To release or not to release...

[sgeos]

No chance you'd get anywhere. If I remember correctly, Spanish has those weird letters (accents and all that; a little bit like our german umlauts (?????? and of course the ?)) - which means you'd need unicode support

French speakers prefer to have accents, but they can get by without them. I suspect Spanish speakers are the same. I suspect many of them would prefer no accents to English.

And even if it was possible technically - it would be quite a lot of work. Just think about the mad minstrel songs, and the fortune cookies.

Bah. This is all translation work that has no technical implications if ASCII is used. Although there is are a fair amount of messages to translate, none of it is difficult translation work and the amount of text is finite.

On the other hand: I think the fact that there is "only" an english version is good, because:
- it keeps us together: there are not *that* many ADOM players; and now imagine there were x forums for y languages...

This is a xenophobic argument. Are you actually afraid of growing numbers of Spanish ADOM players? Are you afraid of them if you don't deal with them now and you would not deal with them on a Spanish ADOM newsgroup?

You'd have to localize monster and item names and so on,

Not a big deal. A full time translator can typically pull off ~2000 words per day, so I imagine monsters and items would be translated more or less instantly. The amount of text is finite, as ADOM is not being actively developed, so I imagine we could use a calculator to get a rough estimate of the amount of time it would take to translate it.

and that would mean that we couldn't really communicate with each other.

Not true. I play games in Japanese and I am used to referencing both English and Japanese material. I have no trouble communicating with English or Japanese players and they have no trouble communicating with me. This is not a problem. (How many languages can you communicate in?)

I fear the translation would never, ever reach the quality of the original thing.

This fear is unfounded. Go out and talk to people. There are people out there who have native bilingual language skills. Have more faith in other people.

The writing is far and away not what makes ADOM special.

It is certainly not impossible to translate. Treating it like the Qur'an in this regard is just silly.

The in game mini-FAQ seems to suggest that multiple language support for ADOM would require serious amounts of work on the source code, rather than just translation of in game text

The thing is, localizable text is a problem that has been solved, and the solution is simple. This should be no harder to implement than any other feature. From the arguments given, the opposition to localization seems to either be lack of knowledge of the solution, or lack of desire for localizations. If the real reason is that localizations are not a priority, I wish that would just be stated honestly instead of trying to pretend that an insurmountable technical limitation is preventing localization. Go see how Google Android apps handle strings (they use a single XML file).

To localize ADOM, the piecemeal messages ("The foo", "attacks", "and", "misses") would need to be replaced with static messages ("#THE_MONSTER() attacks and misses.", "#THE_MONSTER() attacks and hits for #X() damage.", ...) that are dispatched when the final piece of piecemeal version would have been printed. These modifications would not be hard for someone who has the time to go in and permute all possible messages (using message tags, like #THE_MONSTER(), to keep the number of permutations reasonable). Also, all inline strings would need to replaced with string IDs.

[Epythic]

French speakers prefer to have accents, but they can get by without them. I suspect Spanish speakers are the same. I suspect many of them would prefer no accents to English.

Still, that would look amateurish. I get a headache everytimes someone writes... dunno, "Gruesse" (greetings) instead of "Gr??e". A friend of mine types on an english keyboard, nearly all texts he writes have this problem. It's ugly.

This is a xenophobic argument. Are you actually afraid of growing numbers of Spanish ADOM players? Are you afraid of them if you don't deal with them now and you would not deal with them on a Spanish ADOM newsgroup?

Huh? No, I'm not afraid of spanish ADOM players. I'm afraid that the already small group of roguelike-players might get separated by the use of different languages.

I don't want a german translation either.

so I imagine we could use a calculator to get a rough estimate of the amount of time it would take to translate it.

$ strings Adom_winbeta4.exe | wc
24391  159248  971344
$ man wc | head -n 6 | tail -n 1
wc - print newline, word, and byte counts for each file

In any case, one man made it, so it should be possible for another man to translate it. Still, probably a lot of effort. Do you know a "professional translator" who'd be willing to do a full-time job for free? No? So it's gonna take a *bit* longer. And remember, there are quite a number of languages...

Not true. I play games in Japanese and I am used to referencing both English and Japanese material. I have no trouble communicating with English or Japanese players and they have no trouble communicating with me. This is not a problem. (How many languages can you communicate in?)

FYI, I only use three languages (german, english, french). Well, actually, forget about french.
When playing german games I often have the problem that I don't know the english names, and vice versa.

This fear is unfounded. Go out and talk to people. There are people out there who have native bilingual language skills. Have more faith in other people.

Of course there are such people, but face it, they are a minority.

And then, if there are no problems - why make localized versions at all?

It is certainly not impossible to translate. Treating it like the Qur'an in this regard is just silly.

Ok, I got that point.

The thing is, localizable text is a problem that has been solved, and the solution is simple. This should be no harder to implement than any other feature. From the arguments given, the opposition to localization seems to either be lack of knowledge of the solution, or lack of desire for localizations. If the real reason is that localizations are not a priority, I wish that would just be stated honestly instead of trying to pretend that an insurmountable technical limitation is preventing localization. Go see how Google Android apps handle strings (they use a single XML file).

I never said that it was impossible due to technical limitations. But still, they are there and have to be worked around.

Besides:
1. you have a problem
2. you decide to use XML
3. now you have two problems.

To localize ADOM, the piecemeal messages ("The foo", "attacks", "and", "misses") would need to be replaced with static messages ("#THE_MONSTER() attacks and misses.", "#THE_MONSTER() attacks and hits for #X() damage.", ...) that are dispatched when the final piece of piecemeal version would have been printed. These modifications would not be hard for someone who has the time to go in and permute all possible messages (using message tags, like #THE_MONSTER(), to keep the number of permutations reasonable). Also, all inline strings would need to replaced with string IDs.

Thanks, but I know how it's done technically.

You already named one big problem: development time.

Here's another problem that might or might not be related to adom: the use of male/female/plural... word forms, and articles. Not a problem in english, where you simply write "the". But french (le/la/les...) ? I'd probably get it wrong, so I don't give an example.

[Silfir]

How about German? Der, die, das? And the different plural endings? -er, -en, -n, none at all, -e with an umlaut in the word, -er with an umlaut in the word? The totally different sentence structures at times?

I can't really add much more about this. There's a load of translation work needed. It's not impossible to have very widespread localization - Battle for Wesnoth for instance has more languages than one could possible make use of, some included just for novelty. It IS fortunate that most of the people who do translations are people who can't contribute to most of the other aspects anyway. So yeah, if open-source ADOM becomes a reality, a translation project might be something. I imagine sgeos will be first in line to modify all the language code to produce perfect French, Spanish, German, Russian sentences, and someone will figure out how to put other ASCII tables in...

EDIT: I'm playing German Wesnoth right now. I keep groaning at the translation - it's mostly perfect, but sometimes the words they chose for important stuff just doesn't fit right. "Einberufen" for "recall"? It doesn't even make sense, since "Einberufen" is closer to "draft", and you don't draft units with this - you really do recall units you had in previous scenarios with this - you don't draft them! "Ausbilden" doesn't hit "recruit" either - it should really be "Rekrutieren". "Ausbilden" is "train". Training and recruiting are more or less the same, but really only more or less.

And then there's "Waldkenntnis" for "ambush", where they tried to be smarter than the English version and ended up introducing an extremely silly-sounding word instead of just going with "Hinterhalt" like they should have.

This is part of what I fear when it comes to translation quality - these quirks are the result of a lot of translators arguing over different forms of translation and ending up going with strange compromises. With one guy with the final say over the translation ist just sounds a lot more coherent.

EDIT: And rangers are "Waldhüter"? Are these guys serious? What, I dare you, was wrong with "Waldläufer"? ("Waldläufer" is basically a "woodstrider", while a "Waldhüter" is a "woodkeeper". Yeah, like goalkeeper. In fact, goal is "Tor", and goalkeeper is "Torhüter".) They made Aragorn a "Waldläufer" when they translated Lord of the Rings. In just about any translation of any fantasy work that includes rangers, they go with "Waldläufer". And you go with "Waldhüter"? Aragorn is a fucking goalkeeper? You are silly and I hate you all. I think I'm going back to English soon, but not just yet. Haven't amused myself enough

[Sradac]

I say why bother translating. ADOM is fine the way it is, its a game written in english. If someone wants to play it that bad, they better go learn english. Dont want to/ cant? Go play something in your language and leave us alone. ADOM has done fine its entire life being in english.

[sgeos]

Go play something in your language and leave us alone.

Overtly xenophobic. Why would it bother you if someone translated ADOM, played it and left you alone?

This is part of what I fear when it comes to translation quality - these quirks are the result of a lot of translators arguing over different forms of translation and ending up going with strange compromises. With one guy with the final say over the translation ist just sounds a lot more coherent.

My personal philosophy is that for a top quality translation, you want 3 passes done by 3 different people. (This is more for high quality paid translation, but I think some of the same principles apply to free volunteer translation as well.)
Pass 1: Translation from source to target language by a native of the source language.
Pass 2: Retranslation from source to target language by a native of the target language.
Pass 3: Editing by an expert of the target language.

Naturally, all the people communicate and the result of the previous pass is used as the base for the next one. The goal of pass one is to make sure all of the subtle implications in the orginal text are included in the resulting translation. The goal of the second pass is to make sure that translation is fluent with correct grammar, etc. The goal of the third pass is to make sure that the final translation is consistent and of high quality.

So yeah, if open-source ADOM becomes a reality, a translation project might be something. I imagine sgeos will be first in line to modify all the language code to produce perfect French, Spanish, German, Russian sentences, and someone will figure out how to put other ASCII tables in...

Given input from the translators. I'm not familiar with the complete set of requirements for French, Spanish, German, or Russian sentences as I only speak English and Japanese. I do know that Korean grammar is more or less identical to Japanese grammar, and that Chinese grammar is very simple.

$ strings Adom_winbeta4.exe | wc
24391  159248  971344
$ man wc | head -n 6 | tail -n 1
wc - print newline, word, and byte counts for each file

I assume that that means there are ~160,000 words of text to translate? If so, the back of the envelope estimate is that a translation will take 80 full time days or about 640 hours (4 months full time; one third of a year). This is only the translation; it does not include externalizing strings, preparing the files to be translated, reworking the text display system, dealing with characterset issues, testing, or anything else. Note that this is for a single language translation, and each language translation should take about this long.

Note that 640 hours assumes that none of the material has already been translated. There may be some partial ADOM translations out there that have not be released in an executable. Here is a Japanese ADOM wiki that contains translations for many strings in the game:
http://adom.mine.nu/index.php?FrontPage

In any case, one man made it, so it should be possible for another man to translate it. Still, probably a lot of effort. Do you know a "professional translator" who'd be willing to do a full-time job for free? No? So it's gonna take a *bit* longer. And remember, there are quite a number of languages...

This is a defeatist attitude. Yes, I know professional level translators who have translated things they are interested in for free. Yes, they need to coordinate translations with real life so they can not put in eight hours a day. No, I don't know a Spanish or French translator, but if there is enough interest from either of those groups, I trust one is out there.

This is Japanese where some people are talking about the impossibility of making a Japanese version of ADOM because it is closed source:
http://jbbs.livedoor.jp/bbs/read.cgi...1073979503/l50

They even have a translated version of the ADOM manual:
http://www.mars.dti.ne.jp/~pillars/

If, given the source code, some Japanese are confident in their ability to deal with characterset issues, I suspect the European characterset issues would be that much easier to deal with.

Re: ASCII for French or Spanish

Still, that would look amateurish.

The point was that although it is not an ideal solution, it is an immediate solution. Given the current technical limitations of ADOM, a French or Spanish translation is possible (although Japanese is not). I don't know anything about making French or Spanish console applications, but I'm sure somebody out there does. I'm sure somebody could around the characterset issue. I'm also sure this is easy to do if you know the solution. (IIRC, kterm exists under linux for displaying kanji, so I suspect ADOM could be ported to run under kterm for Japanese, and similar things Chinese, Korean and the like.)

Huh? No, I'm not afraid of spanish ADOM players. I'm afraid that the already small group of roguelike-players might get separated by the use of different languages.

Why would a growing number of ADOM/roguelike players be a problem? The English speaking players are playing the existing English version, so localized versions would be for people who are non-English speakers. I don't see what the problem is.

RE: bilinguals

Of course there are such people, but face it, they are a minority.

This serves to weaken the case of foreigners destroying your community even more. Yes, they are a minority, but you only need one person who is willing and able to do the job and you get a translation. It is not a matter of, "how many bilinguals could do the translation", but rather a case of "is there some to do an X language translation". If enough people who speak that language are interested, the answer will be yes, and there is not a problem. If there are not enough people interested to get a translator, then it is not a problem because there is no real interest.

Besides:
1. you have a problem
2. you decide to use XML
3. now you have two problems.

This is an XML-phobic response. My point was that Google Android applications have built in localizability. The problem has been solved. Google pulled it off by externaling the string and putting them in one file. It happened to be an XML file, but the file type is neither here nor there.

Here's another problem that might or might not be related to adom: the use of male/female/plural... word forms, and articles. Not a problem in english, where you simply write "the". But french (le/la/les...) ? I'd probably get it wrong, so I don't give an example.

False. This is a problem in English. The articles "a" and "an" are used depending on whether or not the first letter of a word makes a vowel sound or not. When I was being paid to do this kind of work, we implemented an article system that went hand in hand with the message tag system the game used. Each item and monster had an article and a singular name and a plural name ("an", "apple", "apples"; "a", "goblin", "goblins"; "a", "sheep", "sheep") and messages used tags like #AN_ITEM(), #THE_ITEM(), or #X_ITEMS() to drop the correct language into the message.

"You found #X_ITEMS(p0, p1).", ITEM_APPLE, 0 -> "You found no apples."
"You found #X_ITEMS(p0, p1).", ITEM_SWORD, 1 -> "You found a sword."
"You found #X_ITEMS(p0, p1).", ITEM_APPLE, 1 -> "You found an apple."
"You found #X_ITEMS(p0, p1).", ITEM_SWORD, 3 -> "You found 3 swords."

Naturally, this is only a problem for dynamic parameters because everything else can just use a static translation.

I never said that it was impossible due to technical limitations. But still, they are there and have to be worked around.

No, but people seem to be acting like ADOM is not being localized due to insurmountable technical limitations when this is complete nonsense.

You already named one big problem: development time.

Development time is one of the only legitimate reasons I can think of for abandoning a localization effort. Naturally, if ADOM were open sourced, this problem would disappear. It seems that technical limitations, and a few other excuses are cited as the reason when development time and lack of interest is the real reason.

[Covenant]

I say why bother translating. ADOM is fine the way it is, its a game written in english. If someone wants to play it that bad, they better go learn english. Dont want to/ cant? Go play something in your language and leave us alone. ADOM has done fine its entire life being in english.

I'm English myself, so I have no problem playing ADOM in that language, but I honestly find that statement a little offensive.

EDIT: Had a moment of near-apocalyptic stupidity where I almost volunteered myself to help with translation as I was an expert on one target language (due to academic qualifications). That target language being English.

[Epythic]

Overtly xenophobic. Why would it bother you if someone translated ADOM, played it and left you alone?

Would you please stop accusing people of being xenophobic?
For the reasons I laid out to you. See above. (Heres one of them, in a short form: splitting the community into tiny fractions).

My personal philosophy is that for a top quality translation, you want 3 passes done by 3 different people. (This is more for high quality paid translation, but I think some of the same principles apply to free volunteer translation as well.)
Pass 1: Translation from source to target language by a native of the source language.
Pass 2: Retranslation from source to target language by a native of the target language.
Pass 3: Editing by an expert of the target language.

In the F/OSS world it usually works different:
- write patch.
- have patch reviewed by N other persons.
- apply patch.

Naturally, all the people communicate and the result of the previous pass is used as the base for the next one. The goal of pass one is to make sure all of the subtle implications in the orginal text are included in the resulting translation. The goal of the second pass is to make sure that translation is fluent with correct grammar, etc. The goal of the third pass is to make sure that the final translation is consistent and of high quality.

Problem is, that triples the dev time. Do you have the manpower to do this? Where?

I assume that that means there are ~160,000 words of text to translate? If so, the back of the envelope estimate is that a translation will take 80 full time days or about 640 hours (4 months full time; one third of a year). This is only the translation; it does not include externalizing strings, preparing the files to be translated, reworking the text display system, dealing with characterset issues, testing, or anything else. Note that this is for a single language translation, and each language translation should take about this long.

This is exactly my point. It would be a LOT of work. Work that only a small number of players would benefit from, work that a number of other players would not like to see.

Note that 640 hours assumes that none of the material has already been translated. There may be some partial ADOM translations out there that have not be released in an executable. Here is a Japanese ADOM wiki that contains translations for many strings in the game:
http://adom.mine.nu/index.php?FrontPage

Won't help you, doing the actual translation is not what takes most of the time. I think.

This is a defeatist attitude. Yes, I know professional level translators who have translated things they are interested in for free.

But then, roguelikes are a niche market. The translation would be basically for nothing, wouldnt it?

If, given the source code, some Japanese are confident in their ability to deal with characterset issues, I suspect the European characterset issues would be that much easier to deal with.

I guess you have no idea what you are talking about.

(IIRC, kterm exists under linux for displaying kanji, so I suspect ADOM could be ported to run under kterm for Japanese, and similar things Chinese, Korean and the like.)

kterm is just a terminal emulator. Implementing it in a program is trickier.

ASCII: 1 char = 1 byte.
Unicode: 1 char = 1 number.
Utf-8: 1 char = 1 to 2 bytes.

That just as an overview. You can't just put in a unicode string and be done with it.

Why would a growing number of ADOM/roguelike players be a problem? The English speaking players are playing the existing English version, so localized versions would be for people who are non-English speakers. I don't see what the problem is.

The problem is that currently, players are forced to play the english version, but having different (localized) versions would allow the players to get away from the english version. Bad, bad.

This serves to weaken the case of foreigners destroying your community even more. Yes, they are a minority, but you only need one person who is willing and able to do the job and you get a translation. It is not a matter of, "how many bilinguals could do the translation", but rather a case of "is there some to do an X language translation". If enough people who speak that language are interested, the answer will be yes, and there is not a problem. If there are not enough people interested to get a translator, then it is not a problem because there is no real interest.

First of all, you don't need to be a bilingual person to be able to translate something. I think I'd be capable of doing an English->German translation, given enough time.
But then: find someone who'd do it first. It would be a lot of work, for basically nothing. Time that I could spend elsewhere. But I repeat myself.

This is an XML-phobic response. My point was that Google Android applications have built in localizability. The problem has been solved. Google pulled it off by externaling the string and putting them in one file. It happened to be an XML file, but the file type is neither here nor there.

"-phobic" is your favourite suffix, aint it?
Actually that was intended as some kind of joke, to reduce the tension. Obviously my attempt failed.
[Although I really dislike XML]

False. This is a problem in English. The articles "a" and "an" are used depending on whether or not the first letter of a word makes a vowel sound or not. When I was being paid to do this kind of work, we implemented an article system that went hand in hand with the message tag system the game used. Each item and monster had an article and a singular name and a plural name ("an", "apple", "apples"; "a", "goblin", "goblins"; "a", "sheep", "sheep") and messages used tags like #AN_ITEM(), #THE_ITEM(), or #X_ITEMS() to drop the correct language into the message.

Wonderful. </sarcasm>

No, but people seem to be acting like ADOM is not being localized due to insurmountable technical limitations when this is complete nonsense.

Technical and social limitations. I wont repeat myself again here.

Naturally, if ADOM were open sourced, this problem would disappear.

Open source != free developers.

And heres another point that might be a problem: what does TB think about this? I could imagine that he wouldn't like it...

[Sradac]

not xenophobic at all, but people are making such a huge deal about translating adom, all im saying is it seems to be a heck of a lot more of a job than its worth, there are plenty of non-english roguelikes out there to satisfy people that dont speak/read it. Im not complaining that those ones arent english and demanding they be translated, so no one else should do the same towards adom.

[Covenant]

Im not complaining that those ones arent english and demanding they be translated, so no one else should do the same towards adom.

(Non-English Speaker) Okwonko: 'Wow, this Shakespeare seems like a pretty cool guy (he kills Romeo and isn't afraid of anything) - I'd like to read his works. I'll see if there's a version available in the language of Umuofia... Oh, there isn't. I hope someone translates it soon, I deserve to be able to read them just as much as anyone else.'

Sradac: 'Don't make such a big deal out of it! Just go read some Marlowe plays, it's all the same. It's not worth the effort of translating it - be happy with what you've got.

Look at me, you don't see me complaining about not being able to read... Well, there isn't much I can't read because lucky for me the rest of the world caters to English speakers, so I have a much wider selection available than someone like you. But I wouldn't complain. So don't you complain!

If you really want to read it, go learn the language yourself. It's not like I've ever read or watched anything that's been translated from its original language for my benefit. Japanese films/games? Hell no, I don't play those. The Bible? Never heard of it! Games developed on a different OS than my own but then ported to the one I use? THEY DON'T EXIST!'

Look, the point is, Shakespeare's works have done great all this time not being available in your language - so go away. We don't need your kind here.'

Maybe I'm extrapolating incorrectly, but in my opinion that's entirely how you've portrayed yourself.

Seriously Sradac, I find your comments abhorrent, and entirely xenophobic.

[Epythic]

Seriously Sradac, I find your comments abhorrent, and entirely xenophobic.

I side with Sradac here.

But then, this is becoming a full-blown flamewar. I don't like that. I will not be doing another post in this thread [I hope], and my final comment shall be that I am against and will not support any ADOM translation project, but what others do is not my call.

Note to self: STOP.