• 🏆 Texturing Contest #33 is OPEN! Contestants must re-texture a SD unit model found in-game (Warcraft 3 Classic), recreating the unit into a peaceful NPC version. 🔗Click here to enter!
  • It's time for the first HD Modeling Contest of 2024. Join the theme discussion for Hive's HD Modeling Contest #6! Click here to post your idea!

Ratings System

Status
Not open for further replies.
Level 22
Joined
Feb 6, 2014
Messages
2,466
Why is the ratings system based 5? Isn't that too discrete? Why not based it on 100 and having a rating above 60/100 equals approval. That way, players can really compare the quality because sometimes, even though both resource has the same mod rating, there is still a significant difference between the two.
 
Okay, I see the point. And let's see... DC got 120/100 :D

Map Section : This means that Hive Map Reviewers most likely has to revert to Kobas' system or create something similar to it, since back then it uses a base of 100 (then converts to the base of 5 to fit with the system) which fills this completely, but the current one (where we use our own styles of review) would be harder to use in this environment, unless of course, we adapt our own styles for the new rating system.

Spell Section : it's difficult to rate, even at scale of 5, what about a scale of 100? I doubt it will make life easier. Spell (&System) are hard to review, as being very complex to rate, and not to mention those over 2K long Jass codes or those long, annoying GUI triggers.

Skins, Model & Icon Sections : I have no comment here, as I lack expertise on these areas.
 
Level 22
Joined
Feb 6, 2014
Messages
2,466
Map Section : This means that Hive Map Reviewers most likely has to revert to Kobas' system or create something similar to it, since back then it uses a base of 100 (then converts to the base of 5 to fit with the system) which fills this completely, but the current one (where we use our own styles of review) would be harder to use in this environment, unless of course, we adapt our own styles for the new rating system.

What?? So this was the Rating System before?? Then why change to base 5?
 

Ash

Ash

Level 22
Joined
Aug 27, 2005
Messages
1,684
Yeah, I agree with Ralle here. I don't think it's too discrete, but I do think that a rating of x/100 provides too much noise. I wrote a guide on how I rated things as a moderator, I wonder if you could help provide any more advice?
 
Level 23
Joined
Apr 16, 2012
Messages
4,041
the problem with having only 5 points is that you kind of stereotype maps regarding ratings. For instance, not every 4/5 map is really 4/5, but some people say 3.75/5, but give you 4 because there is no way to say 3.75.

That being said, 100 points is way too much, but maybe going back to 10, and mods using 5 points still(moderation could be set to something like some mods use right now, 5/5 amazing, 4/5 outstanding, 3/5 average, 2/5 lacking 1/5 shit, but maybe not so strict)?
 
Level 22
Joined
Feb 6, 2014
Messages
2,466
the problem with having only 5 points is that you kind of stereotype maps regarding ratings. For instance, not every 4/5 map is really 4/5, but some people say 3.75/5, but give you 4 because there is no way to say 3.75.

That is exactly the problem. It is not always true.
Some have 2.5/5 -> 3/5
some have 3.4/5 -> 3/5
Same rating bit huge difference.
 

Ash

Ash

Level 22
Joined
Aug 27, 2005
Messages
1,684
That is exactly the problem. It is not always true.
Some have 2.5/5 -> 3/5
some have 3.4/5 -> 3/5
Same rating bit huge difference.

Okay, so I'm going to give you a proper answer about why I think the 5-thumb system's pretty good. I am only doing this because I'm in the lab, and don't want to do any work. To begin, and to answer the question you're posing, generally a 2.5 map was still a 2 in my books, and a 3.4 map still a 3. Anyway.

I think the way you're examining a ratings system is wrong. It is not a case of simply saying a /100 is better than a /5, as better/worse judgements are misleading - it is instead two different ways to optimise a system for users. You should really be asking "how can we make this system accessible, reliable and beneficial?" The obstacles you're attempting to overcome will define how you address barriers and, eventually, how the solution is proposed.

The question isn't "is this system better than that system", but "why do I have a rating scale in the first place?" "What is the point in me including this on the site?" Now, I'm not Ralle, but if I led a community of designers, creators and gamers, then I would have a rating scale to try and not only distinguish what makes good different to bad, but also to establish a middleground whereby we can actually judge good and bad. I would have a rating scale to allow users to express opinions, to allow moderators to set benchmarks to, really, allow people to contribute to the community. As a result, we begin to maintain a community by engaging people in the system, the place gets better, creators know what they're doing right and, also, what they're doing wrong. Users gameify the rating system, and so actually providing your opinion is an inherently worthwhile process.

Generally speaking, I accept that the more options you give people the better, as in people will be able to give more accurate feedback. 5 options trumps 3. But this comes as a tradeoff: the more options available, the less content rated. Given the queue of resources in The Hive at the moment (and realistically, ever since I began modding), I don't know how good an idea it would be to really push the boat out there. So speaking on a level of productivity, 5 beats 10.

If we look beyond a content-level and to the users themselves, people will only ever have a certain "mental budget" that they're looking to invest. If you provide more options, you're providing more work. The more work, the fewer decisions. This holds true in many more contexts than just a rating system, too.

And so, given we're only ever trying to provide a general overview in the ratings system but a much more detailed account in both the review and discourse below, five stars, five thumbs, 5 rectally shaped stars are perfectly fine.

giphy.gif
 

Winimasker

W

Winimasker

Hi, granularity is not always good. Rating system can be just simple. :)

It is just like a dice.

Dice may have 3/6/8 surfaces and so on, but it never includes so much of "surfaces" for rolling. Do you know what I mean?

Let's take a metaphor, do you want the website to be super complicated? If the answer is yes, then all users will get confusion, frustration, panic and boredom to super complicated websites which contain an extreme number of mini-buttons which are harder to be clicked.

That's all. Remember, when you make a game, please don't make your granularity to be [highlight]very[/code] high, there are average gamers. Not all gamers are experts (of course :) they can be children :( ), not all gamers are non-professional novices.

If granularity is high in number, it can lead to useless situation (it will seem to don't have any specific usage :( ).

Have you read game mechanics? If you have, then you should know it.

About board game: possibility, granularity, etc.
 
Level 22
Joined
Feb 6, 2014
Messages
2,466
the more options available, the less content rated. Given the queue of resources in The Hive at the moment (and realistically, ever since I began modding), I don't know how good an idea it would be to really push the boat out there. So speaking on a level of productivity, 5 beats 10.

I don't agree on this part. People can easily estimate their ratings and I saw some mod reviews using base 100.
I also think people will not easily rate 100/100 on a map they enjoyed if they have played other maps. People tend to compare games they played.
Example, if they give a map 100/100 and then played another map and enjoyed it too but he/she enjoyed the the first one, people tend to give the second map a lower rating let's say 97/100. In base 5, since they both enjoyed the map, it will be both 5/5 because 4/5 is just too far.

@oJ.r3verFailO, yeah if it is too complicated, how about using /10 instead of /100.
 

Ash

Ash

Level 22
Joined
Aug 27, 2005
Messages
1,684
I don't agree on this part. People can easily estimate their ratings and I saw some mod reviews using base 100.
I also think people will not easily rate 100/100 on a map they enjoyed if they have played other maps. People tend to compare games they played.
Example, if they give a map 100/100 and then played another map and enjoyed it too but he/she enjoyed the the first one, people tend to give the second map a lower rating let's say 97/100. In base 5, since they both enjoyed the map, it will be both 5/5 because 4/5 is just too far.

Unfortunately it seems like your objection is very anecdotal, but I can point you to a comparatively large and wordy body of literature which suggests similar things to what I was saying.
 

Winimasker

W

Winimasker

Shadow Flux said:
Example, if they give a map 100/100 and then played another map and enjoyed it too but he/she enjoyed the the first one, people tend to give the second map a lower rating let's say 97/100. In base 5, since they both enjoyed the map, it will be both 5/5 because 4/5 is just too far.

I just know that not only one guy rates for each resource ((5/5) rating). If there are many users to rate a resource then that resource's rating value will have fractional part.

Shadow Flux said:
@oJ.r3verFailO, yeah if it is too complicated, how about using /10 instead of /100.

O.O I don't know... I just know that high granularity is bad, go to ask Ralle... For instance, if DotA includes millions of items, then only the important and strong items will be focused by players, and so the rest will be ignored.

That is what I mean... -_-
 
Last edited by a moderator:
Level 22
Joined
Feb 6, 2014
Messages
2,466
O.O I don't know... I just know that high granularity is bad, go to ask Ralle... For instance, if DotA includes millions of items, then only the important and strong items will be focused by players, and so the rest will be ignored.

That is what I mean... -_-

But number of items != Rating Scale.

Since hive 2 is coming, an implementation of a larger scale rating (/10) could be appropriate.

@Ash, what I mean is using /10 scale will not significantly decrease productivity and it will provide better accuracy.
The tradeoff is just better.
 
O.O I don't know... I just know that high granularity is bad, go to ask Ralle... For instance, if DotA includes millions of items, then only the important and strong items will be focused by players, and so the rest will be ignored.

That is what I mean... -_-

What on earth does that have to do with rating? Sure, in a DoTA-style game, you want to be able to understand what all of the items are and what they do.

When rating, it's assumed you already know how to count. If you can't count past 5, you have no business rating.

Changing it to 100/100 would just make most people vote 100/100, making it the same as 5/5 as we have now.

Some people will, some people won't. At least then the average would be able to be somewhat swayed by people who want to give a more precise rating, rather than a bunch of 5/5's averaging out to a 5/5.

do you want the website to be super complicated? If the answer is yes, then all users will get confusion, frustration, panic and boredom to super complicated websites which contain an extreme number of mini-buttons which are harder to be clicked.

A rating scale with 1-100 isn't "super complicated". The only way you wouldn't understand how it works is if you don't know how to count.

If someone doesn't want to be precise, they can always continue using x/5 as always, and multiplying their rating by 20 (3/5 = 60/100).

If it's really that big of a deal, you could even have players choose in their preferences whether to use the x/5 or 1-100 rating scale, and that would be how ratings would appear to them, and how they would rate things themselves. If they chose x/5, then the normal 5-thumb scale would be presented for them to click. If their preferences are 1-100, then they would enter that in whatever method you use (typing it into a box, scrolling, or what?) People with 1-100 ratings as their preference would see all ratings appearing at the 1-100 scale (with thumb ratings being converted), players with x/5 would see 1-100 ratings converted into a decimal (70 would be 3.5, 75 would be 3.75, just as average ratings already use a decimal).

This way people can rate x/5 or 1-100, whichever they prefer.

That would be work, though; if you have to choose, I would still strongly advise using 1-100, because people who want to give x/5 can stick to 20/100, 40/100, 60/100, 80/100, and 100/100; people who want to give x/10 can 10/100, 20/100, etc; but people who want to give a 55/100 cannot give a 2.75/5.

I understand how 1-100 can be intimidating to people who are terrified of numbers, though. So how about x/10? That's still small enough for people to develop a mental image of what a resource of each rating should be worth.
 

Ash

Ash

Level 22
Joined
Aug 27, 2005
Messages
1,684
@Ash, what I mean is using /10 scale will not significantly decrease productivity and it will provide better accuracy.
The tradeoff is just better.

Netflix did a study once where they took randomly, and without the user's knowledge, two huge cohorts of people. Group A, and Group B. The first rated films, shows, whatever they would normally do, on the 5 star scale while Group B were able to do the exact same thing on a 10 point scale. What they found from this was that the second group rated significantly less items than the first. Given that we really want more ratings, this is probably our best solution.

Perhaps encouraging better critique will fill the short fall between quality and quantity?

I feel like we're probably reaching an impasse here. I don't think The Hive will change the rating scale we use, and while I think suggesting ways to improve is a good thing, I think it would be really useful if you directed your attention to different kinds of things than grabbing this bull by the horns.
 
Netflix did a study once where they took randomly, and without the user's knowledge, two huge cohorts of people. Group A, and Group B. The first rated films, shows, whatever they would normally do, on the 5 star scale while Group B were able to do the exact same thing on a 10 point scale. What they found from this was that the second group rated significantly less items than the first. Given that we really want more ratings, this is probably our best solution.

Perhaps encouraging better critique will fill the short fall between quality and quantity?

I feel like we're probably reaching an impasse here. I don't think The Hive will change the rating scale we use, and while I think suggesting ways to improve is a good thing, I think it would be really useful if you directed your attention to different kinds of things than grabbing this bull by the horns.

This isn't the same because when people watch Netflix, they are not trying to spend time rating movies; they don't really care. They just want to go on to the next movie, so if it's not as convenient as possible for them to rate, they won't bother.

This is an entirely different situation. Here, people go to the resource already planning to rate it. They are required to post a comment before they can rate it. After going through the time it takes to post a comment, the additional half-second it takes to come up with an x/10 rating in oppose to an x/5 rating is nothing.

Taking a study that shows results of one sort doesn't prove anything. The situation is different. You can always find a study that shows the results you want. There have been so many studies done, of every sort, with different results in each depending on the conditions. One study might say green, the other might say red, to the same question, all based on the differing conditions of the study.
 

Winimasker

W

Winimasker

InfernalTater said:
you could even have players choose in their preferences whether to use the x/5 or 1-100 rating scale, and that would be how ratings would appear to them, and how they would rate things themselves.

You're right actually. I still think that you have read user interface, but...

InfernalTater said:
I would still strongly advise using 1-100, because bla bla bla

You are not right... You never read user interface I guess.
- - - - - - -
Code:
/* Java program coded by
oJ.r3verFailO */
import java.util.Scanner;
import javax.swing.JOptionPane;

public class RatingSystem{
  public static void main(String[] args){
   Scanner scRating = new Scanner(System.in);

   double totalResultRating = 0.0;
   double rateCount = 0.0;

   System.out.print("Please enter a number representing your rating method\n(note: the input must be less than 5): ");

   int denominatorRating = 0;
   denominatorRating = scRating.nextInt();

   if(denominatorRating != 5 || denominatorRating != 100){
      JOptionPane.showMessageDialog(null, "[ Error message: \"The rating method you would like to use\ncannot be other value rather than 5 and 100!\" ]", "Error Reminder", JOptionPane.ERROR_MESSAGE);
   }

   System.out.print("Please enter a number representing your critical rate\n(note: the input must not be negative): ");

   int numeratorRating = 0;
   numeratorRating = sc.Rating.nextInt();

   if(numeratorRating < 0){
      JOptionPane.showMessageDialog(null, "[ Error message: \"The rate you have entered cannot be negative!\" ]", "Error Reminder", JOptionPane.ERROR_MESSAGE);
   }

   else if(numeratorRating > denominatorRating){
      JOptionPane.showMessageDialog(null, "[ Error message: \"The rate you have entered cannot be larger than " + denominatorRating + "!\" ]", "Error Reminder", JOptionPane.ERROR_MESSAGE);
   }

   double resultRating = numeratorRating / (double) denominatorRating;
   
   totalResultRating += resultRating;

   rateCount += 1;

   double averageRating = totalResultRating / rateCount;

   double outputRating = 0.0;
   int baseRating = 0;

   System.out.print("Please enter a number representing the way\nyou would like to look for the rate\n(note: 1 = (#/5) rating view, 2 = (###/100) rating view): ");

   int choiceRating = scRating.nextInt();

   if(choiceRating == 1){
      outputRating = averageRating * 5;
      baseRating = 5;
   }

   else if(choiceRating == 2){
      outputRating = averageRating * 100;
      baseRating = 100;
   }

   else{
      JOptionPane.showMessageDialog(null, "[ Error message: \"Input error!\" ]", "Error Reminder", JOptionPane.ERROR_MESSAGE);
   }

   System.out.println("" + outputRating + "/" + baseRating);
  } //end main method
} //end class RatingSystem
/* Doid you mean
this kind of rating system?
(a simple code/trash -> it is coded simply, not a critical code) 
it is not a completed code, it doesn't have more classes to be implemented
for realistic result. */

Ash said:
quality and quantity

InfernalTater, have you think of these two terms?
 
Last edited by a moderator:

Ash

Ash

Level 22
Joined
Aug 27, 2005
Messages
1,684
@Ash you mean this study?

That would make sense if the user does not require to comment on the resource but THW does. If we are really just looking for more ratings, the comment requirement should be removed.

Nono, so you're missing the point. The ratings are the rough indication, the comments are the elaboration. More ratings = more discussion in comments. Greater rating scales = more cognitive demand = less comments, less rich discussion.
 
Level 22
Joined
Feb 6, 2014
Messages
2,466
Nono, so you're missing the point. The ratings are the rough indication, the comments are the elaboration. More ratings = more discussion in comments. Greater rating scales = more cognitive demand = less comments, less rich discussion.

So if you are elaborating it in the comment, (which requires mental deamnd), you might as well allow that person to give a more accurate rating. If he takes the time to make an elaborated comment, then he will give a better accurate rating though THW does not allow him.
 

Winimasker

W

Winimasker

Shadow Flux said:
If we are really just looking for more ratings, the comment requirement should be removed.

What? (I'm shocked!) :O

There are no comments if there are no rating.

There are no discussions if there are no comments.

There are no resource process automation, resource process improvement and resource process re-engineering if there are no discussions.
- - - - - - -
How can resource developers know and understand why did users rate like that if there are no comments. Oh my gosh! :O

If the comment system is removed, then what are the purposes of posting resources? Are resources posted for download only?

Resources can be discussed by posting a comment so that resource developers can improve it and get understanding from all users who comment.
- - - - - - -
There is the improvement because there is a voice which is meaningful for someone to get understanding and realise it.
 
Level 22
Joined
Feb 6, 2014
Messages
2,466
What? (I'm shocked!) :O

There are no comments if there are no rating.

There are no discussions if there are no comments.

There are no resource process automation, resource process improvement and resource process re-engineering if there are no discussions.
- - - - - - -
How can resource developer know and understand why did users rate like that if there are no comments. Oh my gosh! :O

If the comment system is removed, then what are the purposes of posting resources? Are resources posted for download only?

Resources can be discussed by posting a comment so that resource developer can improve it and get understanding from all users who comment.

You don't get what I'm saying. My post you quoted has a sarcastic tone.
 

Ash

Ash

Level 22
Joined
Aug 27, 2005
Messages
1,684
So if you are elaborating it in the comment, (which requires mental deamnd), you might as well allow that person to give a more accurate rating. If he takes the time to make an elaborated comment, then he will give a better accurate rating though THW does not allow him.

I think we're just going to continue disagreeing. I can see why you care about changing the rating system, but it's a qualitative and bias care with only anecdotal support. I've given you a lot of quantitative research, as well as qualitative reasoning, so you can choose to engage in that and try to work something out, improve based on this information, or just continue to maintain your own.
 
Nono, so you're missing the point. The ratings are the rough indication, the comments are the elaboration. More ratings = more discussion in comments. Greater rating scales = more cognitive demand = less comments, less rich discussion.

I don't believe that would be the case at all. I believe that a larger rating scale for more precise ratings would only provoke more discussion as they don't just have to explain why they came to North America, but rather why they specifically chose Brooklyn (I'm just using this metaphorically).

And, like I said before, anyone who doesn't want to rate with such precision can just continue on determining their rating by x/5, writing x/5 in their comment and multiplying it by 20 to submit the rating (or by choosing x/5 as their default rating scale, if both scales are implemented and users can choose).

I think we're just going to continue disagreeing. I can see why you care about changing the rating system, but it's a qualitative and bias care with only anecdotal support. I've given you a lot of quantitative research, as well as qualitative reasoning, so you can choose to engage in that and try to work something out, improve based on this information, or just continue to maintain your own.

You've given research of studies that have a completely different situation. They don't prove your point at all. You also really have a condescending tone. Assuming that you have a superior intellect and vast knowledge compared to those whom you disagree with is not going to produce a solution. That is not the way to debate with someone. You're not reasoning, you're giving irrelevant research, and saying "See, look, this random person says I'm right, so that means you're wrong." Sure, you found a study that resulted in the x/5 being more efficient for their particular system. I wouldn't be surprised if other studies run by people searching for the opposite conclusion found opposite results. No matter what you're trying to prove, there is always a study somewhere that has come to the same conclusion; it could be the exact opposite of another study under different conditions. That proves nothing.

Yes, a couch potato on Netflix is less likely to rate something if he has to take an extra second to think about it.

THW is a completely different situation, because people often went to the resource with the specific intention of reviewing it. They are also already required to write a comment in order to rate it, so anyone unwilling to take the effort to multiply their 3/5 by 20 would also be too lazy to write a comment, or would write a comment like "5/5 rly good" or "1/5 usobad". You keep complaining about how the effort it takes to factor in a rating at a greater scale, would discourage people from making the effort. Do we even want a rating from someone who is so thoughtless? Isn't the purpose of requiring a comment exactly that, to require enough effort that people who don't really care won't rate it? We have this requirement because we don't want it to be too easy. We don't want a hundred people coming through rating something after only half a second of consideration.

You keep talking about quality versus quantity, and acting as though more numbers in the rating is "quantity", and smaller numbers is "quality". In reality, larger numbers = potentially less ratings with more effort in each rating = quality, and smaller numbers = potentially more ratings with less effort = quantity.

Sure, not everyone will make the effort to rate as precisely as a 64 or 83 for example, but they don't have to. People who produce resources will appreciate being able to finally tell the difference between specific users' opinions of each of their resources. When they do a better job on one than another, they will be able to look at the same users' ratings of one compared to their ratings of the other, to know that overall, those particular users preferred one over the other. With the current system, we can't really tell how our resources compare to each other (our own resources; not against other people's) in the eyes of the community.

x/5 ratings simply don't tell the creator anything. With x/5 ratings, you might as well completely remove ratings and just focus on comments, if that's what you want, because the rating is too imprecise to be helpful to anyone.

No, I'm not suggesting removing the rating scale, for those who have interpretation issues, I was saying that to show just how useless the x/5 system is.
 
Level 21
Joined
Nov 4, 2013
Messages
2,017
All right, time to express my views.
First of all, let me tell you that giving ratings out of 100 has some drawbacks. It will cause a larger number of inaccurate ratings (we already have a lot of inaccurate ratings out of 5 so imagine out of 100...) which will eventually cause imprecise distribution of quality. People will find it quite difficult to give the right rating. In addition, both systems cannot be implemented. Can you imagine that a map that scored 40/100 would be equal to a map that scored 59/100? They are both a 3/5 but the difference between them is too great. A rating out of 5 is effective, simple and somewhat accurate. And no, the x/5 system is not useless. It discriminates well between the quality of maps. Each rating 1/5, 2/5, 3/5, 4/5, 5/5 (maybe also 6/5?) shows what a map is like. There might be a large difference between, for example, all the 3/5 given before but is it more shocking comparing a 40/100 to a 59/100 or an almost 3/5 to a sure 3/5? You already know the answer. Oh, here are what the x/5 ratings stand for:

1/5 = The map is totally unacceptable, lacking and terrible
2/5 = The map is bad and unenjoyable but may have some potential
3/5 = The map is acceptable and enjoyable, may still have several issues
4/5 = The map is great and really enjoyable, may have few issues
5/5 = The map is nearly perfect and very enjoyable with almost no issues
6/5 = The map is impeccable and its quality is so great to astonish everyone

So tell me if you can apply this to ratings out of 100. There is a too massive range that not everyone can really differentiate well between, for instance, a 1/5 and 2/5 or even a 3/5 and 4/5 on a scale out of 100.
Also, more psychologically (hello Ash xD), the human brain likes to have few options to choose from instead of many so I think that if changes are to be made, the rating system should not rise above x/10.
 
All right, time to express my views.
First of all, let me tell you that giving ratings out of 100 has some drawbacks. It will cause a larger number of inaccurate ratings (we already have a lot of inaccurate ratings out of 5 so imagine out of 100...) which will eventually cause imprecise distribution of quality. People will find it quite difficult to give the right rating.

2/5 = 40/100. They don't have to be precise with a 44/100 if they don't want to.

In addition, both systems cannot be implemented. Can you imagine that a map that scored 40/100 would be equal to a map that scored 59/100? They are both a 3/5 but the difference between them is too great. A rating out of 5 is effective, simple and somewhat accurate.

It's called decimals. I mentioned this multiple times already. Our current rating system already uses them for the averages, so a 59/100 total rating (total thumbs at 59, from 20 ratings meaning a potential of 100) shows up as a 2.95/5, while an average of 40/100 shows up as 2/5.

And no, the x/5 system is not useless. It discriminates well between the quality of maps. Each rating 1/5, 2/5, 3/5, 4/5, 5/5 (maybe also 6/5?) shows what a map is like. There might be a large difference between, for example, all the 3/5 given before but is it more shocking comparing a 40/100 to a 59/100 or an almost 3/5 to a sure 3/5? You already know the answer.

It's more shocking to compare a 40 to a 59/100. And what's wrong with that? It should be, there's a big difference between a 2/5 and a 2.95/5 that x/5 ratings without decimals aren't able to express.


Oh, here are what the x/5 ratings stand for:

1/5 = The map is totally unacceptable, lacking and terrible
2/5 = The map is bad and unenjoyable but may have some potential
3/5 = The map is acceptable and enjoyable, may still have several issues
4/5 = The map is great and really enjoyable, may have few issues
5/5 = The map is nearly perfect and very enjoyable with almost no issues
6/5 = The map is impeccable and its quality is so great to astonish everyone

So tell me if you can apply this to ratings out of 100. There is a too massive range that not everyone can really differentiate well between, for instance, a 1/5 and 2/5 or even a 3/5 and 4/5 on a scale out of 100.

1/5 = 20/100 (1-30/100;0-1.5)= The map is totally unacceptable, lacking and terrible
2/5 = 40/100 (31-50/100;1.5-2.5) = The map is bad and unenjoyable but may have some potential
3/5 = 60/100 (51-70/100;2.5-3.5) = The map is acceptable and enjoyable, may still have several issues
4/5 = 80/100 (71-90/100;3.5-4.5) = The map is great and really enjoyable, may have few issues
5/5 = 100/100 (91-110;4.5-5.5) = The map is nearly perfect and very enjoyable with almost no issues
6/5 = 120/100 (110-120;5.5-6) = The map is impeccable and its quality is so great to astonish everyone


There's no difference. If you rate a 77/100, that's very close to a 4/5, which is what it can be interpreted as by people who are afraid of multi-digit numbers.


Also, more psychologically (hello Ash xD), the human brain likes to have few options to choose from instead of many so I think that if changes are to be made, the rating system should not rise above x/10.

Some people's brains prefer fewer options to choose from. Some prefer more. People who want less choices can choose to restrict their own ratings to x/5 by rating 20/100, 40/100, etc, without using the numbers in between. But people who want more choices can't give a 2.5/5.

Perhaps a compromise? Adding a decimal to the ratings so that people can give a 2.5? Or if not tenth-place decimals, allowing people to give half-thumbs?
 

Chaosy

Tutorial Reviewer
Level 40
Joined
Jun 9, 2011
Messages
13,183
All right, time to express my views.
First of all, let me tell you that giving ratings out of 100 has some drawbacks. It will cause a larger number of inaccurate ratings (we already have a lot of inaccurate ratings out of 5 so imagine out of 100...) which will eventually cause imprecise distribution of quality. People will find it quite difficult to give the right rating. In addition, both systems cannot be implemented.

What do you think we are? Some kind of brainless being? if you can't rate something on a scale from 1-100 you don't have a future. I doubt you'll even manage to get a job at MC Donalds.

Can you imagine that a map that scored 40/100 would be equal to a map that scored 59/100? They are both a 3/5 but the difference between them is too great. A rating out of 5 is effective, simple and somewhat accurate.

Just no.

And no, the x/5 system is not useless. It discriminates well between the quality of maps. Each rating 1/5, 2/5, 3/5, 4/5, 5/5 (maybe also 6/5?) shows what a map is like. There might be a large difference between, for example, all the 3/5 given before but is it more shocking comparing a 40/100 to a 59/100 or an almost 3/5 to a sure 3/5?
It's not useless, it's bad. Not very far from useless though, the quality between 3/5 maps and 4/5 maps are usually visible. However the quality among the 3/5 varies A LOT which is pretty shitty.
 

Ash

Ash

Level 22
Joined
Aug 27, 2005
Messages
1,684
I don't believe that would be the case at all. I believe that a larger rating scale for more precise ratings would only provoke more discussion as they don't just have to explain why they came to North America, but rather why they specifically chose Brooklyn (I'm just using this metaphorically).

And, like I said before, anyone who doesn't want to rate with such precision can just continue on determining their rating by x/5, writing x/5 in their comment and multiplying it by 20 to submit the rating (or by choosing x/5 as their default rating scale, if both scales are implemented and users can choose).

Research disagrees. Your suggestion of multiplying an /5 rating by 20 only results in a psuedo-100, quintessentially /5 system.

You've given research of studies that have a completely different situation. They don't prove your point at all. You also really have a condescending tone. Assuming that you have a superior intellect and vast knowledge compared to those whom you disagree with is not going to produce a solution. That is not the way to debate with someone. You're not reasoning, you're giving irrelevant research, and saying "See, look, this random person says I'm right, so that means you're wrong." Sure, you found a study that resulted in the x/5 being more efficient for their particular system. I wouldn't be surprised if other studies run by people searching for the opposite conclusion found opposite results. No matter what you're trying to prove, there is always a study somewhere that has come to the same conclusion; it could be the exact opposite of another study under different conditions. That proves nothing.

Okay, thanks for your opinion on that. When we start a thread on "reasons Ash is a fucking asshole" I'll be certain to PM you and make sure you're involved. As for your issue with research, I understand it's not done directly on THW, but it is certainly applicable to THW. I entirely disagree with your comment that it proves nothing, and feel it is an extremely uneducated issue to bring to the table. If you know of any research that suggests otherwise, then bring it up. That's how it works in the sciences, in the humanities, and in the arts, and I'm pretty sure you're wrong in trying to refute all of them.

Yes, a couch potato on Netflix is less likely to rate something if he has to take an extra second to think about it.

THW is a completely different situation, because people often went to the resource with the specific intention of reviewing it. They are also already required to write a comment in order to rate it, so anyone unwilling to take the effort to multiply their 3/5 by 20 would also be too lazy to write a comment, or would write a comment like "5/5 rly good" or "1/5 usobad". You keep complaining about how the effort it takes to factor in a rating at a greater scale, would discourage people from making the effort. Do we even want a rating from someone who is so thoughtless? Isn't the purpose of requiring a comment exactly that, to require enough effort that people who don't really care won't rate it? We have this requirement because we don't want it to be too easy. We don't want a hundred people coming through rating something after only half a second of consideration.

Finally, something that actually justifies the keystrokes you're taking to write them. So to me, one of your main issues seems to be that you don't think existing research on ratings hits the mark. To address the first point you're pushing, "a couch potato on netflix" really undermines the process that they've gone through to rate something. You watch a movie, sometimes for three hours, and then, having decided you either like or dislike it, give it a rating. I would be willing to bet that for resources excluding maps, very few people spend that much time deciding their rating. I would also disagree that people approach all resources specifically wanting to review them. There's certainly many cases when people just come to download maps, and when I was younger certainly cases when I just turned up to download a resource to use in a map. The overarching concern you raise about people giving bad feedback is commented upon below, albeit briefly.

You keep talking about quality versus quantity, and acting as though more numbers in the rating is "quantity", and smaller numbers is "quality". In reality, larger numbers = potentially less ratings with more effort in each rating = quality, and smaller numbers = potentially more ratings with less effort = quantity.

No, you've read what I'm saying wrong. More numbers = greater accuracy, less numbers = greater quantity. Five beats three, five beats ten. You're throwing too many variables into the mix with a 100 system, so that's not even being considered. Not even WC3C uses anything greater than a five system, and they're extremely detailed in their feedback - I also worked there as a map mod.

Sure, not everyone will make the effort to rate as precisely as a 64 or 83 for example, but they don't have to. People who produce resources will appreciate being able to finally tell the difference between specific users' opinions of each of their resources.

But this is exactly what the comments are there for. The ratings are rough indicators, categorical variables, which allow for an easy classification and easy searchability. The "people who produce resources" are certainly smart enough to read the comments and be able to tell apart different users opinions, right...?

When they do a better job on one than another, they will be able to look at the same users' ratings of one compared to their ratings of the other, to know that overall, those particular users preferred one over the other. With the current system, we can't really tell how our resources compare to each other (our own resources; not against other people's) in the eyes of the community.

x/5 ratings simply don't tell the creator anything. With x/5 ratings, you might as well completely remove ratings and just focus on comments, if that's what you want, because the rating is too imprecise to be helpful to anyone.

No, I'm not suggesting removing the rating scale, for those who have interpretation issues, I was saying that to show just how useless the x/5 system is.

You're holding a false assumption that the ratings are for the creator, and I think maybe this is motivated by your own previous experiences with the resources section. Please check your sense of self-worth at the door, and approach this discussion from a detached standpoint.

Ratings are there to categorise resources into the good, the bad and the ugly. If you want feedback, read the comments. If you want good comments, educate the users.

EDIT:

Also, more psychologically (hello Ash xD), the human brain likes to have few options to choose from instead of many so I think that if changes are to be made, the rating system should not rise above x/10.

Absolutely. I don't even think there's an issue with where it is now.
 
Last edited:
Level 23
Joined
Apr 16, 2012
Messages
4,041
what about the 10 points? :D

Its not so dispersed as 100 point system, you could still go down with mapping x/10 to y/5 if you are lazy bum, and as said, there are quite big differences between worst 3/5 and best 3/5 map, and now it could be 5/10 and 7/10, which is a little bit more noticable difference(if you see 3/5, you think of it as decent map, but not all 3/5 necessarily are decent)
 

Kazeon

Hosted Project: EC
Level 33
Joined
Oct 12, 2011
Messages
3,449
If many users indeed care about that accurate scores just request Ralle to put a textbox before the thumb buttons, which allows user to manually input their rating with certain accuracy and constraints, like 3.89132897 whatsoever, both for moderator and user. I believe computer can calculate average and round things very easily, don't worry.

But still, for who do not care about accurate rating are still allowed to simply hit the thumb button.
 

Deleted member 242951

D

Deleted member 242951

A new resource system may make it difficult for moderation but its cool from the aspects of comparison...
 
Level 22
Joined
Feb 6, 2014
Messages
2,466
If many users indeed care about that accurate scores just request Ralle to put a textbox before the thumb buttons, which allows user to manually input their rating with certain accuracy and constraints, like 3.89132897 whatsoever, both for moderator and user. I believe computer can calculate average and round things very easily, don't worry.

But still, for who do not care about accurate rating are still allowed to simply hit the thumb button.

I don't see any disadvantage using this because it does not replace the old rating system.
 
Okay, thanks for your opinion on that. When we start a thread on "reasons Ash is a fucking asshole" I'll be certain to PM you and make sure you're involved. As for your issue with research, I understand it's not done directly on THW, but it is certainly applicable to THW. I entirely disagree with your comment that it proves nothing, and feel it is an extremely uneducated issue to bring to the table. If you know of any research that suggests otherwise, then bring it up. That's how it works in the sciences, in the humanities, and in the arts, and I'm pretty sure you're wrong in trying to refute all of them.

I never said that all research is wrong. I said that your research didn't prove anything. Stop with your condescending tone, trying to set yourself up as a superior intellect, as though anyone who disagrees with you is simply ignorant of the entire world of science. You are being quite arrogant yourself. Scientists prove their hypothesis (or disprove it) through experiments that try to simulate the conditions they expect. When you show me an experiment that is more similar to the conditions of THW, then I will believe it. But right now, there's a huge difference between your studies, and our situation: In the studies you link, the people being tested are not there to review something; they're there to watch something. In which case trying to get ratings is a matter of making it as easy and convenient as possible. Our situation is different; we don't want as many ratings as possible, we want good ratings. That's why we require that a person write a comment before they rate something. If the extra half-second it takes to type out a larger number, and the extra half-second it takes to figure out what that number means, is enough to discourage them from rating, they wouldn't have written anything worth reading anyway. You keep preaching about science and research, as though you're an expert. If you knew anything about the scientific method, you'd know that any difference between the experiment and what it's trying to simulate can completely change the results. Sometimes the difference could be negligible, and still completely change the outcome. In this case the difference is certainly substantial enough to affect the outcome. The situation is so different in the Netflix study compared to here, that it is useless in trying to determine what the outcome of this will be.

Finally, something that actually justifies the keystrokes you're taking to write them.

You really, really need to learn how to discuss things without assuming that you are superior to anyone who disagrees with you. We're all wrong sometimes, even you, believe it or not. If you act this way all the time, you will inevitably treat someone like shit in the future when they turn out to be right, and you turn out to be wrong. I'm sure it's already happened before. If you try showing more respect for people that take the time to discuss their contrary opinions with you, you'll get more respect in return, but with the undertone of every post from you being insulting, you're lucky I'm even still trying to develope an agreeable solution with you.

So to me, one of your main issues seems to be that you don't think existing research on ratings hits the mark. To address the first point you're pushing, "a couch potato on netflix" really undermines the process that they've gone through to rate something. You watch a movie, sometimes for three hours, and then, having decided you either like or dislike it, give it a rating. I would be willing to bet that for resources excluding maps, very few people spend that much time deciding their rating. I would also disagree that people approach all resources specifically wanting to review them. There's certainly many cases when people just come to download maps, and when I was younger certainly cases when I just turned up to download a resource to use in a map. The overarching concern you raise about people giving bad feedback is commented upon below, albeit briefly.

The only people who actually criticize a movie within their mind to develop their rating throughout the entire movie, are those who are already planning to rate it. People who didn't plan to rate it, and whose ratings are determined by how convenient it is to rate, had just decided what they wanted to rate it the second before they clicked the button. Additionally, not all movies are 3 hours long; that is rare. Generally, they are an hour and a half. Most people who review maps have played for a similar amount of time. If they haven't, it's either because the map didn't have enough content for a game to last that long, meaning that they wouldn't need to spend that much time to examine its content anyway, or because they were too lazy to actually look at it, in which case we don't want their rating anyway. Models don't need the same level of scrutiny because they are one object. You also mention that people may not have used the resource with the intention of rating it. Sure, they may not have. But they had to comment before they could rate it; if they took the time to write a comment, then they more than likely already intended to rate it. If they were just showing up to download it and quickly click the rating, then they're already stopped by the requirement of writing a comment, not by the state of the rating system.

I never claimed that everyone approaches all resources specifically wanting to review them. Turning my posts into all-inclusive generalizations, when that is not what I said, just to make them easier to counter, is quite a lame way to argue. However, many people do go to resources with the intention of rating them. Certainly enough people that resources would still receive plenty of quality ratings, despite making them take two seconds more effort. The way you argue against this, it's as if you think people would have to wait in line for 4 hours and sacrifice an arm and a leg to rate something if the number was a 10 instead of a 5.

No, you've read what I'm saying wrong. More numbers = greater accuracy, less numbers = greater quantity. Five beats three, five beats ten. You're throwing too many variables into the mix with a 100 system, so that's not even being considered. Not even WC3C uses anything greater than a five system, and they're extremely detailed in their feedback - I also worked there as a map mod.

Just because 5 beats 10 in one situation, doesn't mean it does in every situation. If something working in one situation meant it would work in every situation, then we would send preschool teachers to break up wars since it worked in the classroom. Every situation is different, which means that the optimal solution will also vary. I also never stated that x/100 was the only option. I have mentioned other options, such as x/10, and decimal values, in every post. We don't need x/100 specifically; we just need a way to make our ratings more precise.

But this is exactly what the comments are there for. The ratings are rough indicators, categorical variables, which allow for an easy classification and easy searchability. The "people who produce resources" are certainly smart enough to read the comments and be able to tell apart different users opinions, right...?

"Easy classification and easy searchability". Do you know anyone that searches for maps based on their rating? The rating means nothing, since one resource that would have been worth a 3.5, and another that would have been worth a 4.4, will both be worth a 4/5, despite the drastic difference in quality. Yes, people who rate resources can read comments, however you still can't tell precisely what a person would rate something by reading their comment, because everyone has different tones. You might think that someone's 3/5 on your resource would have been a 2.5/5 because of their long list of criticisms, when in reality those criticisms might not have been important enough to them to affect its quality; they might have still thought it was worth a 3.5/5.

Let's see you read a few comments and decide what you think people would have rated something, without their rating, and then see if your prediction was accurate. It probably won't be, because everyone's tone is different, and it can be especially hard to tell how someone feels when you're simply reading text. I highly doubt you'd be able to predict their ratings to a degree of accuracy. It might sound like they didn't like your resource as much as they liked a different one, but they might have liked it more, and took the time to criticize it so thoroughly just because they thought it was good enough to be worth improving as much as possible.

You're holding a false assumption that the ratings are for the creator, and I think maybe this is motivated by your own previous experiences with the resources section. Please check your sense of self-worth at the door, and approach this discussion from a detached standpoint.

Ratings are there to categorise resources into the good, the bad and the ugly. If you want feedback, read the comments. If you want good comments, educate the users.

I never assumed that ratings are only for the creator. But at the moment, that's the only thing they can be used for. They're useless to anyone else, because with the system being so inaccurate, if you judge maps based on their rating, you'll miss out on a lot of great maps. More accurate ratings would help them to be more useful across the board. You also completely ignored my suggestion to allow users to input tenth-place (x.x) decimals in their ratings. Please explain to me how implementing this would adversely affect ratings.

At the moment, it doesn't seem like you're trying to discuss things at all; it seems like you're treating this like a battle instead of a discussion. I have presented plenty of compromises that would satisfy both crowds. My various solutions would provide a huge amount of benefit to the half of the community that wants precise ratings, while causing no drawback to those who do not wish to use the system.

Like I've said many times, you can scale an x/5 up through simple math, but you can't scale an x/10 down (without decimals). Why use something that satisfies one crowd while neglecting the other, when you can use a system that allows both crowds to rate things the way they want to? You don't have to use precision, just because it's available. If we used a decimal system, or a larger scale, people wouldn't have to give a 3.5, or a 7/10. They could continue giving the same x/5 ratings as always. You're basically opposing this simply because it doesn't meet your personal preference. Okay, so you don't want to rate things a 2.5/5. Then don't. Rate it a 2/5, or a 3/5. Why stop someone else from rating accurately just because you don't want to?

If many users indeed care about that accurate scores just request Ralle to put a textbox before the thumb buttons, which allows user to manually input their rating with certain accuracy and constraints, like 3.89132897 whatsoever, both for moderator and user. I believe computer can calculate average and round things very easily, don't worry.

But still, for who do not care about accurate rating are still allowed to simply hit the thumb button.

This is a good idea and is basically the same as one of my suggestions. Although I don't think we need that many decimal places; one decimal place would be enough. Or if not a text box, simply allowing users to click half of a thumb by having their cursor on the left half of the thumb, or a whole thumb with their cursor on the right half of the thumb. That may sound weird, but it really wouldn't be as awkward as it sounds when you try it.
 

Ash

Ash

Level 22
Joined
Aug 27, 2005
Messages
1,684
A whole bunch of stupid shit that I'm not going to waste my time replying to

I still absolutely disagree with what you're saying, and you ignore so many of my points that I'm not going to waste my time replying when I could be drinking more beer. Realistically, there have been countless conversations like this since the Hive's opening, and your crusade now won't make a difference: I sincerely doubt we are going to change our rating scale.

If you would actually like to discuss any of the points I've raised, please address them and stop falling back on your ignorant defence of calling me elitist. When you present an idea that isn't stupid and actually raises a good argument, I'll reply. Until then, please continue to waste your time writing replies that essentially say the same thing because you're too emotionally involved to do otherwise. /Yawn.
 
Last edited:
Level 21
Joined
Nov 4, 2013
Messages
2,017
2/5 = 40/100. They don't have to be precise with a 44/100 if they don't want to.

If the reason of implementing the x/100 system isn't being more precise, then I don't see why even implementing it. An x/100 is supposed to give more accurate ratings so if you say that a 44/100 can be rounded to a 40/100, then why not sticking to the x/5 system? A 40/100 gives a worse impression than a 2/5.

It's called decimals. I mentioned this multiple times already. Our current rating system already uses them for the averages, so a 59/100 total rating (total thumbs at 59, from 20 ratings meaning a potential of 100) shows up as a 2.95/5, while an average of 40/100 shows up as 2/5.

Using decimals would convert anyway the x/5 system to a pseudo-x/100 one. I even think some people do not even appreciate the use of decimals. We might think about this point though, it's not that bad after all. I'd suggest a one decimal place only so that the options aren't massive.

It's more shocking to compare a 40 to a 59/100. And what's wrong with that? It should be, there's a big difference between a 2/5 and a 2.95/5 that x/5 ratings without decimals aren't able to express.

The difference between a 40/100 and a 59/100 clearly means a great difference in quality. To wipe this out, we'd need to make new catergories apart from the 6 common ones we know (Unacceptable, Lacking, Useful, Recommended, Highly Recommended, Director's Cut) while with an x/5 system, even if decimals are used, the difference woudln't be really that shocking.

1/5 = 20/100 (1-30/100;0-1.5)= The map is totally unacceptable, lacking and terrible
2/5 = 40/100 (31-50/100;1.5-2.5) = The map is bad and unenjoyable but may have some potential
3/5 = 60/100 (51-70/100;2.5-3.5) = The map is acceptable and enjoyable, may still have several issues
4/5 = 80/100 (71-90/100;3.5-4.5) = The map is great and really enjoyable, may have few issues
5/5 = 100/100 (91-110;4.5-5.5) = The map is nearly perfect and very enjoyable with almost no issues
6/5 = 120/100 (110-120;5.5-6) = The map is impeccable and its quality is so great to astonish everyone

There's no difference. If you rate a 77/100, that's very close to a 4/5, which is what it can be interpreted as by people who are afraid of multi-digit numbers.

Same thing I said in the paragraph above (that's why I asked how you would express this with the x/100).

Some people's brains prefer fewer options to choose from. Some prefer more. People who want less choices can choose to restrict their own ratings to x/5 by rating 20/100, 40/100, etc, without using the numbers in between. But people who want more choices can't give a 2.5/5.

The human brain is structured to go for the easiest ways of solving problems so an x/5 is more convenient. Consider that the larger the maximum rating is, the greater the probability of giving inaccurate ratings and receiving complaints from users.

Perhaps a compromise? Adding a decimal to the ratings so that people can give a 2.5? Or if not tenth-place decimals, allowing people to give half-thumbs?

We shall think about this point before making it true.

What do you think we are? Some kind of brainless being? if you can't rate something on a scale from 1-100 you don't have a future. I doubt you'll even manage to get a job at MC Donalds.

What are you saying man? If you can't rate something accurately out of 100, it doesn't mean you're brainless. To give the right rating, you must have experience and professionality and those who lack them are not to be considered idiots. It's like saying "If you can't review a map well, you are stupid." False statement of course because not everyone is able to review maps professionally yet this doesn't justify stupidity.
MC... Donalds? I don't know what links an x/100 system with MC Donalds...


To which part? The one where I described the x/5 system? If that's so, let me tell you a few things. You cannot deny that it's simple due to the few options it provides. You cannot deny it's effective because it's been around for over 10 years and it successfully distinguished bad maps from good maps. The accuracy of the x/5 is questionable because I also think that an x/100 would be more exact (naturally given that you can rate accurately) nonetheless the degree of accuracy of the x/5 system is high enough to be utilized in rating maps.

It's not useless, it's bad. Not very far from useless though, the quality between 3/5 maps and 4/5 maps are usually visible. However the quality among the 3/5 varies A LOT which is pretty shitty.

And the quality among, for instance, a 60/100 and a 79/100 wouldn't be pretty horrible as well? Seeing a 3/5 gives a better impression than a 60/100 thus incrementing the downloads of the map. You should also think about this crucial point as well because nobody is willing to upload a map which will get few downloads since few downloads = few suggestions.


I don't know if we'll ever reach an agreement here. We just keep debating each other without even convincing the other side.
 
Level 22
Joined
Feb 6, 2014
Messages
2,466
If the reason of implementing the x/100 system isn't being more precise, then I don't see why even implementing it. An x/100 is supposed to give more accurate ratings so if you say that a 44/100 can be rounded to a 40/100, then why not sticking to the x/5 system? A 40/100 gives a worse impression than a 2/5.
The 100 scale rating is too big and looks horrible so we change the idea to allowing decimal in the 5 scale rating system.


The difference between a 40/100 and a 59/100 clearly means a great difference in quality. To wipe this out, we'd need to make new catergories apart from the 6 common ones we know (Unacceptable, Lacking, Useful, Recommended, Highly Recommended, Director's Cut) while with an x/5 system, even if decimals are used, the difference woudln't be really that shocking.
No need to have new categories, just need to let people rate more precisely. The categories can stay the same. So if you play a map and you think this is better than most 4/5 rating but isn't enough for a 5/5 rating. A 4.5/5 rating is appropriate and people will understand when they see that rating, "ah this map is between recommended and highly recommended" and can assume that this is better than most 4/5 maps but not better than most 5/5 map. Without it, you would rate a 4.5/5 to 5/5 and you will see a significant difference which may lead to false conclusion that a 4.5/5 map is 5 to THW's even though the significant difference is still there.

The human brain is structured to go for the easiest ways of solving problems so an x/5 is more convenient. Consider that the larger the maximum rating is, the greater the probability of giving inaccurate ratings and receiving complaints from users.
But we AREN'T solving problems, we are just rating resources how we think good they are. Some people prefer to give more precise ratings while some don't. Why limit other people who wants to give more precise rating? I don't know why you think a larger rating scale will result to inaccurate ratings. Having larger rating scale basically allow people to give more precise rating. The accuracy of a rating depends on that person's opinion.

What are you saying man? If you can't rate something accurately out of 100, it doesn't mean you're brainless. To give the right rating, you must have experience and professionality and those who lack them are not to be considered idiots. It's like saying "If you can't review a map well, you are stupid." False statement of course because not everyone is able to review maps professionally yet this doesn't justify stupidity.
MC... Donalds? I don't know what links an x/100 system with MC Donalds...
He is not saying if you can't rate map accurately, you are an idiot. More like, if you cannot convert your own 5 scale rating to 100 scale rating, you're an idiot. Btw, working at McDonalds is metaphorically used to represent people who failed at their school (mostly not good at simple math) and have no choice but to work at McDonalds just to have a job. I'm not insulting McDonalds employee, i'm just explaining to you the link between them. That is why Chaosy said, if you are afraid of decimal places or larger numbers, then even McDonald wouldn't hire you.



I don't know if we'll ever reach an agreement here. We just keep debating each other without even convincing the other side.
Probably, we need a third party like TDW.

If you would actually like to discuss any of the points I've raised, please address them and stop falling back on your ignorant defence of calling me elitist. When you present an idea that isn't stupid and actually raises a good argument, I'll reply. Until then, please continue to waste your time writing replies that essentially say the same thing because you're too emotionally involved to do otherwise. /Yawn.
But didn't he said that the points/research you said is not applicable because the situation here is different? He may be emotionally involved, but he does have a point.

If the extra half-second it takes to type out a larger number, and the extra half-second it takes to figure out what that number means, is enough to discourage them from rating, they wouldn't have written anything worth reading anyway.

I totally Agree with this.

Like I've said many times, you can scale an x/5 up through simple math, but you can't scale an x/10 down (without decimals). Why use something that satisfies one crowd while neglecting the other, when you can use a system that allows both crowds to rate things the way they want to? You don't have to use precision, just because it's available. If we used a decimal system, or a larger scale, people wouldn't have to give a 3.5, or a 7/10. They could continue giving the same x/5 ratings as always. You're basically opposing this simply because it doesn't meet your personal preference. Okay, so you don't want to rate things a 2.5/5. Then don't. Rate it a 2/5, or a 3/5. Why stop someone else from rating accurately just because you don't want to?


I totally agree with this too. If people don't want to rate with decimal place, then fine but at least allow other people to rate with decimal place.
 
Level 21
Joined
Nov 4, 2013
Messages
2,017
No need to have new categories, just need to let people rate more precisely.

Oh man, you've just pronounced holy words for me. However, the problem is that not everyone in the hive can do that, this is why an x/100 wouldn't be convenient. A lot of people will hand out even more inaccurate ratings. Let's take an example. Two guys play a map, one gives it 2/5 because there are many issues and the other gave it 4/5 because he likes. Interpret this with the x/100 and what do we get? A 40/100 VS an 80/100. Senseless, really senseless. This is already happening with the x/5 system but when you deal with larger numbers, the impact is greater. The question here is not if you're afraid of decimals or large numbers and if MC Donalds would hire you or not (lol?) but if the new system would overall bring more accurate ratings or not. This is also why I said that new categories would be added because halfways aren't really a good idea. A map that scores 4.5/5 (90/100) would need a whole new category with the x/100 to be sure it goes to the right place.
In addition, if no new categories are added, the same problem of the x/5 will occur. Useful maps (3/5 ones), for example, will have an even higher range of qualities.
 
I still absolutely disagree with what you're saying, and you ignore so many of my points that I'm not going to waste my time replying when I could be drinking more beer. Realistically, there have been countless conversations like this since the Hive's opening, and your crusade now won't make a difference: I sincerely doubt we are going to change our rating scale.

If you would actually like to discuss any of the points I've raised, please address them and stop falling back on your ignorant defence of calling me elitist. When you present an idea that isn't stupid and actually raises a good argument, I'll reply. Until then, please continue to waste your time writing replies that essentially say the same thing because you're too emotionally involved to do otherwise. /Yawn.

As if you said anything different? You repeated the same argument each time as well. But no, everyone else is stupid, and you're smart! You don't make the same mistakes!

We both repeated ourselves, because we were both responding to the same thing every time. I repeated myself because you repeated yourself. When you say the same thing that you said the first time, my response is going to be the same as the first time, because it was true then, and it's still true now. You repeated yourself for the same reason. I don't hold you at fault for it because we both did the same thing, however I do hold you at fault for constantly being insulting, when I did nothing to you but try to discuss our two different opinions in hopes of coming to a compromise that satisfies both parties.

And here you are again, being insulting with every post you've made. I have not once insulted you. You say I am "too emotionally invested". You are assuming that I am emotionally invested because of your own emotional investment. You care so much that you insult anyone with a different opinion. You tell everyone their ideas are stupid, just because they disagree with you. There is nothing stupid about my suggestion. Whether or not x/5 or x/10 is better, is debatable, but to assume that either rating system is stupid, is to be so conceited as to assume that only your own beliefs are of any worth. Do you yell at people who support the other team in sports, and insult everyone that is of a different political party than yourself, too?

And the typical "I'm too cool to care, gonna go drink a beer" response. Yeah, right. I wouldn't be able to post such absurd walls of text if I didn't have your own lengthy (albeit shorter than mine) posts to respond to.

If the reason of implementing the x/100 system isn't being more precise, then I don't see why even implementing it. An x/100 is supposed to give more accurate ratings so if you say that a 44/100 can be rounded to a 40/100, then why not sticking to the x/5 system? A 40/100 gives a worse impression than a 2/5.

I think you misunderstood what I meant. I wasn't advising that people refrain from using the precise numbers such as 44/100. I was saying that using a precise system would not force people who oppose precision to be precise. If they have something against being accurate with their ratings, they can continue using x/5 and multiplying it by 20. They don't have to use all of the numbers available. But having them there would allow people who want to be accurate to be accurate.

Like I said before, though, a decimal system would work well also. I would prefer a decimal system to x/100; x/100 was not my idea, but rather someone else's. I'm not asking for x/100. I'm asking for a more precise rating system, whether it's x/100, x/10, x.x/5; whatever. It just needs to allow a higher degree of precision. You don't have to be precise. But why force everyone to use an inaccurate system just because you prefer not to be accurate? You can always continue being inaccurate as always if you dislike hitting near the mark. I don't care; shoot your arrows at someone else's target if the idea of being accurate is that terrifying. But don't tell everyone else they're not allowed to shoot near the bull's eye just because you don't want to.

The human brain is structured to go for the easiest ways of solving problems so an x/5 is more convenient. Consider that the larger the maximum rating is, the greater the probability of giving inaccurate ratings and receiving complaints from users.

Solving the problem is the issue. Taking the easy way out, but not really fixing anything, isn't going to do anyone any good. While half of people prefer to keep things as simple as possible, the other half would like to have a chance to really get in there and do exactly what they plan to without being restricted to the few paths that are made available to them. People wanting things to be simple is the reason we have terrible facebook games. Go play those. Everything doesn't have to be as simple as possible. If that were the case, we would never have even invented electronics in the first place.

Nobody's asking to allow us to rate things based on the square root of the factorial of an imaginary number. You don't have to be a rocket scientist to figure out how x/10 works.

All we want is a balance between simplicity and flexibility. x/10, or a decimal (to the first value, such as 3.5; not 4.38479284797627868127), is not incredibly sophisticated. If the meter between simplicity and complexity was a meter in length, x/10 would be a few centimeters away from far end of simplicity. It's still extremely simple.

You say that the larger the maximum rating is, the greater the probability will be that people will give inaccurate ratings and receive complaints from users. Sure, people who have the IQ of a hamster will give less accurate ratings. Anyone incapable of rating things on a scale from 1 to 10, can always just rate it from 1 to 5, and double their rating, while others who want precision can take advantage of the system by giving a 7/10. If they can't multiply a single-digit number by 2, they have no business using a computer. You say it would cause more complaints from users. It hasn't happened lately, but I've received quite a few messages from people complaining about my ratings, and a lot of them asked me to give partial ratings, such as 3.5/5, or to increase the scale of my ratings to 1-100. They cared so much that they even sent PM's and VM's.

What are you saying man? If you can't rate something accurately out of 100, it doesn't mean you're brainless. To give the right rating, you must have experience and professionality and those who lack them are not to be considered idiots. It's like saying "If you can't review a map well, you are stupid." False statement of course because not everyone is able to review maps professionally yet this doesn't justify stupidity.
MC... Donalds? I don't know what links an x/100 system with MC Donalds...

You don't have to rate it to the nearest number. If you have trouble deciding what to give something, but you think it's in the 60-80 range, and decide on a 70, and an all-knowing being with supernatural judgement capabilities decides it's worth a 64, that's not an inaccurate rating. But if you can't figure out whether you want to give something a 40, or an 80, you're hopeless.

You cannot deny it's effective because it's been around for over 10 years and it successfully distinguished bad maps from good maps.

It definitely hasn't.

And the quality among, for instance, a 60/100 and a 79/100 wouldn't be pretty horrible as well? Seeing a 3/5 gives a better impression than a 60/100 thus incrementing the downloads of the map. You should also think about this crucial point as well because nobody is willing to upload a map which will get few downloads since few downloads = few suggestions.

This is because, psychologically, the only thing most people are used to seeing a 100-point scale on is their grade point average, and anything below a 70 (or 60, depending on the scale their school uses) is failing, because it's not judging the quality of their product, it's judging how much of the material they've learned, in which case knowing only half of the material would mean the student really hasn't grasped it.

For the first few weeks, people would think the ratings are lower than they actually are. After a while, they'll get used to it and form an idea of what each level means, just like they got a feel for how good or bad their GPA was.

I don't know if we'll ever reach an agreement here. We just keep debating each other without even convincing the other side.

I have tried to agree with you. Several times. I have repeatedly stated that a decimal system or x/10 system could be used instead of an x/100 system, because a lot of people are terrified of numbers the length of a pronoun. I have tried to compromise. I understand that not everyone wants a complex system, so something as complex as 1-100 would not work for everyone. The x/100 was never my idea, it was just someone else's example of a potential solution, and because it was being discussed by others, I discussed it, because it would be rude to become part of a discussion and ignore a particular part of it that I almost agree with just because that wasn't specifically what I had in mind.

How would 3.5/5 hurt anyone? You don't have to use decimals if you don't want to. You can always continue giving a 3/5 or 4/5. But why stop everyone else from improving the accuracy of their ratings just because you want to keep yours simple? If you want to keep yours simple, fine keep it simple, nobody's stopping you. You don't have to rate to highest degree of accuracy that a system allows. If you don't want to be accurate, don't utilize it; continue on with x/5, and scaling it to fit whatever system is used. Even if the ratings were out of 1,000. What difference would it make? A 1/5 is a 200/1000, or a 100/500, or a 600/3000. It doesn't matter what the scale is, nobody's forcing you to use every number available.

If you have a scale that measures weight to the nearest hundredth of a gram, and you're only interested in knowing how many grams something weighs, you don't have to throw the scale away and buy one that's less accurate. Just round it. Big deal. If you only want to know how many kilograms it weighs, then look at the digits starting at thousands of grams, and ignore the rest. But you can't take a scale that only measures to the nearest pound, and use it to measure to the gram. It doesn't work the other way around.

But you and Ash are right about something else, that it's not worth discussing any further. I've tried to compromise with you, and provided several alternatives, but you're not interested in compromising to use a system halfway between what you want, and what others want; you want to force everyone to use what you want.
 
Level 22
Joined
Feb 6, 2014
Messages
2,466
Let's take an example. Two guys play a map, one gives it 2/5 because there are many issues and the other gave it 4/5 because he likes. Interpret this with the x/100 and what do we get? A 40/100 VS an 80/100. Senseless, really senseless.

but 2/5 = 40/100 and 4/5 = 80/100. What's senseless about that? the large difference? If you look at the ratio, they are the same. 2/5 is just REALLY far from 4/5.
 
All we want is a balance between simplicity and flexibility. x/10, or a decimal (to the first value, such as 3.5; not 4.38479284797627868127), is not incredibly sophisticated. If the meter between simplicity and complexity was a meter in length, x/10 would be a few centimeters away from far end of simplicity. It's still extremely simple.
Actually having a 5/5 with decimals will still be more simple than a 10/10 system considering how the base number goes up into the tens.

Considering that some people won't even notice the you could rate with decimals depending on how the system is implemented.

It will end up being somewhat more flexible and a simpler number to read when the resource has the rating averaged out and I don't think people really register much after 2 decimal places anyway.

If we were to have a rating system with decimals I don't think users should be able to have the power to rate with more than one decimal points considering that anything else more could become nuisance for the eye to read.
 
Status
Not open for further replies.
Top