Posted under Tags
BUR #4335 has been approved by @evazion.
create alias komano_aun -> komano_aunn
Although i agree with the argument in topic #14084, "Aunn" is the most common romanization, due to it being the one present in the game's files.
As far as i know Danbooru is the only site which lists her romanized name as Aun.
As such, i believe it would be best to change the character tag in order to follow the romanization found in the game files, until a better one is found.
Username_Hidden said:
As far as i know Danbooru is the only site which lists her romanized name as Aun.
Back when I joined the community, roughly a decade ago, arguments like this were brought up all the time. They were consistently rejected.
It used to cause a lot of friction, but the idea was crystal clear: "For us on Danbooru, the most common romanization is not a priority. We want to know how Japanese names are spelled in hiragana or katakana. If more of us knew enough Japanese, we would not even bother to romanize any names for tagging purposes."
If we were to follow those "classical Danbooru guidelines", using Aunn in tagging would be an absurd idea.
Then again, I know many guidelines have shifted towards "most widely known", "most practical", and the like over the years.
Where is the community standing today? komano_aun and kongara_(touhou), or komano_aunn and konngara_(touhou)? Is there any reason why we would not want the two Touhou character tags to be consistent with each other?
Before making decisions, we should pay attention to what this nn spelling is all about. It is a totally nonstandard, yet fairly widespread means of representing the ん kana. For instance, the original form of the artist tag sen'yuu_yuuji was sennyuu_yuuji, which is potentially a lot more confusing than either Aunn or Konngara.
Also, if "the most widespread form in English-language sources" becomes the ideal, then we surely should use koizumi_junichiro (or even junichiro_koizumi), not koizumi_jun'ichirou, right? How far are we going to take it?
The "classical Danbooru guidelines" were not intuitive for the average fan, but they had one big advantage: if we were capable of tracking down the exact intended spelling and pronunciation of (almost) any given Japanese name, it was impossible to romanize it wrong.
Non-standard kana spellings, such as 眞白ヶよぇ (Mashiroke Yoye) that henmere explained to me eight years ago, caused trouble even then.
Katajanmarja said:
The "classical Danbooru guidelines" were not intuitive for the average fan, but they had one big advantage: if we were capable of tracking down the exact intended spelling and pronunciation of (almost) any given Japanese name, it was impossible to romanize it wrong.
The problem with the previous romanization system is that it was way too rigid, which caused a lot of confusion on quite a bunch of tags and had wrong romanizations. It was (and is) in no way a failproof system.
The most recent revision happened in topic #17011.
Before making decisions, we should pay attention to what this nn spelling is all about. It is a totally nonstandard, yet fairly widespread means of representing the ん kana. For instance, the original form of the artist tag sen'yuu_yuuji was sennyuu_yuuji, which is potentially a lot more confusing than either Aunn or Konngara.
If i remember correctly, ん is supposed to be romanized as "nn" or "n'" if present in the middle of a phrase, and "n" at the end.
The correct spellings would then be Aun and Konngara.
The reason why ZUN wrote Aunn in the game files is unknown - some have been speculating that it was a byproduct of JA IME, which won't let you spell ん without typing two Ns.
Either way, we aren't talking about the correct spelling, but the most widely used.
It's romanized as "Aunn" in the omake, it's not a romanization people had to make up from scratch and it's official, so we should go with that until further notice.
EDIT:
You should really have given a straight link to topic #17000, where the actual discussion about the revision proposal had taken place. There I could see a less rigid system had reached wide active support, including Evazion. Most of my comments below lack potential to change anything, now that I have read most of said discussion.
However, I am not deleting my comment here, for it does include some relevant points not mentioned in topic #17000.
Also, I would like to say something about the values that back up my opinions. While I am certainly leaning towards prescription rather than description when it comes to the rules of written (not spoken!) language, my aim on Danbooru has never been to prescribe; it has been to spread (and obtain) knowledge and understanding.
If Evazion considers easy access for casual fans clearly more important than teaching at least some basics of the Japanese language to the fandom, then it is rather futile of any semi-casual community member to challenge him. I might still do that if I were one of the most prominent tag gardeners here; but I am not.
***
ORIGINAL MESSAGE:
Username_Hidden said:
the previous romanization system (...) was way too rigid
That is a matter of opinion. In my opinion, a rigid transliteration is a good transliteration.
However, I am not saying my opinion cannot be challenged. Flexible systems can be good for people who are just interested in how things are spelled in general in a given context (such as within the wider western fan circles). Rigid systems are good for people who are genuinely interested in spellings used outside the Latin-alphabet world (even if they are not especially interested in learning all the languages of the world; for instance, I have never seriously tried to learn Japanese). Aliases will always be needed, whichever of the two is preferred by the community.
I notice that Shinjidude’s revision in topic #17011 (thank you for the link!) is more of a revision proposal, if very well composed. It has faced no opposition, but it has not stirred up real discussion either. I suppose I should go write a comment.
I would like to stress that I highly value Shinjidude’s input on this site, and would consider it rather stupid to wage an all-out war against his proposals. Over the years, I have supported almost every decision of his, and I hope my differing views on this one point will not be considered a personal attack.
Username_Hidden said:
the previous romanization system (...) caused a lot of confusion on quite a bunch of tags and had wrong romanizations
Wrong romanizations? Such as?
The only example I can come up with off the top of my head is that cch (ecchi, for example) was never purged and replaced with tch (or openly endorsed as an "officially" accepted Danbooru exception). On all other fronts, the "cleaning" work went on steadily for years, towards less and less wrong romanizations, until the most ardent proponents of a "rigid" system went inactive.
Username_Hidden said:
It was (and is) in no way a failproof system.
Uh, how come? Of course it is a "failproof" system for all cases where a standard hiragana spelling is available, if we accept its fundamental goals. For instance, it dictates quite clearly that 士郎 正宗 (しろう まさむね) is to be romanized as Shirou Masamune. This may be counterintuitive to those who are not interested in kana spellings, but one of the fundamental ideas of the system is that counterintuitive results are successes rather than failures as long as they spread knowledge of the correct kana spellings.
Username_Hidden said:
Either way, we aren't talking about the correct spelling, but the most widely used.
Saying "we" is not a very good idea here. You are talking about "most widely used", while I am talking about "correct". Furthermore, I am wondering what is the best way to reach "correct"; I have had my reservations regarding "Danbooru Hepburn" as well.
Once again, it is (and has been) a fundamental debate whether the Danbooru community wants "most widely used" or "correct". Both are valid goals to pursue.
I can see a future where "most widely used" wins. It is a very practical way to follow, and many (most?) members seem to favor it or at least accept it. I do not favor it at all, but I can easily imagine accepting it.
However, "most widely used" can never be failproof by definition, unless we define it as something as simple as "the majority of Google hits right now". Otherwise, "most widely used" will frequently lead to arguments like, "X spelling was always used by manga scanlators until 2007!" "Y spelling was almost exclusively used in the English-language TV anime and OVA releases between 2006 and 2008!" "Z spelling has been used in the official manga translations since 2011!" Of course, if fans want to stay civilized, they can work out such differences.
Username_Hidden said:
If i remember correctly, ん is supposed to be romanized as "nn" or "n'" if present in the middle of a phrase, and "n" at the end.
I have never encountered such an official system, so I have supposed the nn to be a "wild" waapuro romanization. If you are referring specifically to the Gaimushou Ryoken Kitei guidelines mentioned by Shinjidude, I am not familiar with them, for which I apologize; my background is mainly in Wikipedia romanizations and older history books.
Updated
After the release of the HSIFS demo but before the release of the full game, Eternity Larva was tagged here as "etarnity_larva"(sic) due to how her name appeared in the game, despite the fact that Zun obviously intended for her name to be spelled correctly.
I vote we follow that standard, and tag Aunn's name with two Ns.
The bulk update request #4335 (forum #178251) has been approved by @evazion.
BUR #4803 has been rejected.
rename komano_aun_(komainu) -> komano_aunn_(komainu)
Missed in the above update.
The bulk update request #4803 (forum #181384) has been rejected by @tapnek.
