Java-Gaming.org Hi !
Featured games (83)
games approved by the League of Dukes
Games in Showcase (521)
Games in Android Showcase (127)
games submitted by our members
Games in WIP (589)
games currently in development
News: Read the Java Gaming Resources, or peek at the official Java tutorials
 
    Home     Help   Search   Login   Register   
Pages: 1 2 [3] 4
  ignore  |  Print  
  who needs judges?  (Read 17948 times)
0 Members and 1 Guest are viewing this topic.
Offline woogley
« Reply #60 - Posted 2006-01-28 17:46:29 »

morre, I do understand where you're coming from Wink

but we can't deny a game that really does deserve *most* of the votes. in the end the contest is about what is the "best" game according to the gamers. why not vote for the best game - period?

lol @ kev
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #61 - Posted 2006-01-28 17:49:44 »

Haha. Well, judges would be ok with me... but where do we find them? How do we coordinate them? Most importantly - what voting system should they use? Cheesy

Offline woogley
« Reply #62 - Posted 2006-01-28 17:50:24 »

lol
Games published by our own members! Check 'em out!
Legends of Yore - The Casual Retro Roguelike
Offline DonaldEKnuth

Junior Devvie





« Reply #63 - Posted 2006-01-28 17:50:46 »

Morre: Sorry if that ASCII table was to large, you can see it here as a screenshot. I am sure that you understood it already.

Enough about that now, you've got my opinion. I'll lure in the background for a while now.  Grin I'm sure that I won't win which ever system of giving ratings that are used, but getting some people to rate my game would be nice. We will see what happens. Smiley

My 4k 2006 Entry: Sokoban4k
Offline woogley
« Reply #64 - Posted 2006-01-28 17:52:53 »

donald - your game will rate either way, because the categorized system runs infinite (even the 2005 games will able to be voted on). so you might not see the result right away, but over time you'll get a good idea about where it stands. but as far as the 4K winner goes, we need that answer cut and dry
Offline DonaldEKnuth

Junior Devvie





« Reply #65 - Posted 2006-01-28 17:54:41 »

Well that's good enough for me! Smiley Also, be sure to cut and paste that table and use it as a desktop background, it took quite a while to create!  Grin

My 4k 2006 Entry: Sokoban4k
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #66 - Posted 2006-01-28 17:54:55 »

But woogley, I don't think you understand. I can't possibly imagining games deserving *most* of the votes - only a *highest average*. See my point? With an average system, you don't even have to push it the way donald wants if you don't want to - it'd still be more fair Smiley

Just voting for the best game would be interesting... we'd have 3 games with lots and lots of votes, and then leaving the rest with none. That's the way they do in the olympics and athletics championships - decide three winners, give them gold, silver and bronze, and forget about the rest. I don't think this is what this contest is about - it's not about winning, beating the others, it's about showing people what you can do with 4k. The way to go is to make sure every game gets a rating - from the winner, to the game that's in last. Granted, the one that comes in last won't be happy, so perhaps you should exclude the statistics of the last few games or so, but I'd take it if I was last - at least I know where I came. If you just select the top 3 games, then the other developers won't even know how well their games did.

EDIT: I agree with Donald, what I'm concerned about is mainly having my games (that aren't the winners - I'm sure of it) get a rating, no matter if it's good or bad. I just want the criticism. This, you can achieve with an average vote. And in 3 days, no less, just like you wanted to do with the other system. Personally, I can't see any advantages with the top 3 system - please do explain them to me Smiley

Offline woogley
« Reply #67 - Posted 2006-01-28 18:05:54 »

okay, the top 3 system has this advantage: it's FAST. though this may yield a 50%-25%-10%-1% kind of slope for the games. but it's the straight up question "what is your top 3 favorite games?" - no need to consider categories, just what do you enjoy playing the most. period. and I think that's pretty much what counts. sort of "I cant explain why I just love this game! I dont know what it is about it but its great!"

using the other system, yes, we will get more solid statistics rather than having 25+ games at a 1% sort of vote. but, that just isn't doable on 72 hours. it's 5 categories for 30+ games, at least 150 different things to decide on (assuming people will rate EVERY game. which is something you can't guarantee in any public system)
Offline DonaldEKnuth

Junior Devvie





« Reply #68 - Posted 2006-01-28 18:12:23 »

Forgot to answer earlier:
Quote
I think what you're trying to say is: The system wouldn't allow you to vote on the top few games IF you haven't voted on the one with the least votes - vote on it, give it either a good OR a bad vote, decreasing or increasing it's average, and then you're free to vote on the top games, and thus there's really no restriction, just something to give the more obscure games the exposure they need. They won't get better scores this way, just more votes (which isn't better in an average system). Also, if you use this system, there's no point in making the system weighted, since all games will have roughly the same amounts of votes, so it's either weighting, or donald's system, not both.
This is more or less as I meant it, except that I thought it would be based on the overall number of votes, not displaying the names of the games who have the largest amount of votes already. But I like your idea more. A variation of it would be:

You have four votes and a vote can be good or bad:
Vote 1) You have to rate the game with the least amount of votes.
Vote 2) You have to rate the game with the second least amount of votes.
Vote 3) You may vote on any game you like.
Vote 4) You may vote on any game you like.

Anyway, I'm afraid woogley has made up his mind already.

My 4k 2006 Entry: Sokoban4k
Offline woogley
« Reply #69 - Posted 2006-01-28 18:14:02 »

Anyway, I'm afraid woogley has made up his mind already.

not really, I'm not even coding it til mid february anyway. for the moment I'm following the KISS principle though
Games published by our own members! Check 'em out!
Legends of Yore - The Casual Retro Roguelike
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #70 - Posted 2006-01-28 18:29:33 »

Oh, but woogley, you misunderstood me again. I'm NOT saying we divide it into categories, I'm NOT saying you should decide on 150 different things. I don't even know where you got the category idea, because I didn't mention it... what categories were you even thinking of? Smiley

I'm suggesting a system as simple as yours, only more flexible:

1) You see a game.
2) You play it.
3) You like it, and think, "Wow, I like it"!
4) You give it a rating from 1 - 10. In this case, you give it a "10".

This way, you can STILL pick your favorite three games, and give them a 10 - you don't even have to vote on any other games. It's simple, it's fast, and it's possible withing 72 hours. However, if you like, you have the option to vote on every other game as well... "Let's see, let's give this one 8... I don't know why, but hell, I like it. Oh, this one, however, deserves no more than 2."

Simple, fast, fair.


EDIT: The only advantage I can see with your solution is that the statistics might, if we're very unlucky, change after 72 hours so that a new game gets the top rating. This, however, is only natural - you view things differently at different times. And, the one that won the first time obviously was a worthy winner at that time - perhaps it was more fun to begin with, but not so fun in the long run. Fine, so the contest isn't about the long run? Big deal - you didn't intend to consider the long run with the "top 3 favorites" anyway, did you? Smiley

Offline jbanes

JGO Coder


Projects: 1


"Java Games? Incredible! Mr. Incredible, that is!"


« Reply #71 - Posted 2006-01-28 18:30:44 »

* That is the best definition I've ever seen of "begging the question". By that definition, you are correct. I concede the point. Smiley

As for the rest, we can argue the details of each other's points until we're blue in the face. What I really want to know is, what is your EXACT problem? If you don't trust that Woogley is being truthful with us, then say so. Otherwise, I just don't see what your issue is. He's passing the credentials in exactly the same way each of own computers do every time we log into the site. The site can handle the multiple logins too, as I have logged in from multiple computers on several occasions.

I guess why we're annoyed is that you're objecting without any actual objection. Just various vague references about what might happen. Now I'll continue to look at his security, but as it stands now I see no real security issues above and beyond those that already exist.

Java Game Console Project
Last Journal Entry: 12/17/04
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #72 - Posted 2006-01-28 18:32:57 »

jbanes, have you even looked at the other posts? Only you and markus are arguing about security issue. It's the voting system that we're discussing Cheesy

Offline woogley
« Reply #73 - Posted 2006-01-28 18:36:04 »

Morre, I'm fine with people voting for the game quality from 1 to 10, and I dont mind implementing it. either way though there's a good chance there will be some games that nobody votes on.

one issue though. let's compare Miners4K - a good game, versus Warpstar4K - not so good. Let's say like 30 people or whatever vote somewhere between 7 and 10 and averages I duno.. like 9.2 or something. but lets say only one person votes for Warpstar4K and that vote happens to be a 10.

10 > 9.2

would I consider the winner like this?
1  
2  
3  
4  
if (Warpstar4K > Miners4K) {
  if (Miners4K.votes > Warpstar4K.votes) winner = Miners4K
  else winner = Warpstar4K
}
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #74 - Posted 2006-01-28 18:40:19 »

Again you're wrong. Warpstar IS a good game - I find it highly addictive Cheesy

Anyway, that's what the weighting system or donald's system is for. You just implement either donald's algorithm (so that all games get at least a decent amount of votes, perhaps almost equal) or the one that IMDB uses (which is server-side only, so no more work for the voters - it considers the number of votes and takes that into consideration. Also, a game has to get at least, say 10, 20, 100 (I don't know how many it'll be) votes to get considered into the contest, with the weighting technique that IMDB uses.)

Again, I quote IMDB:
Quote
The formula for calculating the Top Rated 250 Titles gives a true Bayesian estimate:

weighted rating (WR) = (v ÷ (v+m)) × R + (m ÷ (v+m)) × C


 where:

  R = average for the movie (mean) = (Rating)

  v = number of votes for the movie = (votes)

  m = minimum votes required to be listed in the Top 250 (currently 1300)

  C = the mean vote across the whole report (currently 6.Cool

You just apply that algorithm and hopefully it'll all work out magically, haha Smiley

Offline jbanes

JGO Coder


Projects: 1


"Java Games? Incredible! Mr. Incredible, that is!"


« Reply #75 - Posted 2006-01-28 18:41:43 »

jbanes, have you even looked at the other posts? Only you and markus are arguing about security issue. It's the voting system that we're discussing Cheesy

Indeed. I've been keeping an eye on it, but you guys are buzzing by so fast it's hard to keep up. So we've sort of ended up with two separate threads. Smiley

I have to side with Woogley on this one. The complex system that you and Donald are suggesting invites Murphy's law. Whatever can go wrong, will go wrong. Thus the "best" system is to allow each person to vote for their favorite game. Unfortunately, it's doubtful that single votes would give enough of a spread to declare anything more than first place. His system of voting for three options is a tried and true method used on many corporate boards for elections. For example, my condo association is incorporated, and has the shareholders - aka the owners - vote for as many candidates as there are positions. The problem is that board members are all equal at voting time. Since we want a clear winner here (instead of the three most popular), the weighting of +3, +2, and +1 make sense.

IMHO, if a simple system such as this is not welcomed by the community, then I would say that Kevglass has the right idea. Go back to using judges. Smiley

Java Game Console Project
Last Journal Entry: 12/17/04
Offline woogley
« Reply #76 - Posted 2006-01-28 18:43:53 »

Morre: yeah but what if 100 people vote a "2" for Warpstar. and 50 people vote "8" or "9" or whatever for Miners.

though Warpstar < Miners, Warpstar.votes > Miners.vote

see that's the concept I'm not clear on how to avoid
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #77 - Posted 2006-01-28 18:46:12 »

Then miners would win. The alghorithm takes that into consideration. The NUMBER OF VOTES do not decide the winner, but if an entry has very few votes, it's score is slightly lowered, or something like that Smiley

Offline woogley
« Reply #78 - Posted 2006-01-28 18:49:01 »

lol, seems a bit complicated. I dont have a final decision yet but I can tell you Im all about keeping it *simple*

this vote is cut-and-dry, which game is more widely thought of as the best. I welcome a sharp slant if there is one. some games are just that good!

your averaged statistics will come in the all-the-time-running category-based voting, so you will get the criticism you want over time
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #79 - Posted 2006-01-28 19:05:33 »

Fine, but I do not think this is complicated. All the work YOU would have to do is implement a very simple maths algorithm, clearly stated in one of my earlier posts, into the script that decides who wins. All the work the viewers have to do is select their favorites, and give them a "10". I'd say being allowed to vote freely on the games you like, and give them either positive (5+) or negative (1-5) votes, would be SIMPLER than forcing them to select three entries from a special page, and then submitting all the votes at once.

Offline woogley
« Reply #80 - Posted 2006-01-28 19:07:13 »

like I said, I'll consider everything. I wish I could continue this convo but a good friend of mine had a family member die sooo I'll chat up with you guys later Wink

thanks for all input!
Offline Markus_Persson

JGO Wizard


Medals: 16
Projects: 19


Mojang Specifications


« Reply #81 - Posted 2006-01-28 19:20:05 »

I agree(!) with jbanes. You people need to post slower. Wink

Play Minecraft!
Offline oNyx

JGO Coder


Medals: 2


pixels! :x


« Reply #82 - Posted 2006-01-28 22:50:12 »

Since voting is a damn complex topic (I mean it), I asked someone with lots of knowledge in that area.

---
If you consider more than two options the lesser_evil spoiler problem arises with single mark voting (aka Plurality or First Past The Post).

This problem does NOT get reduced at all with giving people several points for free allocation (aka Cumulative -- normally only recommended for multiwinner methods and even then it sucks).

But what if options got a score according to the inverse of their rank (without equal ranking except at the end)?  This is the Borda Count. It can give quite an incentive to give a complete ranking. Say you only want to rank 4 of ten candidates. Then the first gets 3 points, the second 2, the third 1. But if you rank them all, the first gets 9, the second 8, the third 7 etc. This doesn't seem to reduce the chance of winning for an option if several similar options are introduced. But consider the following:

Alice, Bob, Carl decide on their favourite movie.
Alice says: "My favourite movie of all time is Fist Of The North Star."
Bob says:  "My favourite movie of all time is Fist Of The North Star, too"
Carl says: "I liek Police Academy it is LOL!!1"

It seems the winner is clear. However Carl remembers he also likes some other movies.
Carl: "Police Academy I is best, Part II is almost as good as I, III is almost as good as II, IV is almost as good as III, V is almost as good as IV, VI is almost as good as V, and VII is almost as good as VI. Fist Of The North Star is almost as good as Police Academy VII. What are your opinions?"
Alice and Bob remark how much they hate all Police Academy movies and both give the following movie ranking:

1. FOTNS
2. PA I
3. PA II
4. PA III
5. PA IV
6. PA V
7. PA VI

Now guess who the Borda winner is 0_o

The chance for an option to win increases in Borda when similar options are in the race. I think this is much weirder than the opposite problem.  There are several versions of Borda but they all suffer to quite some extent from this. It is the most idiotic voting method I know.

It is much better when people are free to give ratings. However with ratings it is almost always best for a tactical voter to only use the extremes of the scale. Even if you have not the slightest idea how others vote! Big scales give an illusion of precision. They tempt people who are honest to have less influence on the outcome than tactical voters. So it is better to have a small scale. Binary is enough. The binary version of ratings is named Approval Voting and is advocated by political scientist Steven Brams.

When somebody first sees the difference between a First Past The Post ballot

MARK ONE OPTION. THE OPTION WITH MOST MARKS WINS.
A ( )
B ( )
C ( )
D ( )
E ( )
F ( )

and an Approval ballot

MARK ONE OR SEVERAL OPTIONS. THE OPTION WITH MOST MARKS WINS.
A [ ]
B [ ]
C [ ]
D [ ]
E [ ]
F [ ]

he might ask: "Isn't that against 'One Man One Vote'?"

Is he against women voting rights?

"Umm no. But candidate A and B are the most popular so almost everybody is voting for one of them. My true favourite is A, so I vote for A. And only A. Somebody else said he likes C so he will mark B and C and that is unfair."

Why?

"Because he has two votes and I have only one!"

Wait a second. You have it easy because your true favourite is somebody who will probably finish first or second. The other knows that his favourite chance to win is slim so he has to compromise by voting someone equal to his favourite. You would prefer to be in his situation?

"Butbutbut the method is crap anyway coz what happens if there are similar candidates? You know candidate E(litist McLefty) who is an elitist leftist has an identical twin F(renchy McLefty) with identical positions. Those that vote for E also vote for F and vice versa which increases the chance that one of them wins. I am glad they are not triplets."

You have it backwards. First Past The Post punishes similar candidates by splitting their support. This lack of punishment in Approval is not the same as giving a bonus to similar candidates like the Borda Count does.

"No you don't understand. I mean I voted for A, okay? Someone else can vote B, C, D, E, F? Those ballots have different weights. The other is five times heavier! That is logic."

Sigh. According to that logic, a ballot that approves all candidates would be the strongest. In First Past The Post, adding one ballot to a tie may break it and adding another ballot may recreate the tie. One can view this as evidence of equal ballot strength. However when there are more than two options, two First Past The Post ballots can not completely cancel each others effect. The two candidates in the close tie get further ahead of the rest. But in Approval two ballots can completely cancel each other and you gave just an example for that.

"But I like First Past The Post more. Approval is ugly and unintuitive. And it forces you to be dishonest."

However unpopular your favourite is with the crowd, you can rely on one thing with First Past The Post: You never have a reason to put the option you hate most at another place than the bottom. If there are several options you hate equally most, those go to the bottom no matter what your tactical concerns are.

This is also true for Approval. It adds the following: You never have a reason to put the option you love most at another place than the top. If there are several options you love equally most, those all go at the top no matter what your tactical concerns are. Approval is symmetrically completed Plurality.
---

Well, I agree (now) Smiley

[See also: http://en.wikipedia.org/wiki/Approval_voting]

edit:

One addition... immediate results shouldnt be shown at all. Regardless which voting system is used. (For preventing tactical voting)

edit2: sense++ Tongue

弾幕 ☆ @mahonnaiseblog
Offline darkprophet

Senior Devvie




Go Go Gadget Arms


« Reply #83 - Posted 2006-01-28 22:54:33 »

Woogly, Its nothing against you or your system, but I wouldn't trust my cat with my password, im afraid it might go tell other cats that and one of them might write it down...

I still think that getting a big amount of judges is a good idea. And letting them use whatever system you come up and let them register on your site. If you decide to want judges this year again, i'll be more than happy to be one...

DP

Friends don't let friends make MMORPGs.

Blog | Volatile-Engine
Offline woogley
« Reply #84 - Posted 2006-01-28 23:06:05 »

Woogly, Its nothing against you or your system, but I wouldn't trust my cat with my password, im afraid it might go tell other cats that and one of them might write it down...

you act like you're giving me the password, but I never see it at all. you trust all of the subnets between your computer and this site including through your ISP. I highly doubt my script would go tell other scripts about your password. as stated before, if you dont trust it, temporarily change your password ..
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #85 - Posted 2006-01-28 23:18:07 »

The approval system works for me, I guess... But I do think it should be modified slightly to give an average between 0 and 1 instead of letting the one with the most votes win. Not a big deal, but I think it'd be better in this particular case Smiley

Offline oNyx

JGO Coder


Medals: 2


pixels! :x


« Reply #86 - Posted 2006-01-28 23:34:51 »

Morre, can you please eleborate with an example. Its sorta unclear what you mean. Do you mean using a scale for the rating or an alternative way for representing the results? o_O

弾幕 ☆ @mahonnaiseblog
Offline Morre

JGO Knight


Medals: 2
Projects: 10


I'm Dragonene on IRC.


« Reply #87 - Posted 2006-01-29 00:54:27 »

Well, I do know what I meant, but on second thought it's silly Smiley
Never mind that last suggestion.

Offline oNyx

JGO Coder


Medals: 2


pixels! :x


« Reply #88 - Posted 2006-01-30 01:45:15 »

Btw we can run the poll with this forum software without any modifications. Well, that is if we use approval voting.

Say there are 50 games at the end. Then there are 50 options, 50 votes per user, the poll exipres in x days and
  • Only show the results after the poll has expired.

And the question could be something like "What were the best games this year?". You could vote for one game, for your personal top 3, top 5 or top 10... or even for all (voting for all or not at all has the same effect).

弾幕 ☆ @mahonnaiseblog
Offline kappa
« League of Dukes »

JGO Kernel


Medals: 78
Projects: 15


★★★★★


« Reply #89 - Posted 2006-01-30 01:57:31 »

of course its a good idea to have an overall best 4k game vote but i think we should also put in other stuff, like sub catogories like, best gameplay, best graphics, best technical achievement, etc etc, and maybe go even further with top platformer, top puzzle game, top shooter, top action etc. (maybe even like top 5)

not only that but we should rate different parts of a game, gameplay, replay value, sound, graphics etc. this way we should get much more accurate view of the contest.

this way alot more of the contestants will be happier with the outcome, as it can be quiet a let down if you put a lot of hardwork into a game and it doesn't get the recogition it actually deserves.

although this may sound like a lot of work on the part of the voters a few quick check boxes, pull down menus should make it quick and easy.

thats just what i think would be best IMO, donno but what do u guys think?
Pages: 1 2 [3] 4
  ignore  |  Print  
 
 
You cannot reply to this message, because it is very, very old.

 

Add your game by posting it in the WIP section,
or publish it in Showcase.

The first screenshot will be displayed as a thumbnail.

xFryIx (55 views)
2014-11-13 12:34:49

digdugdiggy (34 views)
2014-11-12 21:11:50

digdugdiggy (29 views)
2014-11-12 21:10:15

digdugdiggy (23 views)
2014-11-12 21:09:33

kovacsa (46 views)
2014-11-07 19:57:14

TehJavaDev (50 views)
2014-11-03 22:04:50

BurntPizza (49 views)
2014-11-03 18:54:52

moogie (65 views)
2014-11-03 06:22:04

CopyableCougar4 (63 views)
2014-11-01 23:36:41

DarkCart (148 views)
2014-11-01 14:51:03
Understanding relations between setOrigin, setScale and setPosition in libGdx
by mbabuskov
2014-10-09 22:35:00

Definite guide to supporting multiple device resolutions on Android (2014)
by mbabuskov
2014-10-02 22:36:02

List of Learning Resources
by Longor1996
2014-08-16 10:40:00

List of Learning Resources
by SilverTiger
2014-08-05 19:33:27

Resources for WIP games
by CogWheelz
2014-08-01 16:20:17

Resources for WIP games
by CogWheelz
2014-08-01 16:19:50

List of Learning Resources
by SilverTiger
2014-07-31 16:29:50

List of Learning Resources
by SilverTiger
2014-07-31 16:26:06
java-gaming.org is not responsible for the content posted by its members, including references to external websites, and other references that may or may not have a relation with our primarily gaming and game production oriented community. inquiries and complaints can be sent via email to the info‑account of the company managing the website of java‑gaming.org
Powered by MySQL Powered by PHP Powered by SMF 1.1.18 | SMF © 2013, Simple Machines | Managed by Enhanced Four Valid XHTML 1.0! Valid CSS!