Multiple translators v solo

Originally Posted by Cosign

I'm not saying we write crap.

Of course you are not saying you write crap. However, that doesn't mean that you are not writing crap.

[drmchsr0]

Originally Posted by Asceai

Whatever people might say about the 'traditional' model employed by mirror moon, now yandere, amaterasu etc. is that at least they finish and polish their projects before moving on to the next, rather than letting them linger around in a finished-but-not-quite-finished state

There's nothing wrong with the traditional model. All it needs are dedicated people and the right skills.

Now, the traditional model could be improved on, like letting 2-3 translators work in tandem to translate one game (and of course, 2-3 hackers to hack said game, but that might be overkill), which would cover redundancy and even include a mentoring aspect (should it be needed but this is besides the point), but as a whole, it does work, somewhat.

However, the project leader has to pick the right people (skills, personality, whether they'll fit in, for example) for the job.

Originally Posted by Cosign

See, I think that is a misconception. We are so tight-knit and focused on quality quality quality all the time we get tunnel-vision. We start to think that everyone who reads this wants nothing but perfect translations. I bet hundreds of people are discovering and rediscovering visual novels everyday. As long as the English is readable, Casuals don't care.

What I was referring to is the sad, sad state of fansubbing today. There are groups that churn out quality stuff, but on the whole, the casual fan cares not about quality as long as they can somewhat understand it. Heck, they don't even care if the translation is utter crap as long as it's in English. Which leads to subbers going for faster and faster release times until quality is left in the dust.

Originally Posted by Cosign

So within 13 months, they were able to translate 99.999% of 100432 lines of text (or 3.16 Mb). While I won't say it is the most efficient beast I've ever seen, I would still commend it for being an exceptional feat.

Yes. The translation quality was not the most consistent. But all I am asking is, 'Is that really what fans want?'.

From what I tried, the Clannad stuff was borderline unreadable in many places. That's not even accounting for translation accuracy.

While the project moved quickly at first, I think the inability to finish speaks for itself. Obviously, people deemed it impossible to salvage, at least not without someone redoing most of it from scratch.

Nevertheless, I think it was an important step in setting the stage for future translation efforts. People saw why it failed and have launched collaborative projects that take a middle ground. Doing something like Clannad again would be idiocy.

[jyuichi]

Since it was brought up, the Chaos;Head patch was pretty damn terrible. At least they admitted it was alpha but I was glad I could understand Japanese for quite a bit of it. Usually I lend good titles to my friends but the patch was too painful...

It's probably obvious but I am completely opposed to this idea that "any translation is a translation worth playing".

[Cosign]

Originally Posted by jyuichi

It's probably obvious but I am completely opposed to this idea that "any translation is a translation worth playing".

And I agree.

[Moogy]

Originally Posted by Asceai

Saya no Uta, that came infinitely close to completion but never got a final high-quality release

Would just like to clear this up. In January 2009, I released a v0.9 of the patch. It was complete and playable; there were no actual textual issues beyond some typos. There were simply some technical issues to be sorted out (translating some dialogues and fixing the load times, iirc).

In February, I released v1.0, the final version, which fixed the aforementioned technical errors as well as all of the reported typos. So, yes, Saya did get a "finished" release.

Kikokugai's release is also considered finished, though it's a bit of a ghetto "patch," since we forced it into Saya's engine and as such you cannot view any of the game extras.

Hanachirasu is currently in a completely playable state, save for an issue with transparency in the credits and an untranslated rape scene, both of which will be worked out before the final release. Unlike Kikokugai, we actually went through with hacking the Saya word-wrapping into Hanachirasu's own engine, so all of the game's functionality is retained.

...Actually, we didn't translate the minigame in Hanachirasu, so I guess there is that. I still consider it a "finished" release though.

Muramasa should be much the same whenever it eventually gets finished; our approach to N2System has come a long way since the days of C;H. Looks like R11's getting its technical problems cleaned up for a final release at the moment as well. I'm not arguing that there needn't be more polish on some TLWiki releases, mind you, I'd just like to avoid misunderstandings.

Recently one of my favourite manga translation groups, that really worked hard to give a quality release, dropped a certain manga and the new group that has picked it up is awful. Their releases are faster but they full of typos, commas in random places, and worst of all: "the 2 of us".

I am definitely happier to wait longer for a quality product; if that means only having one translator, so be it.

However it's important to remember, consistency does not always mean quality. Two good translators will still translate well. In fact, if the translators are less skilled then I believe having multiple translators may work to the groups favour.

But one translator is probably the best way, and I'm glad so many groups are using this method.