Wiktionary talk:About Persian
Transliteration
editI prefer to use the following system:
- long "a" = â
- short "a" (zabar) (fatha) = a
- خ = x
- ش = š
- ژ = ž
- long "i" = i
- zir (kasra) = e
As for the ezafe, I prefer to place it in the transliteration as e preceded by a - (hyphen). What does everyone else prefer or what do they want to present for the official policy? --Dijan 21:17, 12 April 2007 (UTC)
- I don’t have a strong preference, but I suggest what I put at Appendix:Persian transliteration. Feel free to modify it. —Stephen 23:50, 14 April 2007 (UTC)
- The Wiktionary:About_Persian#Transliteration says that there will be a vote on what transliteration system to adopt for the site, but I do not see a record of any such vote nor do I see any suggestion of when/how it might take place. Any ideas as to when or if this decision might happen? --Mavaddat (talk) 18:46, 29 December 2016 (UTC)
- We use Wiktionary:Persian transliteration (Wiktionary system). I'm not sure there was ever a formal vote, since most Wiktionary editors do not know Persian. We just came to agree on the system shown. —Stephen (Talk) 02:24, 31 December 2016 (UTC)
New addditions
editIt looks very nice. Just a few things:
- In ===Etymology=== section individual language template should be deprecated in favour of generic
{{etyl}}
. Most old folks are still accustomed to old templates, but the newcomers will love the flexibility of{{etyl}}
. - An example with a loanword from Turkish otağ should be replaced with a better one (perhaps from Old Persian? ;), because 99% of all Turkish loanwords in all the languages are in fact from Ottoman Turkish (ISO code 'ota') which was written in Arabic script, not from this after-1928 Turkish with reformed orthography and lexis.
- Example in the ===English-Persian=== section should use
{{t}}
instead of{{t+}}
- Examples in ====Etymology==== section should be rewritten using
{{etyl}}
and{{term}}
- Instead of using explicit categorization with [[Category:Persian nouns]] + reinstating the headword in every POS line, rewrite the examples to use
{{infl}}
with sc=fa-Arab, or maybe create{{fa-noun}}
which should use it implicitly? - Dari and Tajik entries should not be put in ====See also==== section (that catch-all section that people tend to (mis)use for various purposes is likely to be deprecated sometime), but instead mentioned in ===Etymology=== section, as cognates or something. ===Etymology=== and ====Descendants==== sections are the only place where entries should link to other languages' entries. --Ivan Štambuk 10:59, 4 April 2008 (UTC)
References
editHi, I've created {{R:STEIN}}
which can be used to reference comprehensive Persian dictionary compiled by Steingass in 1892 (thus out of copyright) and freely searchable here. However, I see several potential problems with it:
- It transcribes pronunciations that may have shifted a bit by today, especially in the omnipresent Tehrani dialect of "New Persian" which I've read on WP had much innovation in phonology in the last century, as opposed to more conservative peripheral dialects
- Some meanings (probably whole lot of them) are probably only used in the language of literature, and never colloquially, and thus distract from the primary, denotative meaning of the lexeme which should be listed first and emphasized
- Some of the meanings have become dated, obsolete or archaic in the meantime.
Thoughts? --Ivan Štambuk 23:11, 17 June 2008 (UTC)
romanisation of چ
editWhat's the romanisation of چ ? Is it č? Please specify in the body. I have just added the Persian translation of what time is it using this romanisation: sâ'at čand ast? (ساعت چند است؟). I hope it's right. --Anatoli 23:05, 18 January 2010 (UTC)
- Both č and ch can be used. The چ letter basically has the same sound as the ch in the English word "chair". Placebo 16:16, 27 July 2010 (UTC)
Templates
editNo {{fa-noun}}
, {{fa-verb}}
on this page? Then it's out of date, we generally avoid writing the categories out like Persian nouns when a template can do it. Mglovesfun (talk) 20:01, 5 February 2010 (UTC)
Misspelling
editI think it is not good to use 'y' in phrases like 'Jomhuri-ye Eslâmi-ye Irân' because it makes the reader misspell that as 'Jomhuriie Eslâmiie Irân' (i accented like p in 'tappe') though it is practically pronounced 'Jomhurie Eslâmie Irân'; I think it is better to write this way: "Jomhuri-e Eslâmi-e Irân" --C.pazoki 17:03, 5 July 2010 (UTC)
- Sorry?? Mglovesfun (talk) 17:13, 5 July 2010 (UTC)
- He’s talking about the ezafe, a sound that is used to connect nouns to their adjectives but which is not written. It is okay with me if we spell it -e instead of -ye, but how it’s spelled really will have no effect on how Persian students write it. It is elementary grammar and every student knows exactly what it is, no matter how it is spelled. —Stephen 18:13, 5 July 2010 (UTC)
- (Bump) In terms of romanisation I used زبان فارسی (zabân-e fârsi), adding ezafe if it's pronounced. Are there objections? --Anatoli 03:52, 21 February 2011 (UTC)
Spoken words used in casual conversations
editI would like to bring some words into attention which I'm honestly puzzled are worthy of an entry or not. As an example I'd like to use the word باران. This is the formal and correct way to spell it and in formal speeches the word is pronounced "bârân" as the entry says.
But in casual conversations, most people in Iran have developed an "oon" sound so the word would be pronounced "bâroon" instead, seeing as the last vocal is shorter and easier to pronounce. "bâroon" is only informal.
If you had to spell this with Persian letters it would be spelled as بارون.
My question is, do you think بارون, تهرون, ایرون, زندون, etc. deserves their own entry?
I would like to point out that these words are frequently being used on the Internet. If you look up these words on Google you can find quite a few hits on them. بارون comes up with 1,150,000 results and images of rain. ایرونی comes up with half a million hits. قلیون comes up with appo. 50 thousand and lots of pictures of hookah.
So these words are used frequently. They aren't used as much as their formal counterparts but still enough so that there is a chance somebody might run into these alternative spellings.
Furthermore, the Persian Wiktionary has the fa:بارون entry added saying it's meaning is باران.
So my question is, are these words worthy of an entry? What is the best solution to this? Placebo 16:02, 27 July 2010 (UTC)
- Of course they deserve their own entries. If such entries are created, please make sure they are marked as "colloquial" or "informal" and that they have links to literary and/or formal Persian spellings. Take a look at توانستن (tavânestan), تونستن (tunestan) and تونسن (tunessan). --Dijan 03:31, 31 July 2010 (UTC)
- I have tried to create a list with all of the words that came to my mind. There are many more that I will add as I remember them. Here is the list. The six entries that I have created so far are ایرون, ایرونی, بارون, تهرون, زبون, قلیون. Placebo 13:36, 31 July 2010 (UTC)
Entering zero-width non-joiner
editWhat are the methods to enter zero-width non-joiner on a computer? Apart from copying existing examples, I don't see another method. How is it done on Persian keyboards and on popular virtual keyboards? --Anatoli 03:44, 21 February 2011 (UTC)
- It is located in different places depending on the Persian keyboard that you use. Commonly it can be accessed by typing shift-space, or ctrl-alt-2, or ctrl-shift-2. Yours may be different. My Windows XP-Pro uses ctrl-shift-2. —Stephen (Talk) 11:44, 21 February 2011 (UTC)
- Thank you, Stephen. I try to type: خانهام - my house, I saw it written this way in a textbook. Interesting that Google produced heaps of خانهام. --Anatoli 12:10, 21 February 2011 (UTC)
Tajiki Persian is not descended from Iranian Persian.
editTajiki Persian is not descended from Iranian Persian, so I don't know why it is being shown in this way on Wiktionary. Kaixinguo~enwiktionary (talk) 12:23, 15 January 2018 (UTC)
- @Kaixinguo~enwiktionary: This will have to be changed here. @-sche? --Per utramque cavernam (talk) 13:06, 15 January 2018 (UTC)
- OK, I have set "tg" to have the same ancestor as "fa". I have also updated WT:AFA to reflect what was agreed upon and recorded at WT:LT after several discussions (linked there) with knowledgeable editors - that Dari is subsumed, not separate, while Tajik is separate still. - -sche (discuss) 18:41, 15 January 2018 (UTC)
Moved from User talk:Calak
editHey, Calak. Tajik is actually a variant of New Persian, and not descended from Middle Persian (300 BCE – 800 CE). As such, Tajik should be placed under Persian. Thanks!--Victar (talk) 10:42, 15 December 2017 (UTC)
- Perhaps actually placing Tajik, Persian and Dari under a "Classical Persian" level is best. See here. @JohnC5 what do you think? --Victar (talk) 10:51, 15 December 2017 (UTC)
- @Victar: Not my area of knowledge. —*i̯óh₁nC[5] 11:17, 16 December 2017 (UTC)
- @Victar: Hi! Yes, I agree. Iranian Persian (pes), Tajik (tgk) and Dari (prs) should be under "(Classical) Persian" (fas) level [1]. So for "intelligence" we have:
- (Classical) Persian: هوش (hōš)
- Dari: هوش (hōš)
- Iranian Persian: هوش (hûš)
- Tajik: ҳӯш (hüš)
- (Classical) Persian: هوش (hōš)
- So please define "pes", "tgk", "prs" and "fas" in Module:languages/data/3. Thanks.--Calak (talk) 15:13, 15 December 2017 (UTC)
- Calak, if you see the page I exampled, I'm using language code
fa-cls
for Classical Persian. --Victar (talk) 15:21, 15 December 2017 (UTC)- @-sche sorry to bother you again, but I've just noticed this. So apparently this edit (a followup to that conversation) is wrong, and the ancestor has to be moved one step further again. Sorry for the trouble! Pinging @Victar as well. --Per utramque cavernam (talk) 23:56, 21 January 2018 (UTC)
- Just to be clear, you want to go back to Tajik being a descendant of (an etymology-only subvariety of) Iranian Persian? I'll ping @Kaixinguo~enwiktionary, who participated in the WT:T:AFA discussion, to be sure they are on board with that.
On a separate note, Calak, Wiktionary uses two-letter codes rather than three-letter codes whenever the ISO has assigned them, so the code for Persian is "fa" not "fas", and Tajik is "tg" not "tgk", etc.
And based on past discussions (documented at WT:LT), "pes" and "prs" are subsumed into "fa" here, on the grounds that they are mutually intelligible. I could add an etymology-only code for Dari, though. - -sche (discuss) 01:12, 22 January 2018 (UTC)- Sorry, "one step further again" wasn't very clear; but no, not go back. From what I've read above, the ancestor of Tajik should be Classical Persian directly; it shouldn't go through Middle Persian, since the split happened before that. --Per utramque cavernam (talk) 01:21, 22 January 2018 (UTC)
- I know Classical Persian is ancestor of Tajik and Dari, but we should split Dari from Iranian Persian.--Calak (talk) 05:43, 22 January 2018 (UTC)
- Dari is mutually intelligible with Iranian Persian, and written in the same script. We have not had any problems that I am aware of in treating them as the same language. —Μετάknowledgediscuss/deeds 05:46, 22 January 2018 (UTC)
- Yes they have same script but have different phonology. Dari has 8 vowels but Iranian Persian has 6 vowels, beside different words. For example in Iranian Persian شیر/šir means 1.milk 2.lion but in Dari شیر/šir means milk and شیر/šēr lion. Now we can reflect this different.--Calak (talk) 06:12, 22 January 2018 (UTC)
- We already do reflect that, in our pronunciation sections. —Μετάknowledgediscuss/deeds 06:20, 22 January 2018 (UTC)
- Right; this mirrors other languages, including English, where some dialects have different phonology or words, and sometimes even (as in English) some spelling differences. - -sche (discuss) 16:40, 22 January 2018 (UTC)
- There are a couple of nonsense suggestions in this discussion as a whole, it's a bit dismaying to see. @Victar's suggestion was fine. Kaixinguo~enwiktionary (talk) 18:49, 22 January 2018 (UTC)
- @-sche: Just to make it clear, no, that is not what I meant at all about the etymology. By the way, I was referring to etymology sections in entries and 'descendants' sections though I have seen the issue crop up elsewhere. Perhaps this discussion does not belong under the same heading although they are related and thank you for moving them. Kaixinguo~enwiktionary (talk) 19:45, 22 January 2018 (UTC)
- @-sche, Metaknowledge: I am opening a new discussion about Dari Persian transliteration. It is unfortunate timing, but it seems that the issue has now become pressing. Kaixinguo~enwiktionary (talk) 19:45, 22 January 2018 (UTC)
- Right; this mirrors other languages, including English, where some dialects have different phonology or words, and sometimes even (as in English) some spelling differences. - -sche (discuss) 16:40, 22 January 2018 (UTC)
- We already do reflect that, in our pronunciation sections. —Μετάknowledgediscuss/deeds 06:20, 22 January 2018 (UTC)
- Yes they have same script but have different phonology. Dari has 8 vowels but Iranian Persian has 6 vowels, beside different words. For example in Iranian Persian شیر/šir means 1.milk 2.lion but in Dari شیر/šir means milk and شیر/šēr lion. Now we can reflect this different.--Calak (talk) 06:12, 22 January 2018 (UTC)
- Dari is mutually intelligible with Iranian Persian, and written in the same script. We have not had any problems that I am aware of in treating them as the same language. —Μετάknowledgediscuss/deeds 05:46, 22 January 2018 (UTC)
- I know Classical Persian is ancestor of Tajik and Dari, but we should split Dari from Iranian Persian.--Calak (talk) 05:43, 22 January 2018 (UTC)
- Sorry, "one step further again" wasn't very clear; but no, not go back. From what I've read above, the ancestor of Tajik should be Classical Persian directly; it shouldn't go through Middle Persian, since the split happened before that. --Per utramque cavernam (talk) 01:21, 22 January 2018 (UTC)
- Just to be clear, you want to go back to Tajik being a descendant of (an etymology-only subvariety of) Iranian Persian? I'll ping @Kaixinguo~enwiktionary, who participated in the WT:T:AFA discussion, to be sure they are on board with that.
- @-sche sorry to bother you again, but I've just noticed this. So apparently this edit (a followup to that conversation) is wrong, and the ancestor has to be moved one step further again. Sorry for the trouble! Pinging @Victar as well. --Per utramque cavernam (talk) 23:56, 21 January 2018 (UTC)
- Calak, if you see the page I exampled, I'm using language code
- @Per utramque cavernam: that (saying the split happened before Middle Persian) confuses me, since Classical Persian is encoded as, and also explained by other references as, an early variety of
fa
that came after Middle Persian. Wikipedia seems to agree with treating Tajik and (Iranian) Persian as developments of either Middle Persian or (Early New / Classical) Persian, but not anything earlier. - -sche (discuss) 16:40, 22 January 2018 (UTC)- @-sche: Ahah, I'm making a jumble of it all, sorry about that. I thought Classical Persian was the stage before Middle Persian. But I think (?) I've got it now.
- first solution (the previous one): Middle Persian > Classical Persian > Iranian Persian > Tajik
- second solution (the current one): Middle Persian > 1) Tajik + 2) Classical Persian
- third solution: Middle Persian > Classical Persian > 1) Tajik + 2) Iranian Persian
- I'll just let you guys decide what's best. --Per utramque cavernam (talk) 17:57, 22 January 2018 (UTC)
- Guys I talk about reconstruction namespaces not entries! how can we reflect different pronunciation between Iranian Persian and Dari in reconstruction namespaces?!--Calak (talk) 18:26, 22 January 2018 (UTC)
- Please hold important policy discussions like this at WT:AFA, not on an individual user's discussion page. Kaixinguo~enwiktionary (talk) 18:36, 22 January 2018 (UTC)
- Guys I talk about reconstruction namespaces not entries! how can we reflect different pronunciation between Iranian Persian and Dari in reconstruction namespaces?!--Calak (talk) 18:26, 22 January 2018 (UTC)
- @-sche: Ahah, I'm making a jumble of it all, sorry about that. I thought Classical Persian was the stage before Middle Persian. But I think (?) I've got it now.
- @Per utramque cavernam: that (saying the split happened before Middle Persian) confuses me, since Classical Persian is encoded as, and also explained by other references as, an early variety of
- I agree and have moved the discussion. - -sche (discuss) 18:48, 22 January 2018 (UTC)
- @Calak: How would you indicate the pronunciation differences if Dari had its own code? Indicate them the same way, but just using the code "fa" and qualifiers as needed. :) Also, I made "prs" (Dari) an etymology code, so you can write things like
{{cog|prs|سویدن}}
, and add vowel dots or a particular transliteration or{{IPAchar|//}}
as needed. - -sche (discuss) 18:56, 22 January 2018 (UTC) - @ZxxZxxZ, Irman, Dick_Laurent, Dijan, Atitarev, Stephen G. Brown, Kolmiel, LissanX, Rye-96, Aperiarcam Apologies to those whom I have forgotten. Kaixinguo~enwiktionary (talk) 19:26, 22 January 2018 (UTC)
- @-sche: There's been an agreement here thus so far that Dari Persian and Iranian Persian are being treated together, although the issue of Dari Persian-only transliteration has been lurking unresolved in the background, suddenly adding a Dari Persian-only etymology code represents a major shift. Please slow down such changes until the people who have actually made significant edits in Persian have contributed, too. Kaixinguo~enwiktionary (talk) 19:58, 22 January 2018 (UTC)
- Yes, this is all I need. We want to indicate the pronunciation differences, nothing more.--Calak (talk) 20:06, 22 January 2018 (UTC)
- @-sche: There's been an agreement here thus so far that Dari Persian and Iranian Persian are being treated together, although the issue of Dari Persian-only transliteration has been lurking unresolved in the background, suddenly adding a Dari Persian-only etymology code represents a major shift. Please slow down such changes until the people who have actually made significant edits in Persian have contributed, too. Kaixinguo~enwiktionary (talk) 19:58, 22 January 2018 (UTC)
- @Calak: How would you indicate the pronunciation differences if Dari had its own code? Indicate them the same way, but just using the code "fa" and qualifiers as needed. :) Also, I made "prs" (Dari) an etymology code, so you can write things like
- I'm aware of the prior discussion that led to Dari being subsumed into "fa", having started it, hehe. An etymology-only code doesn't represent a major shift, IMO: Dari is still handled under ==Persian==, but now words borrowed specifically from it can be marked. On reconstruction pages, the Dari forms can now be given more easily; but if this prompts someone to implement an even more robust solution instead, like separate "fa-tr=" and "prs-tr=" transliteration parameters to display Dari and Iranian transliterations with their different vowels, all the better. - -sche (discuss) 20:31, 22 January 2018 (UTC)
- OK, OK stop showing off :P By coincidence this Dari Persian issue has come up today at امریکا. @Metaknowledge has already objected to what I suggested about Dari Persian below. I think there have been plenty of discussions to be had about various aspects of Persian, but the main reasons not to have them recently have been that people such as ZxxZxxZ, Irman, Dick_Laurent and Dijan (and perhaps others) have all been absent, so it would be like trying to hold a conversation alone.
- P.S. Personally I wouldn't 'give' 'fa' to Iranian Persian, I thought Iranian Persian and Dari Persian have and equal claim on 'Persian', don't they? Kaixinguo~enwiktionary (talk) 20:54, 22 January 2018 (UTC)
- I object to your method of showing it, and the highlighting going on at امریکا is pretty bad on the eyes as well. I said that it's not a bad idea, but that it's not as necessary as you seem to believe, because we are not having actual problems in documenting Dari Persian. —Μετάknowledgediscuss/deeds 21:11, 22 January 2018 (UTC)
- As per usual, I must disagree. Dari Persian is scarcely being represented at all. It's a complete afterthought. Kaixinguo~enwiktionary (talk) 22:07, 22 January 2018 (UTC)
- I object to your method of showing it, and the highlighting going on at امریکا is pretty bad on the eyes as well. I said that it's not a bad idea, but that it's not as necessary as you seem to believe, because we are not having actual problems in documenting Dari Persian. —Μετάknowledgediscuss/deeds 21:11, 22 January 2018 (UTC)
- P.S. Personally I wouldn't 'give' 'fa' to Iranian Persian, I thought Iranian Persian and Dari Persian have and equal claim on 'Persian', don't they? Kaixinguo~enwiktionary (talk) 20:54, 22 January 2018 (UTC)
- @Calak: Thanks. Do you think you could outline your scheme for Dari Persian transliteration somewhere please when you have time? I take it that any other moves to show Dari Persian Romanisations in the future naturally will follow the same scheme, too. Kaixinguo~enwiktionary (talk) 22:31, 22 January 2018 (UTC)
- Hello everyone! I initially didn’t want to add anything here because I have absolutely no experience in the formatting required to organize these sorts of things, so pardon me beforehand for possibly making a formatting mess here. As I understand, the issue is distinguishing between Iranian, Dari and Tajik dialects of Persian? If so, I agree with making a distinction when it comes to categorizing entries here separately and distinctly formatting for Iranian Persian, Dari Persian and Tajik Persian. I believe a lot of misinformation could be spread by implying Iranian-specific aspects of Persian, particularly when it comes to colloquialism, also apply to Dari or Tajik Persian. If the conversation here is about something else entirely, then forgive my lack of Wiki-savviness because I don’t know what exactly is going on here... (lol). User:LissanX (talk) 23:36, 26 January 2018 (UTC)
Please do not change placement of synonyms without discussion
editSome people have added a new placement of synonyms under each definition line, but they have only done it in a few entries and then left all the others as they are. And sometimes the synonyms they gave were wrong anyway. I believe that this placement of synonyms is a mistake for Persian, as it doesn't suit the future of the way the entries will develop when they (hopefully) show Dari persian and have Dari Persian transliterations as well, which will make entries much longer with two transliterations (e.g. بچه bačče Iran bačča Afghanistan, as a possibility). Anyway, that is only my opion, but if there is a decision to switch the formatting in this way, at least put a notice here informing everyone and edit more than one entry so you are not just leaving 8999 entries out of sync with all the others. — This unsigned comment was added by Kaixinguo~enwiktionary (talk • contribs). at 12:33, 15 January 2018 (UTC)
- I don't agree on any of these points. I don't think we necessarily need to elevate Dari transliterations in the future (although it's not a bad idea), I don't think that your proposed format for doing so is efficient or aesthetically pleasing, I don't think adoption of such a format would impede usage of
{{syn}}
and its kin, and I don't think that entries being "out of sync" in terms of a relatively minor formatting detail is at all a problem. —Μετάknowledgediscuss/deeds 18:49, 15 January 2018 (UTC)
Romanisation of Dari Persian - offer your opinions please
editIt is common practice to put an Iranian Persian-based transliteration for the simple reason that that is what the majority seem to have been most familiar with, so far. There are a handful (~30) of Dari-Persian only words that have no transliteration, because it would be ludicrous to have an Iranian Persian transliteration for a Dari Persian-only word and no Dari Persian transliteration. There has been no native speaker of Dari Persian on Wiktionary so far, let's face facts: there have only been a handful of editors of Persian at all.
I understood and I always believed that there was a common consensus that the default reasoning for the absence of Dari Persian transliteration so far is that no native speaker of Dari Persian had started editing and put forward their own transliteration system, example sentences etc. And there are compelling reasons for it not to be done ad hoc by others. There's a plethora of bad and unreliable resources out there for Dari Persian that are not accurate or confuse Iranian Persian with Dari Persian. Some of them were no doubt produced with the American military market in mind during that era just to make a quick profit and have been worked on by Iranian Americans, probably, resulting in inaccuracy and misinformation. There's even a dictionary (!) of Dari Persian to Iranian Persian that is full of faults, even a respected Iranian publisher cannot get it right [2]. It is a not a simple area to deal with. There are political and cultural considerations to be taken into account as well as linguistic ones.
Though the decision has been made to have Dari Persian and Iranian Persian together, it has to be understood that some Dari speakers could well appear one day and demand to know why they haven't been given, or why they aren't entitled to their own language sections. Dari Persian has been incorporated into Persian here, but it's always been my understanding that it will have equal and fair treatment and that it is alongside Iranian Persian, not 'subsumed' or 'beneath' waiting to be 'elevated', and that is how their inclusion under one language header can be justified. The only reason (as I understood it) for the absence of Dari Persian transliterations, has been a lack of qualified editors, not that is in any way inferior or less significant. I believe the equal representation of Dari Persian is important, not only because that would one day make Wiktionary a useful resource for Dari Persian, but also because the fact is that the one 'variety' of Persian is not superior to another. Moreover, Dari Persian remains closer to the classical pronunciation and it's very beautiful to hear poetry in that accent. As I said, Dari is used by a whole country of people (well, the ones of that country who speak it), it is just as dear and important as Iranian Persian in everyone's hearts.
Perhaps the only reasonble and fair alternative to having transliteration sections for both Iranian Persian and Dari Persian, would be to put every word with a classical Persian transliteration section and Iranian and Dari pronuncations (and others) in the pronunciation section together. I believe that Z devised a scheme for this and it looked good, but it hasn't been applied extensively for unknown reasons.
However, having aired some of these thoughts due to recent developments, it seems that not everyone is in agreement as I had assumed. Please everyone give your thoughts on the status of Dari Persian transliterations which will be, no doubt, more informed than my own. Kaixinguo~enwiktionary (talk) 21:33, 22 January 2018 (UTC)
- I am sympathetic to the idea of elevating classical romanisations, but I presume most users are coming here for Iranian Persian, and I think they would be confused by seeing ē where they expect i, for example. —Μετάknowledgediscuss/deeds 21:46, 22 January 2018 (UTC)
- By the way — since this needs to be said — we are not disrespecting Dari by standardising on Iranian Persian. There are at least 3 times as many native speakers of the latter (probably more like 4 or 5 times as many), and there is vastly more written material (more than 90% of Iranian adults are literate, but only about 30% of Afghanistani adults). By treating Iranian Persian as the norm, we are simply following the facts of what we can use as evidence in entries. That doesn't mean it's the right way to go about this, but it does mean that it is justified and not because one is less "dear in our hearts". —Μετάknowledgediscuss/deeds 22:04, 22 January 2018 (UTC)
- We don't even have Classical romanizations at all; even having them in the pronunciation section would be helpful.
- Also @Metaknowledge I see that you think Kaixinguo's proposed romanization format is unwieldy. How about this: بچه (I. bačče, A. bačča) —AryamanA (मुझसे बात करें • योगदान) 22:09, 22 January 2018 (UTC)
- The approach used at امریکا seems bizarre to me: two separate sections, labelled as having different etymologies(!), for identical headwords (امریکا) and meanings ... two separate sections for the same word? IMO, we can handle it better the way that e.g. words that have more than one gender, or dialectally-different past tenses, or several script forms (e.g. Serbo-Croatian) are handled.
I like Aryaman's suggestion, except that I would spell the words out, like "Iranian:" and "Dari:". But if we do that, it would seem easier to maintain in a standardized/unified way (so we don't have some entries saying "Dari:" and some saying "Afghani:" and some saying "Afghanistan:", etc) and cleaner in terms of the code and machine-readability and reusability of output, to have versions or "masks" of{{l}}
and{{m}}
for Persian that allowed the two transliterations to be input as different parameters, rather than manually typing "Iran:" into the "tr=" as if it were part of the transliteration. In other words, on the model of{{he-l}}
, perhaps have {{fa-l|بچه|I=bačče|A=bačča}} outputting بچه (Iranian: bačče, Dari: bačča).
Less optimally, since it would needlessly require a lot more typing, would be to only make{{l}}
and{{m}}
accept tr1= and tr2= and trgloss1= and trgloss2= (which would also be necessary as a backend to the "masks" above), and (if certain users preferred not to allow language-specific "masks") require people to spell out {{l|fa|بچه|tr1=bačče|trgloss=Iranian|tr2=bačča|trloss2=Dari}}, preferably in such a way that the parameters of trgloss1= and trgloss2= were passed through a module that could standardize "Iran:" and "Iranian:" etc into a single displayed label. - -sche (discuss) 22:53, 22 January 2018 (UTC)
- @-sche: It may seem bizarre to you but the two words have different pronunciations and have entered Dari Persian and Iranian Persian separately. There's no need for the exclamation mark. Dari Persian has taken most of its words for country names independently of Iranian Persian, and the same goes for Tajiki Persian. There's plenty of words that are Dari-Persian only and the same goes for Iranian Persian, it's just that it seems more striking because they happen to be on the same page in this case. Another example along similar lines is مین which means 'mine' (landmine) in each place but has come from French 'mine' in Iranian Persian and English 'mine' in Dari Persian. The pronunciation and etymology is different, but the meaning is the same. But that's actually virtually irrelevant to the issues being discussed here. I would use 'Iran' and 'Afghanistan' rather than 'Iranian' and 'Dari', but I would leave the actual Dari Persian transliteration field empty until some native speakers of Dari Persian started editing, for the reasons I have stated above. Kaixinguo~enwiktionary (talk) 23:10, 22 January 2018 (UTC)
Modification of English-Persian translation layout
editWhat if the English - Persian translation sections were modified to follow the pattern suggested above by Calak?
- (Classical) Persian: هوش (hōš)
- Dari: هوش (hōš)
- Iranian Persian: هوش (hûš*) [*huš in the system used at the moment in entries]
- Tajik: ҳӯш (hüš)
- (Classical) Persian: هوش (hōš)
I think it looks really good. Does anyone have any thoughts on this? Maybe it's been rejected already as an idea. @ZxxZxxZ, Irman, Dick_Laurent, Dijan, Atitarev, -sche, Metaknowledge — This unsigned comment was added by Kaixinguo~enwiktionary (talk • contribs).
- Just so you know, Kaixinguo, if you don't add your signature in the same edit as you add a ping, it won't go through. That's also why when you ping people and then later try to add more people to the template, it won't go through.
- As for the suggestion... I don't think people looking for Tajik translations will think to look for them under P. (But you could make the same argument for Dari and Farsi already, I suppose.) —Μετάknowledgediscuss/deeds 16:41, 23 January 2018 (UTC)
- (And Chinese. —suzukaze (t・c) 18:41, 23 January 2018 (UTC))
- In general, I think translations should be listed under the same name as the entries use (in the L2 header), since that's where anyone who uses Wiktionary for any length of time, and looks at the entries, will then (quite reasonably) look in the translations tables. - -sche (discuss) 19:11, 23 January 2018 (UTC)
Finishing transition to lower-case Romanisations for proper nouns
edit@ZxxZxxZ, Irman, Dick_Laurent, Dijan, Atitarev, LissanX, Metaknowledge There has been a move at some point to change the Romanisations for proper nouns to lower case, but it is incomplete. The entries can't stay as they are with some proper nouns capitalised and others lower case so I am proposing to finish this by making them all lower case (I don't have any personal preference whatsoever). Please let me know if you object. Kaixinguo~enwiktionary (talk) 18:38, 27 March 2018 (UTC)
- Yes, of course, it has to be done. A while ago I requested a bot run but nobody did it. --Anatoli T. (обсудить/вклад) 21:16, 27 March 2018 (UTC)
- OK, it would be good if others would confirm that this is what they want, though. I couldn't find the discussion at all so I thought I would check. Kaixinguo~enwiktionary (talk) 13:35, 28 March 2018 (UTC)
- Also, should this cover usage examples, too? Kaixinguo~enwiktionary (talk) 13:37, 28 March 2018 (UTC)
I am undecided, many scholarly works (though not dictionaries) do use capitalization for transliteration of Persian or other scripts. Moreover, Tajik, which uses Latin as well, has a capitalization system. --Z 13:45, 29 March 2018 (UTC)
- OK I won't do anything until there is a consensus, thank you :). Kaixinguo~enwiktionary (talk) 15:22, 29 March 2018 (UTC)
{ {collapse|content=
@ZxxZxxZ: I'm sorry, but User:Atitarev has gone ahead with the change anyway. I can't not go along with it because he will block me from editing pages again (see:[3]). Kaixinguo~enwiktionary (talk) 22:30, 1 April 2018 (UTC)- @Kaixinguo~enwiktionary: That protection was made back in 2015, long gone now - when it seemed clear that there IS a consensus to keep lower case transliterations only. If you think that this is still being discussed and not agreed on, I won't protect pages.
- The standard at the time was to capitalise romanisations. That is why they were all that way, and why the regular editors of Persian were capitalising them. It was pretty awful of you to block me from editing 'Iran' and I did leave for almost two years after that. Anyway, it doesn't matter now.
- @Kaixinguo~enwiktionary: It wasn't my impression that it was standard. Lower case and upper case were both used by regular used and there WAS a discussion about with most editors agreeing to use the lower case for all Persian transliterations. I'm sorry if I caused you grief then, it wasn't my intention. @Wikitiki89, -sche: Do you happen to remember where that discussion was? --Anatoli T. (обсудить/вклад) 13:14, 3 April 2018 (UTC)
- @Atitarev: Found it: Wiktionary:Beer parlour/2014/January#Capital letters in transliterations of languages that do not have capital letters. --WikiTiki89 14:45, 3 April 2018 (UTC)
- I saw that conversation before when I was looking for how this situation came about and why someone suddenly started changing all the entries. It's a conversation started and conducted by people who don't edit Persian, it's vague and inconclusive and there's only one comment about Persian from someone who has edited it regularly (Dijan comments, but about Arabic). Kaixinguo~enwiktionary (talk) 13:34, 14 April 2018 (UTC)
- @Atitarev: Found it: Wiktionary:Beer parlour/2014/January#Capital letters in transliterations of languages that do not have capital letters. --WikiTiki89 14:45, 3 April 2018 (UTC)
- @Kaixinguo~enwiktionary: It wasn't my impression that it was standard. Lower case and upper case were both used by regular used and there WAS a discussion about with most editors agreeing to use the lower case for all Persian transliterations. I'm sorry if I caused you grief then, it wasn't my intention. @Wikitiki89, -sche: Do you happen to remember where that discussion was? --Anatoli T. (обсудить/вклад) 13:14, 3 April 2018 (UTC)
- The standard at the time was to capitalise romanisations. That is why they were all that way, and why the regular editors of Persian were capitalising them. It was pretty awful of you to block me from editing 'Iran' and I did leave for almost two years after that. Anyway, it doesn't matter now.
- @Atitarev: No, don't apologise, because it wasn't your fault, I think I just thought at the time that I was wasting my time because the transliteration couldn't be standardised, that nothing would ever progress because of that with Persian or it would all always be inconsistent.
- Even if there was a conversation about it (which I missed), it can't have been being followed uniformly at the time. It doesn't matter any more, there just needs to be an agreement now. Kaixinguo~enwiktionary (talk) 13:27, 3 April 2018 (UTC)
- Well, hopefully if there is an agreement, people will also agree to follow it. I don't know why it was difficult to agree on either "â" or "ā" and "ou" or "ow", the capitalisation rules and stick to that policy. I mean, if somebody doesn't like the existing policy, should suggest a change, rather disregard it and use whatever seems nice to him/her at the moment. There's no perfect transliteration for any script, it's a choice (hopefully based on a standard with any required modification), which should be made, be well described and adhered to. I prefer consistency and following rules, even if I may not like some individual choices. --Anatoli T. (обсудить/вклад) 13:52, 3 April 2018 (UTC)
- If User:Qehath would give his opinion then that would be all the main editors of Persian (except for one...). Kaixinguo~enwiktionary (talk) 12:46, 3 April 2018 (UTC)
- Also I just want to point out, that it is years that the Persian transliteration has been causing problems now. People who edited really well have permanently left over it. There is this constant back-and-forth over â and 'a' with macron that I can't type right now. It's more than ten years that this goes on for. Kaixinguo~enwiktionary (talk) 12:55, 3 April 2018 (UTC)
- I mean, TEN YEARS. Maybe more than that! I believe we lost User:Optional, User:Placebo and Ivan Stambuk over it.
@Atitarev I have put that we have de facto changed to lower-case now below in the summary of changes. Add to the list if you can think of anything. User:ZxxZxxZ, User:Qehath and User:Dijan did all the important work on Persian entries, so I would have liked to see an agreement from them really. User:Dijan probably isn't here enough to give his input, though. Kaixinguo~enwiktionary (talk) 23:58, 1 April 2018 (UTC)
- I strongly believe that standard Persian transliteration should all be in lower case. If not, this policy page and WT:FA TR shouldn't be ambiguous about it. --Anatoli T. (обсудить/вклад) 23:35, 1 April 2018 (UTC)
I agree with making transliterations lower-case. It makes no sense, in my opinion, for a transliteration to abide by the grammatical rules of English like it’s a quasi-translation. - LissanX (talk) 01:16, 30 March 2018 (UTC)
I, too, am in agreement to make all transliterations of Persian in lower case. Obviously, this would not apply to the Tajik variants. -Dijan (talk) 01:41, 3 April 2018 (UTC)
- Thanks, Dijan. Of the people who have an opinion then that's more in favour of lower case so far. Kaixinguo~enwiktionary (talk) 12:48, 3 April 2018 (UTC)
I actually don't have super strong opinions on this. Who would have expected that? Anyway it seems like lower case is winning, I'm ok with that. — [ זכריה קהת ] Zack. — 23:44, 3 April 2018 (UTC)
- @Qehath: Thanks 😊, any chance you could act as ambassador to Armenia and find out what they think? Kaixinguo~enwiktionary (talk) 00:16, 4 April 2018 (UTC)
- I have no reason to expect he'd feel differently. — < [ זכריה קהת ] Zack. — 02:22, 4 April 2018 (UTC)
- @Qehath: Would you prefer if the examples are lower case as well? While I was searching for conversations about this, I think you commented in the past that they should be capitalised. Kaixinguo~enwiktionary (talk) 13:33, 14 April 2018 (UTC)
As there seems to be strong feelings surrounding discontinuing and deleting {{etyl}}
I could convert some of the instances of this template whilst I am changing the capitalisation of proper nouns. Is it accurate to convert every instance of {{etyl}}
to {{bor}}
when the Persian comes from Arabic?
Also, some of the sections using {{bor}}
are worded as 'Borrowed from...' and others say 'From...', which is preferable? Thanks. Kaixinguo~enwiktionary (talk) 13:32, 28 March 2018 (UTC)
- @Kaixinguo~enwiktionary, if
{{bor}}
is being used, it should be "borrowed from". I don't know why this is being brought up on this page. --Victar (talk) 15:16, 14 April 2018 (UTC)- It's being discussed on this page because this is the page for discussing the formatting of Persian entries. Whoever it was who altered this template created a big mess. Some entries say 'borrowed from' and others say 'from'. Since a bot changed everything, it is all inconsistent. Kaixinguo~enwiktionary (talk) 15:51, 14 April 2018 (UTC)
- Where does it say that it should always be 'borrowed from'? If that's the case, why was the template altered to stop it showing 'borrowed from', so that now someone has to go and change every entry back, if you're saying it has to be that wording? Kaixinguo~enwiktionary (talk) 15:53, 14 April 2018 (UTC)
- @Kaixinguo~enwiktionary, that's nonsense. This is a general formatting question, not specific to Persian. Persian should not have its own formatting scheme for
{{bor}}
. The removal of the lead text, which I for one was against, was due to people having to use parameters to modify the grammar, usually from "Borrowing of" to "borrowed from". It was decided that the it would be easier to manually type it all in instead. --Victar (talk) 16:03, 14 April 2018 (UTC)- Right, no need to be rude af about it. Kaixinguo~enwiktionary (talk) 16:07, 14 April 2018 (UTC)
- @Kaixinguo~enwiktionary, that's nonsense. This is a general formatting question, not specific to Persian. Persian should not have its own formatting scheme for
Actually, maybe not
editWell, only one person replied here anyway but the fact is I have been in an accident in the snow last month and I thought this was a good chance to do something useful while I am getting better. But I don't need to deal with this kind of thing: https://en.wiktionary.org/w/index.php?title=User_talk%3AGfarnab&type=revision&diff=49230382&oldid=49230190 at the moment. It's still useful if anyone puts what they think on these issues but I will be leaving this for now. Kaixinguo~enwiktionary (talk) 14:21, 28 March 2018 (UTC)
- Yes, it'd be a good idea since etyl is deprecated and we ought to progressive get rid of it. bor is for borrowings, inh for inherited terms and der for whenever we are not sure about which of the former two to use.
- I wish you the very best in your recovery, be it while editing or not
=== Summary of formatting changes in Persian entries ===
@ZxxZxxZ, Irman, Qehath, Dijan, Atitarev, LissanX, Metaknowledge, Calak, Lo_Ximiendo, AryamanA
- All romanisations of proper nouns are transforming to lower case.
{{etyl}}
is being removed.- 'Borrowed from' is becoming 'from'.
- 'Ou' is becoming 'ow' in the abscence of any other agreement, information or discussion. I will maintain a list of words with 'ow' so it can be put back easily if the decision changes.
- Dari Persian transliteration is still outstanding.
Kaixinguo~enwiktionary (talk) 23:12, 1 April 2018 (UTC)
- This discussion has become unfocused and overreaching. It should be broken up into individual points, most of which probably belong in the Beer Parlour, not here. --Victar (talk) 15:19, 14 April 2018 (UTC)
- The discussion is not unfocused whatsoever. The relevant points have indeed been broken down into separate headings. I have explained why I was raising several points at once, it is because I did have a lot of free time to go through each entry and to try and make them more consistent. If I was going to do that, then it was worth trying to get some consensus on these points rather than edit them all once to change the capitalisation, then again to remove 'etyl', then again to change 'ow' to 'ou' or vice-versa, then again to add 'borrowed from' back into etymology sections...do you see my point? Kaixinguo~enwiktionary (talk) 15:48, 14 April 2018 (UTC)
- With all due respect, this whole discussion reads like thought vomit, as I as said, overreaching. --Victar (talk) 16:03, 14 April 2018 (UTC)
- Like I said, rude as fuck. Fuck this. You could have explained it more easily than just complaining. Kaixinguo~enwiktionary (talk) 16:09, 14 April 2018 (UTC)
- So I volunteer to spend my time spent recovering in a lot of pain from an accident cleaning up tedious formatting problems created by other people, and I try to get some consensus on it so I'm not totally wasting my time, and all you can do is be rude and complain? Kaixinguo~enwiktionary (talk) 16:13, 14 April 2018 (UTC)
- With all due respect, this whole discussion reads like thought vomit, as I as said, overreaching. --Victar (talk) 16:03, 14 April 2018 (UTC)
- The discussion is not unfocused whatsoever. The relevant points have indeed been broken down into separate headings. I have explained why I was raising several points at once, it is because I did have a lot of free time to go through each entry and to try and make them more consistent. If I was going to do that, then it was worth trying to get some consensus on these points rather than edit them all once to change the capitalisation, then again to remove 'etyl', then again to change 'ow' to 'ou' or vice-versa, then again to add 'borrowed from' back into etymology sections...do you see my point? Kaixinguo~enwiktionary (talk) 15:48, 14 April 2018 (UTC)
- Well many thanks for your very helpful contribution. Kaixinguo~enwiktionary (talk) 16:14, 14 April 2018 (UTC)
- Kai, is this helping you in your recovery? I do appreciate your edits and hope you make more of them, but I posit you have first a duty with your health. You maybe have more to win at the moment listening to Beethoven's Heiliger Dankgesang than writing to Victar.
- Whatever you decide, I hope you rest and get well soon!
}}
This issue is now settled in favour of converting to lower case [4], with three neutral votes (User:ZxxZxxZ, User:Qehath and me (four if you count V.P.), three in favour (User:Atitarev, User:Dijan and User:LissanX), no opposing votes and no vote from User:Irman. Kaixinguo~enwiktionary (talk) 14:49, 27 April 2018 (UTC)
Example of Tajiki Persian published in the Latin alphabet
editI can't believe I have finally seen it: [5]
Technically speaking one could make entries for some Tajiki Persian entries in the Latin alphabet and use these images as the citations... Kaixinguo~enwiktionary (talk) 11:34, 4 April 2018 (UTC)
- Unfortunately they're all from the same source. --WikiTiki89 14:02, 4 April 2018 (UTC)
@Qehath, ZxxZxxZ, Irman When I'm updating the proper nouns that are place names, I could switch to using {{place}}
which Chinese seems to have adopted. Should I do that or would you prefer to keep the current format of: [[placename]] {{gloss|a city in___}}? Kaixinguo~enwiktionary (talk) 08:14, 20 May 2018 (UTC)
- For example: لندن. @Atitarev as well. Kaixinguo~enwiktionary (talk) 08:25, 20 May 2018 (UTC)
- @Kaixinguo~enwiktionary: The use of templates is encouraged, especially if they are properly and add categrisations but it's not a requirement in this particular case, IMO. --Anatoli T. (обсудить/вклад) 08:39, 20 May 2018 (UTC)
- Only London capital of Uk pronounced landan in Persian elsewhere it would pronounce london, lândon or lânden, so you did well.(Irman (talk) 08:57, 20 May 2018 (UTC))
- @Kaixinguo~enwiktionary: The use of templates is encouraged, especially if they are properly and add categrisations but it's not a requirement in this particular case, IMO. --Anatoli T. (обсудить/вклад) 08:39, 20 May 2018 (UTC)
- I don't...super care. Sorry. (I like inflection stuff) — [ זכריה קהת ] Zack. — 23:48, 23 May 2018 (UTC)
- @Qehath: OK, thank you. I didn't think you would mind either way, but it's good to avoid controversial changes. Kaixinguo~enwiktionary (talk) 11:25, 31 May 2018 (UTC)
- Speaking as someone who is only now learning how to use
{{place}}
, I think it's far superior and can allow for more powerful documentation and categorisation of placenames in Iran. —Μετάknowledgediscuss/deeds 13:54, 31 May 2018 (UTC)- @Metaknowledge: Sorry, I missed your reply. Looking at some of the proper nouns, there does seem to be more inconsistency than other entries, possibly because there's often less reason for people to update them. If you think it's a good idea then perhaps it should be adopted. Kaixinguo~enwiktionary (talk) 18:20, 3 June 2018 (UTC)
- Speaking as someone who is only now learning how to use
- @Qehath: OK, thank you. I didn't think you would mind either way, but it's good to avoid controversial changes. Kaixinguo~enwiktionary (talk) 11:25, 31 May 2018 (UTC)
Standardising transliterations - remaining points
editI don't really know who, if anyone, wants to be pinged about this, but there are some remaining points that are still not standardised across entries; it would be really helpful if you could vote to indicate your preferences:
1.'Ou' or 'ow'
editExamples: روشن as 'rowšan' or 'roušan', دولت as 'dowlat' or 'doulat'.
'ou' (roušan/doulat)
edit'ow' (rowšan/dowlat)
edit- more common.--Calak (talk) 08:49, 6 June 2018 (UTC)
- Support, already the standard in most entries. —Μετάknowledgediscuss/deeds 10:16, 6 June 2018 (UTC)
- It should be written as 'ow'. On a related note, 'o' should also be written as 'ow' when it is transliterating ـُوْ. Some examples include Nowruz (نُوْرُوز), Ferdowsi (فِردُوْسِی) and Khosrow (ْخُسْرُو). LissanX (talk) 01:12, 19 December 2018 (UTC)
2. 'M' or 'n' where ن in the combination نب is pronounced 'm'
editExample: دوشنبه as 'došanbe' or 'došambe'.
'm' (šambe)
edit- yes, indeed!--Calak (talk) 08:55, 6 June 2018 (UTC)
'n' (šanbe)
edit- Support, closer to original script. —Μετάknowledgediscuss/deeds 10:16, 6 June 2018 (UTC)
3. Gemination reflected in all transliterations including without prescence of ـّ in the script
edityes (bačče/tappe/melli/sonnati)
editno (bače/tape/meli/sonati)
edit4. Initial ع reflected in transliteration or not
editE.g. عزیز as 'aziz' or aziz, initial ' is omitted in some dictionaries; both can be found on Wiktionary.
aziz
edit'aziz
edit5. خوا as 'xa' or 'xwa'
editE.g. گیاهخواری.
xa
edit- Support; xw should only be in specifically classical transcriptions, as Calak says. —Μετάknowledgediscuss/deeds 10:16, 6 June 2018 (UTC)
xwa
edit- xw- in classical Persian, x- elsewhere (classical Persian xwaš but new Persian/Dari xoš).--Calak (talk) 08:53, 6 June 2018 (UTC)
Thanks. Kaixinguo~enwiktionary (talk) 17:38, 5 June 2018 (UTC)
- Ping didn't work, trying again:
@Ariamihr, Atitarev, Calak, Dijan, Irman Kaixinguo~enwiktionary (talk) 07:32, 6 June 2018 (UTC)
- @Kolmiel, Metaknowledge, Qehath, Rye-96, ZxxZxxZ Kaixinguo~enwiktionary (talk) 07:33, 6 June 2018 (UTC)
- @Calak Yes, I accept this sound Support. /xw/ exists in classic Persian and some modern dialects such as Persian Kermanshahi (not Kurdish Kermanshahi), Persian Ilami (not Kalhuri Kurdish) and some northwestern Persian dialects of Iran. Ariamihr (talk) 10:46, 6 June 2018 (UTC)
- Overall, if we choose a dictionary/standard to base the transliterations on, my opinion is not important but anyway:
- 'ou' was used much longer and this is what I normally see in dictionaries.
- '-mb-' but no strong opinion on this. My small dictionary use it. Metaknowledge has a point there, though.
- Render geminations. Writing shadde is never required in Persian.
- Use ' to show the spelling and for etymological reasons. BTW, completely ignored in Cyrillic Tajik in initial positions.
- 'xa' if it's pronounced so in this variety. This is way too common in textbooks.
- I also think for transliteration purposes, we should for one standard, not multiple, e.g. formal Iranian, Tehrani but pronunciation sections can show Dari, colloquial, regional, etc. --Anatoli T. (обсудить/вклад) 10:57, 6 June 2018 (UTC)
- @Atitarev: Do you mean a transliteration scheme from a dictionary should be followed instead? Regarding only putting formal Iranian transliterations, I think you know I don't agree, but in any case that should be raised as a discussion point as there are some words with a classical-based transliteration as well as some that try to reflect both Dari Perisan and the Persian of Iran. Kaixinguo~enwiktionary (talk) 13:57, 6 June 2018 (UTC)
- @Kaixinguo~enwiktionary:: I mean if there IS an existing standard we can follow and is acceptable, then we don't have to decide on the above questions. As for Iranian transliterations, it's just my opinion. Persian transliterations, most likely, will never be automated and the other transliterations (Dari, colloquial) are less accessible, and so, more likely to be in error by new generations of editors. --Anatoli T. (обсудить/вклад) 22:38, 6 June 2018 (UTC)
- @Atitarev: Do you mean a transliteration scheme from a dictionary should be followed instead? Regarding only putting formal Iranian transliterations, I think you know I don't agree, but in any case that should be raised as a discussion point as there are some words with a classical-based transliteration as well as some that try to reflect both Dari Perisan and the Persian of Iran. Kaixinguo~enwiktionary (talk) 13:57, 6 June 2018 (UTC)
Thank you to those who did reply. Kaixinguo~enwiktionary (talk) 09:01, 9 June 2018 (UTC)
Persian automated transliteration discussion
editThere is a new discussion on automating Persian transliteration by providing diacritics in the headword - Wiktionary:Beer_parlour/2021/December#Persian_automated_transliteration. --Anatoli T. (обсудить/вклад) 23:41, 15 December 2021 (UTC)
- Did this ever go anywhere? It looks like the proposal would have passed by the 2/3 standard. The current solution seems to be to manually add the transliteration, which can lead to inconsistency. One compromise could be to add the diacritics for the purpose of the transliteration but omit them from the visible display. 70.172.194.25 06:04, 24 January 2023 (UTC)
Two Transliteration Systems?
editPersian Wiktionary does technically have a Dari transliteration system but generally only fallsback to it when an Iranian Transliteration is not available. This is obviously not a big issue for Iranian users but is a pretty big issues for Dari users. With Dari specific words where a Dari Translation system is made up or (when certain edits are insistent on using Iranian transliteration) an Iranian pronunciation is made up, and a transliteration is written which does not match any standard variety of Persian. Since Tajik has its own Wiktionary section, and debates on making Dari separate have not been settled, I think having two transliteration systems would be the best way to go. Since Persian (excluding Tajik) only has two standards, there would only ever be a need for two transliterations max. And I think adopting such a transliteration system would (help) settle debates about separating Dari from the Persian Wiktionary.
For example the کابل wiktionary page could be
کابل • (kâbol; kâbul)
others could be something like:
گل • (gol; gul)
رنگ • (rang)
دیوانه • (divâne; dēwâna)
This can also help ease confusion when a Dari-specific word randomly switches transliteration systems, if a word isn't used in Iran then the translation system could mark that as such
بنگلهدیش • ( - ; bangladēš)
کوریا • ( - ; kōrîyâ)
تیلفون • ( - ; tēlifōn)
هوتل • ( - ; hōtal)
because of the dash, it's somewhat more clear to the user why transliteration system is different, currently Dari specific words usually just switch transliteration with no indication. Which may be confusing.
I also think for words without Dari transliterations it may be beneficial to do the opposite, but that can be discussed.
if there is agreement on creating two systems here is my proposal:
ا • (a; a) | آ • (â; â) | اِ • (e; i) | ای • (i; ē) | اِی • (i; î) | اُ • (o; u) | اُو • (u; û) | او • (u; ō) | و • (v; w)
The Dari translation is based on ALA-LC switching only v > w. The Iranian transliteration is just based on the current one.
I can see some opposition to this so let me make a few points first
- Would it be confusing to the user?
- IMO not having an indication of which transliteration is being used or making up an Iranian pronunciation without indication is more confusing.
- Why would this help prevent Dari from splitting?
- most Dari users I see complain that the wiktionary is a worse experience than just using any other Dari dictionary online. Especially the lack of indication for Dari's vowels in transliterations, requiring Dari speakers to learn some IPA beforehand. This makes it somewhat inaccessible. I think two transliterations is the easiest way to address this and similar the issues without splitting the wiktionary.
- Changes from current transliterations?
- weirdly, all Dari-specific diphthongs are already standardized in the current transliteration system, (ay, aw) but not Dari's short vowels. This would add distinctions for Daris short vowels (and the letter w) and won't actually change much else. There also seems to already be a Dari transliteration system, as I've said, but it's usage is inconsistent as of right now.
- How do other Wiktionary's with this issue solve it?
- Usually by selectively splitting. Like how Arabic entries for specific dialects will have a separate category from Arabic, but discussions in this wiktionary indicates people want to solve this issue but don't want to split. I think this is the best resolution without splitting the wiktionary. Sameerhameedy (talk) 04:35, 24 January 2023 (UTC)
- @Sameerhameedy: Just a question for now. Maintaining two standards would requires two vocalisation system, e.g. دُولَت (dowlat) (Iranian), دَوْلَت (dawlat) (Classical or Dari). (Interestingly, the Iranian diphthong /ow/ is transliterated as "ow" per latest agreement). It would be possible to largely automate transliteration from vocalisations but the idea to provide vocalisations wasn't very welcome, one reason possibly because of the known regional differences. Do you embrace the use vocalisations in the headword, e.g.
|head=
دُولَت,|head2=
دَوْلَت? - There hasn't been a lot of activity in the Persian namespace lately. (Notifying Ariamihr, Dijan, Mazsch, Qehath, ZxxZxxZ): calling just in case. Anatoli T. (обсудить/вклад) 05:22, 24 January 2023 (UTC)
- Also @User1267183728390127891247, Fenakhay, Fay Freak, Benwing2, Erutuon, Tibidibi, Kaixinguo~enwiktionary, if you're interested. --Anatoli T. (обсудить/вклад) 05:29, 24 January 2023 (UTC)
- @Atitarev I saw that proposal but I believe it's different from mine, though I may have read it incorrectly. I believe that proposal was to creat automatic IPA transliterations using voiced Arabic headers, I don't know the technicalities of that, but most Dari dictionaries I've used use latin transliterations for pronunciation. I personally was thinking something like:
- ===Etymology===
- from bla bla
- ===Pronunciation===
- classical IPA: /kɑbul/
- Dari IPA: /kɑbʊl/
- Iranian IPA: /kabɒl/
- Tajik IPA: /kɔbul/
- ===Noun===
- (fa-regional)
- کابل • (kâbol; kâbul)
- #definitions
- (perhaps the other transliterations can be incorporated into the fa-regional template) Sameerhameedy (talk) 15:24, 24 January 2023 (UTC)
- @Sameerhameedy, Atitarev The idea of allowing for multiple transliterations of a given term came up once before and I thought of whether it's technically feasible to implement; probably, but it's a lot of work. Personally I think this can be handled by giving alternative pronunciations (which in fact is already being done). Compare, for example, Hebrew, where Biblical Hebrew and Modern Hebrew have different transliterations (the choice seems to have been made to use modern transliterations, but it's not clear this is helpful for terms attested in the Torah). Benwing2 (talk) 05:54, 24 January 2023 (UTC)
- I support presenting the differences in the Pronunciation section. It would be nice if we do that in the transliteration as well, similar to the Hebrew language, in which we (used to?) mention both modern Hebrew and Biblical Hebrew transliterations, but it would be a lot of work. Following the practice of the Dehkhoda Dictionary, I oppose vocalizing the headword. --Z 12:14, 24 January 2023 (UTC)
- @ZxxZxxZ I may have read your comment incorrectly but I think I agree. All Persian dictionaries I've used show pronunciation with Latin transliterations, not with diacritics. I don't know if voicing headwords would be particularly helpful. I wouldn't be opposed to having transliterations in the pronunciation section, my main issue is that Dari dictionaries use latin Transliteration, rarely IPA. Very few Dari speakers know IPA and the lack of a Latin transliteration makes it pretty inaccessible. I don't really care where I just think a Latin transliteration is necessary for Dari. Perhaps the fa-IPA template could be reworked to automatically make transliterations? Sameerhameedy (talk) 15:24, 24 January 2023 (UTC)
- @Sameerhameedy, ZxxZxxZ: Pity, you haven't warmed to providing vocalisations - I find them useful, even if they are not frequently used for Persian. I think multiple transliterations is a good idea for Persian.
- The regional vocalisations and transliterations (even different systems) can be all shown in the ===Pronunciation=== section, somewhat similar to Korean (4 systems), Chinese (multiple topolects and multiple systems), Burmese (multiple systems) or Thai words (two systems).
- The transliterations can be generated and exposed for modern Iranian, classical and Dari, since the IPA can. Anatoli T. (обсудить/вклад) 22:32, 24 January 2023 (UTC)
- @Atitarev: So I think i misunderstood what you were saying, so I went to check. Do you mean something similar to the transliteration table at 결투 (gyeoltu)? Because I actually think that's a great idea. Though sometimes transliterations are not 1=1 so there would need to be an override. But if you want to include voiced forms of words in the table, my only question is what diacritics could be used to indicate when the pronunciation no longer matches written form? For example in Dari خود used to be pronounced /xwad/ but (because of the partial-collapse of /wa/) is now /xʊd/ which is normally the diacritic pesh اُ by itself. But the inclusion of wâw indicates a long vowel so it's contradictory. I don't know how we would get around that is my concern. But your idea to borrow the transliteration table from Korean is a really good idea. Sameerhameedy (talk) 23:37, 24 January 2023 (UTC)
- @Sameerhameedy: Iranian خود would get no vocalisation (no diacritics) to get "xud", since خُود would mean "xoud", not "xud" but Dari would need خْوَد to be transliterated as "xwad". Pls check Module:fa-translit/testcases, e.g. "کُروز" vs "کُرُوز".
- The Korean pronunciation section shows more than one transliteration of the same word, e.g. Revised Romanization is "gyeoltu" but McCune–Reischauer "kyŏlt'u" (these are different transliteration systems but not regional differences). Classical/Iranian/Dari can work the same way but also look at Chinese, which show regional transliterations and IPA (it's quite advanced). Persian pronunciation sections don't need that level of complexity but they could include IPA, transliterations (and vocalisation) for each of the Persian variety. Anatoli T. (обсудить/вклад) 00:18, 25 January 2023 (UTC)
- @Sameerhameedy:
- Modifying your example, I'd like to see something like this (with corrected links, transliteration methods, etc.)
- Example 1
- ===Pronunciation===
- Classical Persian: IPA: /kɑːˈbul/, transliteration: kâbul, vocalised: کابُل
- Dari: IPA: /kɑːˈbʊl/, transliteration: kâbul, vocalised: کابُل
- Modern Iranian: IPA: /kɒːˈbol/, transliteration: kâbol, vocalised: کابُل
- Tajik: IPA: /kɔˈbul/, transliteration: Tajik Cyrillic: Кобул (Kobul)
- Example 2
- ===Pronunciation===
- Classical Persian: IPA: /dawlat/, transliteration: dawlat, vocalised: دَوْلَت
- Dari: IPA: /dawlat/, transliteration: dawlat, vocalised: دَوْلَت
- Modern Iranian: IPA: /dowlæt/, transliteration: dowlat, vocalised: دُولَت
- Tajik: IPA: /davlat/, transliteration: Tajik Cyrillic: давлат (davlat)
- The vocalisation ("کابُل") could drive (automate) both transliteration and IPA, or just the transliteration ("kā`bul" or "kābul" - input to
{{fa-IPA}}
) with a way to suppress what's not needed. Anatoli T. (обсудить/вклад) 02:59, 25 January 2023 (UTC)- @Atitarev: yes, I think something like that would be really good. Though if this does end up happening the exact layout and placements should be discussed. But I think a multi transliteration system similar to that would be good. I have to say though, I am biased towards the look of the transliteration table used in the Korean Wiktionary. (I kinda wanna copy it) If some vote to make it the standard comes about I think examples of what the layout would look like should be included in any votes. I think that's part of why the previous vote didn't have a lot of participants.
- @Atitarev: So I think i misunderstood what you were saying, so I went to check. Do you mean something similar to the transliteration table at 결투 (gyeoltu)? Because I actually think that's a great idea. Though sometimes transliterations are not 1=1 so there would need to be an override. But if you want to include voiced forms of words in the table, my only question is what diacritics could be used to indicate when the pronunciation no longer matches written form? For example in Dari خود used to be pronounced /xwad/ but (because of the partial-collapse of /wa/) is now /xʊd/ which is normally the diacritic pesh اُ by itself. But the inclusion of wâw indicates a long vowel so it's contradictory. I don't know how we would get around that is my concern. But your idea to borrow the transliteration table from Korean is a really good idea. Sameerhameedy (talk) 23:37, 24 January 2023 (UTC)
- @ZxxZxxZ I may have read your comment incorrectly but I think I agree. All Persian dictionaries I've used show pronunciation with Latin transliterations, not with diacritics. I don't know if voicing headwords would be particularly helpful. I wouldn't be opposed to having transliterations in the pronunciation section, my main issue is that Dari dictionaries use latin Transliteration, rarely IPA. Very few Dari speakers know IPA and the lack of a Latin transliteration makes it pretty inaccessible. I don't really care where I just think a Latin transliteration is necessary for Dari. Perhaps the fa-IPA template could be reworked to automatically make transliterations? Sameerhameedy (talk) 15:24, 24 January 2023 (UTC)
- I support presenting the differences in the Pronunciation section. It would be nice if we do that in the transliteration as well, similar to the Hebrew language, in which we (used to?) mention both modern Hebrew and Biblical Hebrew transliterations, but it would be a lot of work. Following the practice of the Dehkhoda Dictionary, I oppose vocalizing the headword. --Z 12:14, 24 January 2023 (UTC)
- Also, I would like to point out that the Module:fa-translit/testcases that you mentioned doesn't seem to be able to make multiple transliterations. Sameerhameedy (talk) 06:29, 25 January 2023 (UTC)
- @Sameerhameedy: Sure. I am not a template or module guru, anyway. Just presenting the contents I wish to see, not the final display form or layout. Anatoli T. (обсудить/вклад) 06:42, 25 January 2023 (UTC)
- @Sameerhameedy:: Re: "doesn't seem to be able to make multiple transliterations". No, it doesn't, it's not even used anywhere.
- As per technical feasibility, please see User:Benwing2 post above at 05:54, 24 January 2023.
- The Burmese, Korean, Thai or Chinese terms generate multiple codes from one input (based viewed in the expanded mode).
- E.g. အဏုမြူလက်နက် (a.nu.mrulaknak), อาวุธนิวเคลียร์ (aa-wút-niu-kliia), 核武器 (héwǔqì) (just check the Mandarin portion), 핵무기(核武器) (haengmugi) (the choice of words is not intentional, includes my recent edits). Anatoli T. (обсудить/вклад) 08:13, 25 January 2023 (UTC)
- @Sameerhameedy: Sure. I am not a template or module guru, anyway. Just presenting the contents I wish to see, not the final display form or layout. Anatoli T. (обсудить/вклад) 06:42, 25 January 2023 (UTC)
- Also, I would like to point out that the Module:fa-translit/testcases that you mentioned doesn't seem to be able to make multiple transliterations. Sameerhameedy (talk) 06:29, 25 January 2023 (UTC)
- @Kaixinguo~enwiktionary, AryamanA I think you guys may be interested in this conversation as you've proposed something similar beforeSameerhameedy (talk) 15:43, 24 January 2023 (UTC)
Linking words with ؤ and ئ and هٔ
edit(Notifying Ariamihr, Benwing2, Dijan, Mazsch, Qehath, Rodrigo5260, ZxxZxxZ, Sameerhameedy, Saranamd): Hello.
Letter هٔ, used in ezâfe now links to ه, which seems to be the common practice with both Wiktionary and Wikipedia.
But is it right to remlove hamza in all cases? E.g. link ؤ and ئ to و and ي as well (the latter also links to the Arabic ي (y), not ی, not sure it's right).
So, should سؤال link to سوال or to سؤال? Anatoli T. (обсудить/вклад) 04:35, 24 March 2023 (UTC)
- I think سوال should be treated as a misspelling of سؤال (similarly, مساله and مسأله should be treated as the common misspellings of مسئله, in accordance with the rules of the Academy of Persian Language and Literature). The same organization states that ezafe after final unvocalized heh must be represented by a hamzeh over heh, but due to technical difficulties in entering as well as displaying this character, which is testified by the limited usage of this character over the Internet, I support redirecting instances of هٔ to ه. --Z 13:03, 24 March 2023 (UTC)
- Note that there is little difficulty for entering ء, ؤ and ئ in most Persian keyboards, and there is no problem in displaying it. They are also commonly used over the Internet. --Z 13:06, 24 March 2023 (UTC)
- @Atitarev I agree with @ZxxZxxZ. Adding or not adding hamza should still take you to the entry, but the entry itself should have the hamza if it's the correct spelling. Also, can we have یٰ link to ی ? the dagger diacritic is the only way to mark when a ye is acting as an alif, but adding a dagger diacritic seems to cause the ye to be treated as a different character entirely. Sameerhameedy (talk) 06:37, 25 March 2023 (UTC)
- @Sameerhameedy, ZxxZxxZ, Benwing2:
- Thanks. I am afraid, I don't understand 100%.
- ZxxZxxZ says "سوال should be treated as a misspelling of سؤال". So the correct entry should be سؤال. That means the hamze shouldn't be deleted from the link, at least for letter ؤ.
- Based on the further message, redirecting هٔ to ه is correct.
- I think I stuffed up my request to @Benwing2. I have requested to remove links for ALL letters with hamze above but correct is only for letter هٔ.
- Correct: هٔ with hamze: ایالات متحدهٔ آمریکا (eyâlât-e mottahede-ye âmrikâ) links to ایالات متحده آمریکا
- Incorrect: ؤ with hamze: سؤال links to سوال but should link to سؤال
- Please confirm!
- As for یٰ, it's already linking to ی, as I have advised. قُرونِ وُسْطیٰ (qorun-e vostâ) is already linking to قرون وسطی Anatoli T. (обсудить/вклад) 07:40, 25 March 2023 (UTC)
- I agree with 1 and 2. --Z 23:00, 25 March 2023 (UTC)
- @ZxxZxxZ: Thanks!
- Based on this, @Benwing2, could you please partially revert diff, so that only هٔ is replaced with ه?
- Sorry for the confusion before! Just FYI, @Sameerhameedy. Anatoli T. (обсудить/вклад) 10:49, 26 March 2023 (UTC)
- I agree with 1 and 2. --Z 23:00, 25 March 2023 (UTC)
Template:head allows etym languages
edit@Sameerhameedy, Atitarev Just FYI, I've fixed Module:headword and all modules it calls to allow etym languages such as fa-ira
and fa-cls
. Benwing2 (talk) 23:54, 27 November 2023 (UTC)
- @Benwing2 I think adding etymology support for most templates is a great thing, but as for the header template... Sorry I'm not sure how I feel about that. While it does address some issues Dari and Classical Persian have had, I think moving transliterations to the pronunciation section would've been better. Now we have multiple header templates for the same language. Which is basically the same thing as splitting dialects without the benefits of separate L2s or changing what varieties are grouped together. (e.g. if we combined Dari and Tajik instead, Dari headers could generate Tajik spellings). IMO if we want to keep all Persian varieties together (which is probably for the best) it should be formatted differently than other languages, else the benefits of them being together are diminished. - سَمِیر | Sameer (مشارکتها · بحث) 21:55, 28 November 2023 (UTC)
- @Benwing2, @Sameerhameedy: Great, thanks!
- What are the implications, though. We are still using
fa
in all Persian headword templates. They are not modularised. - Potentially, if new templates are added, multiple headwords could be used with appropriate vocalisations and mostly automate dtransliterations.
{{fa-ira-noun|دَقیقِه|pl=دَقائِق|pl2=دَقیقِها|tr=dağiğe}}
{{fa-cls-noun|دَقِیقَه|pl=دَقَائِق|pl2=دَقِیقِها|tr=daqīqa}}
- But I don't think this move is popular. Anatoli T. (обсудить/вклад) 21:56, 28 November 2023 (UTC)
- @Atitarev @Sameerhameedy That is a good point. I agree with you about keeping all Persian varieties together, whatever you think works best for the headwords is fine with me. Let me know if there are other templates that need etym-language support. Benwing2 (talk) 23:07, 28 November 2023 (UTC)
- @Sameerhameedy, @Benwing2:
- I like this template and it would be great if the Persian headword is modularised and is displayed similar to Sameerhameedy's great template: خودکشی (xwadkušī/xodkoši). Let's also get rid of
|head=
and use an unnamed parameter - vocalised Persian. - The vocalisation used is that of the classical Persian, perhaps it can be enhanced to show the Iranian as well, if different (it was done but then you rolled it back), even if the Iranian vocalisation doesn't fully match the transliteration e.g. "خودْکُشی" (xodkoši) (since your already automated Iranian
|tr=
based on the classical vocalisation). If you insist on not showing the Iranian vocalisation, then so be it. --Anatoli T. (обсудить/вклад) 00:41, 29 November 2023 (UTC)- @Sameerhameedy: Also, you know you can use fa-ira, fa-cls, prs in usage examples, with full or partial automated transliterations, too. Ideally, those codes should show the tooltips in usage examples you made in User:Sameerhameedy/example entry. Also, @Benwing2, take a look (with no tooltips, though but with
|q=
)- مَادَرِ خُودْرَا نَمیشِنَاسِی؟ (Classical Persian) ― mādar-i xūdrā namē-šināsī? ― Don’t you recognize your own mother?
- مَادَرِ خودْرا نِمیشِناسی؟ (Iranian Persian) ― mâdar-e xodrâ nemi-šenâsi? ― Don’t you recognize your own mother?
- Anatoli T. (обсудить/вклад) 00:57, 29 November 2023 (UTC)
- I've only substituted "خودْرا" with "xodrâ" in the Iranian, since this can't be correctly vocalised but you can use your magic you already using in
{{template:User:Sameerhameedy/fa-l}}
Anatoli T. (обсудить/вклад) 01:01, 29 November 2023 (UTC)
- I've only substituted "خودْرا" with "xodrâ" in the Iranian, since this can't be correctly vocalised but you can use your magic you already using in
- @Sameerhameedy: Also, you know you can use fa-ira, fa-cls, prs in usage examples, with full or partial automated transliterations, too. Ideally, those codes should show the tooltips in usage examples you made in User:Sameerhameedy/example entry. Also, @Benwing2, take a look (with no tooltips, though but with
- @Benwing2 I still think this example I showed you is the best way to go about this. I think we should have a bot run to remove transliterations from headers that already have fa-IPA on the page, and add the Tajik spelling generated by fa-IPA. Entries lacking fa-IPA should be unaffected.
- Something we should address later though: I also think we should make a Persian link module similar to the one I made, but it should probably hide diacritics tbh. (at least in "fa" links but probably keep them in quotations and usage examples).
- - سَمِیر | Sameer (مشارکتها · بحث) 00:59, 29 November 2023 (UTC)
- @Sameerhameedy: I feel like you have been ignoring me and excluding from the discussion. Not sure how I managed to upset or you just don't like my ideas. It's fine, though. I won't contribute, if you don't want. Anatoli T. (обсудить/вклад) 01:10, 29 November 2023 (UTC)
- @Atitarev I was not ignoring you, note that my last comment was around the same time as your last three. I didn't get around to responding but I did edit some of my comments after reading some of yours. I initially wanted to generate both transliterations in quotations too but seeing your comment decided that maybe just using dialect codes for quotations and usage examples would be better. Quotations are collapsible anyway so the extra room is probably fine.
Also, due to personal annoyances with needing to re-enter information so frequently, I'd rather not require any input in the header. A full Persian entry requires re-entering information in fa-IPA, the header, the Iranian conjugation templates, the Dari conjugation templates, then in fa-regional, which is all a bit absurd honestly. fa-IPA already does all transliteration conversions and has the information other templates need, according to Ben it is possible to allow other templates to use that information, and I think we should look into that. It would not just give us the ease of editing other languages have, but make supporting all varieties of Persian very very simple. If your header suggestion scrapped information from fa-IPA I suppose I'd be fine with it, though I would prefer if Persian headers were more minimalistic given much of that information is already present. - سَمِیر | Sameer (مشارکتها · بحث) 01:32, 29 November 2023 (UTC)- @Atitarev @Sameerhameedy Are you asking about reusing the code itself, or scraping the information from the fa-IPA template? Reusing the code is easy; scraping is also possible but takes a bit more work. I need to work on scraping for Thai and Khmer first, then I can get to scraping for Persian if that's what is needed. Benwing2 (talk) 02:46, 29 November 2023 (UTC)
- @Benwing2 yes I was talking about scrapping information from fa-IPA, in a prior conversation you mentioned it was a way we could get the header to generate the Tajik spelling. But it would be amazing if we could could use it for conjugation templates and other things as well. Currently Persian conjugation template require are separate for Dari and Iranian Persian and Dari requires a romanization for Kabuli and formal and Iranian Persian requires an input for tehrani and formal. In an Idea world we can have 1 conjugation template that fentches all four romanizations from fa-IPA. - سَمِیر | Sameer (مشارکتها · بحث) 03:07, 29 November 2023 (UTC)
- @Benwing2:
- In MY opinion, the headword could look something like this:
- خودکشی (xwadkušī/xodkoši)
- It's almost perfect. + plural forms, etc. whatever goes into the current header.
- Sameer, if I am not mistaken, doesn't want vocalisations and or even transliterations in the headword, he thinks the pronunciation section as it is, is sufficient.
- In my opinion, it's not fair to say that Persian headword should be simple, especially because there are three major varieties, or four if we include Cyrillic-based Tajik. The complexity depends on the language. Arabic and now Urdu have vocalisations and automated transliterations. Why can't we have it for Persian? Because of the complexity or because of the dialectal/historical differences? Edit tools allow to enter diacritics and you can display both classical and modern.
- Can we consider adding vocalisations in the pronunciation section at least (not the respelling but the accurate vocalisations? (Occasional majhūl, irregularly pronounced loanwords are fine, it's no different from Arabic). Anatoli T. (обсудить/вклад) 03:07, 29 November 2023 (UTC)
- @Atitarev adding harakat to the pronunciation section is preferable, I think. But there should ideally be a way to imbed it into fa-IPA, so they are collapsible.
- "Arabic and now Urdu have vocalisations and automated transliterations. Why can't we have it for Persian?"
- Just saying, every Arabic dialect was made a different L2 language and we are trying to not split Persian, so the Arabic approach doesn't really apply here. Also there's only two standard varieties of Urdu, and they're both nearly identical. There's not a lot of languages trying to do be dialect inclusive to the extent Persian is, so there's not a lot of examples to look to. Except maybe Chinese? (Chinese's situation is very different though). Almost every other language on wiktionary would've split, so in my opinion, a unified Persian should have different considerations than other languages on wiktionary. (most of which are split languages or don't have the same problems as persian) - سَمِیر | Sameer (مشارکتها · بحث) 05:11, 29 November 2023 (UTC)
- @Sameerhameedy: Thanks for engaging.
- If you prefer showing harakat only in the pronunciation section, it's fine then, collapsible is also fine.
- Arabic wasn't split here at Wiktionary, it was always the case and the differences are too great to merge. Any attempts to merge failed for these reasons. However, people disagreed even to provide regional pronunciation for all literary terms, which are shared by Arabic dialects. Here, the resistance is based on the dialect pride or self-awareness, or other emotional, not linguistic reasons.
- Different transliterations for Persian varieties is an issue but not that it can't be overcome. You're very close to it. Certain cases require manual
|tr=
, which is also fine. They are not numerous. Anatoli T. (обсудить/вклад) 05:29, 29 November 2023 (UTC)
- @Sameerhameedy: Thanks for engaging.
- @Atitarev Also, I would prefer to minimize the amount of redundant information that needs to be inputted in Persian entries. That being said, if considering Ben will eventually add scrapping from fa-IPA, if the header used the Iranian and Classical transliteration from fa-IPA I guess that's fine, though again I'd still prefer not having transliterations at all. But I'm fine compromising on that so long as the header does not require input and can just scrap information from fa-IPA. - سَمِیر | Sameer (مشارکتها · بحث) 05:25, 29 November 2023 (UTC)
- @Sameerhameedy I agree with @Atitarev that it would be better to put transliterations in headwords (and everywhere else if possible). There's a potential problem with scraping which is that one written word might have multiple vocalizations. I think User:Theknightwho made it possible to figure out where in the page a given template is, meaning that the headword might be able to automatically pick the right vocalization/pronun (assuming there's only one per etym section), but for terms on other pages, this isn't really solvable. I don't know the portion of words in running text that have multiple potential vocalizations but it could be pretty high; it is in Arabic text, for example, at least I think so. It could pick the first listed one in the hope that it's the most common, but (a) you'd have to know this, and (b) it's fragile since people might rearrange the etym sections or put another etym section in the front. One possibility is to put a special "use-this-vocalization" marker next to (or as a parameter to) the vocalization to use when there are multiple on a page and one is a lot more common than the others; then, if none are marked, the vocalization-by-scraping fails. In such a case, you'd have to manually vocalize the word in question (not all words, if there are multiple and only some are ambiguous). That said, it doesn't seem to me to be a huge issue in the meantime to redundantly put the same info in both the headword and the pronunciation section; this is done for other languages, and there are typically new-entry-generation modules to assist in the task (or bot scripts can be written). Benwing2 (talk) 06:08, 29 November 2023 (UTC)
- @Sameerhameedy, @Benwing2:
- I thought it could be difficult but a kind of marker/parameter is a good idea.
- In any case:
- Currently, each Persian entry has a transliteration in the headword. Possibly 95% based on Iranian Persian. Removing it could make quite a few users and editors unhappy now or in the future. Changing to classical can be discussed or put on a vote. If scraping works, it's fine, a bot can remove all manual transliterations in one go. I personally think it should have at least one, a default method. The only language without headword translit. is currently Chinese, which groups all topolects.
- Adding a transliteration is already a manual input, which is error-prone and requires some special characters. The output from
{{template:User:Sameerhameedy/fa-l}}
, eg.
- is perfect for the headword, IMO for the headword. Manual input is only دَقِیقَه, if it can't be read from the scraper. Plurals, comparatives, etc. can be left unvocalised and untransliterated.
- I don't want to be a hurdle to your efforts. If you disagree, please say, which points. Anatoli T. (обсудить/вклад) 06:32, 29 November 2023 (UTC)
- @Atitarev The example under number 2 is nice and i'd be down with that. But are you sure we shouldn't hide diacritics in header? It may be more neutral to list both the Iranian and classical vocalization in the pronunciation section instead of just listing the classical translit in the header.
- Regardless, we would still need a bot to scrap transliterations (at least once) in order to make that change. The good news about doing that though, the header can just use the classical transliteration to generate the Tajik spelling. Instead of asking fa-IPA for the Tajik spelling every page load.
- @Benwing2 if i'm understanding what Atitarev is saying correctly this might be the best way to solve multiple problems we are facing? We won't (theoretically) need any templates to actively scrap from IPA, just some bots in a bot run. And the header wouldn't need to scrap the Tajik spelling in order to show it since it would already know the classical spelling. Both the Iranian and Classical transliterations would be displayed too. Based on what you said earlier, this would be the most feasible option, no? - سَمِیر | Sameer (مشارکتها · بحث) 06:43, 29 November 2023 (UTC)
- @Sameerhameedy: Thanks, glad we agreed on something (transliterations). As for diacritics, I think they are good, educational and useful otherwise. Since classical translit. is displayed first, it won't be confusing re differences. I think if we agree on diacritics on vocalisations, the pronunciation section can stay as is. Anatoli T. (обсудить/вклад) 06:49, 29 November 2023 (UTC)
- @Sameerhameedy: Forgot to mention: I have no objection to add Tajik. I never had.
- Yeah. The headword I suggested is a feasible solution, IMO and won't require a scraper. @Benwing2, keeping you in the loop. Anatoli T. (обсудить/вклад) 06:55, 29 November 2023 (UTC)
- @Benwing2 I have updated the page User:Sameerhameedy/example entry to reflect the layout me and Atitarev have settled on. - سَمِیر | Sameer (مشارکتها · بحث) 00:45, 30 November 2023 (UTC)
- @Sameerhameedy OK thanks! Benwing2 (talk) 01:23, 30 November 2023 (UTC)
- @Benwing2, @Sameerhameedy: Sorry for more pings. Yes, we agreed on the basic headword layout. When you get around, please add it. I can imagine the headword needs to be modularised.
- Since the new design adds some complexity, perhaps inflected forms can be rid of transliterations to make it less cluttered. Besides, it won't be possible or practical to correctly transliterated for each of the variety.
- I also think, it would be great, if the usage example templates can display the Persian variety as in Sameerhameedy's examples. Chinese templates can do that based on the topolect parameter. Please see an example at Template:zh-x#佢哋:
- Otherwise, language specific templates could be created and used, as in Chinese. I know some people are against that. Or just use the language code + e.g.
|q=Iran, colloquial
. E.g. - مُرورْگَرِ وِب (Iranian Persian) ― morurgar-e veb ― web browser Anatoli T. (обсудить/вклад) 01:24, 5 December 2023 (UTC)
- Hello @Sameerhameedy, @Benwing2:
- I guess the Persian headword change won't happen based on diff and the lack of any action/interest? We no longer aim to have the previously agreed format? Anatoli T. (обсудить/вклад) 23:35, 11 February 2024 (UTC)
- @Atitarev I'm just testing something. - سَمِیر | Sameer (مشارکتها · بحث) 23:37, 11 February 2024 (UTC)
- @Sameerhameedy OK thanks! Benwing2 (talk) 01:23, 30 November 2023 (UTC)
- @Benwing2 I have updated the page User:Sameerhameedy/example entry to reflect the layout me and Atitarev have settled on. - سَمِیر | Sameer (مشارکتها · بحث) 00:45, 30 November 2023 (UTC)
- @Sameerhameedy: Thanks, glad we agreed on something (transliterations). As for diacritics, I think they are good, educational and useful otherwise. Since classical translit. is displayed first, it won't be confusing re differences. I think if we agree on diacritics on vocalisations, the pronunciation section can stay as is. Anatoli T. (обсудить/вклад) 06:49, 29 November 2023 (UTC)
- @Benwing2 Well I should iterate I am much more in favor of not putting romanizations in the header at all, the romanizations listed in fa-IPA are more detailed anyway. But 1) since header cannot generate the Tajik spelling, and needs to scrap that data anyway, I suggested it since were already planning on doing that anyway. 2) I was NOT suggesting we do this for links, which I think you were under the impression was one of the things I was asking for, as that would be entirely impossible. 3) Since were already planning on adding scrapping for Tajik I thought it'd helpful to at least allow conjugation templates on Persian entries to do so as well (if possible) It would only work for compound verbs but it would a little bit of hassle, But I suppose it's minor enough to ignore. But the annoyance is not re-entering data, it's re-entering data in multiple formats since our header templates, pronunciation templates, and conjugation templates don't all use the same romanization scheme (most of the time). Using it in every entry is probably a bit too much but we could at least use it add Dari conjugation templates to entries that have the romanization in fa-IPA and the Iranian conjugation template (but not the Dari one). - سَمِیر | Sameer (مشارکتها · بحث) 06:35, 29 November 2023 (UTC)
- @Sameerhameedy I agree with @Atitarev that it would be better to put transliterations in headwords (and everywhere else if possible). There's a potential problem with scraping which is that one written word might have multiple vocalizations. I think User:Theknightwho made it possible to figure out where in the page a given template is, meaning that the headword might be able to automatically pick the right vocalization/pronun (assuming there's only one per etym section), but for terms on other pages, this isn't really solvable. I don't know the portion of words in running text that have multiple potential vocalizations but it could be pretty high; it is in Arabic text, for example, at least I think so. It could pick the first listed one in the hope that it's the most common, but (a) you'd have to know this, and (b) it's fragile since people might rearrange the etym sections or put another etym section in the front. One possibility is to put a special "use-this-vocalization" marker next to (or as a parameter to) the vocalization to use when there are multiple on a page and one is a lot more common than the others; then, if none are marked, the vocalization-by-scraping fails. In such a case, you'd have to manually vocalize the word in question (not all words, if there are multiple and only some are ambiguous). That said, it doesn't seem to me to be a huge issue in the meantime to redundantly put the same info in both the headword and the pronunciation section; this is done for other languages, and there are typically new-entry-generation modules to assist in the task (or bot scripts can be written). Benwing2 (talk) 06:08, 29 November 2023 (UTC)
- @Atitarev adding harakat to the pronunciation section is preferable, I think. But there should ideally be a way to imbed it into fa-IPA, so they are collapsible.
- @Atitarev @Sameerhameedy Are you asking about reusing the code itself, or scraping the information from the fa-IPA template? Reusing the code is easy; scraping is also possible but takes a bit more work. I need to work on scraping for Thai and Khmer first, then I can get to scraping for Persian if that's what is needed. Benwing2 (talk) 02:46, 29 November 2023 (UTC)
- @Atitarev I was not ignoring you, note that my last comment was around the same time as your last three. I didn't get around to responding but I did edit some of my comments after reading some of yours. I initially wanted to generate both transliterations in quotations too but seeing your comment decided that maybe just using dialect codes for quotations and usage examples would be better. Quotations are collapsible anyway so the extra room is probably fine.
- @Sameerhameedy: I feel like you have been ignoring me and excluding from the discussion. Not sure how I managed to upset or you just don't like my ideas. It's fine, though. I won't contribute, if you don't want. Anatoli T. (обсудить/вклад) 01:10, 29 November 2023 (UTC)
- @Atitarev @Sameerhameedy That is a good point. I agree with you about keeping all Persian varieties together, whatever you think works best for the headwords is fine with me. Let me know if there are other templates that need etym-language support. Benwing2 (talk) 23:07, 28 November 2023 (UTC)
Persian headword
editBased on the preliminary aggreement above about the future Persian headword, what should be the CURRENT headword before it's modularised and largely automatically transliterated?
Previously, the majority of entries had no vocalisation and the transliterations were mostly based on Iranian transliteration (a recent change was to change letter ق (ğ) from "q" to "ğ" specifically for Iranian Persian.
Based on Classical Persian:
{{fa-noun|head=اِقْلِیم|tr=iqlīm}}
see automated اِقْلِیم (iqlīm){{fa-noun|head=قَلَمْرَوْ|tr=qalamraw}}
see automated قَلَمْرَوْ (qalamraw)
Based on Iranian Persian:
{{fa-noun|head=اِقْلیم|tr=eğlim}}
see automated اِقْلیم (eğlim){{fa-noun|head=قَلَمْرُوْ|tr=ğalamrow}}
see automated قَلَمْرُوْ (ğalamrow)
@Sameerhameedy, @Saranamd, @Benwing2: before calling on to the community, can you say your preference, so that we can making changes to entries?
The first option will allow to simply enter the full classical vocalisation as the input, for this to happen, the majority of Persian entries will need to get the classical vocalisations:
(Just got a response from Saranamd, it seems the switch from Iranian to Persian is not totally agreed on). Anatoli T. (обсудить/вклад) 08:12, 26 December 2023 (UTC)
- @Atitarev If it's going to be eventually modularized to show both Iranian and Classical values (which is clearly the ideal end goal), I don't really see an urgent need to change the current translit.--Saranamd (talk) 08:21, 26 December 2023 (UTC)
- @Saranamd: If we are changing the vocalisations from Iranian to Classical and there is a mismatch, the bot, which does the job may flag it as an error and I don't want to REMOVE transliteration altogether. Deriving the future headword from the current Iranian transliterations won't be an easy task for a bot.
- The automated transliteration can be a good test in about 95% of cases of what the transliteration AND vocalisation should be.
- If the future headword input is قَلَمْرَوْ, calling قَلَمْرَوْ (qalamraw) is a good test for an editor. Just copy-paste and replace the old translit. with the new one.
- There won't be
|head=
in the future headword but a bot replacement might be just to remove|head=
(leave the value) and|tr=
with its value. Anatoli T. (обсудить/вклад) 08:32, 26 December 2023 (UTC)- @Atitarev Oh, so changing to Classical is an intermediary step to showing both Iranian and Classical values?--Saranamd (talk) 09:09, 26 December 2023 (UTC)
- @Saranamd: yes, the classical vocalisation: قلمرو (qalamraw/ğalamrow) Anatoli T. (обсудить/вклад) 09:31, 26 December 2023 (UTC)
- Ok, I'm fine with that then. But I would actually prefer the vocalization to be hidden in display if that matters, especially for usexes and quotes. Monolingual or bilingual Persian dictionaries generally do not fully vocalize, which makes it different from Arabic AFAIK.--Saranamd (talk) 09:55, 26 December 2023 (UTC)
- @Saranamd: yes, the classical vocalisation: قلمرو (qalamraw/ğalamrow) Anatoli T. (обсудить/вклад) 09:31, 26 December 2023 (UTC)
- @Atitarev Oh, so changing to Classical is an intermediary step to showing both Iranian and Classical values?--Saranamd (talk) 09:09, 26 December 2023 (UTC)
Bot for adding reference templates
editCan we make a bot for reference templates to persian entries? I don't wanna have to manually do it for every single entry, I feel like there should be a better way. SinaSabet28 (talk) 22:08, 20 August 2024 (UTC)
- Pinging Persian editors: (Notifying Ariamihr, Atitarev, Babr, Benwing2, Dijan, Mazsch, Qehath, Rodrigo5260, ZxxZxxZ, Saranamd): — Fenakhay (حيطي · مساهماتي) 22:12, 20 August 2024 (UTC)
- @SinaSabet28, @Fenakhay, @Babr, @Benwing2: Hi. Adding references it may be tedious but they are not required. I would oppose adding broken references, e.g. when a word doesn't exist in the link. I would even suggest a bot to clean broken references. Anatoli T. (обсудить/вклад) 01:22, 21 August 2024 (UTC)