Map Ratings

[WaterKnight]

As a user, I can rarely recognize what maps are good on Hiveworkshop from the ratings. Most users just upvote everything they look into/care to pay attention to. It easily hits the limit and is not comparable to other contents of the same genre any longer, which should be the criteria.

The target scale of 1 to 5 thumbs in an overall-value is very imprecise. If everyone sees the map as a 3.6, they have to go with 4 thumbs. If everyone sees it as a 4.4, they still have to go with 4. I dare to guess that most voters do not initially tend to full extremes, at least not to the negative one because that seems offensive and the upper one could be reached because 4.6 is still a 5.

Also, people around the web that are not fangirling often seem to judge contents from the committed mistakes. A map can have a lot of small and big flaws. The question is if the remainder makes up for it. Tiny maps gain a lot of praise not because they are more worthwhile than big variants but because nobody sees grave mistakes in them. This is why another rating could be introduced that estimates a map's epicness, the map's resulting value/windage for criticizing virtually.

Furthermore, I advise one of the following:

1. Couple a voter's reputation to his/her voting weight. This is because Hive deems the reputation trustworthy to evaluate a user's potency and benevolence and it's more central and intuitive than to throw in a new attribute.

2. Force every vote to refer to a post in the thread in order to prove a bit of reasoning and effort. After all, the gauges of individuals diverge hard.

---

Now to the moderators' side.

http://www.hiveworkshop.com/forums/maps-564/bsa-v1a-ai-245114/

I do not understand the review scorings of this type, which are quite frequent. The formula

Total Score: (Gameplay[x/5] + Terrain[x/5] + Management[x/5] / Total Score[15] * 50 + 50)

(yes, the sum should be parenthesized)

implies that any map gets 50 percent without condition and up to another 50 percent for the designated ratings. What is the advantage of that? This leads to the map here gaining 83% and 4 thumbs, although the average thumb count is 10/15 -> 66% and the resolution is still bad.

You can also see that the moderator acknowledges a special feature under Terrain as "really good". And still grants less points than for gameplay where this is not quite the case. So this does not seem to be a complete reasoning for the rating, only additional hints. That's okay but should be evident that way.

In general are special awards recommendable. If a map possesses an outstanding feature, of course it should be highlighted and that is a corner stone for the reader to grasp a firm idea of what to expect. So maybe emphasize it with a star.

 

[Xonok]

I'll just note that it won't do good to make rep have any actual value other than what people give it.
If it would be used for voting weight, then popularity will be more important than actually knowing what you talk about.

 

[WaterKnight]

That's the general problem with reputation. I only suggested that because Hive already has it and seems to pledge to it. At least it attests a certain experience around the site. Then propose another factor.

 

[Xonok]

It's a tough call. I guess it's the classical question of how to find "worthy" people among the unworthy.
Probably the best results would come if someone moderated the general usefulness of a person, but that is bound to be somewhat biased. It also means that there would have to be more work done by mods, which I don't consider good at all.
My dream about this is that there would be some kind of honor system, like a person can be marked as "helpful" and if they get marked as that often enough and by different people, then it is considered to be true. There could be different kinds of honor, like "Funny" for the off-topic trolls.

However, my approach would still contain the same fundamental issue - There is no way to make sure the system is actually used how it is meant to be.

 

[Orcnet]

The computation that I do on my current moderation are practically made by me, it somehow worked smoothly than just giving a map a flat rate of 3 or 5. Hive did not follow my computation nor any other moderator would do.

Couple a voter's reputation to his/her voting weight. This is because Hive deems the reputation trustworthy to evaluate a user's potency and benevolence and it's more central and intuitive than to throw in a new attribute.

Reputation, you mean by a user's personal information among the community or the glowing badges/gems we have? I'll have to down vote for that idea, I mean it won't work out what if the user who has the most largest rep suggested a map to be rated DC, wouldn't that sound bad? well not all users who have rep doesn't mean it was helpful. I myself have rep received from other users because they're spreading it, think I'm funny, etc.

Force every vote to refer to a post in the thread in order to prove a bit of reasoning and effort. After all, the gauges of individuals diverge hard.

Why forcefully vote it if everyone is just freely to do what ever they want? if your planning for a poll, criticism and enough speculation work is enough to rate a map.

 

[WaterKnight]

Wait, what happened to our private messaging again?

You said you were not the reviewer. (

I never use the letter E(its probably ChaosOrcnet99's moderation

) Have you seen my last response?

Anyways I simply designed the computation system myself, nearly all map mods do not do that, never. I'm the only one who did such change and by so I realized it is quite useful, also I also use this during my student teacher work and much to surprise it was that simple.

As said, it gives the impression that I do not have to do anything and still get an evaluation that is commonly understood as mediocre (50%). Are you disapproving anything that is below this threshold, to simply mark it as failed? If on the other hand one thumb counts as mediocre/half the way there, then the other thumbs are moved together more closely in meaning, so there is less of a difference between them. The resolution would be better then for this higher area, which throws the question if the maps really score that high.

Just asking why you did not pick the more intuitive method of having one thumb as the lower limit from 0%, five as the max to 100%. Well okay, 0 thumbs would be even better for 0%, because it depicts there is nothing good in it but Hive does not allow a rating of 0 thumbs. So every thumb would be worth a range of 100%/5=20%. 0-20% -> 1, 21-40% -> 2, 41-60% -> 3, 61-80% -> 4 81-100% -> 5

[...]a map to be rated DC[...]

DC?

Why forcefully vote it if everyone is just freely to do what ever they want? if your planning for a poll, criticism and enough speculation work is enough to rate a map.

Please elaborate.

Maps en mass here hit the upper limit/shortly below it and thereby seem to be the work of a genius that could not soar higher, when in reality, they may not even be mediocre. A major problem is that a lot of maps only get one or a few votes, probably by friends of the author.

 

[fladdermasken]

Funny how rep has weight enough that you don't want it in OT, but not any at all when you discuss actually making use of it. DISCREPANCY MUCH?

 

[Xonok]

Funny how rep has weight enough that you don't want it in OT, but not any at all when you discuss actually making use of it. DISCREPANCY MUCH?

Not allowing it in OT is preventive, so that we'd have a chance to have any meaning for rep.
It should not be assumed that there always is or isn't.

 

[SA Dashie]

Could always use the "reviewed rating" again.
Ratings that would be approved will be added to the main ratings while all the other votes will be mashed up together in a rating.

Alot of users and mostly new or not active users around the forums usually just give maps a 5/5 because they like it..
Which is also irritating cause there is usually some random that rates ones map 2/5 without reason, so an approve system for voting could help.

Every users posts that contain rating could notify one of the moderators.
Imo with this system it shouldn't be to tedious.

 

[Ralle]

I don't think we should give rep more value than it already has..

I know people don't like to hear this, but it seems like rating systems are inherently broken. Most of the time people give 1-bad or 5-good, nothing in between. I know there are some people who are good at giving ratings in-between, but the majority of people give either 1 or 5 stars.
Here are all the stats:

1    5593    8,93 %
2    2721    4,34 %
3    5488    8,76 %
4    11625    18,55 %
5    37228    59,42 %

I would suggest that we add a good/bad system (like/dislike) sort of like what YouTube has. It seems like most systems are going that way these days anyway.

For directors cut I am planning a special tag in the new system, so we won't lose that.

 

[WaterKnight]

I don't think we should give rep more value than it already has..

What value does it have?

I would suggest that we add a good/bad system (like/dislike) sort of like what YouTube has. It seems like most systems are going that way these days anyway.

Well, those are for simplicity. But I do not think that people will actually straight dislike non-offensive content. So it comes down to how many likes a map gains. The more popular a map is, the chance actually rises that stray downvoting people show up.

You could say that the people shall like/dislike the way that the average comes closer to the mark they want. Ex: The map has currently 95%, the voter likes the map but only as much as a personal 90%, so s/he would click the dislike. But then, who would obey such rules? It also gives more power to the later voters.

I do think that a higher resolution like ten thumbs would be useful because it would make the extremes appear more distinctive, reverent. Also add something like a color gradient or some display more evidently denoting the meaning.

 

[bhusta]

Firstly I second WaterKnights thoughts about Orcnets' voting system, but one has to recognize and respect that it is a shitload of work to objectively review a map. And there are many maps in the pending zone I would never want to touch
The good thing is that he has a system. Sure it is improveable.

What value does it have?

Imaginary ones
It should be some kind of reward for helpful posts but who cares?
I learned that there are more people who can actually help you with low reputation as the once with thousands of points.

I do think that a higher resolution like ten thumbs would be useful because it would make the extremes appear more distinctive, reverent. Also add something like a color gradient or some display more evidently denoting the meaning.

Your ideas are good but think about the members you would give this voting oppotunity.
I am extremly generalizing but I think the majority of members are not qualified to vote like this.

I would suggest that we add a good/bad system (like/dislike) sort of like what YouTube has. It seems like most systems are going that way these days anyway.

This is the best thing you can to in an open social community like this. It is very mainstream concept but it sure has its' reasons.
For example the easynes encourages that the users vote in the first place. It sounds ridiculously to say that the option to choose between 0-10 points is a barrier but it assumes one or two seconds for the user to think about his or her decision. These seconds are the reason why people just vote 1 or 5.

 

[I3lackDeath]

Oh look it's this thread again.

Fascinating how people confuse topic X with something entirely different, in this case "reputation". Would be great if the various colours of Play-Doh could be kept separate as long as possible (this time?).

Additionally, the cascade for topics like this will always be the same, leading down to the same core problem(s).

Objectively, it's only natural most people use either extremes of a palett only.
Anything else requires more energy and more (critical/analytical) thinking.
Is a graded rating system more/less useful/working because of it?
The answer is: no.
The goal's the same - to separate "good" from "not so good" items. So the system will fulfil its function either way.

Authors wanting proper feedback (through ratings) is an entirely different issue, and a site like this is not built towards that goal.
As always, I'd say "take/work with what you get".

[Note that one has already been over the problematics of the current rating system, one may want to (re-)read what Mr. Ralle stated in that regard.]