forum

[Proposal] Spread ruleset draft

posted
Total Posts
259
show more
Moonlit
A summary of the discussion so far regarding:

Mapsets cannot include more than 8 total difficulties of a single game-mode.
What is the reason for this rule?

At the start of this discussion, it was put forward by Ephemeral that:
the fat stack of similar yet subtly (or vastly) different tiered difficulties causes lots of fatigue in the review stage, namely the modders and the BN responsible for providing critical feedback during the modding stage.
In some ways this is a self-limiting issue; mappers will be less likely to get mods on large sets, so are less likely to attempt to rank large sets. Modders and BNs are not obligated to mod these sets either.

And that by forcing a limitation on the number of maps in a set, the set would become more "cohesive" instead of being made to fit spread requirements and then being filled with additional difficulties (such as from guest mappers).

He also notes the problems with this idea, stating that any "displaced" maps that cannot fit in the 8 diff set are forced into a new set, potentially containing no new or interesting lower difficulties (following the logic that easier maps require lower note density and are thus have less potential for rhythmic and placement variety).

I believe this quote from him sums it up nicely:
The difficulties are probably going to be "recycled" into the ranking cycle as new sets which require even more oversight during the review stage by virtue of having the four required ENHI difficulties to accompany them in the first place, often times equally as "uninspired" as the same set they came from. There's only so many ways you can spin the lower levels of difficulty in mapping.
The alternative being these "displaced" maps are never created or never uploaded.


After the sharing and attempted interpretation of some playcount data stratified by map difficulty, Loctav made a point that:
The excerise of increasing the set's amount of difficulties to add *even more Experts* (although they are not even targetted at who actually plays the set) just adds content bloat - a lot of the same, without a distinctive difference.
arguing that as most plays are within the hard/insane difficulty ratings, adding more expert difficulty maps without significant difference is somehow detrimental to the game.

There are multiple issues i take with this statement; first of all how do we determine what is a "distinctive difference" between maps? Many people are divided on the differences (or lack thereof) of Monstrata's Zen Zen Zense extras (Taki and Mitsuha) for example. And are players really bothered by such content, even if it is widely perceived to be not "distinctively different"? The aforementioned set was among the most played mapsets for a significant time. I would suggest that its mostly a problem to the creatively minded mappers and modders who do not like the idea of unoriginal content, whereas players are often more focused on challenge and reward (PP) and are less bothered.

Additionally the use of the term "content bloat" is unclear and seems to be a problem that only the team behind these changes is aware of. Noone has properly defined it and thankfully its use has diminished over the course of this discussion. Still I would like to understand why it is a relevant issue (Explain) from the perspective of the rule creators.

Further in the thread Okorin states:
lowering spread requirements for sets further and further will only prove to be detrimental to especially new players getting into the game, there's a constant supply of new music to be mapped as well as old music to take from so denying someone that is just getting into osu to play their favourite song because you think insanes / extras / hards are more important than complete entry level content is just going to disappoint the new people getting into the game.
I dont think this is an incorrect point. Players should be able to enjoy a variety of music at a comfortable level of difficulty. However the lower difficulties are rapidly surpassed by beginners. Beginners are less predisposed to playing very long maps due to stamina, attention span and the nature of how they play osu (many new players only play a map or two while waiting for other games).
For these reasons perhaps it is possible to be more relaxed on lower difficulties while encouraging their production.

The crux of the issue seems to be that newer players must remain in our focus as we move forward, as these players help grow our community. There is a fear that continued relaxation of the ranking criteria may lead to less and less beginner-level content (easy, normal) and reduced retention of these players.
This seems like a possibility to me, although as others have suggested there is a plethora of beginnner-level content already available. How important is it to keep up with current songs, as it is proposed?

How can we encourage more easier maps to be produced?

Desperate-kun makes a point later in the thread that did not receive much discussion:
Instead of limiting the amount of difficulties people are allowed to make, we should try to think of ways that motivate people to map fewer difficulties
He also suggests:
A possibility would be to allow for higher difficulty gaps between two consecutive difficulties (preferably for longer songs, which would solve some other mentioned problems, too) as long as the spread is linear, to encourage people to make small spreads that still cover a wide range of skill levels.
I believe this idea serves to increase efficiency of the mapping system for producing lower difficulty maps without directly limiting production of higher difficulty maps. This spread widening suggestion could encourage more small sets to be produced where they would normally not due to requiring many diffs to fit a tight spread.

Amaikai notes:
As a thought, would it be beneficial to give something extra for providing entry-entry-level content? Mind you this is optional work, not FORCED which i'm againts of. A carrot approach instead of stick.
A point I agree with. What do you think are potential opportunities to reward or encourage mappers in making lower difficulty content?

Other perspectives on set development

SakuraKaminari presents a unique standpoint somewhat later in the thread regarding the position of less experienced mappers and how they are affected by the 8 map restriction.
By limiting the mapset to 8, you're making it much harder for people like me to get GD slots on ranked sets. For example, let's say a new song comes out and a more experienced mapper picks it up and starts creating a set. Sometimes, these mappers are ok with taking a GD from a newer mapper and helping them to make something rankable out of it. This is super important because not only does it allow newer mappers rejected from the academy (like me!) more opportunities to learn about the ranking process and gain mapping exposure, we also learn a lot about mapping, what's rankable and whats not, and usually being modded for rank is more strict and will likely lead to better feedback.
When a mapper has a wide variety of maps and a well made spread they can afford to work on maps with newer mappers to get them to rankable quality. Given how difficult it is for mappers to break into the mapping environment, this can be an invaluable opportunity that would be hampered by the 8 diff rule.

MrSergio brings up a take on the meaning of spread:
Diffs that do not make up for anything useful in the spread are just a way to get past the former meaning of "spread for a song"
arguing that anything above what is absolutely necessary for each level of skill is going beyond the meaning of a true spread. I disagree with this sentiment myself, as the mapping styles of today allow for many interpretations of a song that may end up around the same skill level. As stated earlier it is challenging to judge what a "distinctive difference" really is, so perhaps it is better to allow these additional interpretations to be created and be judged by players - not modders. After all we are making maps for people to enjoy, not to fit a prescribed schematic.

My thoughts

From what I've seen most people are against the 8 diff limit. It seems to miss its intended purpose (as far as I've been able to discern its purpose) and there are probably better ways to increase the amount of beginner-friendly content through encouragement rather than restriction.
In my opinion a ruleset should be in place to ensure a level of basic quality within a map and set; to ensure that when you download a ranked map you can trust it to work and be organised without wild inconsistencies and errors.

A ruleset should not restrict creativity. We are seeing more and more interesting ideas being fleshed out and standardised by mappers experimenting with novel ideas but unfortunately they are bogged down by the ranking system. The 8 diff limit seems like another such restriction and if it were to go through in any form it could limit the evolution of mapping even further.

Perhaps instead of trying to solve all of our problems at the ranking criteria level, we should investigate other avenues to guide mapping towards a greater audience in terms of both involvement and the spectrum of content produced.
Lust
Songs/maps cannot be modified to reach the minimum drain time. Abusing the 5 minute limitation removes its intended purpose. Types of abuse include:
I will never for the life of me understand this
_Meep_
mayb for every extra diff added after the ENHIX spread or ENHIXX spread,
they need to add another lower difficulty
idk
DeletedUser_4329079

Lust wrote:

Songs/maps cannot be modified to reach the minimum drain time. Abusing the 5 minute limitation removes its intended purpose. Types of abuse include:
I will never for the life of me understand this
Because you are basically abusing the difficulty spread rules to be able to rank the map with a lot less effort. On the other hand, this is just a result of the rule being too arbitrary and the massive difference in the effort needed to make a 4+min mapset and a single marathon difficulty.
_Meep_

Default wrote:

Because you are basically abusing the difficulty spread rules to be able to rank the map with a lot less effort. On the other hand, this is just a result of the rule being too arbitrary and the massive difference in the effort needed to make a 4+min mapset and a single marathon difficulty.
Let's say a song is 4:55
Let's say its a deathmetal song that has the potential to be 8*
Are you going to map all 8 difficulties from 1-8* to cover the spread where all the difficulties are sloppily done and are low effort because you're lazy and you dont want to map so much?
Or are you going to map one extremely good 8* difficulty that perfectly represents the song with every bit of effort put into making it perfect?
I don't see it as abusing when extending is practically done to save time
and Saving time =/= Lazy
Yauxo
Throwing stuff in to desp's comment to encourage less diffs.

<8 diffs -> Bubble#1 Rank
>8 diffs -> Bubble#1 Bubble#2 Rank

More BNs to make up for the amount of diffs/work invested. Would be the total opposite for the "too much work for the BNs" thing tho
DeletedUser_4329079

_Meep_ wrote:

Let's say a song is 4:55
Let's say its a deathmetal song that has the potential to be 8*
Are you going to map all 8 difficulties from 1-8* to cover the spread where all the difficulties are sloppily done and are low effort because you're lazy and you dont want to map so much?
Or are you going to map one extremely good 8* difficulty that perfectly represents the song with every bit of effort put into making it perfect?
I don't see it as abusing when extending is practically done to save time
and Saving time =/= Lazy
I never said the word "lazy" in my comment. It also seems like you didn't read the second part of it, but it can't be denied that editing a song to make it longer is abusing the system, even if it occurs as a consequence of the system being very flawed itself.
Lust
I know a way around this rule and you can be sure to see me exploiting it when this comes into place. Don't enact a rule that will always have a loophole (is it really even a rule at that point?)
Monstrata

Lust wrote:

I know a way around this rule and you can be sure to see me exploiting it when this comes into place. Don't enact a rule that will always have a loophole (is it really even a rule at that point?)
Yea I know lol. To give everyone context as to why this was chosen though, basically this was our "compromise" to not eliminating the "approval" category entirely (meaning all maps required a spread regardless of length). Basically, the idea was that "extending and editing songs to fit the 5 minute rule was considered abuse, and we must prevent this somehow". The solution therefore is "people won't edit songs to make them over 5 minutes if that would still require you to map a spread". However, this idea would (obviously) be shot down. The "compromise" was then to add a rule saying we shouldn't edit mp3's with the intention of extending it.

So really, the discussion boils down: "how much of a problem is extending mp3s to fit the 5 minute length?" ... and "is it a form of abuse we should prevent through the ranking criteria?" We need to ask ourselves how much this "abuse" is affecting ranked, and whether we need to address it, or whether we think it's fine to allow songs like this to be ranked. Of course, extreme cases should be handled case-by-case and we can (hopefully) trust BN's not to nominate stuff like harumachi clover or haitai looped 10 times to achieve approval length.

I'll list a few maps that "abuse" this rule so we have some examples to go by to:

https://osu.ppy.sh/s/404360 <--- An entire hypermeasure (4 measures) is added to one of the kiai sections to achieve 5 minute length.
https://osu.ppy.sh/s/403073 <--- Same song, except I had the BPM reduced by 3 (175 > 172) to achieve 5 minute length.
https://osu.ppy.sh/s/418922 <--- The ending of the song is looped with the intro to create a longer fade out. (The intro and outro are the exact same melody)
https://osu.ppy.sh/s/378183 <--- An entire instrumental section was looped (Adding around 25 seconds to the map).
https://osu.ppy.sh/s/503059 <--- Very subtle editing to add 0.5 seconds to the song.
https://osu.ppy.sh/s/315159 <--- An entireverse + chorus is looped (adding over 1 minute to the song).
https://osu.ppy.sh/s/553906 <--- Ending note is extended to reach 5 minutes.


Anyways, as for me, I personally don't believe the "abuse" is common at all, and we shouldn't be changing rules just to avoid the one case in what, 50? It hurts a lot more than it helps. And I don't believe it really helps anyways. The only positive I see coming from this rule is that mp3's aren't poorly edited or tastelessly done. That, and maybe one of the artists getting angry about having their work manipulated somehow? (Though we have zero cases of this historically, and they're probably be more offended we're offering free downloads of their works).

So what do you guys think? We can probably agree mp3 editing to achieve 5 minute lengths is skirting the rules. But do you guys think it's necessarily something that needs to be prevented through the Ranking Criteria? Do you think it's that much of a problem? Are there alternative means to addressing it?
Endaris

Monstrata wrote:

We have two ideas going right now. The first is to change the rule into a guideline, and the second is just to eliminate the rule entirely.
And the third is to change the procedure of how sets and difficulties move to the ranked section.
I'm aware this is an idea that requires actual planning and effort compared to the first two options but it is arguably the option that would satisfy both sides of the argument the most. And both sides definitely have good arguments. That's why this turned into a debate with repeating arguments and none getting really convinced by the other side.

I simply don't see why one would stop at this point and not question the current technical limitations with the modding system and interface between new website and beatmap servers still in development. There's nothing major coded in that regard yet as far as I'm informed so this would be the best imaginable place and time to implement a change to the way sets work and the way ranking works.
The question in this case would be though, what peppy thinks about this and what I as an interested community member could do to support him there.
Monstrata

Endaris wrote:

Monstrata wrote:

We have two ideas going right now. The first is to change the rule into a guideline, and the second is just to eliminate the rule entirely.
And the third is to change the procedure of how sets and difficulties move to the ranked section.
I'm aware this is an idea that requires actual planning and effort compared to the first two options but it is arguably the option that would satisfy both sides of the argument the most. And both sides definitely have good arguments. That's why this turned into a debate with repeating arguments and none getting really convinced by the other side.

I simply don't see why one would stop at this point and not question the current technical limitations with the modding system and interface between new website and beatmap servers still in development. There's nothing major coded in that regard yet as far as I'm informed so this would be the best imaginable place and time to implement a change to the way sets work and the way ranking works.
The question in this case would be though, what peppy thinks about this and what I as an interested community member could do to support him there.
One of the primary arguments for the 8 diff limit is that sets are becoming overwhelming and create what people are calling "content bloating". Having one Miiro set with 120 difficulties hardly seems like it will solve the problem. Also, I don't believe sets are obsolete in any sense. I don't think there was a traditional idea for a "set" to begin with. People have created GD's since osu first started. So your argument isn't convincing. Sure, adding guess difficlulties after a set is already ranked can get rid of the necessity of making a whole spread to rank a new set (for the same song). But there are so many issues with this. What if the original host doesn't want your crappy GD? What if you think the current ranked set is bad and don't want to be associated with that set? As well, what is the nomination process that goes into it? Do such GD's still go through the qualified section? Also people who downloaded the set previously will have to redownload to get new difficulties that have been added, and is there a system that tells people a new difficulty has been added? This is not a feasible option imo. Let's focus on what we can actually discuss and control in the RC.
Shad0w1and
Oh Kibb gonna hate you for that 0.5 sec Mr. Monstrata XD

and for the case like
https://osu.ppy.sh/s/315159 <--- An entireverse + chorus is looped (adding over 1 minute to the song).

I have to say that a lot of old Chinese songs were made that way lol, they looped the same part 3 times and called it a full ver lol. Personally, I do not see it as a problem. Why a loop by the producer is acceptable but by mapper is not acceptable. Though I do think those looped songs sucks, and it is one of the reasons that I like Japanese songs better.
Halliday

pishifat wrote:

A difficulty’s name must be unrelated to a username. Guest difficulties, however, may indicate possession with its mappers’ username or nickname. (e.g. Guest Mapper’s Insane).
Rip StarrodKirby and his awesome Kirby Mixes :'(

(?)
Endaris

Monstrata wrote:

Endaris wrote:

Monstrata wrote:

We have two ideas going right now. The first is to change the rule into a guideline, and the second is just to eliminate the rule entirely.

And the third is to change the procedure of how sets and difficulties move to the ranked section.
I'm aware this is an idea that requires actual planning and effort compared to the first two options but it is arguably the option that would satisfy both sides of the argument the most. And both sides definitely have good arguments. That's why this turned into a debate with repeating arguments and none getting really convinced by the other side.

I simply don't see why one would stop at this point and not question the current technical limitations with the modding system and interface between new website and beatmap servers still in development. There's nothing major coded in that regard yet as far as I'm informed so this would be the best imaginable place and time to implement a change to the way sets work and the way ranking works.
The question in this case would be though, what peppy thinks about this and what I as an interested community member could do to support him there.


One of the primary arguments for the 8 diff limit is that sets are becoming overwhelming and create what people are calling "content bloating". Having one Miiro set with 120 difficulties hardly seems like it will solve the problem. Also, I don't believe sets are obsolete in any sense. I don't think there was a traditional idea for a "set" to begin with. People have created GD's since osu first started. So your argument isn't convincing. Sure, adding guess difficlulties after a set is already ranked can get rid of the necessity of making a whole spread to rank a new set (for the same song). But there are so many issues with this. What if the original host doesn't want your crappy GD? What if you think the current ranked set is bad and don't want to be associated with that set? As well, what is the nomination process that goes into it? Do such GD's still go through the qualified section? Also people who downloaded the set previously will have to redownload to get new difficulties that have been added, and is there a system that tells people a new difficulty has been added? This is not a feasible option imo. Let's focus on what we can actually discuss and control in the RC.


Huh what.
One of the primary arguments is that such huge sets are difficult to moderate and to guarantee quality.
Also I didn't have the impression that you personally have something against more content and personally I don't think it would play out the way you made it up with 120 Miiro difficulties: Right now we have about 50 on 10 sets so it is reasonable to say that the "bloating" would happen anyway. You could go as far as saying that it is even more confusing because there are so many sets and there is no good overview (which is not the case with 2 or 3 sets imo but 10 plays in a different league...).

You also seemed to have ignored a point I made earlier:
When there are already ranked difficulties present for a set, how big will the motivation for a mapper be to map something that plays similar to an existing difficulty?
If there is only 1 similar one it might still be high. But what if there are 5 difficulties that play similar?
Once the amount of difficulties in a set reached a certain point, mapping an additional difficulty becomes unattractive for the mapper for the following reasons:
  1. It might become harder to find modders because modders have an interest in original content that adds to the game.
  2. The additional difficulty will be discovered later and receive less plays overall if it does not have something unique to offer.


Finally I don't consider your doubts convincing:
What if the original host doesn't want your crappy GD?
Well, why would that matter if he does not have to care about getting mods for it, kicking the mapper's ass to improve it etc? Unlike now, the original set creator has absolutely no obligation to make an investment on the additional GD. If it gets through the review phase (the GD mapper has to make it happen) and is of a quality that is deemed good for ranking then it is good for ranking and should be added to the set. You said yourself a "set" never really existed. Also, I have the impression that you did not read the thread linked in my post as it clearly states that the name of the original set creator wouldn't be on the new difficulty so there would be absolutely no reason to have your good name stained or anything what you might have to worry about.
Last but not least, ranked content is created for the playerbase. If you want to map for artistic purposes and for yourself you can map for graveyard or loved.

What if you think the current ranked set is bad and don't want to be associated with that set?
I don't think there would be something that keeps you from making another set for the song. But it would have to be another complete set as it has to now. The objective is not to keep alternative sets from popping up but to enable mappers to add more ranked content on a controlled(=small) scale that is comfortable for mapper, modder and nominator.

As well, what is the nomination process that goes into it?
Personally I would put them into the same nomination process as approval maps. The good thing about single difficulty mapping and modding is that you can go into detail with mods and try to get the best out of that difficulty. This means that the quality of those additional difficulties would be expected to be high. A qualification phase is mandatory.


@Halliday: As Kirby Mixes are traditionally the highest difficulty of a set, they can use custom naming. Don't think they would get DQd for the naming.
DeletedUser_4329079

Halliday wrote:

pishifat wrote:

A difficulty’s name must be unrelated to a username. Guest difficulties, however, may indicate possession with its mappers’ username or nickname. (e.g. Guest Mapper’s Insane).
Rip StarrodKirby and his awesome Kirby Mixes :'(

(?)
Unless it's a Kirby song 🤔
Monstrata

Endaris wrote:

We have two ideas going right now. The first is to change the rule into a guideline, and the second is just to eliminate the rule entirely.

And the third is to change the procedure of how sets and difficulties move to the ranked section.
I'm aware this is an idea that requires actual planning and effort compared to the first two options but it is arguably the option that would satisfy both sides of the argument the most. And both sides definitely have good arguments. That's why this turned into a debate with repeating arguments and none getting really convinced by the other side.

I simply don't see why one would stop at this point and not question the current technical limitations with the modding system and interface between new website and beatmap servers still in development. There's nothing major coded in that regard yet as far as I'm informed so this would be the best imaginable place and time to implement a change to the way sets work and the way ranking works.
The question in this case would be though, what peppy thinks about this and what I as an interested community member could do to support him there.

One of the primary arguments for the 8 diff limit is that sets are becoming overwhelming and create what people are calling "content bloating". Having one Miiro set with 120 difficulties hardly seems like it will solve the problem. Also, I don't believe sets are obsolete in any sense. I don't think there was a traditional idea for a "set" to begin with. People have created GD's since osu first started. So your argument isn't convincing. Sure, adding guess difficlulties after a set is already ranked can get rid of the necessity of making a whole spread to rank a new set (for the same song). But there are so many issues with this. What if the original host doesn't want your crappy GD? What if you think the current ranked set is bad and don't want to be associated with that set? As well, what is the nomination process that goes into it? Do such GD's still go through the qualified section? Also people who downloaded the set previously will have to redownload to get new difficulties that have been added, and is there a system that tells people a new difficulty has been added? This is not a feasible option imo. Let's focus on what we can actually discuss and control in the RC.
Huh what.
One of the primary arguments is that such huge sets are difficult to moderate and to guarantee quality.
Also I didn't have the impression that you personally have something against more content and personally I don't think it would play out the way you made it up with 120 Miiro difficulties: Right now we have about 50 on 10 sets so it is reasonable to say that the "bloating" would happen anyway. You could go as far as saying that it is even more confusing because there are so many sets and there is no good overview (which is not the case with 2 or 3 sets imo but 10 plays in a different league...).

You also seemed to have ignored a point I made earlier:
When there are already ranked difficulties present for a set, how big will the motivation for a mapper be to map something that plays similar to an existing difficulty?
If there is only 1 similar one it might still be high. But what if there are 5 difficulties that play similar?
Once the amount of difficulties in a set reached a certain point, mapping an additional difficulty becomes unattractive for the mapper for the following reasons:
  1. It might become harder to find modders because modders have an interest in original content that adds to the game.
  2. The additional difficulty will be discovered later and receive less plays overall if it does not have something unique to offer.
Finally I don't consider your doubts convincing:
What if the original host doesn't want your crappy GD?
Well, why would that matter if he does not have to care about getting mods for it, kicking the mapper's ass to improve it etc? Unlike now, the original set creator has absolutely no obligation to make an investment on the additional GD. If it gets through the review phase (the GD mapper has to make it happen) and is of a quality that is deemed good for ranking then it is good for ranking and should be added to the set. You said yourself a "set" never really existed. Also, I have the impression that you did not read the thread linked in my post as it clearly states that the name of the original set creator wouldn't be on the new difficulty so there would be absolutely no reason to have your good name stained or anything what you might have to worry about.
Last but not least, ranked content is created for the playerbase. If you want to map for artistic purposes and for yourself you can map for graveyard or loved.

I don't want some random 0 ranked 0 gd mapper trying to get their map bundled with my set. So yes, it does matter. Just because I don't have to kick the mapper's ass to improve it doesn't mean I'm fine with being associated with it. Also I didn't say a "set" never really existed. I said a "traditional definition of a set" never existed. There is no formal definition for what a set must comprise, and there never was. Also please don't delude yourself into thinking the ranekd content is created for the playerbase. Ranked content is entirely dependent on the mapper first and foremost. osu is driven on community content. The content is entirely dependent on what the mapper wants to map and rank., the playerbase has very little influence on what content they want ranked, unless they convince a mapper to map a certain song, or they become mappers themselves.

What if you think the current ranked set is bad and don't want to be associated with that set?
I don't think there would be something that keeps you from making another set for the song. But it would have to be another complete set as it has to now. The objective is not to keep alternative sets from popping up but to enable mappers to add more ranked content on a controlled(=small) scale that is comfortable for mapper, modder and nominator.

As well, what is the nomination process that goes into it?
Personally I would put them into the same nomination process as approval maps. The good thing about single difficulty mapping and modding is that you can go into detail with mods and try to get the best out of that difficulty. This means that the quality of those additional difficulties would be expected to be high. A qualification phase is mandatory.

Those were all rhetorical questions meant to tell you this system isn't feasible in our current context lol. You can theorycraft all you want, but this discussion honestly belongs in Community Features since it doesn't influence the Ranking Criteria but more the whole Ranking System.

@Halliday: As Kirby Mixes are traditionally the highest difficulty of a set, they can use custom naming. Don't think they would get DQd for the naming.
No. "huge sets are difficult to moderate and to guarantee quality." ?? How do you arrive at that conclusion? That's completely ambiguous and not something the RC is concerned with anyways. Huge sets can be difficult to moderate because they require more time and effort from the modder and BNs involved. But if a huge set has quality issues, it's not the fault of it being large, it's the fault of BN's not being thorough. Do you think sets like Hitorigoto and No Title are huge? I really hope you don't, because their total drain is only about 18 minutes. (12 x 1:30). Stuff like: https://osu.ppy.sh/s/94631 is already 14 minutes of drain just with 4 difficulties. And stuff like: https://osu.ppy.sh/s/500858 is already 22 minutes of drain with 6 difficulties. If anything, full-sized mapsets that have a very even spread are the true "huge sets" because they actually require even more modding time than conventional "huge sets".

It's true the "huge sets are difficult to moderate ant to guarantee quality" is indeed an argument being made here. But just read people's responses. It's clearly a poor argument that has no bearing on map quality. It's not something that should be covered in the RC. It is an issue to consider with BNG rules and BN management if anything. I don't believe the issue is valid to begin with though. I'm only acknowledging that two people in this thread think this is an issue.

Endaris wrote:

You also seemed to have ignored a point I made earlier:
When there are already ranked difficulties present for a set, how big will the motivation for a mapper be to map something that plays similar to an existing difficulty?
If there is only 1 similar one it might still be high. But what if there are 5 difficulties that play similar?
Once the amount of difficulties in a set reached a certain point, mapping an additional difficulty becomes unattractive for the mapper for the following reasons:
  1. It might become harder to find modders because modders have an interest in original content that adds to the game.
  2. The additional difficulty will be discovered later and receive less plays overall if it does not have something unique to offer.
Your first point is just completely wrong. It will not be harder to find modders because the song has already been mapped before. Modders don't care about that as much as you think.

The second point is true, and something that will persist even with the system Loctav proposed anyways. If you are mapping a song that already has 5 ranked mapsets, you are making a choice to map it despite knowing it won't get the plays and popularity you expect. Honestly, if you are mapping the song after that many sets have been ranked, there are only two reasons. 1. You like the song, and you are mapping it for yourself. 2. You think you have an entirely new concept and want to get it ranked. If you picked the first choice, the "unattractive" qualities of mapping another set is something you are already expecting. You aren't mapping it for other people though, you're mapping it for yourself.

If you picked the second choice then you are effectively removing those unattractive qualities through creating unique content for that specific song.

Your system doesn't reflect how the modding and mapping community will view things. You ask these big questions like "When there are already ranked difficulties present for a set, how big will the motivation for a mapper be to map something that plays similar to an existing difficulty?
If there is only 1 similar one it might still be high. But what if there are 5 difficulties that play similar?" and those are valid questions. But your answer is very far removed from how active modders and mappers in the community will view them.

Not everyone is creating maps for the playerbase. Do you really think the playerbase needs 8 different versions of Harumachi Clover? yet mappers continue to create more sets. Maybe you should go ask mappers who create these sets for their opinion and reasoning, rather than assuming they are trying to create original and unique content and/or struggling with apparently having to create similar content. You'd be surprised how different maps actually are once you look through different submissions. It just seems to be a popular mindset nowadays to say "all maps play the same".
dhpenguin7
All I can say about mapping. I believe that a mapper should make 2 diffs at least for a map, because it is otherwise just getting a bunch of gd'ers to do your fill and you doing a short gd. Therefore, I believe at least 2 diffs are needed so that one diff map makers could put in more work, especially for tv size maps.
LwL
Had a random thought while reading the last few pages (I think everything relevant was said, definitely agree with Monstrate/CXu on most things, + what I said in my previous post).

Songs/maps cannot be modified to reach the minimum drain time.
If this goes through (in which case I think approval rules should be modified to be a less abrupt cutoff), I think it should be changed to say "If a Song/map is modified to reach the minimum drain time, spread requirements of the unmodified drain time apply". Just in case someone wants to edit an mp3 or the map because they feel it sounds better or represents the song better.
freddiiieeee
Colored text is mine.

Color Code:
  1. DISAGREE
  2. agree
  3. Everything else.

pishifat wrote:

Spread


Rules

All rules are exactly that: RULES. They are NOT guidelines and may NOT be broken under ANY circumstance. Let's see... let's see...

  1. Single-mode mapsets must include a reasonable spread of at least two difficulties. The lowest difficulty must be at least a Normal which complies with their respective mode’s difficulty-specific ranking criteria. This should say at most a Normal instead because the phrase "at least" can imply that a Hard difficulty can be used as the lowest difficulty, then people would get an idea that Insanes can be the lowest, and so forth.
  2. Hybrid mapsets without osu!standard difficulties must include a reasonable spread of at least two difficulties per mode. The lowest difficulty must be at least a Normal which does not break any difficulty-specific guidelines. Not really into the "two difficulties per mode" thing, but can't think of a better solution, really.
  3. If a hybrid mapset includes osu!standard difficulties...
    1. A reasonable spread of at least two osu!standard difficulties must be included. The lowest difficulty must be at least a Normal which does not break any difficulty-specific guidelines. Seems Good.
    2. Converted difficulties must form a reasonable spread. For example, a mapset with Easy and Normal osu!standard difficulties and an Insane osu!catch difficulty is not permitted. One or more additional difficulties may need to be added to fill the gap. It's alright, but it could be worded better; by converted difficulties, if you're meaning "osu!standard :arrow: CtB" and not something like "Mapset as is, difficulties between all game modes must not have large gaps in difficulty", then I can see the gap needing to be filled, but if the latter, then just leave the difficulties as are regardless of jumps like 2* standard :arrow: 4* Taiko
    3. Any two osu!taiko or osu!mania difficulties must be arranged in a reasonable spread. The lowest difficulty must be at least a Hard.
      Same wording issue as 1st rule.
    4. One osu!catch difficulty may be included. It must be at least an Insane difficulty.
    Same wording issue as 1st rule and rule above.
  4. Mapsets must have a minimum drain time of 30 seconds. This ensures each ranked map has a practical play-time. That's obvious.
  5. Marathons must have a minimum drain time of 5 minutes. This excludes especially long mapsets from requiring a spread of difficulties.
    This could be lowered, IMO. There are a great variety of songs with a length between 4-5 minutes that show so much potential without a need for a full spread.
  6. Songs/maps cannot be modified to reach the minimum drain time. Abusing the 5 minute limitation removes its intended purpose. Types of abuse include:
    1. Lowering a song’s BPM
    2. Looping portions of a song
    3. Adding sounds before/after a song begins/ends The first three are debatable because maybe the mapper fiddled with the song a certain way, liked it that way, and wanted to map that edited version.
    4. Extending spinners/sliders over inaudible sounds I agree; what's the point?
    5. Manually removing breaks Totally should stay unrankable, hands down.
    I'm polarized by the "no song modification" thing in general; I like it because it helps to combat such a petty way to cheat the ranking system, but I think that song modifications provide experimentation and variety all around. Personally, I'd lean more toward this rule if the minimum approval drain time were around 4 minutes instead, but if it is to stay at 5 minutes, then I'd be more against this rule.
  7. Mapsets cannot include more than 8 total difficulties of a single game-mode. The highest difficulty of a game-mode is not required to fit within a reasonable spread, so long as no levels of difficulty are skipped. Definitely not. This would essentially split many difficulties of a mapset (ex: 20 diffs into 3-4 different mapsets), which is exhausting to the GD mappers and maphost as they may split their work among each maps, leading to checking different threads for the mods instead of one, and for BNs as they'll need to icon check more maps instead of one. That last sentence is iffy, because it essentially means "Holy SH*T, I can make an 8 star difficulty right after my 5.5 star difficulty because of this!"; no don't hit those people with the "common sense" hammer, because it'll lead to a big mess in the threads, which could lead to more work from the moderation team.
  8. Excluding a mapset’s hardest difficulty, a difficulty’s name must accurately indicate its level of difficulty. Conventional difficulty names vary between modes, but any set of clearly progressive difficulty names can be alternatively used. Additionally, a mapset’s hardest difficulty should not use a name misrepresentative of its difficulty. Unnecessary rule as difficulty naming is up to interpretation of the mapper.
  9. A difficulty’s name must be unrelated to a username. Guest difficulties, however, may indicate possession with its mappers’ username or nickname. (e.g. Guest Mapper’s Insane). No problems here.
  10. Additionally, a mapset host cannot indicate possession in a difficulty’s name. (e.g. Mapset Host’s Insane). Metadata conflicts caused by mapping a song multiple times are an exception. Kind of obvious, considering difficulties without a name are automatically assumed the host's.
  11. A mapset host must have equal or more drain time mapped than any guest difficulty mappers. This is to provide credit where credit is due. Drain times for collaborative difficulties must be listed in the creator’s words for via storyboarding.
First sentence is obvious.
Second one needs some clean-up. "for via storyboarding." Wonder what that means? The mapper might not want a storyboard every time they host a collab, so maybe they should just put respective total drain times worked on in the "space provided".

Guidelines

Guidelines may be violated under exceptional circumstances. These exceptional circumstances must be warranted by an exhaustive explanation as of why the guideline has been violated and why not violating it will interfere with the overall quality of the creation.

  1. Avoid incomprehensible username combinations to indicate possession of a collaborative guest difficulty. If it’s unclear whose usernames are combined, simplification is recommended. True.
  2. Avoid difficulty names with descriptive elements not clearly related to a guest difficulty mapper or a level of difficulty. (e.g. Mapper’s Tragic Love Extra). A mapset’s hardest difficulty may use free naming, but clear and appropriate relation to its song is recommended. Once again,
    up to interpretation of the mapper.
  3. Usernames indicating possession of a guest difficulty should be consistent between multiple mapsets. Varying nicknames for one user makes interpreting who created a difficulty confusing. I think so, too.
  4. Avoid unicode characters in a difficulty’s name. These can cause errors with the beatmap submission system and problems for certain users when appearing in chat.
Don't know too much about this, so you tell me! :D

Well, that's all the feedback I can provide for you. Hoping for the best for the future of osu!mapping! :)

Once again, this draft is not the final result, as we need the feedback of the community first before getting it officially bumped into the wiki! It will be up to discussion for two weeks and close on the 18th of June! Also I'd like to reap my kudosu reward pls thx :P
dsco
Mapsets cannot include more than 8 total difficulties of a single game-mode. The highest difficulty of a game-mode is not required to fit within a reasonable spread, so long as no levels of difficulty are skipped.
I'd like to strongly push for this to be increased greatly or to be removed entirely. All this does is discourage grand ideas such as these beloved sets: 1 2 3 4 5
Should this rule pass I urge it to be raised to 10 though I do not believe this to be a sensible rule at all as I do not see its aim other than to reduce the workload for people checking maps, which shouldn't come at a cost of hindering mapsets.

Songs/maps cannot be modified to reach the minimum drain time. Abusing the 5 minute limitation removes its intended purpose. Types of abuse include:
I do not see the purpose for this rule. I think this would be much more sensible as a guideline and not a hard wall. Should a mapper take a 4:45 song and loop a section to make the map rankable, who is losing out here? I do not believe any person is being offended should this happen, and the community nets another marathon map which are the most well-liked and respected types of maps in the game. 37 of the 40 maps with the highest user rating in the game are marathons.
I think this should be a case-by-case guideline with bn's best judgement, as anything that discourages long songs from being mapped is a bad thing imo. if it can be done tastefully, i have trouble understanding why this could be an issue.
i'd also like to mention that this is a bit of a slippery slope as it would include any song compilations, combinations of songs, etc.
cisphobia
Songs/maps cannot be modified to reach the minimum drain time. Abusing the 5 minute limitation removes its intended purpose. Types of abuse include:
  1. Lowering a song’s BPM
  2. Looping portions of a song
  3. Adding sounds before/after a song begins/ends
  4. Extending spinners/sliders over inaudible sounds
  5. Manually removing breaks
I agree with dsco; this should be a guideline rather than a rule. If an mp3 file is a few seconds or smth away from a full 5 minutes, and the mapper does not want to create a full spread for a 4:45 min. long song, for whatever reason, whether it be laziness or the want to stray away from a laborious spread, they have 3 choices: find a new song to map that's shorter for a spread/long enough for approval, shorten the song to a length they feel is fine (which, imo, i hate when song lengths are cut, for example yuudachi no ribbon & another yuudachi no ribbon xd), or edit the mp3 to fit 5 minutes. Editing the mp3 is pretty much the choice to go for, especially if the song is 4:50 to 4:59 min. in length. It should be up to those that actually bubble & qualify the maps that decide whether or not an mp3 edit is appropriate/abuse or not; because there are definitely cases where it would be insanely inappropriate, like this being looped however many times to be 5 minutes long.
The only people losing out on a ~5 min. song not having a spread are those that cannot play the difficulty that is being approved; either way, under the guidelines right now exists this:
The song should not be too long. Aim for 3 minutes maximum; anything longer gets tiring.
I'm sure that many that play normal or even hard difficulties would struggle with a normal/hard map with a drain time of ~5 minutes, whether the map is approved without a spread or ranked with a spread. The whole mp3 edit rule should be a guideline, as it should be flexible to determine what is abuse and what is avoiding unnecessary spread.



also,
Mapsets cannot include more than 8 total difficulties of a single game-mode. The highest difficulty of a game-mode is not required to fit within a reasonable spread, so long as no levels of difficulty are skipped.
this serves no purpose aside from decreasing work load for modders/those that bubble/qualify, which shouldn't be a valid reason due to the fact that absolutely no one is required to mod/bubble/qualify a map that has a large amount of difficulties -- other maps exist that they could help with less checking required.
however, i do feel that the "highest difficulty not required to fit spread" section is fine, as long as the spread includes easy/normal up to an extra. cant be having 8 stars with a spread of normal to a 5 star insane. a 5 star insane in a spread that requires an extra would probably warrant an ~6(?) star extra, to keep the spread linear.
CXu

dsco wrote:

I do not believe any person is being offended should this happen, and the community nets another marathon map which are the most well-liked and respected types of maps in the game. 37 of the 40 maps with the highest user rating in the game are marathons.
Just for the record, but that statistic is probably skewed. You can only rate a beatmap if you pass it, and with only one difficulty on marathons, most people won't be able to pass it. People who will play the map with nofail to pass it are more likely to like the map than dislike it, since those who dislike it wouldn't bother playing through the map with nofail just to rate it.
Okoayu
closing this for revision then i guess we have more than enough content to go from
Topic Starter
pishifat
after an unnecessarily long wait, the draft is now up for revision once again! over the next two weeks (until August 31st), feel free to discuss anything concerning about the proposed subsection of the ranking criteria on the first post of this thread!

---

there were quite a few changes from the last round of revision, but here's the stuff most of you care about:
  1. rules stating "The lowest difficulty must be at least x" have been changed to "The lowest difficulty cannot be harder than x". there was a lot of confusion about how to interpret these, especially for hybrid mapset rules. hopefully it's understandable now
  2. the "Songs/maps cannot be modified to reach the minimum drain time" rule has been removed. while exploiting songs/maps to reach 5 minutes is still a problem, we've come to the conclusion that there's no practical way to enforce this rule and breaking it is often unnoticable. excluding it entirely seemed like the most realistic choice
  3. the "Mapsets cannot include more than 8 total difficulties of a single game-mode" rule has been removed. there are reasons for this rule, which is why it's been proposed on two occassions now, but mapping community members find it to be a problem. the 16 pages of discussion here explain why most people consider its benefits minor and its downsides substantial, so the rule is now gone

for the smaller changes and reasons why certain things were not changed, refer to the box below!

individual responses
quotes are paraphrased

>Hydria: calling key counts different modes is weird
according to mania players, this isn't too unusual apparently

>everyone: “At least” doesn’t work wording-wise. Applies to all of its uses
adjusted to "cannot be harder than x"

>Sieg: Shouldn’t the “which complies with their respective mode’s difficulty specific ranking criteria” line be the same in these following rules?
yes, applied

>Haskorion: Mention that E/N or N/H are minimum required spread
should be implied already since rule says "must include 2 difficulties" and "one cannot be harder than a normal"

>mangomizer: “The lowest difficulty [of each mode] must…”
yes

>Shad0w1and: Make # of difficulties required scale with length more smoothly t/432739
while it sounds cool in theory, setting time limits will likely cause more concern than the current marathon rule already does. the rule about modifying songs to reach approval limit is gone, but we still think it's better not try to use exploits to make maps above x minutes just to remove a bit of the workload

>ruruchewy: make marathon time limit lower
needs to be a line set somewhere, and 5 minutes is going to be that. (has been discussed in like 5 other threads in ranking criteria archive, so details kind of redundant)

>everyone: the rule about modifying songs to reach minimum drain time has too many holes
rule is now gone

>everyone: 8 difficulties is unnecessarily limiting
rule is now gone

>Hydria: Top difficulty of a mapset should allow any name
current rule only forbids misleading stuff, like calling an extra diff "easy." if that wasn't clear, it should be now

>everyone: "Drain times for collaborative difficulties must be listed in the creator’s words for via storyboarding." has a typo
typo was fixed, then sentence was removed beacuse people will lie about drain time if they're breaking the rule anyway

>UndeadCapulet: If guest difficulty creators are okay without mapset host credit, it should be okay
point was discussed a lot, though it ended up not changing. regardless of whether or not a mapset host is okay with it, the mapset will still be labeled as theirs on listings/profiles etc so they need to have some level of accountability, and drain time is the method we've agreed is most fair for that (despite it still sucking)

>Phantomhive: Shouldn’t accountability be calculated with object count instead of drain time?
Endaris explained this already, but different object types make that impractical

>Natsu: is “Pantsu” diffname ok?
yes!

>Hydria: "A mapset’s hardest difficulty may use free naming, but clear and appropriate relation to its song is recommended." is redundant
yes, removed sentence from second guideline

>Hydria: What about name changes?
they are a valid reason to break the third guideline

>UndeadCapulet: If mapper doesn’t care about credit, it shouldn’t matter.
point is to avoid misleading/unnecessary name stuff, since it makes actual names have less meaning. added clarification in second sentence of guideline

>Yuii-: Just make unicode guideline a rule if it can break the game
as is with in-game chat, excessive unicode is the problem, so there's times where unicode is/is not ok

>Xexxar: Difficulty naming for Extras is arbitrary. should use free naming
oko addressed this in the thread you wrote about this actually

>Cyclohexane: Make wide spreads acceptable, here are pictures
your concern is more of a mode-specific thing. rules here don't define what's acceptable for easy/normal/hard etc, sorry:(

also:
difficulty names are italicized for easier reading
glossary entires mostly start with "A"
extremely minor wording adjustments
Nao Tomori
what happened to the top diff doesnt need to follow spread thing =( that was more or less completely separate from the 8 diff thing, is it still being considered?
Amaikai
So nothing done on cumulative effort needed to rank 3-4minute maps being unreasonably larger and strict 5minute rule still in place? neat.
Monstrata

Naotoshi wrote:

what happened to the top diff doesnt need to follow spread thing =( that was more or less completely separate from the 8 diff thing, is it still being considered?
There are some rules associated with that too, like there still having to be a spread. You can't have a E N H spread and then a 9.0 star diff slapped on the end. The discussion basically resulted in the "minimum" difficulty being an Extra (E N H I X >--- Ultra) but we rarely will see something like this anyways because not many people are actually capable of mapping difficulties so much higher than 5.25 that they actually "skip" a spread tier", and even fewer are capable of ranking.

5.25 > 6.75 can still be seen as a reasonable spread. so you should really only see the rule (if we had included it) being used for a 5.25 > 8 star spread or something.

I think somewhere in the discussion we just decided it wasnt a bad idea but not really something we'd see. It doesn't allow 7-9 star diffs to become any more rankable than they were. You wouldn't be able to quote it and say "well this rule says I can rank my 9 star diff". It just allowed a diff in between to be skipped, but the highest diff still goes through normal nomination procedures and doesn't receive exemption
Xinnoh
thread seems to be way less active than before because no one is aware of it .-.

pishifat wrote:

  1. Any two osu!taiko or osu!mania difficulties must be arranged in a reasonable spread. The lowest difficulty cannot be harder than a Hard.
  2. One osu!catch difficulty may be included. The lowest difficulty cannot be harder than an Insane.
This would imply that there's a maximum of two taiko/mania diffs and only one ctb allowed. "One or more ctb diffs may be included" is fine

Another thing, regarding the second half of that, it really doesn't make sense that something like this is technically rankable

In my opinion, that needs to be revised to mention that the Ctb spread should still be reasonable after the std are converted

I also still think the current situation for maps that are 3-4.99 minutes long is really bad. I don't see why a normal or hard diff can't be skipped on longer sets, since by now there are more than enough low diffs for any new player to play on. If low diff was only required on 3> minutes, TV size is popular enough that they'd still bring enough low diffs for new players
then again this second part was added after the qat had their meeting so I probably didn't really add anything ;-;
Amaikai
>Shad0w1and: Make # of difficulties required scale with length more smoothly t/432739
while it sounds cool in theory, setting time limits will likely cause more concern than the current marathon rule already does. the rule about modifying songs to reach approval limit is gone, but we still think it's better not try to use exploits to make maps above x minutes just to remove a bit of the workload
Can someone clarify why exactly this will "likely cause more concern" than current rule for flat time limit? I mean "we still think it's better not try to use exploits to make maps above x minutes just to remove a bit of the workload" makes it sound like it was best among the worst.

Monstrata wrote:

Naotoshi wrote:

what happened to the top diff doesnt need to follow spread thing =( that was more or less completely separate from the 8 diff thing, is it still being considered?
There are some rules associated with that too, like there still having to be a spread. You can't have a E N H spread and then a 9.0 star diff slapped on the end. The discussion basically resulted in the "minimum" difficulty being an Extra (E N H I X >--- Ultra) but we rarely will see something like this anyways because not many people are actually capable of mapping difficulties so much higher than 5.25 that they actually "skip" a spread tier", and even fewer are capable of ranking.

5.25 > 6.75 can still be seen as a reasonable spread. so you should really only see the rule (if we had included it) being used for a 5.25 > 8 star spread or something.

I think somewhere in the discussion we just decided it wasnt a bad idea but not really something we'd see. It doesn't allow 7-9 star diffs to become any more rankable than they were. You wouldn't be able to quote it and say "well this rule says I can rank my 9 star diff". It just allowed a diff in between to be skipped, but the highest diff still goes through normal nomination procedures and doesn't receive exemption
Is this convention going to be listed somewhere for reference? Since:

Reasonable Spread: A mapset without drastically large differences between difficulties as dictated by difficulty-specific rules and guidelines.
Leaves a lot of leeway for interpretation.
Vacuous
I think it's really nice that they removed the "no editing the audio file to reach minimum drain time". But I did see where they were coming from, I think it's a lot better as a guideline so if BNs see a map that's say, Harumachi Clover looped 5 times they can choose to not bubble it without having the mapper say "but it didn't break any rules.
TL;DR it's better as a case-by-case guideline
UndeadCapulet

Vacuous wrote:

I think it's really nice that they removed the "no editing the audio file to reach minimum drain time". But I did see where they were coming from, I think it's a lot better as a guideline so if BNs see a map that's say, Harumachi Clover looped 5 times they can choose to not bubble it without having the mapper say "but it didn't break any rules.
TL;DR it's better as a case-by-case guideline
i mean
they can still just not bubble it
Vacuous

UndeadCapulet wrote:

Vacuous wrote:

I think it's really nice that they removed the "no editing the audio file to reach minimum drain time". But I did see where they were coming from, I think it's a lot better as a guideline so if BNs see a map that's say, Harumachi Clover looped 5 times they can choose to not bubble it without having the mapper say "but it didn't break any rules.
TL;DR it's better as a case-by-case guideline
i mean
they can still just not bubble it
yeah but there are those annoying people that don't understand BNs can just choose not too and complain because of something that most people would agree with the BN on
xxdeathx
I'm pleasantly half surprised that you guys actually listened and removed the shitty 8 difficulty limit rule.
Topic Starter
pishifat
august 31st was 3 days ago wasnt it
Topic Starter
pishifat
we're back! this draft will be open for one more week of revision, ending on September 24th!

---

little feedback this time, so only a couple changes:
  1. added "or more" in two rules as Sinnoh suggested
  2. clarified how inactive/disagreeing guest difficulty mappers are treated with the addition of the last rule


other feedback is addressed in the box here:
SPOILER

quotes are paraphrased

Sinnoh: In my opinion, that needs to be revised to mention that the Ctb spread should still be reasonable after the std are converted
>this rule already exists. it's the second thing under "If a hybrid mapset includes osu!standard difficulties..."

Naotoshi: what happened to top diff spread gap?
>Monstrata explained on the thread

Amaikai: can we have lower spread requirements for 3-4 minute maps in the same vein as marathons? why were they rejected from previous round of revision?
>people exploit 5 minute limit to make songs/maps pass threshold, setting similar things for 3-4 minutes makes that happen more in different situations

Amaikai: what are spread regulations?
>explained on difficulty-specific sections of the ranking criteria (not what this draft is for)

Vacuous: can we make editing songs a guideline?
>not really practical when the quality of an edited song has so many variables/is not easily detectible. the example you gave would never be ranked regardless of a rule/guideline being in place, so i don't think that's much to worry about



if no major changes pop up in the next week, we'll transfer this draft to the wiki!
Nao Tomori
Hi
Regarding the diffname - username thing (diffname cannot be related to username)

Can we please clarify if this is applying to any and all diffname or just ones in which that diffname would be illogical. I ask this because this rule was implemented to stop people just putting their username as the top diff and calling it a day (lots of older Chinese mappers did this). However I believe that it should be allowed to use a username-esque diffname if the diffname would be allowed if it weren't related to a username - aka, on a guilty kiss map diffname "kisses" would be fine, a musaigen no phantom world map diffname "shadow land" would be allowed, etc etc. Atm it seems unnecessarily restrictive for no real reason.

My suggestion is to change the rule to say that, while the top diff does not need to use a difficulty level, it must be related to the song in some way.
tatatat

pishifat wrote:

we're back! this draft will be open for one more week of revision, ending on September 24th!

---

little feedback this time, so only a couple changes:
  1. added "or more" in two rules as Sinnoh suggested
  2. clarified how inactive/disagreeing guest difficulty mappers are treated with the addition of the last rule
other feedback is addressed in the box here:
SPOILER
quotes are paraphrased

Sinnoh: In my opinion, that needs to be revised to mention that the Ctb spread should still be reasonable after the std are converted
>this rule already exists. it's the second thing under "If a hybrid mapset includes osu!standard difficulties..."

Naotoshi: what happened to top diff spread gap?
>Monstrata explained on the thread

Amaikai: can we have lower spread requirements for 3-4 minute maps in the same vein as marathons? why were they rejected from previous round of revision?
>people exploit 5 minute limit to make songs/maps pass threshold, setting similar things for 3-4 minutes makes that happen more in different situations

Amaikai: what are spread regulations?
>explained on difficulty-specific sections of the ranking criteria (not what this draft is for)

Vacuous: can we make editing songs a guideline?
>not really practical when the quality of an edited song has so many variables/is not easily detectible. the example you gave would never be ranked regardless of a rule/guideline being in place, so i don't think that's much to worry about


if no major changes pop up in the next week, we'll transfer this draft to the wiki!
I don't see any changes on the draft.. ??
Topic Starter
pishifat

tatatat wrote:

I don't see any changes on the draft.. ??

oops, should be good now


Naotoshi wrote:

My suggestion is to change the rule to say that, while the top diff does not need to use a difficulty level, it must be related to the song in some way.

yeah, we'll discuss that
Pho
A mapset host must have equal or more drain time mapped than any guest difficulty mappers. This is to provide credit where credit is due.
Does this also include collaboration marathon maps? I had a discussion on Spaghetti's Boogie map about a year ago (p/5243137) concerning this issue where he mapped less drain time than one of the guest collaborators, justifying it with the fact that he was managing the whole project. Maybe some clarification on that part would be nice.

A mapset host and guest mappers can make changes to their respective difficulties as they wish. If there is a disagreement between both, the mapset host must delete the guest contribution upon request. If a guest mapper cannot be contacted, they will be assumed to agree with any changes.
Shouldn't there be a minimum waiting time included before the host is allowed to make changes to the map? Unless it's very clear that the guest mapper is inactive (e.g. offline since a year ago) I think there should be.

Just my two cents.
Amaikai
[quote=Pishifat]
Amaikai: what are spread regulations?
>explained on difficulty-specific sections of the ranking criteria (not what this draft is for)
[/quote]
Weight was on "drastically large differences" not what each specific difficulty is supposed to contain. Wording used isn't objective and I wouldn't know if (easy),(normal),(hard),(insane)5, 6.5, 8 spread is okay just by reading the rule. So let me ask with different wording:
Are there plans to make document of what bn's are looking for when they browse through maps that are NOT included in rules or guidelines but are standard practises instead? Something like "Introduction to BNG", "BNG and You", "Is my map ready for BN senpai".

As for spread:
I would have hoped to reduce incentives to abuse the system instead of saying "we can't have do that because A is abused. If we make B it will also get abused". I see that as shallow reasoning, it hasn't been tested how changing 2 step system 3 step system affects abuse aka mp3 stretching. Does abuse increase because it's now possible to edit 2:40 songs to 3min and do less strict spread? OR does abuse decrease because 4:40 songs fit on less strict spread and incentive to make it 5minute is less? Or does the amount of abuse stay same? What effects this abuse has to players, mappers, bns?  I find having a gut feeling as justification for rule to be unacceptable.

There were also other alternatives suggested on thread, why would, for example, why these be no good on fundamental level?

Reference table

  • Difference between lowest and highest difficulty in star rating is calculated.
    This difference and song length is matched on 2 dimensional table to see number of required difficulties.
    Mapper maps that amount of difficulties with linear/gradual spread.
    Reference table is absolute, it's up to mapper if he wants to extend the mp3 or nerf highest difficulty to map 1 difficulty less.
Supply and demand

  • Number of difficulties within specific range on ranked mapsets are counted
    When there is oversupply of specific difficulty range, new mapsets can exclude that difficulty range from their set
    When there is undersupply of specific difficulty range, bonus incentives to map specific difficulty range.
    Bonus: You can append difficulties to already created sets later on to fill in the gaps.
Difficulty pack

  • Beg peppy to allow different songs in same mapset
    Concept is to have set with multiple different songs in it
    Set has difficulties which are around same range
    Packs can be for beginner, advanced, insane, extra, ultra
    Only X amount of sets for Y range can be ranked each month
    Exceptional sets can go to loved if they don't get ranked
    (Get some BN collaboration to decide which packs get ranked)
Pool of maps

  • Have contest to map single song.
    Make spread for mapset from entries, pick argueably best for each range
    Rank it, exceptional single difficulties go to loved
    In case of extremely popular song, make 2 sets from the map pool
    (Get some BN to handle this)
LowAccuracySS
Gonna go through your suggestions on these and state at least why I wouldn't want them. My answers are in red.

Amaikai wrote:

Are there plans to make document of what bn's are looking for when they browse through maps that are NOT included in rules or guidelines but are standard practises instead? Something like "Introduction to BNG", "BNG and You", "Is my map ready for BN senpai". The thing is, that opinion can change from map to map. There's no real defined thing BNs would be looking for as all they do is hold a map to the ranking criteria and add normal mod suggestions. Adding this wouldn't honestly help and make the process even more confusing!

I would have hoped to reduce incentives to abuse the system instead of saying "we can't have do that because A is abused. If we make B it will also get abused". I see that as shallow reasoning, it hasn't been tested how changing 2 step system 3 step system affects abuse aka mp3 stretching. Does abuse increase because it's now possible to edit 2:40 songs to 3min and do less strict spread? OR does abuse decrease because 4:40 songs fit on less strict spread and incentive to make it 5minute is less? Or does the amount of abuse stay same? What effects this abuse has to players, mappers, bns? I find having a gut feeling as justification for rule to be unacceptable. The only thing reducing the approval-length time does is cause people to time-strech to 4 minutes instead. I get that there might be a mapper who wants to map a 4:50 song but doesn't want to go through the hassle of mapping a full spread, but it's just tough luck to be honest. Your gut feeling has no stance in an argument and I could say I have a gut feeling that we should make the approval length 10 minutes instead but nobody would agree with that because I have no proof that the amount of abuse would be lower. Lowering it to 3-4 minutes would also be more harmful to mappers because it advocates for low-effort, TV size sets and it makes mappers who do map songs over 3-4 minutes put less effort into songs over 3 minutes because "oh, I only need to map a single diff for this song". 5 minutes is enough to put effort into creating a map worth a single difficulty and anything below it is still short enough to map normally. In the end it comes down to this- do you want to put in the effort or not? If you don't, go ahead and map TV size. If you do, go map those 4 minute songs with full spreads and over 5 minute songs.

i'll get to the rest later
Amaikai

[ Space ] wrote:

Gonna go through your suggestions on these and state at least why I wouldn't want them. My answers are in red.

Are there plans to make document of what bn's are looking for when they browse through maps that are NOT included in rules or guidelines but are standard practises instead? Something like "Introduction to BNG", "BNG and You", "Is my map ready for BN senpai".
The thing is, that opinion can change from map to map. There's no real defined thing BNs would be looking for as all they do is hold a map to the ranking criteria and add normal mod suggestions. Adding this wouldn't honestly help and make the process even more confusing![/Quote]

You are missing the point. Bns are diverse bunch and each have their own tastes, I agree with that. But doesn't change how there are too many unsaid things NOT listed on guidelines or rules that you need to figure out on your own.

I would have hoped to reduce incentives to abuse the system instead of saying "we can't have do that because A is abused. If we make B it will also get abused". I see that as shallow reasoning, it hasn't been tested how changing 2 step system 3 step system affects abuse aka mp3 stretching. Does abuse increase because it's now possible to edit 2:40 songs to 3min and do less strict spread? OR does abuse decrease because 4:40 songs fit on less strict spread and incentive to make it 5minute is less? Or does the amount of abuse stay same? What effects this abuse has to players, mappers, bns? I find having a gut feeling as justification for rule to be unacceptable. The only thing reducing the approval-length time does is cause people to time-strech to 4 minutes instead. I get that there might be a mapper who wants to map a 4:50 song but doesn't want to go through the hassle of mapping a full spread, but it's just tough luck to be honest. Your gut feeling has no stance in an argument and I could say I have a gut feeling that we should make the approval length 10 minutes instead but nobody would agree with that because I have no proof that the amount of abuse would be lower. Lowering it to 3-4 minutes would also be more harmful to mappers because it advocates for low-effort, TV size sets and it makes mappers who do map songs over 3-4 minutes put less effort into songs over 3 minutes because "oh, I only need to map a single diff for this song". 5 minutes is enough to put effort into creating a map worth a single difficulty and anything below it is still short enough to map normally. In the end it comes down to this- do you want to put in the effort or not? If you don't, go ahead and map TV size. If you do, go map those 4 minute songs with full spreads and over 5 minute songs.
1. Read what I wrote again with thought. I claimed keeping current 5minute rule is based on gut feeling.

2. Also why it should be "tough luck" if you want to map 3-4 min map? Thats flatout encouraging people to map shorter or longer.

i'll get to the rest later[/quote]
Don't bother and just read appendum.

As appendum: I don't expect the 5 min rule to change this late in the rule draft proposal and i'm mainly pointing out my frustration about sticking to same old "working" method of doing things and ignoring / not providing reasons why alternatives weren't viable. Hoping 3-4 minute maps get some love later on in next rule revision or in form of getting ranked by more flexible means.
Topic Starter
pishifat
aaaand closed again
Topic Starter
pishifat
i can't bubble threads on the new forum, but i'm here to announce that this draft will be amended soon!

in response to the last short round of revision, we clarified naotoshi's and pho's concerns in their respective rules. the additions are in bold:
  1. A difficulty’s name must be unrelated to a username. Guest difficulties, however, may indicate possession with its mappers’ username or nickname. (e.g. Guest Mapper’s Insane). Words that happen to be usernames are acceptable within difficulty names as long as they relate to the song.
  2. A mapset host and guest mappers can make changes to their respective difficulties as they wish. If there is a disagreement between both, the mapset host must delete the guest contribution upon request. If a guest mapper cannot be contacted for a month, they will be assumed to agree with any changes.

considering these were both minor changes, we'll be moving this over to the wiki rather than going through another round of revision. for future changes to this subsection of the ranking criteria, check out this thread: https://osu.ppy.sh/community/forums/topics/644446
Okoayu
lmao the topic was still locked, unlocked topic
Lorkee
creativity \o/
UndeadCapulet
couple quick things:

rc wrote:

A mapset host must have equal or more drain time mapped than any guest difficulty mappers. This is to provide credit where credit is due.
current wording of this disallows gd'ers from mapping intros/outros/break that the set owner didnt map if the set owner only mapped one difficulty. theres no real reason to disallow this kind of thing, since it just boils down to different song interpretations.

rc wrote:

Single-mode mapsets must include a reasonable spread of at least two difficulties. The lowest difficulty cannot be harder than a Normal and it must comply with its respective mode’s difficulty-specific ranking criteria.
i think this was just carryover from how the pending section functioned? thats been patched tho, and this seems an unneeded restriction. there is a not-insignificant portion of mappers that like mapping rly slow songs, who have been required to add an excessively easy difficulty to their mapset just so they can rank their lowdiffbased set. i think its better to just say "the lowest diff of a set must be Normal". theres no reason to force a set to have two easyicon difficulties like we're doing rn
Pachiru

UndeadCapulet wrote:

A mapset host must have equal or more drain time mapped than any guest difficulty mappers. This is to provide credit where credit is due.

current wording of this disallows gd'ers from mapping intros/outros/break that the set owner didnt map if the set owner only mapped one difficulty. theres no real reason to disallow this kind of thing, since it just boils down to different song interpretations.
I agree with UC's opinion, everyone has different perception about a song, some might prefer to map a part that other don't. I think that the rule about the amount of diff made by the mapper compared to the GDers is enough in my opinion. No need to restrict mappers creativity by keeping a rule that is not really important in my opinion.
J1NX1337

rc wrote:

Additionally, a mapset host cannot indicate possession in a difficulty’s name. (e.g. Mapset Host’s Insane). Metadata conflicts caused by mapping a song multiple times are the only exception.


"Metadata conflicts caused by mapping a song multiple times are the only exception."
I find this sentence a bit vague and potentially misleading from the perspective of someone who hasn't been beforehand aware of the issue/bug referred to here. Maybe reword it to something like "Metadata conflicts caused by a mapset host mapping multiple mapsets of the same song with identical difficulty names are the only exception." for the sake of clarity.
Okoayu
they can just map one more diff or swap hosts with someone that has more mapped than they do, if they truly dont care about putting that much effort into their thing i dont know man

as for the metadata conflict thing, fair enough
Topic Starter
pishifat
thread moved because finalized amendment
Please sign in to reply.

New reply