I don’t get it. Is the joke that i18n for CLIs is unimportant? Or is this an earnest post that just so happened to get posted under humor? I wish I had the source for the image.
The joke is that it’s hard to tell if this is a joke because the lines between good intentions, corporate jargon, and feasibility have been blurred beyond recognition both here and in the real world.
It’s also funny that after all these years, i18n is still a mess. Moreover, even if translations are standard in GUIs and documentation, for some reason, everyone is okay with defaulting to English for the oldest form of computer interaction.
Also, the joke is whatever you want it to be. Follow your dreams.
that’s because as a non-english native, it’s WAY easier to have the whole computer in english. obscuring options and error messages (by translating them) makes it much harder to fix problems.
I hope this is a joke because the Arabic translation is so wrong. It’s also confusing because Arabic is written from right to left so it’ll just create a mess. The translators are using “letter case” and translated it literally to Arabic. The word used doesn’t mean “letter” as in a letter in the alphabet but “letter” as in what you send in the post office. These are totally different words in Arabic.
It’s somewhat difficult to translate, because Arabic doesn’t have the concept of case in letters. Usually you can use “حروف صغيرة” or ”حروف كبيرة” which literally translates as “small letters” and “big letters” when referencing other languages. For the general “letter case” you can use “حالة الأحرف”. So it’ll be something like : تجاهل حالة الأحرف.
So here you substitute الرسالة for the correct word الأحرف to mean “letters”
Dumb question but your comment got this into my head: in your response, since it’s mostly English and LTR, are the Arabic words in your response read right to left?
Yes it’s always read right to left, which can be confusing when you combine English and Arabic. When you reach the Arabic word or sentence you jump to its beginning which is the first Arabic letter to the right, read it from there to the left, and then continue to the next English word when you’re done.
Also this is why unicode has codepoints signifying where to switch between right to left and left to right writing, so that letters can be correctly written “forwards” in the underlying file format (first letter written first) for both writing systems and also rendered correctly for both writing systems on display
Spanish is also wrong, this one means “ignore-letter-size”. I’m not sure if there is an official correct way to say in a short manner, I would say “ignorar-capitalizacion” but I think it’s just a barbarism.
I have to use a German API with weird halftranslations and ultra long names, due to bad model generation. Something like getPersonAntragsPersonAdressDetailEintragList().
Unfortunately, it makes sense, since many of the terms have a very precise legal meaning and can’t be unambiguously translated.
The first release of WSL(2) 1.0 (this versioning alone is worth another post here, but let’s not talk about it) have its CLI –help message machine translated in some languages.
That’s already evil enough, but the real problem is that they’ve blindly fed the whole message into the translator, so every line and word is translated, including the command’s flag names.
So if you’re Chinese, Japanese or French, you will have to guess what’s the corresponding flag names in English in order to get anything working.
And as I’ve said it’s machine translated so every word is. darn. inaccurate. How am I supposed to know that “–分布” is actually “–distribution”? It’s “发行版” in Chinese and “ディストリビューション” in Japanese.
At last I had to switch my system language to English to set a WSL instance up. From then on I never use any display language other than English for Microsoft products. Sometimes “translated” is worse than raw text in its original language.
PS: for the original post, my stance is “please don’t make your software interface different for different languages”. It’s the exact opposite of the author has claimed: it breaks the already formed connection by making people’s commands different.
It’s the CLI equivalence of scrambling every button to make sure they are placed differently in different languages in GUI. I hope this sounds stupid enough so that no one will try it.
A not-so-stupid way that I can think of is to add a “translation” subcommand to the app that given any supported flags in any language it converts them to the user’s language. Which is still not so useful and is not any better than a properly translated documentation, anyway.
The Microsoft Office installer has translated “Office downloads” (as in office is downloading now) to the plural form in Swedish, so it reads grammatically incorrectly as if there’s multiple downloads going on. Very professional, lmao
Try using Excel in another language than English. You have to hope someone, that speaks your language had exactly the same problem as you, because all the formulas get translated and Excel doesn’t recognize the English version when your language isn’t set to English.
So true… Not only sometimes it makes it hard to find the translated function name (especially since they are adding a lot of new functions) but there are quite often longer… For instance in French you go to a simple ifs to si.conditions…
@hstde@Spore Even better, the alphabetical index of function names was generated in English first and then translated, meaning the documentation looks like a scrambled mess in any other language because it is alphabetized according to what the English equivalent would be. #excel
Just wait until you’re working with different time/date formats, like, god forbid, sharing such documents to someone who has their Windows time/date format set differently than you have.
we should all standardize on Esperanto. Not because it is good, but because regardless of which language you know, Esperanto is the last choice, and thus the only equal choice.
Critical security hole a few months before was due to localized variables. And again and again in the past. Aside from countless other issues with batch and powershell scripts because of localized variables.
Translation of developer utilities themselves is the final layer of hell. I’m not hearing anybody out about this kinda stuff - after microsoft decided to TRANSLATE THE EXCEPTION MESSAGES IN .NET WITH NO WAY TO BYPASS IT making them unclear, unusable and ungoogleable, I realized what a terrible idea it is to fragment developer knowledge by language.
That’s why I always select “English (US)” when installing an OS or creating an account online. No bad or missing translation, no mangled UI because of longer words, and of course easily searchable error messages.
Yeah the only drawbacks is I later have trouble giving remote support to family members because their shit is in another language so I don’t know what does the option specifically say hahaha
Yeah, that’s why unlocalize.com exists (or… existed? Dunno, seems down from here!) Or you can have used the official Microsoft Language Portal… until they removed it and replaced it with the worse …microsoft.com/…/microsoft-language-resources (but it’s still usable, I guess…)
Interesting choice to romanize Japanese. Now you have to figure out which romanization system to use (I was surprised を was romanized as o and not wo). But I do get it, I guess, because you have to wonder it would only use Hiragana or mix Kanji in:
大文字と小文字を無視する
だいもんじとこもじをむしする
Well, for the sake of being international, we should just use Katakana everywhere. That’s the sanest suggestion (who’s with me?):
ダイモンジトコモジヲムシスル
Of course, you’re kind of screwed on a TTY, since they don’t generally render unicode…so let’s go back to figuring out which romanization system to use.
It doesn’t really mean anything on its own. It’s romanized as “Shi”. If you know your Japanese, you’ll know “Shi” is how you pronounce 死; or “Death”. The word is not usually written in Katakana, though. There’s also ツ, which is romanized as “Tsu”.
This looks like the final layer of hell. Your coworker writes their scripts in another language and now you have to decipher what the hell they mean. Who has a problem woth English for development tools, etc.? It’s really not a monumental task to learn it, and I’m not even a native speaker.
Even if everyone is using English, there will be cultural differences. I used to work at a company which had a lot of indian externals working on their code base. Whenever I had to work on a mainly Indian developed project i had to get used to how they wrote things. Usually things where named a bit different. Not by much, but enough tho throw me off a couple of times before i got used to it.
IMPORTANT: I am not shitting on how they used English, merely pointing out that they used it differently from how i would have expected.
In this case they were still using English, with minor differences. Imagine one of the Indian externals writing an internal script that utilizes the Indian localisation. You’d have to whip out a translator or dive into the docs for a tool which you may have already used countless times and know how it works when instead, they could have simply learned the English arguments for the tool.
Nothing against people not being native speakers of English, I’m not one either. I just think that this creates more problems than it solves.
I agree with you, that even the devil would run away from localised scripts.
Just pointing out that even if everyone is using English, there will be differences. These differences can make it hard enough - no need for more stuff on top.
Wouldn’t it be easy to convert the code to any language if this was the case though? Any human-language programming is already an abstraction, so why can’t a programming language be abstracted to more than one human language? Literally just swap the command words out for words in the other language, seems like something modern IDEs can trivially do if a language like that existed.
Also, non-English speaking countries exist. Some have actually developed programming languages in their own language so the idea of non-English programming isn’t exactly unheard of. There is no reason that code that won’t be edited by English speakers should always be written in English, it’s not like it’s the one perfect human language for interfacing with computers or anything. So I suspect that should this become a reality, you wouldn’t even notice anything’s changed unless you live in a non-English speaking country. Either way your company can still always just require code to be in English, the same way companies have requirements for formatting and software design philosophies.
May I introduce you to the concept of Microsoft Excel?
One time, someone from HR asked me, if I could help them with an Excel formula. So, I quickly looked up how to do something like that in Excel, adapted it as needed on my laptop, then sent it to them. And well, it didn’t work on their system, because I coded it in English, whereas their OS was in German.
Yep, this sort of behaviour translates to Windows paths also. Why would they name a directory “C:\Users\Example\Desktop”, when they can replace “Desktop” with a locale-specific name, which is not just a link to “Desktop”, but a completely different directory which breaks any scripts expecting “Desktop”.
Was there a thought process applied here at all? Worse still is that many of these localised paths are actually lies. They still use the original developer version in order to not break compatibility with programs, but refuse to admit it in the explorer. It’s maddening.
FUCK whoever thought translating Excel formulas was a good idea. It is the most infuriating shit. Everything I learnt in English is now useless, without googling every fucking function every single time. Fucking idiots.
Add comment