Poll on Final Distribution Options for S&W

[Krikit]

Okay. The eBook is awesome, and I'll just have to wait for more of it to be released, but after we get some translations out.

*need more* ^_^

[TheGiftedMonkey]

Like I said focus more on getting it to work good and translating. The text size should be good enough for the majority of the readers so I wouldn't put much focus on it.

[barbsicle]

Like I said focus more on getting it to work good and translating. The text size should be good enough for the majority of the readers so I wouldn't put much focus on it.

yeah TGM, most errors are cleared and that would translate (no pun intended) to it working pretty much good. now my priority is translation, and i'll set aside a bit of time each week to improve on this. i just went into mad concentration on the day seaghyn posted. now that it can be used as good comparison against the pft, i'm much more slack on the other issues. in fact, i might even need help with them. lol.

yeah that's about it.

Barbs.

[Smidge204]

Questions:

At what point is the work considered "done" and ready to be "released" ?
What action should be taken if there are edits or other changes after "release" ?

There has been an unwritten policy that Baka-Tsuki does not actually release anything because of continuing edits. As of now there is no criteria for a project to be "done" (other than licensing) and any released/distributed materials would become out of date somewhat quickly. Even the Haruhi series, which has been "done" for many months, has recently seen revisions with translation errors fixed.
=Smidge=

[Krikit]

Hmm...good point. I guess what could be done then is a forum opened in the S&W project for the links to download the eBook, or PDF, or whatever might be released, so that new updates could be continuously inputted. Then, whenever something gets fixed or changed, just edit the original post and put in the newest release. The Project will always be on the wiki, so it isn't really like the eBook or PDF needs to be downloaded, it's just a clean, nice asthetic layout of the last revision of the S&W page, for whoever wants a more entertaining read...at least that's what I think.

[Guest]

The finalized version could be made into an ebook or wiki text and then printed to as a pdf. This would work with Adobe Professional or for free with CutePdf. By the way, I'd like to thank the relevant parties involved with these translations. As most with people, I wish that I could help with the translation, but do not have the necessary skill. One thing that might be interesting would be for the translator(s) to create an audio book version that non-Japanese readers could use to transcribe into text. The process would probably be faster and more distributed.

[barbsicle]

One thing that might be interesting would be for the translator(s) to create an audio book version that non-Japanese readers could use to transcribe into text. The process would probably be faster and more distributed.

well either i didn't get you there or you're making me confused. haha.

we, as translators, do the work through text. this is not anime, we're not doing fansubs. by making an audio book version, we'll need people to dub. and transcribing the text is just going back to square 1 no? not to mention i already had so much trouble with the ebook. audio book......can't think of it even in the long road. haha.

well at least that's what i thought of what you were trying to imply. correct me if i'm wrong. heh.


Barbs.

[Pavick]

Perhaps I should have said audio recording. What I meant by audiobook wasn't anything that would be in final format. It seems that it would be faster for a translator to read the untranslated text out load in English while recording with a microphone. It wouldn't need to sound good, but explanations could be added while reading out load so that the recording could be transcribed and then edited. There would be 3 phases instead of 2, but the first most important phase (the translating) could be done with less effort/more quickly.

[Krikit]

So, let me see if I get where your coming from.

The Japanese (or in this case Chinese ^_^) Translator would look at the chinese/japanese text, and then translate it into english into the microphone. In other words, take a minute pause or so, and then say the sentence in english. The continue for each sentence. Pause to read silently it in Chinese/Japanese, and then say out loud the translation?

Then after a section was done, the translator would only have to listen to the audio recording and type the english text into a word editor? I think I can kind of understand the concept, but do you think it would be faster to do it this way? Or am I missing something? I haven't ever tried it, so it might work out to be faster, but then again I can't quite read Japanese/Chinese yet...I'm just the editor =).

Again though, was this what you meant? Or would it be the translator would read the translation into an audio recorder, then send that recording to me, the editor, and I would listen to it and copy the english into a word editor??

[Pavick]

The trick is that the translator only would need to work on the first phase (translating). If you have a dedicated transcriber (the one that writes the text based on the audio recording), then the translator can spend more of his/her effort translating instead of writing and coming up with just the 'right' way to phrase things (as long as explanations are given the transcriber could spend the time to think about how to write it best). Then everyone including the original translator (if he/she so desired) could help edit the text. I'm sure that there are programs that let you record from your microphone straight to mp3 format. I'll look for a good one (hopefully free/shareware) later today. The resulting audio file could be send via instant messenger or torrent. The whole process of having edited text posted on the wiki would probably take more human hours this way, but should take less translating hours.

BTW: Audacity seems to be a good candidate for recording: http://audacity.sourceforge.net/

[Krikit]

Hmmm....hey barb, you want to try it, even if only for a page of text? Just to see how quickly you can translate a page? And then I'll work through editing it and writing it, with whatever notes you put in?

If it doesn't seem feasable after a page trial, we can drop it (sorry Pavik) ^_^, but if it does come out to be a workable way to translate faster......^_^

give it a shot?

[barbsicle]

Perhaps I should have said audio recording. What I meant by audiobook wasn't anything that would be in final format. It seems that it would be faster for a translator to read the untranslated text out load in English while recording with a microphone. It wouldn't need to sound good, but explanations could be added while reading out load so that the recording could be transcribed and then edited. There would be 3 phases instead of 2, but the first most important phase (the translating) could be done with less effort/more quickly.

The trick is that the translator only would need to work on the first phase (translating). If you have a dedicated transcriber (the one that writes the text based on the audio recording), then the translator can spend more of his/her effort translating instead of writing and coming up with just the 'right' way to phrase things (as long as explanations are given the transcriber could spend the time to think about how to write it best). Then everyone including the original translator (if he/she so desired) could help edit the text. I'm sure that there are programs that let you record from your microphone straight to mp3 format. I'll look for a good one (hopefully free/shareware) later today. The resulting audio file could be send via instant messenger or torrent. The whole process of having edited text posted on the wiki would probably take more human hours this way, but should take less translating hours.

BTW: Audacity seems to be a good candidate for recording: http://audacity.sourceforge.net/

well after a long talk with seaghyn, i have a few further questions (and also explanations) to you.

First and foremost, I translate by "ear". Yes, I am an audio-orientated person. So when I translate; 1) I read the untranslated text out loud/it plays in my head, 2) I translate the text and again I read the translated text out loud/it plays in my head and 3) I decide whether it sounds grammatically/structurally correct. So, if I'm not wrong, what you have posted up there about translator records --> transcriber+whoever else writes the text out while the translator prepares the next few lines has been, as stated by me above, already done. Essentially I am "recording" my own voice in my head, translating, and transcribing. The only thing that's different is I'm not translating the 2nd line as I'm figuring the phrasing for the first (of course I'm not superhuman) and to me that appears to be what you're trying to target to save time.

Secondly, a new problem could come in while transcribing. Chinese, like Japanese, has to undergo almost a complete rephrasing to make it sound correct as English. Therefore, a transcriber who does not understand the language in translation will rephrase according to English grammar rules (or whatever they're called). This can, most importantly, distort sentence meaning. And when the translator looks at the edited text he will go "this isn't what it means" and go in one big circle just to end up rephrasing the line himself. Simply, to rephrase right, one has to understand the language itself and not just go by English knowledge. You also mentioned that I could try to explain what I want it to sound like and such, but in my opinion, that kind of throws away whatever time we were trying to save.

Thirdly, is timing. Let's say if(and just if) my transcriber lives in America. That would put an approximately 12 hour time gap between us. So basically when he's sleeping I'm talking and when he's transcribing I'm sleeping. Although really this doesn't seem of any bad on the outside, I assume we will carry this out assuming there will be a constant stream of text being output by the translator. But on the other hand, I could have had a busy day and he would be sitting down there done with the 30-so little lines I put out for the day and waiting for the next few to come. On the other hand, I could be putting out so much text that even if he does commit some time at night to transcribe, the work would just pile up, and at the end of the chapter I'd probably just go back and transcribe my own text anyway.

I told seaghyn that by using a voice comm program like Ventrilo(http://www.ventrilo.com/), the translator speaks while the transcriber does the work spontaneously. This requires both to be online at the same time (which re-iterates point 3), but it does away with 2 steps, which are; a)the need to record, and b)to distribute the audio file. We're going to try it later and see how it works(after my lunch that is).

So yeah, that's it. I'm just putting these questions forth to you so I can determine, by your answer, if it's still feasible or not.


Barbs.

[Pavick]

The reason why I made this suggestion was because it appears that reducing the translator's workload would be beneficial to the translator and the overall project itself. If it were decided that there should be a dedicated transcriber(s), then two possible ways for the translator to save time would be:

a) Read and record the text spoken out loud in English. If a line or phrase has an obvious translation, then the translator could say that. Otherwise, the translator could just explain what should be conveyed. Your second point would be relevant here. Even after grammatical restructuring, one still needs to come up with precise phrasing while writing text that when edited would be in final form. As long as the grammar is restructured in an unambiguous way and the overall content is preserved, the context of the story should allow the transcriber to achieve good quality.

b) The translator could do essentially the same thing as in (a) while using voice chat with the transcriber. In order to save the translator’s time, it could be a good idea to let the translator read out loud while the transcriber asks for clarifications as needed. Instead of having the translator wait for the transcriber to type, it might be a good idea if the transcriber used Audacity with the setting 'What you hear' to record the discussion as a reference.

In regards to your first point, the translator should save time in both cases because his/her concentration wouldn't be divided between translating and typing. Also, less time would be spent on how things should be phrased precisely. As long as the transcriber understands the overall content of each line, the end quality should be pretty good.

I'm not sure which method would save more of the translator's time, but method (b) would prevent the translator from getting ahead of the transcriber, which may be desirable to satisfy your third point. I don't think it would necessarily be a negative thing if the translator got ahead of the transcriber. To maintain consistent style, ideally only one transcriber should be used. However, it should be significantly easier to recruit more transcribers as necessary. They could divide the work among sections and/or chapters. A scenario like this would allow a faster release time in the long run. Another thing to consider is that even if the translator isn’t able to translate for some period of time, the transcriber could still be working as long as there was material left for him/her to transcribe.

Good luck! Regardless of you decide, thanks for spending the time to translate. I just think it would be nice to save you some time.

[barbsicle]

The reason why I made this suggestion was because it appears that reducing the translator's workload would be beneficial to the translator and the overall project itself. If it were decided that there should be a dedicated transcriber(s), then two possible ways for the translator to save time would be:

a) Read and record the text spoken out loud in English. If a line or phrase has an obvious translation, then the translator could say that. Otherwise, the translator could just explain what should be conveyed. Your second point would be relevant here. Even after grammatical restructuring, one still needs to come up with precise phrasing while writing text that when edited would be in final form. As long as the grammar is restructured in an unambiguous way and the overall content is preserved, the context of the story should allow the transcriber to achieve good quality.

b) The translator could do essentially the same thing as in (a) while using voice chat with the transcriber. In order to save the translator’s time, it could be a good idea to let the translator read out loud while the transcriber asks for clarifications as needed. Instead of having the translator wait for the transcriber to type, it might be a good idea if the transcriber used Audacity with the setting 'What you hear' to record the discussion as a reference.

In regards to your first point, the translator should save time in both cases because his/her concentration wouldn't be divided between translating and typing. Also, less time would be spent on how things should be phrased precisely. As long as the transcriber understands the overall content of each line, the end quality should be pretty good.

I'm not sure which method would save more of the translator's time, but method (b) would prevent the translator from getting ahead of the transcriber, which may be desirable to satisfy your third point. I don't think it would necessarily be a negative thing if the translator got ahead of the transcriber. To maintain consistent style, ideally only one transcriber should be used. However, it should be significantly easier to recruit more transcribers as necessary. They could divide the work among sections and/or chapters. A scenario like this would allow a faster release time in the long run. Another thing to consider is that even if the translator isn’t able to translate for some period of time, the transcriber could still be working as long as there was material left for him/her to transcribe.

Good luck! Regardless of you decide, thanks for spending the time to translate. I just think it would be nice to save you some time.

well i have to say i almost agree with most of your points. however, this does not solve most of the questions i put forth to you.

it's true that the main purpose of this is to save translation time and the transcriber acts as the translator's "typewriter". personally, i take quality over speed. so here i shall state some conclusions me and seaghyn arrived to after finishing 3 lines (yes it's little, but do read on).

in all, 3 lines took about 40 minutes. These 3 lines were a very long one, a medium sized one and a short one. needless to say, we shouldn't base everything on these 3 lines. but it has brought about some doubt to its efficiency.

1) Conflict
This is the main thing that is using up whatever time we're trying to save as well as manpower, with a multiplier(as there are 2 or more thinking). I do agree with you that the transcriber must be dedicated, but after working along for a bit with seaghyn, I realize half of our discussion/interrupting each other is because of conflicts. Like I've said earlier on, rephrasing is very different if you're going according to Chinese/Japanese or if you're going according to English. From Chinese/Japanese point of view, you take a sentence, and restructure it almost completely, and leave it at that. For English, though, people not only rephrase to make it sound right, they also omit certain phrases that may seem redundant because they sort of don't fit into the phrase, or perhaps change a word. This, was like I said earlier, distorting meaning, and I've stopped seaghyn many a time during those 3 lines to clarify and request for him to add this or that word (no offense, buddy).
Also, when I hear seaghyn read his sentence out to me, I feel kind of weird (again, no offense). This is due to the fact that everyone has their own style of writing. I write my translation out in my way, he writes it out in his way. So to confirm his transcription was accurate, I have to go through the sentence again, simply because I'm not used to the style. I understand this will change with time, so yeah.
Thirdly is a problem that I predict will happen if we use method (a). Simply, I record and send, the transcriber transcribes this chunk of text. The next day after audio-translating I have some spare time, so I go look at what my transcriber has done and I realize line xx does not match because my transcriber has omitted something. So we have to get into a discussion and time is, once again lost as my transcriber will be writing on the forum/im instead of transcribing. Unless, of course, we leave all the discussion to the end and in one span of time, correct everything at once. But here's where, if we followed conventional translating, I would be on the next chapter as seaghyn edits. So really, we can't get very much ahead of time. In the end we would still be discussing about the conflicts. After all transcriber vs. transcriber (if there are 2) discussion doesn't really help. The translator understands the core text so he would be the one to consult. Which, after this long roundabout statement, just ends up with the translator going back to sort things out and saves no time as a result.

2) Context
After translating lines with seaghyn, he and I agreed that to transcribe, one would need to understand the context well. Of course, context is in which all the conflicts arise. One way a transcriber could understand the context better is by watching the anime or reading the previous chapter. But, the anime is slightly modified (especially with Chloe who plays a big role for the Church) and some parts of the book are not present in the anime. Either way, in my opinion the translator probably knows the context best (as we're going through the thoughts and feelings of each character, main or not, as they perform their role). So, once again in the roundabout way, the translator will have to go through each sentence again to verify the correct context, example being importance or precedence of subject. Sometimes to get the phrasing right, the transcriber might twist the structure such that xx is more important than yy, which results in the sentence not implying what it's trying to mean. Therefore, the translator has to go through it all over again, and yeah, you get the roundabout.

3) My Side
Yes, and this is the problem generated by me. When I was reading out the translated text, it was already rephrased. Seaghyn rephrased a few and that didn't sway any sentence meaning so it was generally fine. But there we have the problem. Basically half the transcriber's work is done, and I'm speaking out what I was just about to type. It's kind of hard to speak on purpose in broken English, so my mind automatically rephrases and I speak. The precise phrasing thing, if I'm not wrong, should be what's done in editing already (otherwise besides spell checks and major error spotting, the editors would have nothing to do either). So basically in the end I'm still doing all the work and seaghyn is doing his work (which is editing) on the spot. Which once again, brings us back to square one.

as for the rest, they're minor problems (like seaghyn screwing up and asking me on im to TYPE what i was trying to say) and don't really affect the pace. so in conclusion, if we do continue with this, i see it as i'm pushing the time i use for typing to seaghyn and in place i translate more lines, which by the end of the chapter, we get into one big discussion and settle everything at once. but even after i think hard on it, the time we could save/the faster we could put out one chapter looks questionable. and, as always, quality overrides speed. i'm sure us translators are more comfortable seeing what we type as well (rather than someone else's style, once again reiterating) and using our style, we can make faster discussions during editing phases and hurry on with our next lines of translation.

TL;DR: i respect your suggestion and thank you for it, but at the moment (unless you hit me with something i didn't realize even up till now) i don't really see the amount of time it could save to being very significant. nevertheless, the trial lines were a half-success. so for me (judgment is away in korea btw) i'll continue with my normal course of typing whatever i translate. nevertheless thanks for your ideas and i hope this one doesn't die out just yet.

P.S. this thread was supposed to be on final distribution, not the translating process. so yeah, we kinda got off-point. heh.


Barbs.

[Krikit]

Sorry Barb, going to clog the "FINAL DISTRIBUTION POST" one last time ^_^

1) Conflict
This is the main thing that is using up whatever time we're trying to save as well as manpower, with a multiplier(as there are 2 or more thinking). I do agree with you that the transcriber must be dedicated, but after working along for a bit with seaghyn, I realize half of our discussion/interrupting each other is because of conflicts. Like I've said earlier on, rephrasing is very different if you're going according to Chinese/Japanese or if you're going according to English. From Chinese/Japanese point of view, you take a sentence, and restructure it almost completely, and leave it at that. For English, though, people not only rephrase to make it sound right, they also omit certain phrases that may seem redundant because they sort of don't fit into the phrase, or perhaps change a word. This, was like I said earlier, distorting meaning, and I've stopped seaghyn many a time during those 3 lines to clarify and request for him to add this or that word (no offense, buddy).
Also, when I hear seaghyn read his sentence out to me, I feel kind of weird (again, no offense). This is due to the fact that everyone has their own style of writing. I write my translation out in my way, he writes it out in his way. So to confirm his transcription was accurate, I have to go through the sentence again, simply because I'm not used to the style. I understand this will change with time, so yeah.
Thirdly is a problem that I predict will happen if we use method (a). Simply, I record and send, the transcriber transcribes this chunk of text. The next day after audio-translating I have some spare time, so I go look at what my transcriber has done and I realize line xx does not match because my transcriber has omitted something. So we have to get into a discussion and time is, once again lost as my transcriber will be writing on the forum/im instead of transcribing. Unless, of course, we leave all the discussion to the end and in one span of time, correct everything at once. But here's where, if we followed conventional translating, I would be on the next chapter as seaghyn edits. So really, we can't get very much ahead of time. In the end we would still be discussing about the conflicts. After all transcriber vs. transcriber (if there are 2) discussion doesn't really help. The translator understands the core text so he would be the one to consult. Which, after this long roundabout statement, just ends up with the translator going back to sort things out and saves no time as a result.

Yeah, this is pretty much how conflict went. For one there was background noise where I use the internet, coupled with a small bit of static from Barb's CPU, it makes words hard to hear, and I had to end up asking Barb to repeat this or repeat that, then when he wants to say something I have something to ask, and it's "So you/What do...oh, sorry, go ahead...okay, it was ....." like that. Then what I wrote was changed a bit because I do have my own style of writing, how I think something should sound, and I don't have the original text. Kind of a flaw on my part. The communication takes time, and Barbs right, when he speaks it, it's pretty much all already in his head. Transcribing isn't as simple as barb reading word for word the chinese text in english while i transcribe. the order of words, structure, etc makes it so he has to figure out what it means, and by that time he's got the way he wants it to sound already.

2) Context
After translating lines with seaghyn, he and I agreed that to transcribe, one would need to understand the context well. Of course, context is in which all the conflicts arise. One way a transcriber could understand the context better is by watching the anime or reading the previous chapter. But, the anime is slightly modified (especially with Chloe who plays a big role for the Church) and some parts of the book are not present in the anime. Either way, in my opinion the translator probably knows the context best (as we're going through the thoughts and feelings of each character, main or not, as they perform their role). So, once again in the roundabout way, the translator will have to go through each sentence again to verify the correct context, example being importance or precedence of subject. Sometimes to get the phrasing right, the transcriber might twist the structure such that xx is more important than yy, which results in the sentence not implying what it's trying to mean. Therefore, the translator has to go through it all over again, and yeah, you get the roundabout.

Yes. Currently I am unable to read the Japanese text. If I was able to read the Japanese text, and had read the novel, then decided to translate, it might at that point be faster to transcribe what barb says...maybe. But with all the interjections, questions, background noise, etc, a considerable amount of time is taken up in Barb saying something, and then me repeating my "fixed" version, to make sure how it sounds is still kept in context. Sadly time isn't that well saved. This is good in theory, but it would work best under these conditions:
-Barb was reading an English book, one that I didn't have.
-Barb and myself had good mics, no background noise, and clear speech
-I was fast at typing (which I am ^_^)
-So Barb would read the book to me, and I would quickly write that text, to have a "copy" on the computer.

That's really the way this would work the best, and be faster than Barb reading the english text, typing it in, forgetting where he was in the book, and scanning, reading the next part, typing, etc.

3) My Side
as for the rest, they're minor problems (like seaghyn screwing up and asking me on im to TYPE what i was trying to say)
TL;DR: i respect your suggestion and thank you for it, but at the moment (unless you hit me with something i didn't realize even up till now) i don't really see the amount of time it could save to being very significant. nevertheless, the trial lines were a half-success. so for me (judgment is away in korea btw) i'll continue with my normal course of typing whatever i translate. nevertheless thanks for your ideas and i hope this one doesn't die out just yet.
P.S. this thread was supposed to be on final distribution, not the translating process. so yeah, we kinda got off-point. heh.

Barbs.

Haha, sorry for that...but seriously, there were these screaming kids that wanted their mom, I just couldn't hear you....plus it didn't help i was a bit confused on some of the words...haha, my bad my bad....wait til I get internet in my apartment, and it's quieter =)

Yeah, I think this is probably not going to work out. But for the record Pav, we did give it a serious effort, it's just under the conditions we have, it doesn't save time, right now we actually lost time....40 minutes for 3 lines =)

Okay, so that's that. Now back to the final distribution options? PDF, eBook, then another one.

==================================================================================================================================

What is the general consensus out there of an audio book? Pav did make me start thinking, and I'm wondering if I voiced an Audio Book, what you would think?
Before you say "No! Your voice SUCKS!" ^_^, there is also the possibility of adding in the audio to the text on the eBook, and you can click an audio button to listen to the text. Check out Deckard Cains History of Sanctuary for the new Diablo III site. Sort of like that. Any thing you guys out there have to input?

[EDIT by Barbs so as not to clog it another time. http://www.blizzard.com/diablo3/world/lore/journal.xml is what we're aiming to achieve with a slight hope that the audio will be implemented too, if it's in majority favor.]

-Krikit/seaghyn