Wiktionary talk:About Hungarian

Definition from Wiktionary, the free dictionary
Jump to: navigation, search

Dictionary[edit]

I found this dictionary to be very helpful in my research have a look and let me know if it helped you or what you think about it. http://www.free-dictionary-translation.com

I looked at it, the way it returns words looks similar to SZTAKI or VIDRA dictionaries. There is no pronunciation, explanation, declension, usage notes, etc. But it contains more Hungarian words and expressions than Wiktionary. What is your research area and how was the free-dictionary helpful to you? --Panda10 15:21, 14 December 2008 (UTC)

e-ë distinction[edit]

Something that I think is missing from the Hungarian entries is the information about the e-ë distinction.

What is it[edit]

There are many Hungarian people (I don't know percentage) who say different sounds for the written letter e. A more open variant and a more closed variant (written ë if the distinction is important). And this doesn't happen at random at all! So when it happens bears information.

Why this[edit]

This is a feature that many Hungarian dialects (don't just think of little villages) possess. Most other dialectal pronunciation is predictable, but since the e and ë sounds are both written e, outsiders cannot possibly predict it from the written form. So it would be a real information.

How to implement[edit]

First of all it should only appear in entries of words containing the letter e.

It could appear in the pronunciation section as a list item. egy would be:


Pronunciation[edit]



The size might be small:



Pronunciation[edit]



The words could link to a wikipedia page or to a no-yet-written Appendix here.

Putting it to the IPA line and using []-brace (specific) IPA would not be a good idea because the exact sound is different in each dialect. The e in Dialect1 might be the ë in Dialect2. They would look like the following in openness order: ë (D1) < e (D1) = ë (D2) < e (D2). Some dialects say ö in place of ë. So it is not really the pronunciation issue that is interesting to us, but the distinction itself. Which one is which, and then it becomes predictable for each specific dialect what the exact pronunctiation is.

What do you think? Qorilla 14:02, 9 July 2009 (UTC)

I am always for improving Hungarian entries. But with so few editors and so little time, we need to set some priorities and decide what's important. Also, it is simply not known who uses the Hungarian section of this dictionary. Would the e/ë distinction be helpful to an average learner? Or are we creating our entries for the linguists who would love to know about fine distinctions in the language. Unfortunately, I don't have enough knowledge about e/ë to contribute and I think the Hungarian section is still very much in need of general development, adding more words, translations, templates, better structure, the representation of the information, correcting mistakes, etc. Your contribution so far is very valuable and I wonder if you'd like to continue along this line. Would you be interested in creating a project page, similar to Wiktionary:Project - Spanish it could be called Wiktionary:Project - Hungarian where the projects would be listed. I have a list of things to do, which is mostly general and could be applied to any of the foreign languages here. We could start the project page with a simple list, than add as we go. Here it is. Let me know what you think, what's missing and if creating a project page makes sense at all.
  • Create entries
    Create main entries for each red link in Index:Hungarian
    Create entries for requested words in Wiktionary:Requested entries:Hungarian
    Clear Category:Tbot entries (Hungarian)
    Add verb forms and noun forms (SemperBlotto has a bot)
    Rethink the structure of noun form entries, the {{hu-inflection of}}, especially when a series of suffixes are present. E.g. kertemet - should it simply say accusative of kertem then at kertem it would say possessive of kert or should it say possessive and accusative of kert?
  • Templates
    Create declension/conjugation tables to cover all types
    Review the function of existing templates and improve, simplify, clean, standardize if possible
    Add functionality to {{hu-compound}} to allow for compound types: organic (szerves) - non-organic (szervetlen), subordinative (alárendelő) (subjective - alanyos, objective - tárgyas, determinative - határozós, adjunctive - jelzős, xx? - jelentéstömörítő) and coordinative (mellérendelő) (valódi mellérendelő, word duplication - szóismétlés, twin words - ikerszók); each can be jelölt and jelöletlen.
  • Categories
    Improve the structure (categories) and make sure each existing entry is added to appropriate categories
    Improve the visibility of hard to find pages, appendices, index
    Add {{hu-case}} to each verb that takes arguments
  • Etymology
    Review existing etymology sections and correct if needed
    Add etymology to each entry
    Decide what goes in the ety section for compound words and derived words
  • Example sentences
    Add example sentences to each main entry
    Add example sentences to each verb form and noun form entry
  • Pronunciation
    Record audio for every word, but at least for the most common
    Add e/ë distinction
  • Pictures
    Add pictures from Commons, if there are none, create and upload your own work
  • Definitions
    List the available English context labels with their corresponding Hungarian translation so editors can use it appropriately
    Add context lables when appropriate
    Add transitive/intransitive label to verbs
  • Patrolling
    Review anon edits in Hungarian entries

--Panda10 22:40, 9 July 2009 (UTC)

These quite match up what I have in mind! Kinda, you took the words out of my mouth :D My general plan is that we can go along doing stuff the way we did, and besides that, the rethinking could go. That is the thing I'd really like to do, rethinking and reorganizing and making things more logical and coherent. (I am fighting with conjugation at the moment. Hungarian conjugation is pretty complicated but I'm making progress.) I also have plans for some new Appendix pages. And having a bit more organized Derived terms section. Rethinking the treatment of suffixes. Seeing what can be automated or semi-automated and how. I will do these as my time allows. I am also in some of these but only in early phase. The Project site will be really needed but maybe not now. And I still feel a nice reader-portal could be also useful, but English Wiktionary doesnt seem to have a place for it. But I think we are in no hurry as this is just a kindof hobby. This text may be a bit incoherent, I changed it a few times and it's late night so forgive me. Besides may I ask in what time-zone you live? Your edits seem to come in odd times :) I'm just curious. Qorilla 23:13, 9 July 2009 (UTC)
Great! If you need feedback or input about structure, etc. let me know. I live in the ET time zone, it's 7:50 pm. --Panda10 23:50, 9 July 2009 (UTC)
I'd like to update you that the morphological etymologies are very helpful for Hungarian students. Though it would be even better if there was a visual distinction between derivation (which change POS) and inflection/declination. OrenBochman 10:23, 9 April 2011 (UTC)

Templates[edit]

I have most conjugation templates finished. There are quite many. I tried to make fewer but I was warned that the many if's needed for that will make huge server load, so I separated them. As there are 3 vowel-harmony types, and 5 templates for each, I ended up with 15 templates, which still use if's for some flags but much less. Plus an extra for the front vowel, the tegyek, vegyek-type. Then those that must be handled individually because they are irregular, like jön, megy, van. Uploading all will not happen at once as I need to test them to be sure. So I will upload them it the sandbox first, test it, then put it to its place then go for the next. Qorilla 08:56, 12 July 2009 (UTC)

That's fine. I am not worried about the number of templates. What's important is to maintain a clear, constistent template naming convention and parameter sequence. For example, a while ago you added a new functionality to hu-pos to handle út, kút and their derivatives which is great. However, the parameters for these types of words have to be entered differently. Normally, the first parameter is the head word, then any stem changes. For út, the order is reversed: hu-pos-atok|ut|j|jai|longstem=út, the consistent order would be hu-pos-atok|út|j|jai|shortstem=ut. This is what I mean when I say let's try to be consistent, this will make the usage easier. Would you mind listing the template names here and what they do? For example: hu-conj-ok-ik, hu-conj-ek-ik, hu-conj-ök-ik - handle the conjugation of verbs ending in ik, etc. Were you able to keep similar parameter settings as in the existing templates? Do the new templates use the same hu-conj base template? You could also upload one of the templates, add the documentation, so we can discuss. --Panda10 14:02, 12 July 2009 (UTC)
I don't really understand the issue with kút. Most times one has to write the shortened stem to the first place. See for example kerék, hatalom. That is the standard way. Most words do not use the unchanged, longer stem, only the short one. So if you want to be consistent with decl (which is not so important as it is another family of templates) then we need to skip an empty place for the normal stem. Like kerék would be {{hu-pos-etek|whatever.as.this.wont.even.be.seen||ei|shortstem=kerek}}, which complicates the things too much and is not easy to handle (and needs a mass bot change). I think the first parameter should be the one that is used to form the normal possessives (with the possible stem change), and for those few exceptions that also use the long stem, it needs to be provided extra. I thought about it then and I found this as the easiest way. What do you think? Qorilla 17:24, 12 July 2009 (UTC)
There is no issue. It's a matter of different views. I was hoping that we can figure out a way to make the parameter sequence more consistent in all Hungarian templates, to make the editing experience easier. If that is not feasible, than it's not feasible. --Panda10 17:50, 12 July 2009 (UTC)
pos not even consistent with decl today, as per kerék. I think the current way is effective, as most times the possessive only uses one kind of stem. Qorilla 18:11, 12 July 2009 (UTC)
There are some exotic types that are not yet fully tested (weird stuff like melegszik - melegedjek), so I need a bit more testing time before finished. Hungarian conjugation is really complicated. But I'm near the finish. Qorilla 17:37, 12 July 2009 (UTC)
No rush. We can always create a unique template just for exotic types if there are no other similar ones. --Panda10 17:50, 12 July 2009 (UTC)
Alright. I just want to let you know that I'll go for a short vacation tomorrow so I will only continue next weekend. Qorilla 18:11, 12 July 2009 (UTC)
Thanks. Enjoy your vacation! --Panda10 18:17, 12 July 2009 (UTC)

Suffix template[edit]

I think we should rework the hu-suffix template:

  • allow multiple suffixes
  • allow translations and alternative representations using {{term}}
  • handling linking vowels inside the template, that is we shouldnt type "linking vowel" each time, so changing or formatting will occur at once at all places.

I made a draft at the Sandbox. It is not changed often so I suppose you will see this:

This is the template sandbox, a page for experimenting with editing template pages. Feel free to try out your skills at formatting here by editing this page. Alternatively, you may want to read through How to edit a page for explanations.

NOTE: Any content added to this page may be deleted in twelve hours or less. Do not use this page for anything that you want to keep.


somethingáloms

parameters:

{{hu-suffix|álom|t1=dream|os|t2=-like|link2=a|k|t3=-s; plural}}


This is of course not a normal usage, it just demonstrates the capabilities, normal usage will be a lot shorter. Qorilla 18:11, 20 July 2009 (UTC)

I agree with the three bullets, but the resulting etymology line appears a little too crowded and not easy to capture/comprehend with one look. Here are my thoughts:
Base word
  1. When the base word is not a valid entry, wouldn't it be better to say From the shortened stem of álom instead of saying álm? I don't want to mislead the learners stating that álm means dream.
Yes, álom should be preferred in an ety. section
Linking vowels
  1. We should determine how to handle the linking vowels. Should we always separate them from the suffix? And dump all variants into one huge category? Wouldn't it be more interesting to see all words with -os in one category? Not to mention that linguists always talk about suffix clusters (képzőbokor), e.g. -s/-os/-as/-es/-ös. So maybe we should list the pros and cons of each method and decide.
  2. If we decide to always separate the linking vowels, instead of adding "linking vowel", we should create an entry for -a-, -e-, -o-, etc. and explain that this is a linking vowel and list all the suffixes where this may be used. In this case, -a- would become a blue link in the etymology. Eliminating the "linking vowel" phrase would reduce clutter in the ety line.
I vote for never separating any linking vowels. Linking vowels just make things more complicated in diplay and for learners as there are many of them and suffixes differ on which ones they accept. However, we should mention at the entries like -es that it can also be considered a linking vowel + -s. The only problem with this is that we have to explain the exact same meaning at all the variants.
Translations
  1. Is it necessary to translate each element? Without translation the line would look cleaner. Users can always look up the meaning of the suffix on its own entry page. Adding tranlations will increase the workload and the possibility of mistakes. I do think that translations are important when the base word has many different meanings and it is necessary to clarify in the etymology which sense is used.
Suffix translations should be used sparingly (only in obvious situations). But as we can optionally translate in hu-compound, the option could be left here too. Most (almost all!) etymologies will only contain one suffix.
Categorization
  1. How will the template categorize the words when multiple suffixes are present? The new template puts the word into a new category, Hungarian words suffixed with -k, not sure if that was intentional. Plurals should go to Hungarian plurals.
  2. The "Hungarian words suffixed with -xx" categories should not be used for declined/inflected verb/noun forms.
  3. For noun forms, the categorization is done by hu-inflection of template, so if you use hu-suffix in the ety line, it should have the nocat=y parameter. The noun form categories are better defined at this point than the verb form categories.
We could introduce a pos parameter, so we could tell to categorize not into Hungarian words suffixed with xx, but Hungarian pos suffixed with -xx. For special cases like the plural, we could use nocat and do the categorisation explicitly on the page. If multiple suffixes are attached, categorisation happens according to the las one attached (even in this form that I wrote) Qorilla 23:35, 20 July 2009 (UTC)

--Panda10 22:53, 20 July 2009 (UTC)

Summary[edit]

  1. Modify the template to accept new optional parameters but leave the current parameters intact, so current entries that use those parameters will not be affected.
  2. Always use a valid base word, even if a shortened stem, a v stem etc. are suffixed. An explanation can be added "From the v stem of ...", "From the shortened stem of...". Question: What to do when the base word does not exist? Should we create an entry for them with Particle pos header?
  3. Do not add alt1
  4. Add translation parameters t1, t2, t3 ... tn (where n = the number of max suffixes + 1)
  5. Decide the maximum number of suffixes and add code to handle it.
  6. Do not separate linking vowels. If an unlinked text is absolutely needed in the middle of the word for any reason, the current link parameter can be used.
  7. Categorize the entry based on the last suffix.
  8. Add a pos parameter to allow categorization by part of speech since many of our suffixes are specific to a part of speech.
  9. A pos parameter will also help separate words into separate categories where a single suffix has multiple meanings (e.g. -ul). --Panda10 00:26, 21 July 2009 (UTC)

Test[edit]

This is the template sandbox, a page for experimenting with editing template pages. Feel free to try out your skills at formatting here by editing this page. Alternatively, you may want to read through How to edit a page for explanations.

NOTE: Any content added to this page may be deleted in twelve hours or less. Do not use this page for anything that you want to keep.


somethingálomos

parameters: {{hu-suffix|álom|t1=dream|os|t2=-ly|pos=adj}}

As to the question about nonexistent stems, I think we should not create an entry. We should note in the Etym section that the word has an ending but it is not suffixed to an existing word (with better wording but the sense should be this). A use of an entry about an alone-nonexistent stem would be to collect the derived terms from it. That should be done in Related terms sections of the "derived terms" themselves, if we don't create that entry about the non-existent stem.

Another thing: I often see words that have an ending, but the etym section just states it originates from some proto-root, but does not address how that root was changed when it became Hungarian. We should link to the suffix page somehow, like saying in the etym section "with the suffix -xx". Qorilla 12:52, 22 July 2009 (UTC)

I started adding the 'pos' parameters and making the categories. When it's finished the 'word' categories should be deleted. Qorilla 18:39, 23 July 2009 (UTC)
Ok, but cleaning out one category at a time would be a better method than one word here and there. I am concerned that this would lead to chaos. --Panda10 00:38, 24 July 2009 (UTC)
Yes. I just experimented around a bit by creating some new categories. I think we should link the categories somehow together. Like the cat. of -os adjectives should have a link in the intro section to the -as, -es etc. Another thing: cleaning out -ság/-ség would really need some bot work. If we have to do it manually then we rather shouldn't. Qorilla 09:08, 24 July 2009 (UTC)

Vowel harmony template[edit]

We type many times: "Used in fron vowel words. The back vowel variant is -xx, the rounded front variant is -yy etc." It would be effective to create a template for this. And I think it does not belong to the definition part. I think we would need a Grammar notes section similar to the Usage notes, where we could explain clearly grammatical features, like how the suffix changes the stem (lenghtening a,e or not, explaining linking vowels, vowel harmony variants, how the created word can be further suffixed (like Józsiék will be suffixed Józsiékat, not -ot as normally would be) etc.) The usage notes are better for really how to actually use it in life. Like when to say it, whom to say it etc but not such exact things as firm grammar rules. What do you think? Qorilla 12:59, 22 July 2009 (UTC)

I like the idea of a template, it should make it easier to add these type of comments and the text itself will be more standardized. I read the BP discussion about the Grammar notes. Using the available Usage notes is not that bad in my opinion. We should not stop adding valuable content just because we think that some of the headers are not precise. --Panda10 23:02, 22 July 2009 (UTC)
That's true. We will use the Usage notes as there is no better choice. Qorilla 23:41, 22 July 2009 (UTC)

Examples[edit]

We need consistent use of example sentences and example words. There are more versions used currently. My suggestions:

Using words[edit]

A macska egeret fogott. - The cat caught a mouse.
  • Hungarian sentence in italic, with the word bolded (with the suffixes if any)
  • English sentence in non-italic, with the corresponding word bolded (with suffixes if any)
  • Words not linked unless there is a special reason.
  • The separator could be a normal - hyphen, or a longer —.

Using suffixes[edit]

Could we use {{term}} here? Then:

lép (to step) → lépés (step)
  • The → can just be - but the → better expresses derivation.
  • The suffix may be bolded:
lép (to step) → lépés (step)

-Qorilla 15:14, 25 July 2009 (UTC)

Both sound good. --Panda10 15:18, 25 July 2009 (UTC)

Categorization with hu-verb[edit]

Should we continue the current categorization using {{hu-verb}}? The categories are not suffix categories, but words ending in xxx categories, useful for similar conjugation. User:PalkiaX50 made a change in the hu-verb template which emptied Category:Hungarian verbs ending in -ít. Now all the verbs that were there are in the non-existing Category:Hungarian words suffixed with -ít. --Panda10 22:30, 29 July 2009 (UTC)

If the categorization is to be used, the category should be named "ending in..." not "suffixed with...". A suffix on Wiktionary is a POS category for an item used to form a new word from ddition to an existing word, and not an inflectional ending. --EncycloPetey 14:57, 31 July 2009 (UTC)
Sometimes the base word is still used, sometimes it has been lost through the ages to leave us only with the suffixed forms. But they are still suffixes. Also, there is no substantial difference in Hungarian between endings and suffixes. I also don't like saying "words ending in -xxx" as that might include words that just randomly happen to end in those letters having nothing to do with the suffix itself. Qorilla 15:10, 12 August 2009 (UTC)

Verb prefix template[edit]

I made a template {{hu-verbpref}} which can be used to generate the derived verb-prefixed forms in the Derived terms section. The template is long but uses no nested ifs. If you like it, I'll begin to add it to entries. Qorilla 15:15, 12 August 2009 (UTC)

I like it. My only comment is that we will have to enter the prefixes every time, but maybe there is no other way to do this. The Hungarian wiki uses {{hu-coverb}}, it's a copy of theirs. I assume you saw it? --Panda10 22:35, 12 August 2009 (UTC)
That is not the standard way terms are listed in the English WT. With the Hungarian style it is a one-liner separated by pipe-characters, but the normal lists use bulletted terms in columns.
We have to specify all prefixes, how else would you specify which prefixes are appropriate and which are not if you don't specify it? The difference is that with the Hungarian style you have to write ...|be=y|meg=y|..., but then it's too much for no purpose. That's why I made it so that we can simply write ...|be|meg|... So the main differences are the appearance and the choice of unnamed parameters as opposed to fixed prefix list (named parameters), from which we choose by writing =y. Qorilla 23:01, 12 August 2009 (UTC)
Ok, thanks. --Panda10 00:35, 13 August 2009 (UTC)
Okkay, I'll start using it. Thanks for your comments by the way! Qorilla 13:34, 13 August 2009 (UTC)

Conjugation templates[edit]

Now after a year I think I have some time to revisit the conjugation templates that I made last year. The basic system would be:

We should have many separate templates so we don't overload the servers with many if - else stuff. The templates should be named so that it's easy to know which is which, and when to use which: the names should not use linguistic expressions like 'rounded' etc. rather practical names, see below. There are certain forms which help classify the verb into categories, and can say many things about the verb, not just the directly mentioned form (for example knowing the past suffix is -ett as opposed to -t, we can know some other forms too, not just the past).

What I'm presenting here (templates, flags, options), is already written in .txt files on my computer, it just needs some effort to fine-tune it.

The verbs are given to the template so that the last letter (Hungarian letter, so dz is counted as one) is separated, which is not used in all templates, but in many it is, so to be coherent let's separate the last letter, like né|z

The first level is the separation according to vowel backness, which appears universally in first person singular (nézek/ütök/állok). So the first part of the template should be hu-conj-ek for néz.

The next level is how the past is formed (-t or -.tt). For néz it is -ett, for kér it is -t. Thus kér would use hu-conj-ek-t. For néz hu-conj-ek-ett. Kér's line of templates ends here, but there is a special case markable with a flag that the past participle uses -ett (like írt - írott in the -o- case, in -ett I can't find any now). For néz we must continue because there are more categories inside.

Next level is how the second person singular is formed: -sz (lépsz), -esz (rengesz), -el (nézel). In each case there is an option (flag) if the past definite conjugation uses -ette (vetsz - vetette, but kergetsz - kergette; zengesz - zengte but intesz - intette; nézel - nézte, but edzel, edzette)

So néz would be hu-conj-ek-ett-el|né|z.

There are some universal options/flags: intrans: intrasitive, so definite forms do not appear. nosubj: there is no subjunctive (felszólító) form (siklik). ik: the verb has -ik ending. For ik-verbs: ek-only: if the eszem form is not used, only the eszek. I can't find in my notes to which verbs this applies, but there are some which are not -em even though they are -ik verbs.

Okay, these were the rather regular verbs. Let's see the special ones. We have those changing the stem. There are basically two in each vowel backness category: those with -ottam (ugrottam) and those with -tam (sodortam). Those with -ottam are all -ik verbs if I'm not mistaken. There are two categories in -tam verbs according to the form sodrok - sodort vs. túlzok - túlzott. These are quite annoying situations as one might also say túlozok, so we need an option that I called 'variants', which whould list both separated by or.

Then we have those having -gyek in subjunctive (vigyek, legyek, ...) This is already online. -gyak is not needed as only igyak uses it, and it can be manually done.

Then we have those -szem verbs (fekszem, igyekszem). These need an option whether -ett is used in past or not: feküdt, but melegedett. We also need option vő-only which is for verbs that use the igyekvő form only, and not the igyekező (fekvő, but not feküdő), and the opposite flag: no-vő (no melegvő). This also implies having variants in other conjugations: igyekezek - igyekszem, but not feküdök.

Some verbs are so special that they need unique templates: hisz, jön, megy, van.

I think I've written about everything. It looks like quite a mess at first, as I didn't take the time to write a comprehensive guide to verbs now as that would fill even a book, I just wanted to show how I organized the verbs to minimize if-else decisions, but also keep a consistent usage, and protect the number of templates from w:combinatorial explosion .

Sorry again for not explaining all very accurately, but this is really a large topic. What do you think about this system? Qorilla 13:47, 19 July 2010 (UTC)

I started uploading them to my userspace, so I can show how they work: User:Qorilla/Conjugation templates Qorilla 20:56, 19 July 2010 (UTC)

Hi Qorilla, good to see you back. I looked at the examples and my biggest problem is to figure out which template to use. I wonder if there is a simpler way of naming the templates. How about picking a representative verb of that group and name the template hu-conj-<verb>, for example for the verbs ending in -szt the template would be hu-conj-fejleszt? Can we think about this a little more? It will be very hard to change the system after it's all done. Thanks for your work, looking forward to the templates. --Panda10 22:14, 19 July 2010 (UTC)
Hi, I made a lot of thought about it. The problem with your suggestion is that each of these templates covers many types of verbs even with such a big number of templates. That's why I also added options/flags for smaller differences. There is no way to select one verb for each template because the categories of verbs has many levels (a tree-like category system), and verbs have features which can be "on/off" in many combinations.
I'll make a guide about how to select the template, as it may be not obvious at first sight but I think the easiest way to recogize it, is writing the discriminating features (the representative suffixes) explicitly in the name of the template. These are easily checked by speakers of Hungarian.
All the front vowel templates are now on my experimental page, I can't think of a verb that would not fit in one of those. But there are some (quite rare) combinations that simply don't have (at least I haven't found) a representative among front vowel verbs, only among the analogous back vowel ones. Qorilla 23:29, 19 July 2010 (UTC)
I believe you put a lot of thought into it. I would be interested to see the selection guide. Just one more idea: Have you thought about naming each cell and using the names as parameters only if there is an exception in that particular cell to the otherwise standard conjugation? --Panda10 23:42, 19 July 2010 (UTC)
I'm writing the guide at the moment. What do you mean by standard conjugation? Is intesz, érsz or nézel the standard one? I think there is no easy and elegant way to categorize Hungarian verbs as they are not real categories (no speaker would say ér is a separate category to int although the latter uses a binding vowel in conjugation; but for example in Finnish there are good separable conjugating categories that even natives know of). In Hungarian these are rather just sound fitting/accomodating rules. Qorilla 23:54, 19 July 2010 (UTC)
Now the guide is there. If it looks complicated at first sight: Hungarian verb conjugations is in fact complicated, I guess that's why it has been left out here at Wiktionary (as opposed to the easier possessive and noun cases). Qorilla 00:58, 20 July 2010 (UTC)

The templates are now all on my userpage! I wrote a longer explanation with many examples. See: User:Qorilla/Conjugation templates. If you think it is acceptable, then I will start moving them to the entries. Qorilla 18:44, 15 August 2010 (UTC)

I've found the following typos:
  1. {hu-conj-ok-vowelout-tam} túlzot should be túlzott
  2. túlzot volna should be túlzott volna
  3. {hu-conj-ok-vowelout-tam} kanyargot should be kanyargott
  4. kanyargot volna should be kanyargott volna
  5. {hu-conj-ok-vowelout-ottam} siklaniük --> siklaniuk
  6. ugraniük --> ugraniuk
  7. {hu-conj-ok-vowelout-ottam} Future participle bomlandó, it's a valid form and it should be there
  8. bomlaniük --> bomlaniuk
  9. {hu-conj-ök-vowelout-öttem} Future participle ömlendő, it's a valid form, although sounding archaic
  10. ömlötték form is missing, it's valid and correct
  11. {hu-conj-szem} igyekezendő is a valid form, although archaic
  12. {hu-conj-szom} nyugodt should be nyugodott in past indefinite 3rd person sg. (belenyugodott, not belenyugodt)
  13. nyugodt volna --> nyugodott volna
  14. alapulandó would be valid but archaic
  15. Hogy helyes: viendő vagy vivendő?
  16. Adverbial participle is víve
  17. ríjük --> ríjuk
  18. fővő or fövő?
  19. fővendő or főendő or fövendő?
  20. fővök --> fövök
  21. fővünk --> fövünk
  22. fővöm --> fövöm
  23. fővöd --> fövöd
  24. fővi --> fövi
  25. fővitek --> fövitek
  26. fővik --> fövik
  27. nyűni similar spelling problems as above

--Panda10 21:22, 15 August 2010 (UTC)

Thank you. Sorry for the typos, they remained when I "converted" the front vowel templates to back vowel and I missed some vowels that should have been changed.
  • bomlandó is valid, you are right.
  • What does ömlötték mean? ömlik is an itransitive verb with a quite passive meaning.
    Have you search for it in Google? There are only a few hits. My example: A folyók körülömlötték a földet. Not used very often.
  • In what way can igyekezendő be used?
    Please google it. A few archaic example will be returned.
  • nyugodott: You are right, and this is a strange verb. For example with áll, the case is exactly reversed. The participle is állott and the past is állt, here the participle is nyugodt, the past is nyugodott. I think this is a word-specific thing, or else we need to think of other verbs that work similar. If it's unique then we may need to create its own template. But I have to mention that megnyugodt is also used, and if I say it many times (as an analogy to aludt), I can convince myself totally that it is the normal way. Although I "know" nyugodott is the more used variant (and I use it normally too). See for example (found with google): "Amire a falunépe megnyugodt, hogy már fél órája csak a békák énekelgetnek a patakparton".
    I still say nyugodott is the correct usage. "Amire a falunépe megnyugodt" can be either a typo or a dialectal usage.
  • viendő: usually the verbs which form their past with -tt as opposed to -vett, don't use -v- in the -endő form: tesz - tett- teendő, lesz - lett - leendő, vesz - vett - veendő. I think only eszik - evett - (meg)evendő, iszik - ivott - (meg)ivandó (although my grandmother says ett and itt instead). Your linked page brings hívanó as an example, but it is of course a totally different case as hívandó is formed from hív, which is not in the tesz, vesz, visz, hisz, lesz, eszik, iszik group. He may confuse it with hisz, for which I say hiendő: For example because hiedelem also has no -v-, although it could have it, like jövendő - jövedelem. Google fight: átviendő: 2600, átvivendő: 522. This is a strange situation where the speakers also feel they are unsure and use this or that by gut feelings as these are rare verb forms. We could perhaps give both forms in the table (although for one, I never say vivendő).
    OK.
  • vive: I think only e gets é. It can be observed with other suffixes, like: tesz - tétel, vesz - vétel, eszik - étel; but iszik - ital (not ítal), visz - vitel (not vítel) and tesz - tét, lesz - lét, but hisz - hit (not hít). I checked google and both are in use actually, and víve is even more frequently written (átvive 1350 - átvíve 4770 - better to check with an accented verb prefix, so foreign words don't get counted).
    Have you used this link before: [1] - although I've seen mistakes in their forms before, I thought this might be something we should take into consideration.
  • I think fő, nyű, rí don't shorten their vowel. "Olyan meleg volt, azt hittem megfővök". "fognyűvő manó". In google the long megfővök has more hits than megfövök, although that doesn't really prove much as both have just a couple of hundred. Maybe both alternatives should be given? Qorilla 22:06, 15 August 2010 (UTC)
    These short verbs may need their own specific template instead of a generic one.
Please see my comments after yours. I think you can go ahead and implement them. Thanks. --Panda10 22:33, 20 August 2010 (UTC)


Making it simpler[edit]

I'm not really satisfied with the system I made, and Panda10 also wrote it's not easy enough to choose the template. But until now I saw no possibility to make things simpler because of the primitive template syntax. However, now I came across the Portuguese conjugation template, which uses a clever solution. More "layers" in the logic. That's an idea that enabled me to unify into one hu-conj-ok template the following ones of my previous system:

  • hu-conj-ok-t
  • hu-conj-ok-ott-sz
  • hu-conj-ok-ott-asz
  • hu-conj-ok-ott-ol
  • hu-conj-ok-vowelout-tam
  • hu-conj-ok-vowelout-ottam

There is a base template which gives the table itself, a middle template that gives the forms, and the outside template which inspects the word and helps the middle template.

Look at User:Qorilla/Easier conjugation templates for examples. I wrote some short explanations there. Qorilla 00:09, 19 September 2010 (UTC)

Now the -ek and -ök versions are also done. Qorilla 16:23, 19 September 2010 (UTC)

Sorry for the long delay. Will it be huge amount of work to implement this? I know you added the previous templates to a large number of verbs. --Panda10 00:44, 8 March 2011 (UTC)
Hi again. Now I put up the templates on the User:Qorilla/Easier conjugation templates page. They work. I have overwritten the hu-conj-ok template, so I replaced all occurences of the old hu-conj-ok. The others are not affected directly, they are independent templates, so they can be changed later. Please delete the hu-conj-ok-old, hu-conj-ek-old templates and the hu-conj-ok/doWork template which are not needed any more. Documenting these changes will take a bit longer. Qorilla 22:53, 15 April 2011 (UTC)

Uncategorized hu entries[edit]

There have appeared at the list of uncategorized pages a large number of Hungarian entries, which seem to appear there by virtue of having lang=Hungarian in a template instead of lang=hu. I assume this is because of some reform in the internal operations of the templates. DCDuring TALK 12:20, 2 March 2011 (UTC)

Can you give an example? Where is this list of uncategorized entries? Qorilla 15:22, 3 March 2011 (UTC)
Special:UncategorizedPages. --Yair rand (talk) 15:23, 3 March 2011 (UTC)

Verb Complements[edit]

I am a currently studying Hungarian, and have found Wiktionary an invaluable resource. I think there is an omission involving the Hungarian verb entries which may not be obvious to the native editor. Briefly it is that the linguistic complements aka vonzat are rarely listed. This is an area where Wiktionary can really outshine paper and other online dictionary.

  1. The verb entries do not show which verb complements vonzat they require. For example vár typically requires a direct object and should be presented as vár valamit/valakit.
  2. This information is not regular - i.e. it can only be outlined in a grammar and requires their full details within the lexicon.
  3. Without this information a non-native Hungarian learners will:
    1. be unable to make well formed sentences (using vár without an accusative object). Also which conjuration should be used?
    2. find it even harder to select between the nounces of meaning created by the unintuitive verb prefixes.
    3. mix and match incorrect verb-noun-case combination. (which of beszáll, leszáll, kiszáll, felszáll, go with auto,taxi,vonat,villamos,lépcső and what is the appropriate case ending cluster (ba/-be;-ból/-ből; -ra/-re etc.)

look at száll and beszáll for some free form Verb Complements and consider the following recommendation.

Prescriptions[edit]

Here are different level of activity to verb complement omissions.

  1. The Hungarian verb citation template {{hu-verb|vár|valamit}} should to be augmented to provide a complement summary and existing verb migrated alternatively complement summary needs to be added {{hu-verb-comp|valamit}} so as to indicate their available complements. I.E. vár valamit.
  2. The complement summary template should take parameters to indicate
    1. Mandatory complement
    2. Optional complement should be indicated by () - give something (to someone)
    3. Multiple complements options when available they should be indicated separated by a comma. Though a / can also be used to indicate a variation in one of the comlement options.
    4. In the template documentation different examples are given
    5. In case ending which of the complement should take. Note currently ({{hu-case|ra}} or {{hu-case|ba}}), are in use which is confusing since it places the itself into the case ending category instead of the complement.
  3. Special emphasis should be on verbs where Hungarian 'deviates from English usage'. (This will benefit both Hungarian ESL students and foreign HSL students)
  4. For some verbs complements are further complicated by being related to 'specific noun'. These restrictions are best explained in the usage notes section.
  5. if and when example sentences are provided they should aim to illustrate correct complement use

OrenBochman 09:05, 9 April 2011 (UTC)

Please take a look at tartozik. It has several different complements and each has an example. You said using hu-case is confusing. I'm not sure why it is more confusing than using the Hungarian words valamihez, valakihez, valamibe, valakinek, valamivel, valamire instead of just the endings without "valami". Yes, verbs are still not a very developed area, but there is only two or three Hungarian editors and so much material. --Panda10 12:23, 9 April 2011 (UTC)
I have examined the current system and prepared a rudimentary template to go along with it. If it is acceptable, ({{hu-case|ra}} etc templates can and should be adjusted so that the categories and subcategories align. I have used rohan as a test case. Regarding the shortcoming/advantages of the current/new system.
  1. Currently in 98% of the verbs there are no complements indicated (since there are no examples).
  2. The categories provided by the current template are confusing because they are misleading - they place verbs into a category of words which take some case ending. However the verbs and adverbs are cannot take these while nouns adjectives and pronouns can. My recommendation is to change the name of the categories to indicate that it is the complement or not to use a category for this at all (The template:hu-case can be quickly migrated to a more accurate nomenclature. The groups require more work/thinking, but see the third item in the next point.
  3. Regarding the question about the merit of valamire is over just the ending. There are three advantages to using the long form though they come into play for annotating verb and not the example sentence.
    1. It allows an editor to indicate that a verb requires an object as a complement - valami, an accusative object valamit, that the object is person, valakit or someplace valahol.
    2. The currently implemented options are in the Template:hu-verb-comp. I have tried to make things clearer by indicating in English what these Hungarian complements mean in English.
    3. This format allows one to provide semantic parent categories to the morphological ones that are currently in use which will be much more meaningful. i.e. vhova as a parent for -ba/-be, -ra/-re etc...
  4. regarding the question raised on my talk page, (how will the user know which case ending to use if we only tell him the rohan takes a vhova complement.
    1. if it is the same as English he already knows
    2. if it is different but regular, he should have to seek the answers in a Hungarian grammar or an appendix.
    3. if it is irregular or confusing then further lexical information is needed as stated in the original proposal above.
  5. in view of these points the new template should enable editors to provide users all the information they require 90% of the time. OrenBochman 10:54, 10 April 2011 (UTC)
Hi Oren, thank you for the detailed explanations. I think we both have the same goal, to make Wiktionary useful to Hungarian learners. I am looking forward to seeing your changes implemented. It might be good to work out the details on a separate test page using the verb tartozik since it has multiple complements illustrated with examples sentences. I looked at your test at rohan and here are my thoughts:
  • The hu-verb-comp template should be placed on the definition lines, not on the inflection line where hu-verb is. The English Wiktionary has some standards that each participating language has to follow. This dictionary keeps the inflection line for verb forms (the most important ones). The hu-verb template does not have this functionality today, but eventually, it will.
  • The Hungarian words vhova, vmi, vki, etc. should be placed after their English pair (to somewhere, something, someone) and not before them. For example, the first definition of tartozik would be: belongs to someone or something (valakihez / valamihez)
  • The category name Category:Hungarian words taking -hoz/-hez/-höz can be changed to Category:Hungarian verbs taking complement vhez where vhez means valakihez or valamihez. But this can wait since you are building the parent system above this.
  • For the new category names you are proposing, I would use Category:Hungarian verbs with complement vhova. The Hungarian part of the category name would be at the end to facilitate better sorting.
--Panda10 13:03, 10 April 2011 (UTC)

I think this should only be used if it is a grammar question, not a semantic one. "Rohan" for example takes vhova because of its semantics, not because of Hungarian grammar. Tartozik on the other hand uses it grammatically (though such a distinction is not always clear). It should be in the definition line, as different senses may take different complements. I think we can leave simlpy the suffixes. Distinguishing between valaki and valami is not a good idea, because almost always both can be used. But the current system is a bit verbose too, as it lists all forms of the suffixes. Maybe the "vhez" form could solve this but it is not a conventional notation so it would not be understandable for the readers. I can not think of a much better version than the current. Qorilla 17:26, 16 April 2011 (UTC)