Improving reader ←→ tl groups communication :rejected:

[Mangolala]

3 suggestions to improve information for tl groups about readers and for readers about tl groups (or individuals).

1. A way to signal interest in a manga for would-be translators.

For readers if would be a "I'd like this to be translated" button, maybe with additional info like "because it was dropped / because the current tl is bad / because no one has touched it yet"

For translators it would be a page with the info presented in different ways to help them make their choice: not just manga with the highest interest but also interest filtered by genre, time since last interested reader, motivation (e.g. because it was dropped), publication status, licencing status etc.

So new translators could easily find that one-shot that would be easy for them to start with while pleasing a lot of people, or know on which manga to start after filtering by genre/publication status.

2. More details requested on the translation at upload time

This would include level of Japanese proficiency (veteran, beginner, MTL), source language (often enough it's a second hand translation and not from the original language).

It should be a couple of radio buttons so not too burdensome, but for uploaders managing many manga from different groups we can imagine a system with presets.


3. Scanlation quality scoring

The ability to score not only the manga itself, but also the scan, cleaning, typeset, translation quality in order to give a better at-a-glance idea about the actual reading experience beyond the synopsis and overall score. As this is a website dedicated to fan translations, I feel like it makes sense to judge these elements as this is what the scanlation teams are in control of, rather than the story. This info could be aggregated and displayed on the tl group pages as well.

 

[schlo]

my concern is that #3 might drive some away. For the ones doing it for fun, it might be come less enjoyable if their being rated and eventually leave.

I think instead maybe helping groups improve might be a better idea. Maybe a subforum off the groups section called improvements or helpful hints. Might even have some threads permanently tacked at the top, maybe a English Translation page, Spanish Translation, Cleaning page... These pages would be limited to 1-3 sentence hints. I mention this because I see, what to me is some simple stuff passing through. One example, that surprises me, is I see misuse of the pronouns he and she. The translator is referring to a male character and uses she or the reverse female and use he. It makes me think they don't completely grasp the pronouns. There are some others that I've seen that a simple one or two sentence rule/question would indicate how it should be translated. Something where they can find common mistakes or rules and quick and easy guides. If it takes more then 3 sentences start a separate post. The goal being the tacked threads should be easy to scan and read.

 

[Halo]

> 1
Let's systematize begging! It's not like people who can actually read these works can make a better choice.

> 2
I didn't realize scanlation was a job. Perhaps we should start asking them for their CV and recommendation letters from previous employer?

> 3.
Love this one. This is how you get another Great Exodus of MangaDex.
99% of readers won't even bother to help out with some basic proofreading, but would gladly shit on translation like it's their calling in life.

 

[justforthelulz]

The real reason Moses led his people out of Egypt was because the Pharaoh was super critical of his translation choices.
"What shall I do unto this people? They be almost ready to stone me _{(for daring to translate 'baka' into 'idiot')}.”

 

[Holo]

#3 is a slippery slope to scanlation policing, which we don't do.

 

[Mangolala]

I had typed a lot but lost my comment. To summarise my response: the criticism on the 3rd point, which is shared across your @schlo @Halo and @justforthelulz's comments, is fair. It's the weakest proposition that may lead to the most unintended consequences.

To respond to the rest of @Halo's uselessly smartass-y comment, #1 and #2 are all about formalising information that many Tlers already share/ask in their TL notes or comments ("hey guys it's just a MTL" / "will continue if this seems popular, tell me in the comments").

#1 specifically is about providing more information and giving additional tools to TLers looking for their next work. Sounds odd to be against more information. You'll be able to ignore it even more easily than the current comments "begging" for someone to pick up a work.

 

[Mangolala]

Thanks for the response @Holo, I agree with your and the others' feedback that it's actually not a good idea. What do you think of #1 and #2?

 

[Halo]

@Mangolala 👮‍♂️7

#1 and #2 are all about formalising information that many Tlers already share/ask in their TL notes or comments

Which is their choice. "Formalizing" it doesn't leave much of a choice. In your description of #2, they would feel pressed to include some information that can lead to unnecessary judging of their work.
#1 can discourage translators who are either slow or taking a break, it's sounds like a popularity contest for next sniping targets. The "because the current tl is bad" reason is just a way to openly shit on existing translations. The "because no one has touched it yet" is fine, but this is what MangaUpdates' wishlist is for.

 

[Mangolala]

@Halo well at this point you're just criticising a potential implementation of the idea that would be harmful. I gave examples that would obviously need to be discussed/fleshed our further before implementing the feature.

For #2 It's easy to put a line of text saying that sharing the information is optional. As for #1 you can put some rules to avoid this. Mangadex has already some time-based rules, like the need to wait 6 months before uploading the scans of a disbanded group (which is why Mangastream scans have all been uploaded the past few days). Nothing stops the team from implementing the feature in a way that does not encourage sniping.

 

[MavB_Ver]

@Mangolala
#2 Sorry if I'm blind but where in your first post implies that this information will be optional, is it the "radio buttons"? My impression on this suggestion is like, where is the "nah, I don't want to check either of them"? Also, you should at the middle option between beginner and veteran, not everyone is either of them.

 

[schlo]

@Mangolala

I had typed a lot but lost my comment.

doh, hate when that happens.

As far as 1 and 2, I'm not sure how useful they will be vs work required by the md team. Which would seem to be a big issue.

For 1, how many teams are actually having trouble figuring out what they want work on next? Maybe a few, a simple post can solve that with no work. Admittedly I don't translate, but if I were I would first be looking at stories I like not what someone else wants.

For 2, I"m not sure how many would use the info other then for curiosity sake. Some manga have multiple groups working and quite a few readers, myself included, are reading the first one that comes out whatever the quality. I believe a second hand translation, while less desirable isn't like to affect the choice of many people.

 

[Mangolala]

@MavB_Ver my bad I should have added more details. Generally a feature request is rarely implemented as is. You first define the goal, the outline of what it would look like, then when it's validated you dig into the details of the implementation. I expected answers to be "it could work, but only if its optional", instead of "you didn't explicitly say that its optional, must be that you want to force people". Same for the "veteran, beginner". I agree with you, I just didn't think it useful to try to create a categorisation right now, those it's not essential to understanding the idea. Maybe the categories would be more wordy, like "I translate manga regularly" vs "I just started translating manga", vs "I am confident, but not bilingual" or whatever.

As I said in previous comments the aim for me is to give the possibility to share info that is already being shared by Tlers, but "hidden" in TL notes. So that is relevant for the Tlers who want to share this info. If they don't want, that's fine.

@schlo for #1 I think it's less about solving a dire problem than adding an additional convenience. Of course anyone can use google and browse top reddit threads and check wishlists. But MD is in a good position to add another possibility, which is to conveniently browse by # of translation requests while filtering with other criteria. That would be neat for people who, for example, are trying to find a first manga to translate but want something which would have a large enough audience to have a community. Of course, many Tlers don't care, or have a backlog, or have different strategies. But it's also true that the popularity metric is important for many of them, and that mangadex wants to be TL-team friendly. This feature would help with both criteria.

For #2 I think that there are more people than you think who care about quality, especially because there are sometimes several teams working on the same manga, or even the same chapters.

It's true that for ongoing manga, most of the time people following the updates just read the first chapter that is out. But are you saying that you only read ongoing manga and are always up to date with the latest chapter? There is a huge part of the manga reading audience that reads Manga with some delay (see all comments like "I just binged this and I need more!") or start manga that are already completed. For this audience the info would be valuable. It may even help MD decide which translation to show you when there are multiple options. Most people go with the default. Why not try to make it the best possible experience?

Lastly, it would also allow Tlers who want it to share that info in a structured way.

Overall I think that those features would differenciate even more Mangadex, as a fan translation website, from random aggregator websites. Instead it would make it more about the community of Tlers and readers, by matching their existing usage patterns. I even think that some more work could be done on the pedagogy side for all those people who think that mangadex is an official publication website (many do!) or who don't understand the work behind Scanlation. But that's for another discussion.

 

[schlo]

@Mangolala as far as using google and browsing I didn't think they would do that. I figured since they are translating manga, they are reading it and already have some that aren't translated and that they like. I couldn't tell you the names since while i will recognize a group name I don't focus or follow groups but I know a couple of them at least have brought manga here on their own not at someones request.

As far as 2, no I read a mix. I guess we just have different views on the number of people that are going to focus on quality. I think many would prefer quality but are going to go with what's first. As far as catching up or reading manga with many published chapters, binge reading ..., I would guess many are like me. They find something, open the title click on the 1st chapter or wherever they left off and start reading. If there are multiple groups I will likely go with the group I recognize or pick one start reading and if as I start reading it is unreadable I will go try the other group. Unless you put all the info you are talking about in the chapter list I won't see it. Barring something odd, I'm not going to stop and jump to a group page to see their skills.

I'm not against adding functionality, for me it is about resources, bang for the buck and time to implement the whole system. I don't think this gives enough bang for the buck.

 

[TalesForgotten]

I think that 1&2 are a solid starting point as conversation pieces for features that are doable and would help some part of the community. Realistically it is up to the team here at mangadex to decide whether it's worth the dev time and how a feature similar to this would look once implemented. The original comment was phrased as a suggested idea and while it is good to get feedback from the community about how we the users would like the feature to look I think it's unfair to shoot it down based on arguments like it's not worth the dev time unless you yourself are a mangadex dev with the power to make decisions about added features. The worth of this idea is entirely up to the team to decide after they've considered how it would be implemented and what it would do to improve the user experience. I think it's far more helpful to give constructive ways to improve an idea and make it worth more "bang for your buck" than to discount it because it wouldn't be personally helpful to you.

P.s. @schlo Although I quoted your comment specifically I intend this for all commenters to read and think about. My only goal is to encourage a healthier more thoughtful approach to discussing these suggestion threads and leave the feasibility question in the hands of those with the power to make the final decision after seeing the idea as it has been refined by the community.

 

[Halo]

@Mangolala Even if it's optional, #2 still introduces a way to rate and categorize translations, albeit by translators themselves. If someone decides not to disclose that information, they would appear overall worse next to a translator who marked themselves as a "licensed Japanese". It can be used to justify sniping and make the scene to appear more demanding than it needs to be. Plenty of translators will be using this feature jokingly, choosing the least appropriate option. etc
It's one thing when they do it in their credits, but MangaDex collecting that information would carry a very different connotation. Besides, if you care about quality so much, you should be making your own judgement rather than trusting whatever teams have to say about the quality of their work.

#1 ManagaDex is not a database, it doesn't have every single title, nor it host every single translation there is. What will happen if someone makes a request to translate something that is already being translated, just not uploaded on MD? Or is licensed in their language? Dunno, this sounds like a moderation headache. But if it gets implement in a way that doesn't affect existing translations, I have nothing against it.

 

[Mangolala]

@Halo

Trolling

I think the question is less "would some people abuse the system?" (because people will, and they do, when you look at joke descriptions, joke tagging, troll uploads or hate rating), but rather "would trolls defeat the point of the system?".

For #1, one way of reducing trolling would be to reduce the validation that troll gets. When I troll the score, I immediately see the score dropping. When I troll the description, I know people will see it right away etc. The amount of tl requests a manga get doesn't have to be a first level information, meaning that MD could make it needed to select "more info" or something like that to avoid the instant gratification of trolling it. It doesn't not remove, but massively reduce trolling.

For #2 I see little reason for TLers to purposefully troll their self assessment, especially negatively. Why would they want to demotivate people from reading their work? They could do it if they know that people will read it anyway (they're the only TL) in which case it doesn't matter for the reader. If done in a way that is aggravating (for example falsely marking your translation as better than it is to get priority), then

[ul]
* at worst we get a situation like now where you try the alternative translation when you see that the currently selected one is not as you expected
* at best the community flag the discrepancy (the same way that you would flag a vote bombing for example) and mod can give it a look.
[/ul]

peer pressure

If someone decides not to disclose that information, they would appear overall worse next to a translator who marked themselves as a "licensed Japanese"

I think that unless some info shared by the TLer is really misleading, the vast majority of people will not think twice about this info, especially if it's presented in a neutral way. The point is to have this info appear in the overview page of the manga: a careful presentation should make it about "this chapter is MTL" rather than "this translator uses MTL".

quality

if you care about quality so much, you should be making your own judgement rather than trusting whatever teams have to say about the quality of their work.

I think we loop back to our initial disagreement: I don't think more info is a bad thing, even if I, specifically, could do my homework on my own. There are many who don't care about many of MD features but we still have them. If this feature leads some people to start reading more about translation quality, that is a win for the community. If that can help MD automatically present the non-MTL translation instead of the MTL one that was done just to get people interested in the manga, that is a win (people would be free to switch as they already are). If it leads people to think "ah that was a great manga, but it apparently was an MTL, so I'll buy the official publication to see the difference" that is a win. All those wins represent a small group within the community each time, but it still makes for a feature that has more pros than cons I think. And most of all, by exposing more info about fan TLs in a way that can be both informational and pedagogical, it adds value to Mangadex as a platform for fan TLs.

Mangadex is not a database

Agree! But I'm not sure why you would mention that. The point is to have that information for the titles on MD, not as a general public information database. It's aimed at the TLers who are also part of the MD community.

What will happen if someone makes a request to translate something that is already being translated, just not uploaded on MD? Or is licensed in their language?

The same than what happens when someone does it in the comments section of a title: nothing! Why would mods be involved? A TLer looking at MD now and seeing a title with no TL will check if a group is not TLing the manga elsewhere before committing. That will stay true with that system.

 

[Plykiya]

@Mangolala There are more series than there could possibly ever be scanlators to do them. Finding series to translate aren't hard, finding people who are free to translate them is. They're typically already doing what they want to be doing, not just sitting around free.

As for #2, it'd be hard to prove and groups could lie or overestimate themselves so not feasible.

 

[Mangolala]

@Plykiya I understand your point, but following your reasoning if the only way to find a new series to translate would be to go through an unordered list of a thousand titles, it would be fine because "finding new series isn't hard, look, a thousand titles in front of you!".

What's missing from your statement is that translators do rely on different ways to find new series to translate, because just looking randomly at a list doesn't cut it. Maybe they follow popular new series on jp websites, maybe they get recs from their friends, maybe they check baka-updates, maybe they find a new serie by chance. Whatever they way, there is filtering process to find new series.

So #1 is about providing them an additional pathway/filter to discover series they may not have heard of. This filter would be unique to Mangadex and strengthens MD's positions as a space friendly to TLs.

That's it. There was no claim that this would be the only way for TLs to find new series, as it's obviously not.

As for #2 I don't think that the lie/overestimating would matter so much as long as the system is designed properly. At the most basic, the feature could be a "MTL" tag applied on a chapter basis, which would reduce the issue to a yes/no: is it MTL or not. There's not reason to lie about that unless you want to devaluate your work. And if someone hides it, then it's no better than the current situation.
So overall an improvement in terms of information for the readers. But I understand that it would only bring uncertain benefits so I won't argue further.