 |
|
|
|
Welcome to the Invelos forums. Please read the forum
rules before posting.
Read access to our public forums is open to everyone. To post messages, a free
registration is required.
If you have an Invelos account, sign in to post.
|
|
|
|
|
Invelos Forums->DVD Profiler: Contribution Discussion |
Page:
1 2 3 4  Previous Next
|
|
Proposal for romanization of names of Japanese actors and actresses for the database |
|
|
|
|
|
| Author |
Message |
| Registered: March 14, 2007 |  Posts: 1,328 |
| | Posted: | | | | Another post on discussion of romanization in Asian titles in the Contribution Rules Committee Forum Quote:
Quoting Peter von Frosta:
Quote:
I somehow missed that one.
Well, nobody except a local would be able to distinguish if this is a phonetic interpretation or if this word/phrase has indeed been integrated into the local language. Problem is, it seems there are not too many chinese, japanese, malay, thai etc. users in here...
Peter,
A western word being integrated into the local language would be exception rather than the norm. If I had to put a number on it, 95 to 98% of all western words used in a title for Asian movies would be non-localized words. But it's not too difficult to figure out which is which and perhaps it doesn't matter. I will explain.
There are 5 broad instances when international titles are shown on DVD cover or screen with Chinese/Japanese/Korean movies for their local release. These international titles are mostly in English for whatever reason. I quickly scanned through my R3 Korean DVDs (Korea locality) for ones that had titles in both Korean and Western words.
They are:
1. International Title is a direct or close translation of the local title
Cheongun - Heaven's Soldiers
Noksaek Euija - Green Chair
Ssaumeui Gisul - Art of Fighting
Nabi - Mr. Butterfly
Gyeongmajang Ganeun Gil - The Road to the Racetrack
Gangweondo-eui Him - The Power of Kangwon Province
Ttuleoya Sanda - Dig or Die
Nappeun Namja - Bad Guy
Namgeuk Ilgi - Antarctic Journal
Hwal - The Bow
Misulgwan yeop Dongmulweon - Art Museum by the Zoo
Salin-eui Chueok - Memories of Murder
Chingu - Friend
Naesaengeh Gajang Areumdawun Iljuil - la plus belle semaine de ma vie (a French International Title)
2. International Title is romanization of Local Title
Chunhyangdyeon - Chunhyang
Ssaulabi - Saulabi
Swiri - Shiri
Eoudong - Eoudong
3. Local Title is phonetic translation of Western words (in this case, I would argue that International Title is the real original title)
SSeokeul - The Circle
Malaton - Marathon
Deiji - Daisy
Taepung - Typhoon (Taepung is Korean word for Typhoon, so not a strict phonetic translation)
Beulu - Blue
Syo Syo Syo - Show Show Show
Sadeu Mubi - Sad Movie
Hapi Ero Keurismas - Happy Ero Christmas
Welkomtu Dongmakgol - Welcome to Dongmakgol
Keulaesik - The Classic
Bon tu Kil - Born to Kill
Telmi Sseomding - Tell Me Something
Inteobyu - Interview
Tukabseu - Two Cops
Oldeu Boi - Oldboy
Yeseuteodei - Yesterday
Naechyureol Siti - Natural City
Oasiseu - Oasis
Pon - Phone
4. International Title and Local Title are completely different
Yeogyosu-eui Eunmilhan Maeryeok - Bewitching Attractions
Buhwal - Revenge
Muyeong-geom - Shadowless Sword
O Sujeong - Virgin Stripped Bare by Her Bachelors
Dalkomhan Insaeng - A Bittersweet Life (Titles are completely opposite in meaning)
Juhong Geulssi - The Scarlet Letter
Hyeongsa - Duelist
Chinjeolhan Geumja-ssi - Sympathy for Lady Vengeance
Woechul - April Snow
Neoneun Naewunmyeong - You are My Sunshine
Binjip - 3 Iron
Jambokgeunmu - She's on Duty
Injeongsajeong Bolkkeoteopda - Nowhere to Hide
Silla-eui Dalbam - Kick the Moon
5. Mixture of Western Words and Local Words in a Title
Palweoleui Keurismas - Christmas in August (this is a direct translation although Christmas is used in both the local title and international title)
Gongdong-gyeongbi guyeok JSA - Joint Security Area (Korean title is a direct translation of Joint Security Area and JSA is an acronym)
I think depending on the circumstances, what we put into each title field would be different. In case of type 3 above, I do feel strongly that the International title should be the main title and romanized local title do not make any sense, so it doesn't matter if it's a localized word or not. Only time it comes into picture is with type 5 above. Frankly Palweoleui Keurismas looks wrong. In this specific instance, if I had to choose a local title, it would be written as Palweoleui Christmas.
| | | | My Home Theater | | | | Last edited: by xradman |
| | Registered: May 2, 2009 | Reputation:  | | Posts: 490 |
| | Posted: | | | | I'm bringing this thread back to life for a question. What's the status of this nowadays and was there ever a decision? I'm about to contribute some Japanese profiles which have all credits in kanji (sorry if I'm using the wrong term, but I'm by no means an expert in Japanese language). Currently, there's a "transcription" of the names, but it's inconsistent and IMO incorrect. For example, a writer has the credit "Kôji Suzuki (as Koji Suzuki)", while his "real" name in Latin letters would rather and more commonly be "Kōji Suzuki". However due to the limitations in the Invelos software (damn I wish there were none at least for characters within the ASCII system!), if you try to input "Kōji Suzuki" it becomes "Koji Suzuki". The circumflex symbol ("^") is not really used for generally transcribing Japanese and looks more like a "hack" to "simulate" the "ō" (etc) character. For this reason, I would personally rather have just standard A-Z characters for Japanese names. Who's onboard? Or is this a non-topic meaning it's rather arbitrary? I guess the same applies for other languages as well, which has characters that aren't in A-Z format, but with a few European exceptions. Heck you can't "even" use Cyrillic, full Czech or Polish characters.  If you ask me, this is an unacceptable software limitation. | | | | Last edited: by MikaLove |
|   | T!M | | Profiling since Dec. 2000 |
| Registered: March 13, 2007 | Reputation:  |  Posts: 8,736 |
| | Posted: | | | | Quoting MikaLove: Quote:
Or is this a non-topic meaning it's rather arbitrary? The unfortunate fact here is that very few users are interested in this at all. And those that are, seem to have given up on the hope that any kind of consensus can be reached. Quote:
Heck you can't "even" use Cyrillic, full Czech or Polish characters.
If you ask me, this is an unacceptable software limitation. That is extremely unfortunate, indeed. But it doesn't seem like Invelos is going to address that problem anytime soon - or even ever. So however unacceptable it may seem, we pretty much don't have any other option than to accept it. |
| | Registered: May 2, 2009 | | Reputation: | | Posts: 490 |
| | Posted: | | | | Thanks for replying. I understand that we can't expect virtually anything more to be improved with or implemented in the software or how the DB handles foreign characters. I have tried submitting some cast with Czech "exotic" letters, but they turn into unreadable symbols. E.g. for "Hannibal Rising"; Jaroslav Psenicka (credited as Jaroslav Pšenicka) The "š" turns into a small "█" in the software. I notice that when you copy/paste the character is there, but even worse than this the system bypasses some characters completely. This goes for writing overviews as well. I'm repeating myself, but it just feels so limited and amateur-ish. The question is how we should handle romanized Japanese names. Since all other topics are a dead horse. The CLT is full with in my opinion garbage/nonsense where people have haphazardly submitted romanized name versions like in my previous post. Most stem from how e.g. IMDb chooses to display names like "Kōji Suzuki", with a macron over the "o", as "Kôji Suzuki", with the circumflex. That makes no sense, since in French, the "ô" has one sound and in the romanization of Japanese, the "ō" sounds completely different. The languages has absolutely no relation and don't even both belong in the Indo-European language group. Still, these name versions have made it here and for many names, the "common names" are the "bastardized" romanizations of Japanese. Since many movies may have written credits on e.g. covers, booklets and other types of inlays/sheets that come with a disc, that can't be seen as a homogeneous source of documentation, because these will definitely vary. The CLT becomes a little like the saying "tell a lie enough times and it becomes the truth". As an example of geographical variations and standards, in Sweden we have our way of transcribing Cyrillic/Russian and for the English speaking there's another. AFAIK however most of Europe use the same transcription as we do in Sweden for Russian. I think what it all boils down to is what works best for us and for the system at hand. We definitely need to come to some sort of consensus how to write names, once they are at least found and identified. It could be easier with Japanese since there shouldn't be too many variations or deviations. That is again why I think we should make all at least Japanese names have English keyboard A-Z letters only, regardless of what. Minimize the pain of maintaining the DB.   |
| | Registered: March 18, 2007 | | Reputation: | | Posts: 6,461 |
| | Posted: | | | | It turns out that Japanese has a more or less "official" method of romanization. It is called "Romaji". You can "romanize" Kanj, Hirigana and Katakana, such that any Western speaker can simulate the pronunciation. I agree the IMDB spelling of Koji Suzuki is wrong. I am not even sure that the spelling that includes the long "o" character is even correct. But if it is, the romaji equivalent is "ko-ji." The hyphen is used in lieu of a special character to indicate the previous character is elongated. So, regardles of his actual name, romaji provides the solution: it is either "koji" or "ko-ji".
So, Japanese is actually the easy one.
Back when I was living in Denmark, I recall that Danish has some similar convenient methods, such as using "aa" instead of the character with the little circle on top, pronounced in English as long "o". I think there were others. | | | Thanks for your support.
Free Plugins available here.
Advanced plugins available here.
Hey, new product!!! BDPFrog. | | | | Last edited: by mediadogg |
| | Registered: May 2, 2009 | | Reputation: | | Posts: 490 |
| | Posted: | | | | Quoting mediadogg: Quote:
It turns out that Japanese has a more or less "official" method of romanization. It is called "Romaji". You can "romanize" Kanj, Hirigana and Katakana, such that any Western speaker can simulate the pronunciation. I agree the IMDB spelling of Koji Suzuki is wrong. I am not even sure that the spelling that includes the long "o" character is even correct. But if it is, the romaji equivalent is "ko-ji." The hyphen is used in lieu of a special character to indicate the previous character is elongated. So, regardles of his actual name, romaji provides the solution: it is either "koji" or "ko-ji".
So, Japanese is actually the easy one.
Back when I was living in Denmark, I recall that Danish has some similar convenient methods, such as using "aa" instead of the character with the little circle on top, pronounced in English as long "o". I think there were others. I think there's a difference between Scandinavian languages and Japanese, regarding more than just relationship (if there were any). The so called macron over a vowel in Japanese does what you describe it to do. To elongate the vowel. But in Scandinavian language, the letters Å, Ä/Æ, Ö/Ø can be both short and long. It's not to romanize the languages either to "transcribe" them to aa/ae/oe/, since it's not necessary. You may see it happen that way, but I feel it's more like a substitution when the character is missing on someone's keyboard. But it's also very rare to see. Irregardless, the fact remains that yes, Japanese is at a glance easier to "bypass" and come to an agreement about than for e.g. Korean or Chinese names and their parsing. I have seen the Japanese vowels with the macron countless times and I think it's commonly understood as a guide how to pronounce the vowel. But the software will not recognize that character but removes the macron. That should be a decider. The other much bigger problem we have is how do we document cast and crew from credits fully written in Kanji/Hiragana/Katakana? I would say it's impossible and simply not feasible to even try doing, for our database. There must be, in the credits or accompanying the credits for the disc in question, a (reliable, verifiable) romanization of the Japanese crew and cast for it to be entered. The fact is, and why I'm waking this thread up now, that I have contributed all the profiles in the DB for Ringu, Ringu 0: Basudei, Ringu 2, and Rasen with crew and cast, and which I am basing on the complete Arrow Video Limited Edition collection, which comes with a very nice booklet that lists crew and cast, along with of course checking the in-movie credits manually for each disc in the collection, because those take precedence if there are any variations/deviations. I am quite adamant that this is absolutely the way it must be done. I have gotten a lot of "yes" votes for my contributions, with one exception for one profile and from one member here, stating the existing data is "correct". But the fact is that I have checked all profiles, virtually, and there is no documentation anywhere from where the credits are taken. And yet still, if we were to approve of this data, only people knowledgeable enough about Japanese could verify it, and even further identify errors and make changes. Naturally this should be a big no-no. I'm sorry for writing a lot, but this is rather crucial. Otherwise we could just ditch "foreign/exotic" data altogether. At least the way I see it. |
| | Registered: May 2, 2009 | | Reputation: | | Posts: 490 |
| | Posted: | | | | So I have just, after much and long work researched the common name for Joji Iida and here are my thoughts after this "adventure", that has actually kept me up nights: This should also matter for and affect all romanized Japanese names in the DB, because that was always the purpose of this research.Unless the in-movie and on screen rōmaji of credits in kanji are using the so called "Kunrei-shiki or Nihon-shiki romanization", which do use the circumflex ("^") for Ô and Û, we must always assume and use the most widely spread "Hepburn romanization", which uses the macron ("¯") for Ō and Ū. In the end this will be bypassed by the software and turn into O and U.And for credits purely in kanji, we have no choice but to also use the Hepburn romanization. Note that local profile variations (if there are any) which may use arbitrary subtitles as romanization are not valid for contribution! The romanization must be the same for all profiles of the same title, otherwise we would have a very strange and confusing situation, regarding the CLT. Also, if there is a contribution of credits that are purely in kanji but still the crew and cast are noticeably "large" or complete, there must be some sort of verifiable documentation here, where the romanization is, well, documented. Maybe a pinned (LOL...) topic for this. I hope this makes sense and is clear to everyone and that we can agree. | | | | Last edited: by MikaLove |
| | Registered: May 2, 2009 | | Reputation: | | Posts: 490 |
| | Posted: | | | | Example of invalid vs. valid credit for Joji Iida (as George Iida): Invalid: Battle Heater (Kotatsu) (1990) <-- the correct credit is Joji Iida  Valid: Another Heaven (Anazahevun) (2000) <-- the correct credit is Joji Iida (as George Iida)  This is why the first example – of arbitrary, localized rōmaji in the form of subtitles – is invalid (a picture says more than 1,000 words):  However: I think, actually, the only thing that could and maybe even should be argued here, is that because of the naming system where you call someone by last name, first name, (which is sort of the same as in e.g. Hungarian), then shouldn't we honestly enter the names like that into the DB as well, 100% consistently of course? I.e. 飯田 譲治 = Iida Jōji --> Iida Joji. I simply feel that the way I have been making the argument for this, has still been based on "translation" and "interpretation"... Please make your voices heard! | | | | Last edited: by MikaLove |
| | Registered: March 14, 2007 | | Reputation: |  Posts: 79 |
| | Posted: | | | | DVDP is not the only IT system that struggles with the macron, and it has become common practice to use the circumflex when macron is not available - also for Hepburn romanization. Using circumflex instead of macron is a much better option than removing it altogether. When you remove the diacritic you're basically giving the cast/crew member a different name.
So while I agree wholeheartedly that we should select a single romanization scheme I cannot support the removal of circumflexes from the database. |
| | Registered: May 2, 2009 | | Reputation: | | Posts: 490 |
| | Posted: | | | | Quoting Exiled: Quote:
DVDP is not the only IT system that struggles with the macron, and it has become common practice to use the circumflex when macron is not available - also for Hepburn romanization. Using circumflex instead of macron is a much better option than removing it altogether. When you remove the diacritic you're basically giving the cast/crew member a different name.
So while I agree wholeheartedly that we should select a single romanization scheme I cannot support the removal of circumflexes from the database. I think the "fault" lies much more in the software than in my suggestion. It will mess up the CLT if we, in fact rather arbitrary, select the circumflex to be the savior for this situation. I don't see why we should give Japanese names a "free ticket" like this when there are similar situations for other characters that go missing in translation. Czech and Polish most of all. Two wrongs doesn't make one right. This is a DB where we strive for the highest integrity of it. And the circumflex is still an IMDb import more than anything. The way I see it is that until it is fixed (which will be never since no one is here to change things), we have to accept the fact that the system "eats up" certain characters. The circumflex also looks very non-Japanese if you ask me. Again, if a person is credited with the circumflex (practically never) but is then credited without it, this would mean that the person is always credited with a circumflex. And this is what I mean will break the CLT for Japanese names. And most of all, we will ignore what the rules say: that we must credit names EXACTLY as they appear. Not "approximately". | | | | Last edited: by MikaLove |
| | Registered: May 19, 2007 | | Reputation: |  Posts: 5,715 |
| | | Registered: May 2, 2009 | | Reputation: | | Posts: 490 |
| | Posted: | | | | And with that, "Ringu 0: Basudei" is now officially the original title for this database.
This place does not have room for fictitious minds and make believe. But we are here to do our work and to A. follow the rules and B. operate in accordance with the system and its software. Regardless of its (rather unfortunate and rigid) limitations.
I am taking this approval as a catalyst for change. Since there is no fruitful discussion going on here anyway, because no one takes part of it.
I have worked my backside off for this the last few weeks with very little in return, so I will do what I think is right. However I have gotten more insight from a few other users.
I am still contemplating how to actually document credits fully in kanji without subtitles or anything else to refer to.
I still have doubts about how we should credit names in kanji, however.
My proposition is definitely Family name/Given name, whenever this is the on-screen credit (which should be always).
If I don't get any feedback, I will go ahead and make a "test contribution" for a profile. | | | | Last edited: by MikaLove |
| | Registered: May 26, 2007 | Reputation:  | | Posts: 2,879 |
| | Posted: | | | | I agree with the Family Name / Given Name, especially since the Japanese government in 2019 asked Western media to print their names in this fashion going forward (even if most don't seem to be listening). | | | If more of us valued food and cheer and song above hoarded gold, it would be a merrier world.
-- Thorin Oakenshield |
| | Registered: May 2, 2009 | | Reputation: | | Posts: 490 |
| | Posted: | | | | Great! Does this mean that Japanese credits are always in this format? Or actually, that this is how names are written everywhere for every occasion? Japanese is very inaccessible to most people. For me it's just hieroglyphs at best. I think we shouldn't [have to] refer to e.g. booklets, however "official" and trustworthy they are when contributing credits. I don't believe in "shortcuts" of this measure. We shouldn't deviate from neither the rules nor from the way of contributing actual credits, despite the fact that it's in an isolated language. For this reason I would propose some sort of vocabulary as reference, here. Which could explain all the different roles in both Japanese (kanji and maybe rōmaji) and English. Maybe this is already available somewhere, but I think it would be better if it was a static part of the forums here. It would also naturally help if there was a way to import credits to text format on a PC. I imagine though that this is wishing for too much, most of all since this place is almost completely dead. My vision is regardless to be able to contribute 100% Japanese profiles like any other Western movie and profile, without any sort of 3rd party information. Oh, and other than that, if there would ever be a change or update in the software, it has to include all available Latin based characters.
At first glance, I think that this chart is complete.Secondly on this wishlist, I would not personally mind if there was also support for the Russian alphabet, since this language is so common and since there are quite a few Russian titles in the DB. Either way, I will at least myself look into one or two Japanese titles and see if I can get a contribution through, with this new name parsing (Family Name / Given Name). EDIT: I enlarged the text above because this is something that is really starting to get on my nerves; that the database supports so few characters so you can't submit credits, titles, overviews properly for way too many languages!For a database of this caliber it is simply intolerable! I can't believe we have accepted and looked past this for so many years. How can we follow the rules and contribute everything EXACTLY as it is typed, if the system doesn't even allow it? | | | | Last edited: by MikaLove |
|
|
|
Invelos Forums->DVD Profiler: Contribution Discussion |
Page:
1 2 3 4 Previous Next
|
|
|
|
|
|
|
|
|
|
 |