The how to edit thread

Posted in

#101 by yorhel
2018-01-30 at 16:36
@100: Make sure to add a trailing slash to the URL:
http://kuma-puro.b.dlsite.net/
#102 by barfboy
2018-01-30 at 20:50
Thanks much.
#103 by leery
2018-02-02 at 11:37
Is 東京 one of those words we must romanise as the English 'Tokyo' and not its accurate romanisation of 'Toukyou'?

Same question for 京都 I guess. There are four instances of Kyouto and two Kyotos in the database. And one Koube, two Oosakas, etc.
#104 by beliar
2018-02-02 at 17:01
@leery: That's exactly the question I asked Yorhel once upon the time. Let me unearth the thread... Ah, here it is: t6246.
#105 by leery
2018-02-04 at 09:31
I suppose there's no standard about the romanisation of censored (bleeped out) characters.

I mean, stuff like ○学生 which could stand for 小学生 or 中学生 or 大学生 or something else. The characters × and ● are sometimes used for bleeping out too.

Many times these special characters get carried over to the romaji, which just feels wrong to me. Sometimes they even get romanised with the letter 'O' based on visual resemblance. Sometimes they get ignored in the romaji entirely.

(Also, there's the case of titles like Boku no xx wa Ryousei-tachi no Tokken Desu! where they get replaced with 'xx' for whatever reason... Then there's Chikan Play Densha no Naka de OO ya XX na Koto which gives me an eyesore currently.)

I've been replacing them with '***' recently. Or just '*' when it's only a single letter. Is that fine?

(I'm not going to go on an edit spree about this issue though.)Last modified on 2018-02-04 at 09:47
#106 by rusanon
2018-02-04 at 09:53
Please do not use *, its forbidden symbol for Windows and special symbol for Linux. I re-name VNs in collection according to VNDB to keep everything manageable, and its annoying when vndb title is stuffed with symbols you can't use in path name.
#107 by leery
2018-02-04 at 09:54
So is ?

Do you propose stripping out all the question marks out of titles?
#108 by rusanon
2018-02-04 at 09:59
I propose at least not adding even more of them when there's no such symbol in original.
There are alternatives as you said yourself: keeping Unicode ● or × or using X. Its 2018, no need to stick to ASCII.
#109 by barfboy
2018-02-12 at 14:36
So, this teacher finds his big titted student bound to a desk with her panties off and legs spread open asleep. Tsukue Otome.

He can help her or he can molest and rape her. Problem is she did it to herself because she has a crush on her teacher and is pretending to be asleep. She's hoping he rapes her and even comes to his aid if someone walks in and thinks he's raping her claiming they are a couple and were just playing.

He doesn't know any of that. Just that he's come across an apparently sleeping girl bound to a desk with her panties off and her legs spread open. Thinking she's being bullied by a classmate he wonders if he should rape her or not. Basically on which party does a rape tag occur the perpetrator or the victim?
#110 by barfboy
2018-02-12 at 14:43
Oh, leery, I have a question very similar to that.

Official website for Chibi Pire! ~Kyuuketsu Shimai to Ecchi de Bitch na Dousei Seikatsu~
The girls are 11○ years old and 12○ years old

link

Now obviously they're "110" and "120" but how do I have that point come across on their character pages
link
When I tried to write that it got changed back.
#111 by 707
2018-02-12 at 15:06
^To be honest if you didn't point that out I would've thought it's a Japanese period (。)

Maybe change it to something more noticeable that they're 12 (but with another digit) years old?
12(0) years old or like what you typed, "12○ years old"
#112 by sakurakoi
2018-02-12 at 15:07
@109 I'd argue that Rape Attempt and Fake Rape should apply respectively (and probably be marked as spoilers), albeit the description not reading like that, at all even but the terms could not fit any better+there is no other term remotely close. He attempts to rape her and she is not actually raped for she wants sexual intercourse i.e is actually not raped=>fake.

@110, the exact age is pointlessly kept ambiguously for a good reason so let's just leave it at that, they could be very well be 119 and 121, or more likely even since two siblings are rarely 10 years apart unless there are maybe more than 2 (but even then the next closest...) but anyway, the user who changed it got deleted so you should be free to use ○ (not the "Japanese period" or whatever it is called, pretty sure they are different, the circle ○ and period。oh, and 0, O and Ö) or replace it with X in case it can not be displayed properly.

and yes, X and O is usually what is used in the west for a wildcard/keeping something ambiguous for whatever reason which we should not question, with the latter only being when more than one is needed. Welp, it is kinda everywhere, ain't it? Over there the circle and cross is simply its own symbol.
#113 by barfboy
2018-02-13 at 03:55
I'm sure it's not that they're not somewhere between 110-119 and between 120-129 years old and more that they're 11 and 12 years old and erectolip is being cheeky getting underage girls past the censor.

For the same way hamham soft gives the girls 'seating numbers'
link
despite the fact that two girls share THE SAME SEATING NUMBER (11)

or how Concept uses 'favorite number'
link
despite the suspicious fact that Kei wishes her favorite number was 18 but it's still only 12 ... how can your favorite number not be whatever you want? When it's how old you are.

Times they be a changin
NekoNeko soft previously tried getting their heroines from White ~Blanche Comme la Lune~ listed on their official site as 11 years old but were forced to remove it after release
finally got a heroine admit she was 8 years old within the text of the game
Ruri no Kasane ~Imouto Monogatari~
But still get a wide release and games have started dropping the 'all girls are over 18' label

We'll see where the future goes. Anyway, long post. Thanks for the input
#114 by eacil
2018-02-13 at 05:37
Erectlip is a troll. I don't read them anymore because it's stupid (in a non fappable way) even though I was really tempted by this one.
They are vampires and they used censorship to troll again by turning the censorship against itself by making a 110 something years old vampire look like its age says 11 years old... what she is physically. It is that simple.
It doesn't mean they are 11 or 110 but can really be between 110 and 119 (in what way does it prevent the joke to work?) and you have no way to know that until it's said in the script, so, no, it's not obvious they are 110 and you have to keep the censorship until real proof. Edit: what you did.Last modified on 2018-02-13 at 05:46
#115 by leery
2018-02-13 at 06:16
and yes, X and O is usually what is used in the west for a wildcard/keeping something ambiguous for whatever reason which we should not question, with the latter only being when more than one is needed.

I've never seen 'O' used here. I'd usually do "insert a word here" kind of marking (which the Japanese ×× essentially is) with an underscore.Last modified on 2018-02-13 at 06:18
#116 by entra
2018-02-13 at 20:12
Not that it really matters much, but what time zone are release dates supposed to be in? Sticking to the release date in Japan for Japanese games is reasonable but what are people doing for games released on globally on digital stores? UTC? If it's UTC, I guess technically all download releases for any language should be UTC too.

Edit: About the censorship, * usually looks silly if it's more than one character in a row except for really well known words. In that case, I think ● , ×, or X is better (assuming they are ignored by search)Last modified on 2018-02-13 at 20:16
#117 by leery
2018-07-14 at 08:07
tl;dr: I'm going to edit some titles with 'shortened loanwords'.

d5:

For English (or other foreign) words in a non-roman script, the original word should be used as long as it is the intended effect.

from anidb's link (omitting some irrelevant things):

Spell in original language where possible

[...]
スカトロ(ジー) (katakana) / すかとろ(じい) (hiragana) ('sukatoro(jii)', 'scatolo(gy)' (English)), Japanese usage is a short form of "scatology", and is consistent with the definition of scatology. Spell it as 'scatolo' for the contracted version (it's just an abbreviation of the loanword) and 'scatology' for the full version.

What to do with Wasei-eigo terms, and invented words using other languages (ie Wasei-xxx)

[...] In general, spell in the foreign language if it makes sense to do so. If it does not make sense, transcribe the reading directly without trying to map to the foreign language.

Examples:

[...]

アパート ('apaato', 'apart(ment)'), Japanese usage is a short form of "apartment house". Spell it as 'apaato' as it does not readily map to existing words in English with the meaning of "apartment".
アルバイト (arubaito, 'arbeit' (German)), Japanese usage means "work or job", and is borrowed from German. Spell it as 'arbeit' in the transcription title, as it readily maps to German.
バイト ('baito', '(ar)beit' (German)), Japanese usage is a short form of "arbeit". Spell it as 'baito' as it does not readily map to existing words in German with the meaning of "arbeit".

I'm going to go on an edit spree sometime in the future, going to kill all the titles with abominations such as Apart etc. in them, for example Chiccha Love Apart.

In short, I'll do a strict romanisation of the katakana.

- アパート becomes 'apaato' (not 'apart')
- アプリ becomes 'apuri' (not 'appli' or 'app')
- コス (from コスプレ 'cosplay') will become 'kosu' and not 'cos', e.g. v9719
- ティー (from ティーチャー) will become 'tii', not 'tea' (e.g. v12353)
- ヤリサー becomes 'yarisaa' (not 'yaricir' -- a non-existent word)

- ドル (from アイドル 'idol') will be left as '-dol' because it's so common. Although I have absolutely no idea what a 'Locodol' is supposed to be...


The first (anidb) rule I quoted I very much disagree with. It also seems to conflict with the バイト example. Is サポありカノジョ supposed to be 'Suppo Ari Kanojo' or something? I'm going to use 'Sapo' to sync it with the six other entries with サポ I think...

Are there any other words I should pay notice to?


Consider this an advance warning. Speak up now if you are against this.

Incidentally, I still think v13914 should be 'takuto'.Last modified on 2018-07-14 at 08:43
#118 by shinnew
2018-07-14 at 08:37
Great, now the DB will be even more inconsistent because you are not going to bother changing the release titles like the hundred previous times you changed a VN title.

Also, アプリ is consistently rendered as "Appli" on AniDB. link link link link link link
#119 by leery
2018-07-14 at 08:49
I'm not entirely fond of AniDB's scheme. Last I checked, they romanised パンツ as 'pants'.

I did ask for an automated sync between VN and release titles a full six months ago, before I made hundreds of edits. I honestly do not think there would have been much if any fallout from such a feature.
#120 by yorhel
2018-07-14 at 09:06
I did ask for an automated sync between VN and release titles a full six months ago
That's a shitty excuse to leave the database in an inconsistent state. Either you do something well or you shouldn't do it at all, IMO.

Is it possible to reliably do these title transformations in SQL (e.g. with regexp_replace)?
#121 by leery
2018-07-14 at 09:21
Fine, I have problems with my wrists if you must know, and would like to keep the labour needed to a minimum.

I figured you'd implement this sort of automation with Multi, actually.
#122 by savagetiger
2018-07-14 at 10:23
コス should still be 'cos' because it's obviously shortened from cosplay.
Same with アプリ, 'appli' an obvious abbreviation.
And change the releases too, it's just a bit of copy paste.
#123 by eacil
2018-07-14 at 10:51
A locodol is another trash concept: a local idol. There is an anime with the same name which tried to ride on the Love Live! success and I almost killed myself after one episode.

I can't say I agree to change all those loanwords to their unrecognizable forms. You see "cos", you know what it's about, you don't have to think about what could that mean, like the weird portmanteaux otacir/yaricir which are making you scratch the head. You spot the origin right away.

Why are you applying this general rule to the specific Japanese use of abbreviations and contractions in the first place? It's not like we have any english use for "cos" and it's not like the word cosplay will suddenly transform into "kosu" if you cut it in half.

This anidb rule is weird considering the arbeit and baito inconsistency. Two really different transcriptions for the same word. It's like they had this baito everybody knew but didn't know what to do with it considering it screws everything. It's baito most likely because no one knew German in the first place and "bei" doesn't even sound like バイ in most languages. It looks to me more like a list of exceptions like gyaru (you want to change that too?), baito, pantsu, and... and... I guess other words from wasei non eigo which got transcribed as is for some reason, than the need to match the original meaning for something, you know, Japanese? I don't see the meaning to be relevant to the transcription. You don't know Japanese then no one has to explain to you what those wasei eigo mean. It's for ease of reading, that's all.

Maybe it's because not many exceptions come to my mind. Until now, I thought the rule "everything (minus some exceptions based on practice) which is a loanword should be transcribed according to the language it comes from" was working on vndb?Last modified on 2018-07-14 at 13:29
#124 by beliar
2018-07-15 at 11:31
I disagree with Leery on this one. We have a saying in my country: "Don't look for earthworms in your asshole". You are trying to invent a consistency by turning recognizable abbreviations of loanwords into unrecognizable kana renditions. If アプリ is a short form for application, then it should also be transliterated into a short form of application - appli works well. The same with the other short forms like コス.

And don't even start me on that 'takuto' nonsense.

Reply

You must be logged in to reply to this thread.