Rating groups an the quality of their work

[Bainhardt]

The point would be to punish groups doing a horrible job.

Well, as you have come forward again to explain yourself more plainly, I will say equally plainly: I do not agree whatsoever that this is a good reason for what you have suggested.

Also the current system encourages being the first and the fastest

I still don't understand what exactly you mean when you say this. There are some groups that can release excellent work every day working on daily manga. There are other groups (like myself) who release their work totally erratically but with consistent quality. And there are likely some groups who jump on projects a bit faster than they ought to and release sub-par work because they didn't work carefully. What majority is it you are seeing, and where are you seeing it, that you have such this strong an impression there's predominantly crap scanlation being released?

Admittedly, I don't read a ridiculous amount of manga, but I do check things from the front page going up almost every day and rarely do I see something so bad I think "these people shouldn't even bother." I can't even think of any recent examples, but I will accept that it's possible I simply haven't seen them.

Anyway, I'll close by saying again that I can't imagine in what world this kind of review system should take priority over an education or tutorials section of the website where scanlation groups can make lessons for others to reference, be they using images + text or video formats. If you believe reviews would encourage groups to strive to improve because you assume all poor scanlators want to improve, then just skip the review part entirely and provide resources for doing so.

edit: If the discussion is shifting to focus more on retooling the Group Likes system or something simple along those lines, that is something I could see myself supporting.

 

[Teasday]

Frankly, the groups that want to improve most likely read the comments they get. I see the utility in a more direct and simple method of communicating quality feedback (especially since it would probably encourage it), but so far chapter/group comment sections have done a decent enough job at it.

One potential method of making the group likes more visible would be a little box above a manga's chapter list that includes all the groups that have released chapters for that manga (and in the selected languages) as well as display their like count and the like button. Appending the like count to each instance of the group names inside the chapter lists would look insanely cluttered, so that's not really an option. However, doing that is a possibility inside the reader.

I haven't made any final decisions, so comments and further suggestions about this are very welcome.

 

[Tikibo]

I'm not going to bother trying to ask what kind of "series of privileges " you mention because I have a feeling that answer you are going to give is going to be one of those "Give me your work for free. You will gain fame" answers you get if you even touch the art market.

I think you didn't understand what I said. I mean, any reader of a FREE scanlation cannot be a consumer, and unless you give ANY arguments (even on PM) without magical "series of privileges" I won't change my mind and you probably won't change yours. I totally agree with your next point: by giving a donation you don't become consumer because you still don't buy those releases. You are only supporting their work, and that's fine, but if you give them money for trash quality...


About better visibility of groups buttons, they need to have more distinct colors/opacity. It's definitely not obvious what they do. I also think that a simple tutorial for people who sign up could help readers to find useful but "hidden" features. Or putting big "help" button with useful functions of reader etc. Even if it just links to manga tutorial (which is a bit old btw).

 

[Zephyrus]

@Teasday

Frankly, the groups that want to improve most likely read the comments they get. I see the utility in a more direct and simple method of communicating quality feedback (especially since it would probably encourage it), but so far chapter/group comment sections have done a decent enough job at it.

One potential method of making the group likes more visible would be a little box above a manga's chapter list that includes all the groups that have released chapters for that manga (and in the selected languages) as well as display their like count and the like button. Appending the like count to each instance of the group names inside the chapter lists would look insanely cluttered, so that's not really an option. However, doing that is a possibility inside the reader.

I haven't made any final decisions, so comments and further suggestions about this are very welcome.

After reading the suggestions and debate in this thread, I've personally decided that there's not really any way that we can implement a scoring system that people actually care about or won't abuse.

I completely agree with Tea that the combination of Group Likes and Group page comments do a pretty good job of reflecting how well a group is doing. Tinkering with the Group Like feature is probably going to be the best path moving forward.

And just to make a general comment: I don't believe in punishing groups unless they are releasing content in bad faith, i.e. troll releases. At the end of the day scanlators, for the most part with those few exceptions that are looking to make a quick buck, scanlate for free. If a group is open to constructive criticism and genuinely wish to improve, then they'll take reader feedback into consideration. If they don't, the only thing that punishment will do is probably drive them away from the scanlation scene entirely. If you want professional level scanlations, you'll need to support the official English release as soon as it's available in your area.

 

[DANDAN_THE_DANDAN]

^^^

And yeah, volunteers can get sued for messing up so badly but does scanlators ever do such a bad job that a punishment to a degree is necessary?

 

[littleoni]

@Zephyrus
Well, actually a way to solve all of my problems would be to hide all of the releases of a certain group (not per work, but completely). It would be a personal decision that cannnot be imposed on others, abusing it would be pretty much impossible, harassing a group would be impossible, etc. Then again, a system like that is not necessary, it would just help a lot to a person who likes to browse a bit off the beaten road series (because longer and more popular works usually get picked up by a serious group at some point).

@DANDAN_THE_DANDAN
I don't have a good example at hand, I don't usually follow the cases where it gets criminal. Most translations are not criminal level bad, but compared to a person doing a really good job someone doing a bad one gets the same treatment and recognition. Over time, why would anyone bother to actually do a good job? If one person can do a crappy job in the same amount of time than a team takes to do a decent one, and then get the same reception, why would you bother putting together a team? My main complaint is not the bad quality (which is also an important issue), but the fact that the current system is encouraging it.

@Bainhardt
After spinning it through I don't think the rating idea is any good anymore. While it's ideal form would be a solution to all problems, the ideal form is unreachable. I had originally thought an imperfect system would make do, but after thinking it throught the mentioned alternative would make much more sense, and would probably be a lot simpler.
I have taken a look at the kind of works you take on, and correct me if I'm wrong, but you are not "competing" with anyone else to do them. In that case, sure, there is no "discouragement" to do things the right way and take your time. As soon as someone intending to release the fastest crap possible shows up, your releases would stop mattering at all, people would read the first one showing up in their follows, they would maybe complain, maybe not, and would either keep following the bad quality releases or drop the series because of the bad quality releases: we don't have the option to ignore them, they clog our follow list, and there is nothing we can do about it. We also don't have a fast way to differentiate the releases of each group, so when the slower but better one does release their version, it's often lost in the sea of trash. In this situation, the slower group's only actual option is to drop the translation and move somewhere else where they don't have to compete, so the worse one "wins", and that's my main concern.

 

[DANDAN_THE_DANDAN]

@littleoni

Over time, why would anyone bother to actually do a good job?

Um, actual passion and pure diligence?

The rest of the argument is fair game but you're forgetting that most scanlators that stay are the ones with passion for their hobby/job. Otherwise, a simple ban would do.

 

[Teasday]

_{@littleoni posted:}

Over time, why would anyone bother to actually do a good job?

Pride in yourself and your work, for one. I honestly care about the quality of my work on a personal level a bit more than I'm interested in other people's opinions about it, so if something doesn't meet my own standards I'm not going to put it out - or if I do, I'll regret it later.

I realize some people care about about the quantity than the quality but I can't stand it

 

[Bainhardt]

@littleoni Thank you for explaining yourself in a way I can understand.
You'd be correct in that I am not competing with anyone else on any of the projects I pursue (in fact, I chose most of them based on that). And while I hold myself to what I would describe as "moderate" levels of quality, I have always accepted that it is possible someone else may come along and do a better job. A lot of the time I think I would actually welcome that, because it would make me feel I could then move on to some other project that is receiving no love, lol

Now, in the opposite direction we have what you've described, which would be if someone releases work faster than I do but at significantly worse quality. You've said something that I totally get, and I actually agree points to a niche, but very real, problem some users may have:

we don't have the option to ignore them, they clog our follow list, and there is nothing we can do about it.

We can follow manga series and we can follow scanlation groups, but we cannot exclude anything beyond specific language releases. I agree that if some people could get use out of being able to filter scanlation groups out of their follows in some manner, then it would be a worthwhile feature to implement. But I don't think it would need to become any more interpersonal than that, and simply remain an option on the user-end like how the language filter is.

 

[littleoni]

@DANDAN_THE_DANDAN @Teasday
The fact that some people do have passion and pride doesn't change that some don't. I have pretty much given up on the idea of rating the groups, it doesn't make much sense, and would not solve much more problems than it would cause. Maybe I should make another suggestion (but I don't really want to open another thread for the same purpose) for the option to just exclude certain groups from your follows feed, so anyone can chose to ignore the bad quality/no passion releases. Currently there's no way to "punish" the "wrongdoers" easily: if I don't want to read their releases, I have to sort by group and ignore them in my follows list, while paying attention not to let the releases that actually concern me go unnoticed. The feedback in the comments section will also be given by people in the same position: they will either complain about the bad quality, congratulate the faster release, or say nothing at all. Almost no one will say anything about the later but better version because they have not seen it because it's being ofuscated (unintentionally, but ofuscated nonetheless).
@Bainhardt
That's exactly what I'm suggesting (right now, at lease, since I have realized the ratings thing was plain dumb): the possibility to block certain groups from the follows feed and from showing up on the series page (not neccesary, but helpful), either on a work by work basis or global to the whole site (I do think that work by work is a better option, but having both should not hurt if it's not too hard to implement).

 

[DANDAN_THE_DANDAN]

@littleoni

Actually, the idea of "blocking groups" of which you don't want to read translations of is a really good idea.

Also,

The fact that some people do have passion and pride doesn't change that some don't.

Be warned, I'm gonna be going on a tangent here.

I often would observe online communities and compare them with animal habitats, just out of curiosity. I found out that users form an "ecosystem" similar to that of of "survival of the fittest", with a few differences of course.

In this case, "fitness" is determined by what the community wants. In the case of MD, it's quality translations.

Long story short, what you're saying is so minute that it is ignorable because bad translations will "die out" since good translations "are the fittest".

And here's the long explanation.

Imagine a good and passionate scanlator with a lot of pride and always pour love into their work. They would have "high fitness". How? The community will spread their name, increase their morale, support and criticize them until they eventually improve to become a master of their hobby.

Now imagine a bad scanlator who does to see how money they can make; they choose unpopular mangas to scanlate, they don't proofread their work and have a decent grammar. Of course, people criticize them but since they have no emotional connection to the community, they never read the comments. Even with very little words to their name, the only readers who would read their translations are new ones or the curious. This is only the best case scenario; the worst being haters spreading their name in bad faith, making their reader base less and less.

Obviously, they won't make much money with little readers. After their curiosity is satiated, they are satisfied while also disatisfying the community. Their leaving was quiet and people were glad that they were gone. This is one of the many ways a "less fit" scanlator can perish.

Plus, if worst comes to worst, they'll just get banned, like the recent case with that name whom I already can't remember.

TL;DR
The problem of "punishing bad scanlators" will solve itself out naturally.

 

[Sokers]

?

 

[AccomplishedList1]

what right for user or reader to "punishing bad scanlators"? can't you just ignore them or you just some kind of ignorant ppl? not all LQ group taking money/want some donation. so far most of them was BIG GROUPS that usually Ask for donate,for fast release. I don't get it wth need such as rating scanlator> did you pay? did you invest your time as they did? if you unsatisfied with LQ you can always pm or comment, point out if they need improve TL,TS,RD.. or help author by buy raws or official eng TL. ????

 

[littleoni]

@AccomplishedList1
That filosophy is exactly the root of the problem. Just because they have invested their time in doing it, they are to be praised, and no criticism can be thrown their way. Well, that's wrong. You have taken the name "translator" you are expected to translate. If you do it poorly, then you can be called out for it. If you just want praise, then do it properly, if you want to improve, then improve, do your best. Running a text through machine translation and then making it public as if it was your translation is kinda deserving or criticism and maybe punishment, since you are probably confering the wrong message because "your translation" is not quite correct; you are also flagging the series as already being translated, so no one else will pick it up; if your translation is correct, but you do a bad job at redrawing, or cleaning, or whatever, you are presenting the series as being low quality and giving people a wrong impression and maybe discouraging them from ever picking it up. No matter what your intention was, if you "invest your time" and release a poorly done thing, no matter if you intend to get money out of it or not, in the end you are hurting the original work, you are hurting the author and you are hurting the current and potential fans of the work.

 

[PimpToad]

Have scanlating (and boy is this a stupid word) grown to such a point that groups are actually competing (wasting their time) on the same manga? And enough groups are actually doing this that people can actually block groups they don't like from perceived slights in quality?

In all seriousness, why is this even a point of consideration? Leaving aside the groups that shamelessly try to monetize their work, I'd like to think this game is done by people who do this for free and mostly out of pocket. Nevermind that the barrier of entry to get into the game is measured only by the amount of hours you're willing to put into it, the vitriol levied against "bad" releases is in no way valid as any grunt can pitch in to make it better.

Do bad translations exist? Sure they do, but how do you even know it's a bad translation if you can't read and comprehend moonrunes? If you can? Then you can help TLcheck!

So yeah I'm simply of the opinion that if you have the time to provide unsolicited "critiques" of someone else's work, then you certainly have the time to help make the work better. If you don't actually want to invest the time to improve the work? Well do kindly keep your mouth shut and simply enjoy the work you're reading for free.

 

[Teasday]

_{@littleoni posted:}

The fact that some people do have passion and pride doesn't change that some don't.

I didn't mean to imply otherwise. You asked a seemingly rhetorical but pretty absolute-sounding question, I answered it literally as a counterpoint.

The fact of the matter is that, as should be obvious, MangaDex doesn't want to actually discourage scanlators from scanlating. I understand your point but "punishing scanlators" for lower-quality work than their competitors, as you put it, is counterproductive to the aim of fostering a growing scanlation community. Even if it put some low-effort groups down a few deserved pegs, I think it's better to avoid creating systems that would collaterally slam awful scores on people new to the hobby still doing their typesetting in MS Paint just because they hadn't learned any better yet, making them quit before they properly even start. I share your frustration about these comic sans speedscans sniping better releases and grabbing all the comments, but it seems to me that you're exaggerating both the harm they cause and their prevalence to some degree. To my knowledge, few groups get away with google translating their releases without any public criticism.

Speaking of sniping and comments, here's a galaxy brain idea: would it be better if every distinct chapter got only a single shared comment section, regardless of how many groups released an alternative? ?

 

[littleoni]

@PimpToad
Well, you are talking a bit of an utopia, I think. On one side, yes and no: there are a lot of groups, and some intentionally try to compete over the more popular but not the most popular works. I started this thread due to one specific "fight" that happened over a series I was reading and that forced me to pretty much drop it unless I'm willing to tolerate worse quality translation. It's not a very common ocurrence and it's not threatening the whole scanlation (I agree, the word is stupid) business.
And about other groups pitching in and fixing things and whatnot ... Has it happened in the past? Well it sure has, sometimes, not too often, almost never. The main reason would be that the "good willed", "proper", "quality" scanlators tend to pick up works that are not being already translated, no matter if the current translation is trash. There is a sense to this: you broaden the array of aviable translated manga, and you don't enter an unnecessary conflict with whoever is already translating it.
Also don't come to me with the "you are reading it for fee" BS. Please, understand how the real world works. I do pay for it because someone is receiving money for it. Whether the money is given directly or not is not the issue. I pay for it with ad clicks (if there is ad revenue), which is money from an ad company, which comes from a customer, who is paid by the ISP, who takes my money. I pay collectively via donation, no matter if I directly donate or not (at this current timeframe, I do; at some I didn't, the situation changes): the collective of people eligible to become donors extends throgh exposition and there is a chance of them becoming one, chance that materializes through repetition (statistic distributions of probability). In the case when there is no money involved, then the person doing it is doing it either for fame (which they should earn doing a proper job) or to get better (in which case they themselves wish to do a proper job). And let's not forget that the question should never be about money because legal issues (taxation and liscenses).
TL;DR: It's not that it's too common of an issue and it's massively causing trouble, but there is no reason for it to be causing trouble at all.
@Teasday
That's one of the reasons why I have given up on the rating idea. It was dumb, and it would not solve anything really. What I'm proposing right now is the option to personally block a group (or the releases of a group for a given work) from my follows list so they don't ofuscate the releases of a maybe slower, maybe better, or maybe just my preferred group.

 

[DANDAN_THE_DANDAN]

@littleoni

What I'm proposing right now is the option to personally block a group (or the releases of a group for a given work) from my follows list so they don't ofuscate the releases of a maybe slower, maybe better, or maybe just my preferred group.

Just make a new thread.

 

[Teasday]

_{@littleoni posted:}

That's one of the reasons why I have given up on the rating idea. It was dumb, and it would not solve anything really. What I'm proposing right now is the option to personally block a group (or the releases of a group for a given work) from my follows list so they don't ofuscate the releases of a maybe slower, maybe better, or maybe just my preferred group.

Yeah I think that's on the todo list

 

[MellowV2]

I find this idea stupid because even scanlators cannot agree on one common standard. There's bound to be disagreements.

For example
1) choosing of different fonts
2) subbing or dubbing sxf
3) keeping or translating japanese honorifics
4) redrawing or filling in transparent text bubbles
5) hyphens or no hypens
6) etc etc

We can't use official scans as a guideline either because some are really good while some are really bad as well. I wouldn't mind a "MTL" or "bad english" tag though.