhet, l. én atyámnak átkoztai (Érdy-kódex: 6b) és a szépasszony főztje (Simonyi 1907: 199), ezzel szemben a vizsgált szerkezetben nem lehet -j-s változat, l. ravasz átkozta szél és a szépasszony főzte húsleves.

So I'm sorry but now I'm undecided about how to handle this phenomenon. Since the changes you recommended above would require a lot of work, we better think about it twice. Maybe we need more time and research. Panda10 (talk) 20:37, 26 September 2020 (UTC)Reply

@Panda10 On the quotes from István Jakab and Péter Nádasdi: they don't really belong here because the forms that they speak about work like adjectives, not like nouns. (The only analogous example is szépasszony főztje, although this -j- is atypical.) We have to consider what kind of suffixes they can take: holtomat, tudtommal, elmúltával, vizsgáltakor etc. definitely indicate nouns (especially the possessive, which practically excludes the adjective sense). You can also check how you can paraphrase them: halálomat, tudomásommal, elmúlásával, vizsgálatakor – nouns. The morphological and the semantic aspects lead to the same result.

In a sense, I can agree with A. Jászó that we can speak of főnévi igenév, just like in the case of -ás/-és nouns and the other types I mentioned (vitel etc.). After all, in terms of their origin, they derive from verbs, and in terms of their function, they act like nouns. Therefore, főnévi igenév may be an apt term in a broader sense. The current categorization just puts somewhat more emphasis on the fact that they have become nouns, and that's why we've been categorizing them as verbal nouns. I suppose you agree that “noun-like verb-derivative” (főnévi igenév) and verbal noun is not an enormous difference.

A. Jászó's other statements (on jövet and -ván/-vén) are intriguing but luckily don't really concern our topic, where the terms act as nouns. Two authors of Magyar grammatika, Klára Lengyel (p. 89) and Borbála Keszler (pp. 315 and 317) equally speak of noun-forming suffixes when describing these -ta/-te forms (and I didn't find any other reference to it in the volume), so I think we can be assured of their assessment. A. Jászó used a different wording and a different approach (that of the form, rather than that of the function) but her conclusion on these terms was still basically the same. Adam78 (talk) 22:14, 26 September 2020 (UTC)Reply

I'm sorry, I think I was wrong in my conclusion. Based on what I found in Zaicz (2006), especially in the entries of ‑t ~ -tt¹ and ‑t¹, it still seems that we could leave (for definite verb forms with -ta/-te) the past-tense form, the verbal participle, and the noun under the same etymology (in practice, often only one out of the latter two exists, either the verbal participle like vizsgálta or the noun like tudta), and (for indefinite verb forms with -t/ott/ett/ött) the past-tense form, the past participle, and the adjective (sometimes there is a noun as well).

I also found an online version of A magyar nyelvtörténet kézikönyve (see the tab Digitális tananyag), the relevant parts seem to be 3.2. Morfématörténet and 3.3. Szófajtörténet.

• pp. 145–146. in the former: Az állapot és az állapothoz vezető út fonódik egybe a befejezett melléknévi igenév és a -t jeles múlt esetében. Láttuk: ez a múlt idő az igenévből alakult ki, állítmányi helyzetben. A két grammatikai jelenség között máig szoros a kapcsolat. Szövegkörnyezet híján például „A hal főtt” mondatban nem tudjuk megállapítani, hogy a főtt múlt idejű ige-e vagy befejezett melléknévi igenév – de ennek a hétköznapi kommunikációban nincs is igazán (vagy csak ritkán van) jelentősége. Bár kontextus nélkül eldönthetetlen, hogy állapotról vagy az oda vezető múlt idejű cselekvésről van-e szó, a jelentések között olyan szoros a kapcsolat, hogy az alaki egybeesés nem zavarja a megértést.
• p. 157: Bizonyosan az alapnyelvre csak egyetlen igeneves szerkezet vezethető vissza, ez a (madár)látta (kenyér)-szerkezet (tagolása: igető + igenévképző + személyjel) (…) Mindkét szerkezetben a -t igenévképzőt találjuk, ez tehát a legrégebbi (noha a rokon nyelvi funkcionális megfelelések -m képzőt tartalmaznak). Ez a -t képző jelenik meg a holta-típusú igenévben is, melyben a személyjel miatt a főnévi jelleg domborodik ki (vö. HB. hadlaua choltat).
• as a side-note, an interesting bit on the same page: A -ni képzős igenév („főnévi igenév”) az ősmagyar kor terméke: egy nomen actionis szerepű (a mai -ás/-és-nek megfelelő) -n nomenképző és -i latívuszrag összekapcsolásával alakult. A nomenképző a főnévi jelleget hangsúlyozza, a latívuszrag viszont a határozói szerepet.

So on the whole, maybe no such fundamental change is needed, after all. Adam78 (talk) 16:20, 27 September 2020 (UTC)Reply

Okay, thanks for the link. So how should we present the etymology of -ta/-te? (In my previous comment I called it the "structure" of -ta/-te.) It doesn't sound like it's the same for the past-tense suffix and the verbal participle. The noun-forming suffix is derived from the verbal participle. Panda10 (talk) 16:57, 27 September 2020 (UTC)Reply

Hello Adam78/archive1,

Really sorry for the inconvenience. This is a gentle note to request that you check your email. We sent you a message titled "The Community Insights survey is coming!". If you have questions, email [email protected].

You can see my explanation here.

MediaWiki message delivery (talk) 18:48, 25 September 2020 (UTC)Reply

Moved to Wiktionary:About Hungarian/Participles. Panda10 (talk) 16:08, 1 October 2020 (UTC)Reply

When you click on the -a/-e possessive suffix link in an entry where you previously added the id2 parameter, does it jump to the correct place? When I test this, it jumps to the correct place for a fraction of a second and immediately jumps down two or three times and I have to scroll up to see the blue-highlighted section. Panda10 (talk) 17:21, 10 October 2020 (UTC)Reply

@Panda10 I've never had this experience. For me, it's always worked fine. Adam78 (talk) 17:59, 10 October 2020 (UTC)Reply

I've just tested on an Android device and it works fine there but on Apple devices (both mobile and desktop) it makes these extra jumps. Panda10 (talk) 20:22, 10 October 2020 (UTC)Reply

@Panda10 Do you think we can still make use of senseids? I'm planning to ask a bot owner to insert "id" parameters into a specific list of words that I prepare, sorted by the sense (see e.g. the different senses of -ik or -ok). Do you have anything against it? (Have you managed to solve this problem since?)

We might opt for inserting the given "id" parameter only for those that do not use the given suffix in the first or primary (majority) sense. For example, there are some 200 words using -ok in the verb suffix sense, as opposed to some 3,800 forms using it as a noun suffix, so the latter could be viewed as "default". However, I'd prefer adding "id" to all senses that are specifically defined in the entry, as someone might change the order of senses later in the future for whatever reason and there would be less distraction with a precise link. Adam78 (talk) 00:35, 16 October 2020 (UTC)Reply

Sure, we can use senseid, go ahead with your plan to add it to all senses. I will ask others about the issue. It might be just my computer. Panda10 (talk) 16:23, 16 October 2020 (UTC)Reply

Hi. When you add something like rövidesen/hamarosan bekövetkező it creates a red link for that slashed phrase. Shouldn't you add two separate translations? Equinox ◑ 18:47, 29 January 2021 (UTC)Reply

@Equinox I invariably do that within seconds; I just don't want to stop my workflow (since some translations consist of a single term, others, of multiple ones), and I think it also makes it easier for others to see my changes in the page history (this way I don't need to describe it in my next edit summary when I only add brackets, a few extra bytes), even if these temporary red links are misleading, I admit. Adam78 (talk) 18:52, 29 January 2021 (UTC)Reply

Hi. Adding 'matter' under 'chemistry' triggered the creation of a ton of new categories for random languages. I'm not sure this is the correct thing to do; it's not obvious to me that matter and chemistry have much to do with each other. If there are subcategories of 'matter' that relate to chemistry, IMO you should add those subcategories directly to 'chemistry'. Benwing2 (talk) 20:17, 20 February 2021 (UTC)Reply

@Benwing2 Hi. It's an interesting point. If we only look at the name of the category, "Matter", one might say it belongs to physics just as much as to chemistry, but if we give it a closer look, with respect to the actual scope of its present subcategories, I think it can be seen that they all belong to chemistry (perhaps with the exception of "Subatomic particles‎", but this latter actually belongs to "Physics" more than "Matter"). These are the subcategories in question: Acids, Chemical elements‎, Drugs, Dyes, Explosives, Gases, Inorganic compounds, Ions, Liquids, Metals, Minerals, Natural resources, Organic compounds, and Poisons.

To your idea of adding certain subcategories directly to "Chemistry": in that case I'm afraid "Matter" itself would become partly redundant and somewhat pointless, and the distinction between the categories directly included in "chemistry" and those that are not may be fairly arbitrary. On the whole, treating "Matter" as a meaningful unit on its own still seems more feasible. Do you think it would be worth discussing this question with others at Beer parlour? In fact, I'd suggest that we delineate the category "Matter" more in accordance with its current content, which has a great deal of overlap with chemistry. Adam78 (talk) 21:14, 20 February 2021 (UTC)Reply

Yes, I think we should bring this up on the Beer Parlour. "Matter" on the whole sounds very vague and it seems very strange e.g. to put "Drugs" under "Matter". Maybe getting rid of it and moving those categories above to "Chemistry" and putting "Subatomic particles" directly under "Physics" is the right thing to do. Benwing2 (talk) 21:30, 20 February 2021 (UTC)Reply

I created an entry for the Hungarian noun form inged in order to create an entry for the Maranao noun. --Apisite (talk) 02:35, 21 February 2021 (UTC)Reply

@Apisite Thank you! :) Adam78 (talk) 02:58, 21 February 2021 (UTC)Reply

How could the word triolet be translated into Hungarian? --Apisite (talk) 17:16, 24 May 2021 (UTC)Reply

@Apisite I've added it to the translation box in the English entry. Adam78 (talk) 17:19, 24 May 2021 (UTC)Reply

@Adam78: Thanks, now I added epigramma. --Apisite (talk) 01:38, 25 May 2021 (UTC)Reply

@Adam78: And triolett, too. --Apisite (talk) 07:35, 25 May 2021 (UTC)Reply

Now how about sestina? --Apisite (talk) 15:35, 25 May 2021 (UTC)Reply

@Apisite Thank you. I've added its translation. Adam78 (talk) 14:22, 27 May 2021 (UTC)Reply

@Adam78: I wonder, why sestina rather than *szestina? Anyway, thanks. --Apisite (talk) 21:45, 27 May 2021 (UTC)Reply

@Apisite Because it's treated as a foreignism in this case (rather than an established loanword), so the s is to be pronounced as sz in Hungarian. (The new orthography is fairly permissive about such cases, e.g. chat and cset are both OK.) However, I've found occurrences of szesztina as well. On the other hand, Szestina only seems to occur as a surname. Adam78 (talk) 07:06, 28 May 2021 (UTC)Reply

I think the Arcanum version is the same as Szókincsháló.hu/szotar. For that, there is an existing template {{R:TotfalusiNagyEty 2001}} but the Szókincsháló website no longer works. At least it did not work today. We might have to switch to Arcanum. Lots of words are using the Szókincsháló site. Panda10 (talk) 21:47, 28 May 2021 (UTC)Reply

@Panda10 I think it could still be used via Internet Archive, but unfortunately not directly, but only if a table and a module are created for it (like the ones for {{R:ErtSz}}), containing the codes for the terms, e.g. 11645 for triolett. Creating this database is not much work (saving the list of terms as HTML pages for each letter of the alphabet, then converting it) but the mismatch between PAGENAME and PAGENAME+HomonymCode should be handled (e.g. tűz1 and tűz2 on the T page). Another method could be extracting the list of codes from Arcanum. It's the same format as ÉrtSz, so the application might be easier, but creating the database is a bit more work. Or, we might ask their contact person or website administrator if they're planning to fix the problem. Would you like to try this latter way first? Adam78 (talk) 22:28, 28 May 2021 (UTC)Reply

I don't see much hope for them fixing it if they archived it. How much work is creating a list from Arcanum? Can I help? What was the method you used before? Panda10 (talk) 17:03, 30 May 2021 (UTC)Reply

@Panda10 My method may not be very efficient but that's how I can do it. You open the list of entries, starting from A, then open View Source (possibly Ctrl+U, depending on browser), turn on line wrapping with the checkmark at the very top, copy the page section which contains the links and their inscription (following the line <div class="list-group mb-3">, which you can find each time as a starting point, possibly with Ctrl+F), starting with <a class="list-group-item list-group-item-action" href="/hu/online-kiadvanyok/Lexikonok-magyar-etimologiai-szotar-F14D3/a-a-F14E1/a-F14E2/">a</a>, the last such line on each page followed by </div>. Then paste these links into an empty plain-text file (e.g. in Notepad), for each page, then save the result. You may want to zoom the code page to a small size like 50% or 33% so that you don't need to scroll much when copying.

When the link list is ready, this part of all links <a class="list-group-item list-group-item-action" href="/hu/online-kiadvanyok/Lexikonok-magyar-etimologiai-szotar-F14D3/t-F^#^#^#^#/ ("^#" indicating any number) needs to be bulk-replaced with nothing (e.g. in Word), this element: /"> with a tabulator (^t), and the closing </a>^p with nothing again (^p denoting the empty new line). Now you have two columns, but actually the term should come first and the code second, so the two columns can be swapped in Excel.

Also in Excel, you need to make a third column with a condition for whether column "A" contains a number, so that we can separate homonymous terms from the rest (they need to be listed in two distinct databases, cf. {{Module:R:ErtSz/data}} and {{Module:R:ErtSz/homonyms}}). This and this are some pieces of help for this function (or if you use a Hungarian-language Excel, you can find their equivalents e.g. here). When the function is inserted to the third cells of all rows, you can sort them by this function (column "C") and, second, by the entry itself (column "A").

In the end, the database of the homonyms needs to be rearranged in a way that the first column should contain the homonymous term and the other columns, the disambiguated links, no matter how many there are (e.g. tus tus-1-50AFB tus-2-50AFF tus-3-50B02 tus-4-50B04 tus-5-50B0A in the database of ÉrtSz.). I can't think so many steps ahead but I suppose I'd manage it somehow. – To whatever extent you can and are willing to help, it'll be appreciated. :) Adam78 (talk) 12:17, 31 May 2021 (UTC)Reply

We should divide up the alphabet to avoid double work. If you started it with A, I can start with Zs and go from there. But first we should create Module:R:TotfalusiNagyEty 2001, Module:R:TotfalusiNagyEty 2001/data and Module:R:TotfalusiNagyEty 2001/homonyms, so we can copy the prepared list immediately to the appropriate subpage. Let me know if this will work for you. Panda10 (talk) 15:50, 31 May 2021 (UTC)Reply

@Panda10 All right. Say, I'll do it until the end of "K". Although I suggest we could use a shorter name, like "TotfEty" – if not for the modules, then at least for the template itself. Adam78 (talk) 17:03, 31 May 2021 (UTC)Reply

We need to add 2001 in the name, Tótfalusi István has another dictionary dated 2005. I tried the method with Zs using only Excel:

1. Copy the links to Excel (column A).
2. Select Data -> Remove duplicates (this will remove the empty lines, every second line was blank for me).
3. Select Text to columns, delimited, use > as a delimiter. There will be a new column B.
4. Replace </a with nothing - column B now contains the word list. Insert a new column A and move C to A.
5. Text to columns again, delimited, use / as a delimiter
6. There will be several new columns. Delete the ones that will not be used and keep the last one. So what we get is in column A zsába, in column B zsaba-F4572. We need to merge these two with a space in between. Is this what we want? Panda10 (talk) 17:23, 31 May 2021 (UTC)Reply

@Panda10 I think we need a tab character between them, otherwise it's fine. Good job! You can see what format is eventually needed in {{Module:R:ErtSz/data}} (a big file); I've fixed its link above. Adam78 (talk) 17:32, 31 May 2021 (UTC)Reply

I added the module and the template with the name you recommended, the data for Zs and tested it at zsakett. So far so good. The only problem was that when I copied the data from Excel to the data subpage, each line was surrounded with quotes, I have to figure out how to avoid that. I won't be able to delete them manually for larger data sets. :) Panda10 (talk) 18:38, 31 May 2021 (UTC)Reply

The homonyms don't work. No sure why. Try zaj, that's the only data so far in the database.Panda10 (talk) 20:39, 31 May 2021 (UTC)Reply

@Panda10 I think I've fixed it. (I compared the code with the one for ÉrtSz. and one line was missing from the end.) – I'm sorry I couldn't proceed more today; I was a bit busy. I'm very glad to see that it eventually works!

By the way, I think we should include the etymology link in the References section, as a footnote to the information given at Etymology, rather than as a link in Further reading, because the latter already references two dictionaries in many cases (ÉrtSz. and Nagyszótár) and a third link there wouldn't help orientation, while it's actually needed as a reference anyway. Adam78 (talk) 21:42, 31 May 2021 (UTC)Reply

@Panda10 As you can see, I progressed a bit more, so I included those under L–Ty as well. U/Ú, Ü/Ű and V are still left to you, in case you're working on them. In terms of the Excel file, I inserted the function "=SZÁM(SZÖVEG.KERES("1"; A1))" (adjusted to the respective rows) in column C, searched for "IGAZ" values, then changed the one or two (sometimes three or four) cells under them to "IGAZ", depending on the number of homonyms, so as to separate them. Then I sorted the table by columns C and A, added the function "=(BAL(A1;HOSSZ(A1)-1))" to remove the number from the end, added "=B1" to E1, "=B2" to F1 (and so on in the same row if there were more than two homonyms), then copied the terms and codes to another sheet, sorted it by B, deleted the rows with no codes, copied the result into Word, replaced "^t^p" with "^p" as many times as was necessary to eliminate useless tabs, then finally inserted the results into the database of homonyms. Adam78 (talk) 15:17, 1 June 2021 (UTC)Reply

Thank you so much! I definitely could not have done it so quickly. I will start working on U/Ú, Ü/Ű and V now. The last step will be the old template replacement. Panda10 (talk) 16:03, 1 June 2021 (UTC)Reply

There is a Lua error at zaj. Panda10 (talk) 16:36, 1 June 2021 (UTC)Reply

Resolved. The tab characters were missing. Sorry about this. Panda10 (talk) 16:44, 1 June 2021 (UTC)Reply

@Panda10 It's OK. I think we've completed it. Could we move this new template to {{R:TotfalusiNagyEty 2001}}? In that case, we wouldn't have to replace any occurrences. {{R:TotfEty}} could stay simply a redirect, useful as a shortcut. Its target is the same (even if at a different website) so we could have implemented it in the original template (it was out of safety that we didn't). I think replacing all the links wouldn't make sense. We need to check them anyway, at least randomly here and there, no matter what their name is. The modules could be kept under the existing names. Adam78 (talk) 23:17, 1 June 2021 (UTC)Reply

I made the change. We might have to watch Category:Pages with module errors just in case. Maybe for homonyms? Not sure how those were handled previously. Panda10 (talk) 17:30, 2 June 2021 (UTC)Reply

Please check the above category. There is a growing number of errors related to the above change. It looks like it's for entries where the actual entry is not in the dictionary and another word is indicated as a parameter. Panda10 (talk) 17:53, 2 June 2021 (UTC)Reply

@Panda10 Maybe we could use the other (TotfEty) version of the template (currently not in use, only a redirect) like we use the old-fashioned {{R:BarcziOrszagh 1962}}, which does allow for a parameter with an alternative name for linking. Or even better, we could create a named parameter for this purpose, like "|t=kohó" ("t" for "term") and somehow insert this parameter in the new template: if it has a value, then PAGENAME gets replaced by that. Adam78 (talk) 18:29, 2 June 2021 (UTC)Reply

I corrected all the errors related to Hungarian in the category, just wanted to make sure it's cleaned up. I noticed that I did not copy all the V's, sorry about that. I won't be able to do it today. As for an additional parameter, not sure what to do. This is no longer a simple template, but a call to a module. We can think about it some more. Panda10 (talk) 18:49, 2 June 2021 (UTC)Reply

@Panda10 I finished the V's, no problem. I tried to improve the module but I haven't really succeeded. I've contacted Erutuon for help. Adam78 (talk) 20:27, 2 June 2021 (UTC)Reply

@Panda10 Thanks to his help, now we can add an alternative name to terms if they're different from PAGENAME, and what's possibly even better, we can even implement this change in {{R:ErtSz}} so that we don't need to use {{R:BarcziOrszagh 1962}} wherever PAGENAME is different for any reason (e.g. zsűri is given in this dictionary as zsüri). What's more, he implemented the handling of cases when we'd like to link to a specific sense of a term not identical to PAGENAME, like mi² from minap (and a minap). It's a great step forward! (I've reverted these deletions of yours and added the right parameter to the template wherever it was missing.) Adam78 (talk) 23:17, 2 June 2021 (UTC)Reply

Hi Adam. Just a heads-up that categorising terms with multiple etymologies by parts of speech (or at all) will no longer be necessary. It was decided at RFDO to avoid these manual categories, which are unlikely to be useful to readers and incomplete unless derived from a database dump. You don't have to do anything, but the categories should no longer be added to entries. Let me know if you have any questions. —Μετάknowledge^{discuss/deeds} 00:52, 10 October 2021 (UTC)Reply

@Metaknowledge Hi and thanks for the message. I understand it won't be maintained or populated in the future but the existing categorization can be preserved, I hope, can't it? See the matrix I created for the relevant Hungarian categories. People's bona fide work that appeared fine for quite some while (with the effective assistance of an administrator) can't just be annulled based on the view of a handful of people at a particular point in time.

I rarely if ever used anagrams here but I didn't object to their existence (although they're featured more prominently than categories). Two-letter and three-letter words could also be argued against, perhaps more easily than multiple etymologies.

Multiple etymologies reveal quite a lot of patterns about the regularities and inner workings of a language, like the load on particular phonetic structures and endings have. It can also help language learners notice the most important distinctions. It's a pity that these aspects escaped the attention of those present at the discussion. If users of certain languages find it cumbersome, they could opt out of this categorization; enabling them to do so could be a meaningful goal.

What's more, terms with multiple etymologies (or to put it simply, homonyms, which could have been a more fortunate name to underpin its existence) provide links between distant concepts just like rhymes, which people (understandably) don't intend to eradicate from Wiktionary.

Wiktionary is based on connections, not only hyperlinks but also the "what links here" function, which cannot be easily replicated elsewhere. It is also based on connections between form and meaning. Connections between meanings are elaborated via definitions, synonyms and other -nyms; connections between form are provided by means of rhymes and multiple etymologies. It's an inherent part of a multi-faceted, flexible content-management system like Wiktionary, which people spent so much time to improve and adapt to the diverse sorts of links between terms in different languages.

If this aspect does get deleted ultimately with all its subcategories for every language, I'd like to have a copy of the present content of the Hungarian-language categories (or in the worst case I'll archive them myself) as a set of user subpages of mine. RSVP. Adam78 (talk) 01:58, 10 October 2021 (UTC)Reply

While I don't have any particularly strong opinions, I was the one who carried out the deletion of the multiple etymologies categories, as the consensus appears to be to delete them. I agree however that the categories are more hassle than they're worth; they must be manually maintained (or with bots, but that hardly can be relied upon either) and don't really serve that much lexicographical purpose (the name "multiple etymologies" is also misleading as the words don't have multiple etymologies, they are distinct words that happen to be homoglyphs). It's certainly possible to back up the existing category contents into userspace before deletion if you so wish, but to my understanding the categories have to go as that was the outcome of the RFD. — surjection ⟨??⟩ 10:43, 10 October 2021 (UTC)Reply

What is the rule for including inflected forms in rhymes? I've noticed that you removed rhymes from the noun form entries but added it to the verb form entries (e.g. jössz). I think we should include all non-lemmas for consistency. Panda10 (talk) 21:01, 2 January 2022 (UTC)Reply

@Panda10 You're right, it's good to lay it down. I add them if they're unique or quasi-unique (like jössz for a second-person rhyme, because of the total assimilation, cf. köszönsz) or sometimes a handful of common examples, but I don't add them (and I remove any existing instances) when all other instances can be inferred by having the suffix itself at the top of the list (e.g. -tek for mentek, kértek etc.). Otherwise the individual lemma endings would be lost in the sea of the superfluous, redundant repetition of suffixes on thousands of ordinary words.

The latter doesn't apply to suffixes that form different rhymes depending on the stem (e.g. -t at the end of vowel-final nouns). In this case I think the best solution is to include a template (e.g. for the rhyme "-ót": "inflected forms with this rhyme can be obtained by adding -t to nouns ending in -ó", linking to the page with the nouns/verbs of the given rhyme if possible) and maybe a few (say, five or so) common examples.

The second-person singular subjunctive creates many different forms and often these are the only instances of a given rhyme, so some examples and the above template will suit best, I think. The case is similar (in terms of rhymes) with the second-person singular indicative present and the third-person singular past forms (-t): they create fairly unique forms that cannot be inferred from the suffix, because of the resulting different consonant clusters. (These inflected forms often don't have a full-syllable suffix that could be linked on a rhyme page as a reference for all the instances.)

What do you think? Adam78 (talk) 21:34, 2 January 2022 (UTC)Reply

OK, I agree that the categories would be cluttered up with noun forms. I assume this also means no rhymes for the possessive forms. Panda10 (talk) 17:49, 3 January 2022 (UTC)Reply

@Panda10: Indeed, I propose no rhymes for the possessive forms, with the exception of those third-person singular possessives whose endings create new rhymes after consonant-final stems, e.g. könyve will create a -ve rhyme (as opposed to könyvem, könyved, könyvünk etc., automatically inferred from the rhymes -em, -ed, ünk etc.: no need to list them individually). On the other hand, e.g. meséje doesn't need to be listed as long as the suffix -je is linked at the top of this rhyme page. So I think the same rule could apply as to the other suffixes (as I described with the example of -t above). Adam78 (talk) 18:23, 3 January 2022 (UTC)Reply

I was checking these words and the gloss on grófnő is very confusing, until I checked grófné. Think there's a more clear wordage here? Vininn126 (talk) 23:24, 8 February 2022 (UTC)Reply

@Vininn126 please check it now. Their difference is like that of queen consort and queen regnant; I added these terms to clarify them although I'm not sure they're used with countesses. Adam78 (talk) 12:13, 9 February 2022 (UTC)Reply

Hey, I was just wondering how Hungarian rhymes work. Vininn126 (talk) 11:00, 27 February 2022 (UTC)Reply

@Vininn126 Consonant-final words are arranged first by their last vowel, then by their last consonant (or consonant cluster); vowel-final words are arranged by their last syllable (a consonant, the last if there are several, plus the final vowel). Adam78 (talk) 20:42, 1 March 2022 (UTC)Reply

Hi Adam, this adjective form is added to Category:Hungarian adjectives suffixed with -ik. We usually don't add form entries to categories. I've just created legrosszabbik and I wonder which direction we should take. Thanks. Panda10 (talk) 20:20, 23 September 2022 (UTC)Reply

@Panda10 Is it a form entry? It has a declension on its own. I think it's a similar case to causative (-at/-et) or passive (-atik/-etik) verb forms, which have their own conjugation tables, so they are viewed as lemmas for the purposes of this Wiktionary, even if their meaning is not defined on its own. – I'm open to the opposite view as well, considering them as forms, but then their forms will be forms of forms. Can we eliminate this contradiction somehow, in a consistent way? Adam78 (talk) 20:43, 23 September 2022 (UTC)Reply

There are other cases where a form has its own declension. Possessive forms, comparative forms, superlative forms. We can make it an adjective entry, not an adjective form entry. Does it make sense? Will the category be of any use? Panda10 (talk) 20:47, 23 September 2022 (UTC)Reply

@Panda10 The possessive, comparative, and superlative forms are already listed in the section Declension (or in the head itself). The -ik forms are not currently listed as part of the declension. They should be listed either systematically as part of the declension, or one by one for each adjective concerned among the Derived terms, similarly to the causative of verbs. I'd say they're more like part of the regular paradigm, so they could be treated like possessives, comparatives and superlatives, but then they should be added to the relevant tables everywhere just like -é(i) was added to the noun declension tables. However, we should also check Magyar grammatika. I can do it in a few days. Adam78 (talk) 21:02, 23 September 2022 (UTC)Reply

OK, thanks for checking. I'm hesitant about adding it to the declension table. It will be an additional empty row for all the main entries, and will only be filled for comparatives and superlatives probably with a special parameter. We have to think about this. Panda10 (talk) 21:07, 23 September 2022 (UTC)Reply

@Panda10 The suffix -ik is described (pp. 188–189, 190 in the 2017 edition) in the chapter A névszók jelei, along with the marker -k of the plural, the heterogeneous plural marker -ék, the possessive marker -é, the possessive suffixes -m/-d etc., and the comparative and superlative markers -bb and leg-. It means that they should be treated the same way as we treat the other markers (inflectional suffixes), that is, as part of the general declension, rather than individually created derived forms. It's actually understandable because it can be freely created for any comparable adjectives. – In fact, I think once I already started working on a modification of the template but there was some hitch that prevented me from completing it. (The empty rows can be most probably eliminated if these rows are only displayed if there is a parameter that enables them.) Adam78 (talk) 16:37, 1 October 2022 (UTC)Reply

Sorry but I'm having a hard time wrapping my mind around this. The last time we wanted to make changes to the module was for -ja/-je, -juk/-jük in a partitive sense discussed in Template_talk:hu-infl-nom#partitive_3rd-person_possessive_(genitive)_forms_of_comparative/superlative_adjectives. At the time, I wanted to solve the problem by adding a possessive table (see legszebb, reverted) because all forms are possible, not just legszebbje/legszebbjük. Now comes legszebbik, which also can have a complete set of declension (legszebbiket, legszebbikünk). So I'm not sure how many rows we need to add to {{hu-infl-nom}}, what should they contain (labels, etc.) and I definitely don't know how to hide them. Panda10 (talk) 18:37, 2 October 2022 (UTC)Reply

I'm uncertain about the etymology of this word. It seems it's egyszerű + -södik since there is no egyszerűs. I couldn't find a -sodik/-sedik/-södik suffix cluster in Zaicz. A few more similar words: sokasodik, sűrűsödik, korszerűsödik, észszerűsödik. Panda10 (talk) 20:08, 29 September 2022 (UTC)Reply

This suffix cluster is mentioned in the Hungarian Wikipedia article Szóképzés a magyar nyelvben, the provided example is németesedik which sounds more like németes + -edik to me. (Németes kifejezéseket használ.) But anyway, it does exist. Panda10 (talk) 17:53, 30 September 2022 (UTC)Reply

@Panda10 For németesedik, the actual meaning (usage) of the resulting word has to be considered first of all. If it means ’to become characteristic of the German language or the German people,’ then németes + -edik may be okay. However, if it means that some group of people becomes partially or entirely German over the course of time, I think német + -esedik is the actual derivation, because they don't only assume the characteristics of Germans but they become ones. (If it can have both meanings, then both derivations may be valid.) To my mind, the latter makes more sense, with respect to the meaning of the resulting word, doesn't it?

The derivation of another similar suffix cluster, -(s)kodik, may also be ambiguous, like katonáskodik. It probably comes from katona + -skodik rather than katonás + -kodik: even though katonás is an existing word, the semantic relation from the latter is doubtful if not impossible; it is much more direct and self-evident from katona, with respect to the meaning of the resulting word.

As far as Magyar grammatika is concerned, in its subchapter Névszóból igét képző képzők (denominális verbumképzők) in A szóképzés, the suffix clusters -sít (államosít), -lkod(ik)/-lködik/-lkedik (gazdálkodik, büszkélkedik, szűkölködik), -skod(ik)/-sködik/-skedik (atyáskodik, vendégeskedik, ügynökösködik), -sul/-sül (állandósul) are listed among single suffixes, without any particular mention of being clusters (page 315 in the 2017 edition). I think we can add -sodik/-södik/-sedik to this list, with a view to the existing instances that underlie the proposed derivation.

Another relevant point is on pp. 312–313:

Sokszor nem könnyű eldönteni, hogy két képző egymásutánjáról vagy egy összetett képzőről van-e szó. Biztosan összetett képzővel van dolgunk azonban, ha közbülső (egyképzős) származék nem létezik. Az összetett képzők között tartják számon a következőket: -ható/-hető (választható, élvezhető); -hatatlan/-hetetlen (megoldhatatlan, elvégezhetetlen); -ságos/-séges (túlságos, lehetséges); a -kás/-kés (bohókás, édeskés); -sít (államosít, önállósít); -sul/-sül (állandósul, létesül).

Előfordul olykor az is, hogy ugyanaz a képzőcsoport bizonyos származékokban összetett képző módjára viselkedik, más hasonló szerkezetű származékokban viszont továbbképzésről van szó (bár a két típus elhatárolása nem könnyű). Összetett képző van például az önállósít, remetéskedik szóban, továbbképzés történt azonban az élesít, ügyeskedik származékokban.

Adam78 (talk) 16:13, 1 October 2022 (UTC)Reply

Thank you very much for the detailed explanation. I updated the related entries. Panda10 (talk) 18:02, 2 October 2022 (UTC)Reply

What do you think about this: e-nyelv.hu: Az egyikőtök (másikótok), hármótok, melyikőtök, valamint négyőtök, ötőjük stb. formákban az -ó/ő képző (nem igei személyrag kötőhangja). Simonyi Zsigmond szerint: kollektív számnévképzőnek látszik (Tüzetes magyar nyelvtan, 533.). Should we change the etymology for these forms? Panda10 (talk) 18:58, 2 January 2023 (UTC)Reply

@Panda10 It should definitely be mentioned (see the referred source in full here) as the first etymology, but I'm not convinced it's a generally accepted analysis, as shown here (look up kollektív). I think we should better admit this uncertainty in some way in the wording (along with mentioning a possible relation with the ending of kettő). The existing etymology might be kept as a second option, stating that its long vowel makes it unlikely. If we add the above analysis, then the entry of -ó/-ő could be updated accordingly. Adam78 (talk) 20:38, 2 January 2023 (UTC)Reply

Thank you for the references. Here is another one, Laczkó Krisztina: Gondolatok személyjeles számneveinkről és névmásainkról. Emlékkönyv Benkő Loránd hetvenedik születésnapjára. Szerk.: Hajdú Mihály–Kiss Jenő. Budapest, 1991. 399–403 mentioned in e-nyelv.hu. Panda10 (talk) 21:05, 3 January 2023 (UTC)Reply