This proposal comes from the concern that there should be an exception in romanizing 1番 in context.

Links for context:
beatmapsets/2208241#osu/4676294



+---------------------------------------------------------------------------------------------------------+

To explain, in the general rules of romanizing numbers, you are supposed to romanize a number as a number.

Yet the concern comes where - yes, technically you do not want to change a number but 1ban will be read wrong by anyone who doesn't know Japanese.

(Lyrics websites have a tossup between １番 & 一番 but it is still read いちばん https://www.uta-net.com/song/356891/ - https://www.kkbox.com/jp/ja/song/OpGE5rcQzUmSZSDFA4)

A person with no Japanese context would easily misread 1番 as "one-ban" -- which is my main issue.

+---------------------------------------------------------------------------------------------------------+

Anything other than 一番/１番 is generally fine and I would draw the line here for standardization.

In the context of this song's title, "1ban kagayaku hoshi"・１番輝く星 -- 1番 is not used as a counter for an object, it is still used as the general understood word for ichiban - which is "first".

I want to argue that Ichiban in general is understood as "First place" "number 1" "best; most​" etc... rather than a numerical modifier
whereas something even close like ２番 does not hold as much importance because after the first, any number proceeding it is a numerical counter ( number + counter modifier 台、号、個、番　）.

Again -- In the context of this title, Ichiban is used as a (word) even though the stylization is different (1ban) the general meaning of the usage is still Ichiban (the word).



+---------------------------------------------------------------------------------------------------------+

I strongly believe that this also applies to other usages in the future.

Some examples of usage I can think of is:

1番最高　＝ "the greatest" would be weird if you romanize it as 1ban saikou because it would be... I dont even know how you would make 1ban saikou palatable in English...

１番上　＝　"uppermost / topmost" and again, I cant think of an equivalent when it would be 1-ban ue (one-ban ue)

１番近い ＝ "closest"　→ Ichiban Chikai v.s. 1-ban chikai (?)

Etc...

+---------------------------------------------------------------------------------------------------------+

To summarize: Ichiban should be accepted a romanization of 1番 as it holds the general meaning of "first place". (In the context of the song it acts as a modifier of the word kagayaku = Bright"est" star)


Thank you for your consideration.

I'm not a fan of the proposal, for a couple reasons.

First one being something you agreed to before: you are supposed to romanize a number as a number. It was the artist's idea to style it as such. Romanizations should carry over said stylistical choices. Whether a non-japanese reader can understand it should not be a concern. Under that logic, wouldnt the best possible romanization be a translation to a more commonly spoken language? That's more akin to a localisation rather than romanisation.

Secondly, I'm not really a fan of having exceptions on top of exceptions. Especially for something already confusing enough like romanisations are for those who do not dominate the language. Proposals should aim to simplify it, not make it more convoluted or full of conditionals.

I agree with the basic premise here, but I'm not really sure why we would stop at just ichiban?

Why not allow any counting words to just use the counting word's full reading in romanisation?

七番目 / 7番目 => Nanabanme etc

I think forcing 1-ban 7-banme to look kinda stupid while also obstructing how to actually read the thing (aka current) is kinda worse as far as the basic goal of providing a romanised title is concerned?

idk what the "hepburn specs" say about this but i think this makes sense because the entire meaning of the counting words in japanese is lost to anyone who doesn't already know lol

I agree with Oko. The reason I capped it at ichiban was because I was weighing in what RandomeLoL stated.

"Secondly, I'm not really a fan of having exceptions on top of exceptions. Especially for something already confusing enough like romanisations are for those who do not dominate the language. Proposals should aim to simplify it, not make it more convoluted or full of conditionals."

I was discouraged from pointing out more due to this type of mindset.

+----------------------------------------------------------------------------------------------------------------+

This shouldnt be the case, this is why we have rules and a platform for rules to be changed (like this forum and the purpose of it) in the first place.


To add; a user I asked wrote:

"numbers should be romanised as such when it makes sense to
like in 山田君と7人の魔女
i would romanise as 7-nin
but if its used as a superlative like 一番近い like u said in ur post
then ichiban as a word is much better"

another user wrote:

"would jp people actually read 1番 as ワンばん
i feel that it's like english people reading "2nd" as "second" instead of "twond""

another user wrote:

"in (redacted's) message i legit read it as oneban
subconsciously
like
this is a good change i think"

+----------------------------------------------------------------------------------------------------------------+

Not tackling this issue would make romanizations for cases like this, not only look awful for people who understand both languages but also make it more confusing for people who do not speak the language as they would be reading it incorrectly therefore being a disservice to the greater population of non-japanese speaking people.

(as stated: "Especially for something already confusing enough like romanisations are for those who do not dominate the language")

(issue where the general consensus of people I have asked subconsciously read 1-ban as "one ban")

This seems overly complicated and doesn't add much benefit imo

i dont think this kind of details belongs to rc, cuz it would end up being too much wall of text

but i do agree with the issue and it should be romanised as Ichiban, i think we should just rely less on rc and more on native language speaker for this kind of special case

reason why i say it should stay as `Ichiban` is because thats how the lyrics is, she says it in the song at 52 sec where chorus starts, and also Official english translation is Ichiban from Crunchyroll

https://www.youtube.com/watch?v=MBM7qyH-YDY


I don't think RC currently says anything that number should be romanized as number, and Ichiban isn't a number anyways, its an adjective.

going by current RC we shouldn't end up as 1Ban or 1-ban, so no change is needed for RC.

For future reference, if you want make a change to RC, you should first point out the exact sentence in current RC which u want to change, and all related clauses

I don't agree with this proposal.

1. The original metadata uses explicitly 1番, and to be faithful to that original metadata, 1-ban should be used. If it was written like 一番 then Ichiban would be used. As Randome said it seems to be a stylization choice done by the artist, so there's no need to "correct" it in any way.

2. Modern Hepburn does not have any rules telling us to romanize numerals in the way they're read. So by standardization they should be left as is.

3. Nothing is stopping you from putting ichiban in the beatmap tags, if you worry about not being able to find the map in a search query.

4. RC doesn't really require this type of specificality, and definitely not for one specific word like this. Even if it's agreed on upon, it should include broader cases and not be limited to いちばん. For example, allowing something like "60年" to be romanized as "Rokujuunen" and such other numerals in various languages too.

I am unable to grep (Ctrl-F) any usage of the word "number" or "numeral" in the main Ranking Criteria, so I think that the "romanize arabic numerals to arabic numerals" rule is more an unspoken practice than a clearly written rule. While I am in favor of "Ichiban" over "1-ban" for this particular case, I don't think this mandates an amendment.

Forgot to update this but after speaking to Nao he came up with:

"When appropriate, numbers can be romanized into their original language or left as is"

This discussion is now closed / resolved. Thank you all for the input.