deg said: its because the moderators are not the developers of even owners the moderators here are all volunteers They're still a part of the staff that works for MAL. Be it administrators, moderators or any other role given an administrative privileges of the website can all be grouped as a staff. |
Seitaro11Feb 12, 2020 6:53 AM
 |
| lol people asking technical details of the new system they will not answer that obviously to avoid being exploited |
| Honestly, I'd love to see a more extended list of the series that were most affected by this change. Other than that, great job guys! |
"At some point, I stopped hoping." |
deg said: Timz0r said: With a fresh start we meant wipe all scores down. Implement new system. I don't mind in re-adding my 700+ scores. But I bet MAL staf are too stubborn to consider this. MAL scores will stay a joke even when the week has passed. lol i choose the lesser evil that is this new system rather wiping out everything every what few years? and you are just inviting more vote brigading with a totally clean state of the scores anyway That's when MAL mods have to step in to stop salty fanboys. It's their (volountary) job afterall. If they don't like their tasks what their doing, they are free to leave their job behind after all. I've moderated for a couple of years a huge RuneScape fansite and I've done the same. I just quitted when I had enough. |
| Sorry, due to licensing limitations, this message is unavailable in your region. Please come drink tea, eat cake and procrastinate at the Cute Girls Doing Cute Things Club. We have simulwatches! \o/  |
Seitaro11 said: deg said: its because the moderators are not the developers of even owners the moderators here are all volunteers They're still a part of the staff that works for MAL. Administrators, moderators or any other role given an administrative privileges of the website can all be grouped as a staff. still you make it sound like the moderators are also part of the developers group that implemented this new system @Timz0r lol the reason for this new automated system is to avoid manual labor more and you expect the staff to do more manual labor? even you will quit in no time if most anime will have vote brigading going on because of the fresh start you want |
Timz0r said: You might not mind but 99% of the userbase probably woulddeg said: fresh start? isnt that whats being done here with the all recalculation going on with the new system? With a fresh start we meant wipe all scores down. Implement new system. I don't mind in re-adding my 700+ scores. But I bet MAL staf are too stubborn to consider this. MAL scores will stay a joke even when the week has passed. |
  |
Psajdak said: deg said: No. I am just kidding.am i reading this right? making alt accounts to manipulate the rankings is fine for you? wut? Still, this new system won't help either, it will just pose a challenge, and make things fun who actually go out of their way to make alts in the first place. The only real solution is to turn off the ranking for a while, until people stop being dicks, and then put them back on. If people continue being cunts AGAIN, then you remove a few high ranking titles, not just rankings, but simply entire titles, until... That will most likely be bad for many good members, but unfortunately, for a bad illness, bitter medicine is needed. Also, delete every account not used for three years or so, not really because of freeing nicknames, but because those are like rotten apples still hanging on a tree. Their scores too mess up with rankings. Unfortunately people will always "be dicks" as you put it. It's just human nature. |
Psajdak said: deg said: No. I am just kidding.am i reading this right? making alt accounts to manipulate the rankings is fine for you? wut? Still, this new system won't help either, it will just pose a challenge, and make things fun who actually go out of their way to make alts in the first place. The only real solution is to turn off the ranking for a while, until people stop being dicks, and then put them back on. If people continue being cunts AGAIN, then you remove a few high ranking titles, not just rankings, but simply entire titles, until... That will most likely be bad for many good members, but unfortunately, for a bad illness, bitter medicine is needed. Also, delete every account not used for three years or so, not really because of freeing nicknames, but because those are like rotten apples still hanging on a tree. Their scores too mess up with rankings. that will sacrifice more users that wanted to score their anime, the lesser evil is this new system that sacrifices few legitimate users while punishing the vote brigade plus we do not know the technical details of how the new system works and we might not find out sooner or later too and that is good already, the more the exposure of exploits got delayed then the better |
enJ0Yable said: From what I see SAO's score barely changed at all. Though I don't know if it's been updated yet (they said it'd take a number of days before the whole website got updated after all)Could you maybe provide information how many percent of accounts have been "blacklisted" by this system, per anime? So for example, how many of the 1.170.000 MAL members that rated SAO season 1 still contribute to its score? That would help seeing the impact on the site and/or on specific anime... |
| |
| I think my previous post was a bit too lengthy for people to spot my question which I put there at the end, so I'll say it again. My Hero Academia's scores (all seasons) have been dropping kinda unnaturally/drastically fast over the last few months as I've seen, so has it also been affected by the downvoting problem? If that's the case then can we hope to see some adjustment in the scores of the MHA seasons as well? |
| "There's no such thing as a painless lesson, they just don't exist. Sacrifices are necessary. You can't gain anything without losing something first. Although if you can endure that pain and walk away from it, you'll find that you now have a heart strong enough to overcome any obstacle. Yeah... a heart made full-metal." ~Edward Elric |
| This is an excellent implementation for a difficult problem to solve. Props, MAL! Finally the people saying you didn't care can eat shit. |
Timz0r said: deg said: lol i choose the lesser evil that is this new system rather wiping out everything every what few years? and you are just inviting more vote brigading with a totally clean state of the scores anyway That's when MAL mods have to step in to stop salty fanboys. It's their (volountary) job afterall. If they don't like their tasks what their doing, they are free to leave their job behind after all. I've moderated for a couple of years a huge RuneScape fansite and I've done the same. I just quitted when I had enough. Uhm, do you realize how much "salty fanboys" this suggestion could create? Not even highly paid staff members are willing to tend to the enormous meltdown that would happen from this. If the staff already has their hands full on an incident like this, how would they handle the entire userbase complaining about the system reset? |
 |
| Thank you very much for this change. We expect integrity and reliability from other people, so it's wonderful to see that from the people behind the website. |
Have suspicions, not faith. |
| @Kineta: As someone who has closely followed the troll voting over the years I would like to emphasize that the results from the detection system have not been able to catch a large number of the trolls. Since 2019, Gintama° dropped from 9.14 to 9.04, LOGH dropped from 9.11 to 8.99, Gintama' from 9.10 to 9.03, Gintama from 9.06 to 8.97, CLANNAD AS from 9.05 to 8.96, Gintama Movie 2: Kanketsu-hen - Yorozuya yo Eien Nare from 9.05 to 8.94. Gintama. from 9.04 to 8.96. The drops were quite glaring and were in patches as I have revealed to you the way around to do such things. These scores were almost stable for years prior and only were heavily affected since the illegitimate accounts report option was taken away. It's obvious that the new system failed to restore the original scores before the trolls found the way around I provided to you through PM. Anyways, if this new system prevents from any large scale manipulations in the future than it's very appreciated. Thank you, the moderators and the development team for the hard work. |
| Stay Home and Wash Your Hands. Protect Yourself. Protect Your Loved Ones. Protect Your Community and Help Defeat Coronavirus. |
| I've read the OP twice now and I guess all this is going over my head. Wouldn't it be simpler to just not have the scores count for the All-Time lists for shows that are currently airing, or to just not count as much? There is a seasonal list too and if something is brigaded and trolled it would only really affect that list. It bothers me alot that people seem to want to try to read people's minds in their votes. People should have a right to express their vote for something the way they want to, even if its illogical or silly. If a bunch of people want to change their votes up or down so be it. Why even let people vote if your just going to change the algorithm because your personally don't like the results? I know this has been in the works for a while, but as someone who has used MAL almost since its creation this has been something that people have complained about the entire time. I can certainly agree with the war on fake or troll accounts. As much as I am a supporter of people's right to vote how they want, its pretty obvious if the account was created and the only show rated is the one being voted up or down. |
 |
Timz0r said: deg said: fresh start? isnt that whats being done here with the all recalculation going on with the new system? With a fresh start we meant wipe all scores down. Implement new system. I don't mind in re-adding my 700+ scores. But I bet MAL staf are too stubborn to consider this. MAL scores will stay a joke even when the week has passed. The only way this has to not be the absolute worst idea ever suggested is if some kind of measure is taken to send all the user list data to the user so they can add their scores easily or even automatically, and even that is an unnecessary chore and bound to create issues and additional and redundant work. Why the fuck would this site risk its database and why the fuck would an individual want their personal ratings to disappear. |
| There is discrepancy between score shown in actual series page and seasonal page. Case in point Bang Dream 3rd season is 7.89 on seasonal page but on the show page the score is 7.74. |
 Kickstarter for Rokujouma is fully funded. Good work everyone. Lets wait for the result of our hard work together. |
| @345EdwardElric Yeah you said it and it turned out that i was wrong, a lot of people were sure that the top 10 would change massively but they all got a small boost in ratings, other than Gintama 2015 and LotGH but even that is not as big as i had imagined, i thought for sure that they would get at least 0.1% boost. Well at least there are no more what ifs, the top 10 is the top 10 and every score is accurate. |
| Are you going to release how the new scoring system will work to the public? |
     |
Ericonator said: Are you going to release how the new scoring system will work to the public? obviously not lol to delay exposing the exploits, same with reddit and facebook they never publicly release their algorithms on their karma and likes system |
Cladocera said: borderliner said: Simple math could easily point out that one 6/5/4/3/2 vote has much more impact than one 9 vote to the average score that was near 8. And remember this is weighted score so there're many other factors considered as well. I'm not arguing that's what happened, just saying it's a possibility. But can't just outright denying it like you did either.Nope, that's not the case. Strongest trend (outside of 10/10 and 1/10) is for 9/10, trending out the scores I make the likely figure 7.62 And that is completely ignoring the many legitimate users (myself included) who decided to rate this a 10/10 not because of Nux or anyone else but because "that's what we're allowed to do!!" I believe I already responded to the second part. But if I may add: I don't exactly know anything behind the scene, but it's highly likely that they did not ignore all the 10's (or else the score would be lower), so I don't think that many legit users would be affected. Even if some do, it's a collateral damage that everyone has to accept. Then again, I'm just a regular user. All I can do is observing the situation and represent possible situations that may or may not happened based on said observation. It goes without saying I don't consider them as fact, so I hope others could also do the same with theirs. I'm not just pulling figures out of thin air here ;-) I have the stats for all scores 1/10 - 10/10 from day one of this anime. I also have very good stats on how the ratings of shows generally trend out over time. And actually it looks like the figure for IR is now revised to 7.60 so perhaps this is being sorted :-) |
| Quantum ille canis est in fenestra |
| Okay so that is something that can be done.. but being able to use decimal scores not? |
 |
Ardanaz said: jjohn333 said: It's a shame that even when the vote brigading(like in ishuzoku reviewers) increases the awareness of the show it actually also damages its reputation and can put off new potential viewers and/or enrage others so they wont watch the show with open mind but blinded with anger, so in the end it can do more harm than good, as well as it upsets the community Well its interesting to see the changes in scores, glad to see gintama a bit more up, definitely deserves it Pretty much what it did for me lol I love ecchi shows and this whole mess has put me off from watching it Imagine caring this much about MAL scores. Smh. |
| It's still has a plot loop I found, which I can't mention for obvious reasons, and why didn't you just make any account that existed less than a year then its score becomes invalid until the user logs in again after a year? which I highly doubt a troll user would remember his password. |
The word elitist is stupid since it's just used by people who can't defend their favorite shows and use it on people who criticize their favorite shows.  |
| I was gonna snap until I see that it is not because its an "ecchi" (or psuedo hentai) anime that had controversies Sad that this has happened tho, nobody really takes MAL ratings seriously and it would have been cool to see a cultured anime like Ishuzoku Reviewers get the no.1 spot. On the real tho no way the ratings drop after ep4 as it was the best ep so far <however this is just my, and a few others, personal opinions |
HaXXspetten said: Just like Kimetsu no Yaiba was upvoted by bots from 8.50 to 8.90, right?Cosmic_Penguin said: Symphogear XV was upvoted by bots from like 7.80 to 8.50 iircWait, people actually brigade for Symphogear and Aikatsu!? Looking at RedditAnimeList which is reddit accounts linked to MAL accounts and thus not subject to botting you see the same trend for Symphogear: http://www.redditanimelist.net/anime.php?anime=32843 |
| Also, I would like MAL to keep the rating system or the algorithm from being revealed to the users/public. Transparency is a good thing but if it harms you (as proven by the previous system in place), it's better to keep some information hidden. |
| Stay Home and Wash Your Hands. Protect Yourself. Protect Your Loved Ones. Protect Your Community and Help Defeat Coronavirus. |
GreenClock said: It's still has a plot loop I found, which I can't mention for obvious reasons, and why didn't you just make any account that existed less than a year then its score becomes invalid until the user logs in again after a year? which I highly doubt a troll user would remember his password. I love this solution |
jjohn333 said: It's a shame that even when the vote brigading(like in ishuzoku reviewers) increases the awareness of the show it actually also damages its reputation and can put off new potential viewers and/or enrage others so they wont watch the show with open mind but blinded with anger, so in the end it can do more harm than good, as well as it upsets the community Well its interesting to see the changes in scores, glad to see gintama a bit more up, definitely deserves it Really? From my perspective it seems like the increased awareness brings in a fair share of fans (mostly for meme purpose) too |
| I have my theories on the way the new system works and I must say, while this is a great change I am still rather skeptical to the reliability of it. I've seen similar situations be solved before on other sites and ended up finding loopholes anyway. Now, it doesn't really matter with me since I am a law abiding citizen B-) but some people will surely catch on sooner or later, and what is even more worrisome in my mind is whether this may disregard legit votes from people who just happen to give an unusual score at the wrong time. |
JokucFeb 12, 2020 8:10 AM
It's only a passing thing, this shadow. Even darkness must pass. A new day will come. And when the sun shines it will shine out the clearer. |
| So out-of-spite 1s on Ishuzoku Reviewers from legit accounts still count, huh? I understand it's slightly countered by 10s from legit accounts, but 1s have more weight than 10s... What about excluding votes from people who haven't seen more than half of all episodes for anime that finished airing? It wouldn't solve problems for other finished anime that are targets of upvoting and downvoting, but it would eventually solve the problem with a large share of the 10s and 1s for IR. Since you have decided to keep the new rating system non-transparent to prevent anyone from manipulating with the score again, I hope that you have at least properly tested it. |
 |
matteas said: So out-of-spite 1s on Ishuzoku Reviewers from legit accounts still count, huh? I understand it's slightly countered by 10s from legit accounts, but 1s have more weight than 10s... 1s are also included in this. What about excluding votes from people who haven't seen more than half of all episodes for anime that finished airing? It wouldn't solve problems for other finished anime that are targets of upvoting and downvoting, but it would eventually solve the problem with a large share of the 10s and 1s for IR. This is already the case. A user must have seen 1/5th of a show for their score to count. |
| Nice to see the vote briganding getting negated. I am surprised FMAB score increase post update, but I guess the 1/10 reviews must have lowered it prior to the update. |
 |
Jack said: Yeah we're definitely the ones stopping you from being a MAL mod, otherwise you'd have no problem ;p Oh it's you! I remember you! F for my pretty well thought out Clannad - JoJo recommendation. If I'm honest: I would if only I had the spare time, really... If you wanna fight weaponized autism, you gotta use weaponized autism. It's just that I'm already using my weaponized autism towards paleontology. ;-( But really joke's aside; some staff here really need to look at the mirror once again and have a reality check. Mods and administators need to admit they make mistakes awell and should apologize whenever possiblle. Shymander should not have putted that sentence in calling Nux Taku an asshole. That was a mistake on his part. |
| Sorry, due to licensing limitations, this message is unavailable in your region. Please come drink tea, eat cake and procrastinate at the Cute Girls Doing Cute Things Club. We have simulwatches! \o/ |
Ardanaz said: matteas said: So out-of-spite 1s on Ishuzoku Reviewers from legit accounts still count, huh? I understand it's slightly countered by 10s from legit accounts, but 1s have more weight than 10s... 1s are also included in this. But if that 1/10 is from a totally legitimate user who is let's say a fanboy of an anime that could be overtaken, then I think it's very difficult to decide that it's an invalid vote. Does the new system also try to address and counter such behaviour, some kind of mass up/downvoting from legitimate accounts? I think that would be very difficult and if there is a countermeasure for that in the new system, then I'm impressed. Anyway, I hope that the situation settles down soon and both sides, fans of IR as well as people who fight for legitimate MAL scores will be happy with the results in the long run. What about excluding votes from people who haven't seen more than half of all episodes for anime that finished airing? It wouldn't solve problems for other finished anime that are targets of upvoting and downvoting, but it would eventually solve the problem with a large share of the 10s and 1s for IR. This is already the case. A user must have seen 1/5th of a show for their score to count. Oh, cool. I didn't know it's been that way already. rohan121 said: Nice to see the vote briganding getting negated. I am surprised FMAB score increase post update, but I guess the 1/10 reviews must have lowered it prior to the update. Several people reported that FMAB's score was 9.22 before all of this happened. |
_kurapanFeb 12, 2020 8:22 AM
|
| That's some awesome news. Pretty good showcasing of the more specific shows that were affected by it. The scoring system was in need of that update to avoid those influences from trolls and others. After all MAL is a database site and it's really important to avoid that kind of bias. Thanks to all the staff involved in the process ^^ |
 |
Timz0r said: Jack said: Yeah we're definitely the ones stopping you from being a MAL mod, otherwise you'd have no problem ;p Oh it's you! I remember you! F for my pretty well thought out Clannad - JoJo recommendation. If I'm honest: I would if only I had the spare time, really... If you wanna fight weaponized autism, you gotta use weaponized autism. It's just that I'm already using my weaponized autism towards paleontology. ;-( But really joke's aside; some staff here really need to look at the mirror once again and have a reality check. Mods and administators need to admit they make mistakes awell and should apologize whenever possiblle. Shymander should not have putted that sentence in calling Nux Taku an asshole. That was a mistake on his part. A mistake and incorrect are two different things, neither of which applied to what Shy said about the youtuber. |
Jack said: Timz0r said: Oh it's you! I remember you! F for my pretty well thought out Clannad - JoJo recommendation. If I'm honest: I would if only I had the spare time, really... If you wanna fight weaponized autism, you gotta use weaponized autism. It's just that I'm already using my weaponized autism towards paleontology. ;-( But really joke's aside; some staff here really need to look at the mirror once again and have a reality check. Mods and administators need to admit they make mistakes awell and should apologize whenever possiblle. Shymander should not have putted that sentence in calling Nux Taku an asshole. That was a mistake on his part. A mistake and incorrect are two different things, neither of which applied to what Shy said about the youtuber. Except calling someone an asshole, even justified as it was, doesn't help matters any. All it did was paint a target on his back. But we should probably get back on topic. |
Roevhaal said: Just like Kimetsu no Yaiba was upvoted by bots from 8.50 to 8.90, right? Uhm, not sure what you mean there, pretty sure that the vast increase in Kimetsu no Yaiba's score started happening since episode 19, which evidently created an enormous amount of hype amidst fans & blew up a good part of the anime community with its quality. It was after that when KnY's score surged from 8.50 to 8.60+ as it breezed into Top 20; with every upcoming episode after that, the scorce started increasing steadily to ultimately reaching a peak of 8.97 when the final episode aired. The rise in score was fast, but still not unnatural as in mostly increased with each & every episode, it was more due to how well-recieved most of the episodes since episode 19 were & due to strong seasonal hype. KnY's score has since been dropping from 8.97 (it's 8.88 now) with the hype dying down gradually, so everything about its score increase feels pretty natural. |
345EdwardElricFeb 12, 2020 8:46 AM
| "There's no such thing as a painless lesson, they just don't exist. Sacrifices are necessary. You can't gain anything without losing something first. Although if you can endure that pain and walk away from it, you'll find that you now have a heart strong enough to overcome any obstacle. Yeah... a heart made full-metal." ~Edward Elric |
More topics from this board
» MAL Yearbook 2024: Brainstorming Sessiontingy - Dec 23 |
12 |
by LittleStar
»»
6 hours ago |
|
» New Page! Manga Adapted to Anime ( 1 2 )Kineta - Dec 9 |
55 |
by cpebora
»»
Today, 2:15 AM |
|
» MAL Game "Fantasy Anime League" Opens for Fall 2024 ( 1 2 3 4 )Kineta - Sep 12 |
157 |
by tokomata
»»
Today, 1:40 AM |
|
» MAL Secret Santa 2024 ( 1 2 3 4 5 ... Last Page )Kineta - Nov 17 |
258 |
by DeadlyRaven
»»
Yesterday, 10:23 AM |
|
» [Update May 24] Interest Stacks: New feature for custom theme lists and challenges ( 1 2 3 4 5 )Kineta - Apr 6, 2022 |
209 |
by okwhynot
»»
Dec 22, 6:36 AM |