Understanding through Discussion


Welcome! You are not logged in. [ Login ]
EvC Forum active members: 79 (8897 total)
Current session began: 
Page Loaded: 03-20-2019 1:21 AM
211 online now:
Dredge, Meddle, Minnemooseus (Adminnemooseus), vimesey (4 members, 207 visitors)
Chatting now:  Chat room empty
Newest Member: WookieeB
Post Volume:
Total: 848,463 Year: 3,500/19,786 Month: 495/1,087 Week: 85/212 Day: 1/14 Hour: 1/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Prev12
3
Author Topic:   Should Simple be Suspended
roxrkool
Member (Idle past 971 days)
Posts: 1493
From: Nevada
Joined: 03-23-2003


Message 31 of 40 (85355)
02-11-2004 10:13 AM


I think we should also seriously consider placing a limit on the number of *veterans* replying to newbie posts. Maybe the first 3 people to respond get to stick around. If a veteran decides to drop out, they could post that they are leaving and the next person can take their place.

It really gets to be a cluster-f*** when we have 10 veterans all hammering a newbie.


    
Mammuthus
Member (Idle past 4549 days)
Posts: 3085
From: Munich, Germany
Joined: 08-09-2002


Message 32 of 40 (85359)
02-11-2004 10:29 AM
Reply to: Message 28 by Percy
02-11-2004 9:37 AM


I think the problem with this is that some people generate more responses than others. The opposing side will likely give low scores to the person posting even if the discussion is rather interesting. Can you imagine Syamsu giving me a high rating for any post in any thread no matter what I say? Or Stephen? What I had in mind still leaves the decision up to the Admins i.e. who goes from debating newbie to old pro with the priveleges and responsibilities that moving up incurs...screw up once you get those privileges you get kicked back down by the Admins. The rating system by members might be a nice way of indicating to the Admins that a problem is arising.

Hey Dan, your last post was really short..I give it a 2


This message is a reply to:
 Message 28 by Percy, posted 02-11-2004 9:37 AM Percy has not yet responded

Replies to this message:
 Message 33 by Dan Carroll, posted 02-11-2004 10:36 AM Mammuthus has responded
 Message 35 by PaulK, posted 02-11-2004 10:43 AM Mammuthus has not yet responded

  
Dan Carroll
Inactive Member


Message 33 of 40 (85360)
02-11-2004 10:36 AM
Reply to: Message 32 by Mammuthus
02-11-2004 10:29 AM


Bah. I knew I should have stuck in a couple paragraphs about Eliza Dushku, or taken a slam at Hambre or something.

Gotta give the people what they want...


This message is a reply to:
 Message 32 by Mammuthus, posted 02-11-2004 10:29 AM Mammuthus has responded

Replies to this message:
 Message 34 by Mammuthus, posted 02-11-2004 10:38 AM Dan Carroll has not yet responded

  
Mammuthus
Member (Idle past 4549 days)
Posts: 3085
From: Munich, Germany
Joined: 08-09-2002


Message 34 of 40 (85361)
02-11-2004 10:38 AM
Reply to: Message 33 by Dan Carroll
02-11-2004 10:36 AM


No problem..I will use Stephen ben Yeshua H-D posterior plausibility analysis to raise the plausibility from 0.6 to 0.61 that Hambre gets a 1 for his next post for bringing up Kid Rock again.
This message is a reply to:
 Message 33 by Dan Carroll, posted 02-11-2004 10:36 AM Dan Carroll has not yet responded

  
PaulK
Member
Posts: 14747
Joined: 01-10-2003
Member Rating: 1.8


Message 35 of 40 (85362)
02-11-2004 10:43 AM
Reply to: Message 32 by Mammuthus
02-11-2004 10:29 AM


There's a particular problem with creationists. Some of them will tend to rate anything they agree with as "excellent" no matter how poorly reasoned. On the other hand there is someone here who classifies posts he can't refute as "yada".

I woudl suggest that a working system would somehow need to take into account the reliability of the person providing the score.


This message is a reply to:
 Message 32 by Mammuthus, posted 02-11-2004 10:29 AM Mammuthus has not yet responded

    
PaulK
Member
Posts: 14747
Joined: 01-10-2003
Member Rating: 1.8


Message 36 of 40 (85364)
02-11-2004 10:47 AM
Reply to: Message 28 by Percy
02-11-2004 9:37 AM


Another thought
An overall scoring system has a couple of problems - firstly few people are going to be motivated to rate a post unless they feel reasonably strongly about it. Secondly any numeric system is going to include an unavoidable subjective element.

I would suggest a good/average/bad system as the only score with the possiblity of adding some sort of qualitative element (what is "good" or "bad" about the post).


This message is a reply to:
 Message 28 by Percy, posted 02-11-2004 9:37 AM Percy has not yet responded

    
Adminnemooseus
Director
Posts: 3879
Joined: 09-26-2002


Message 37 of 40 (85760)
02-12-2004 11:16 AM


Simple has been exiled from the Main Topic Forums
quote:
Hi, Simple!

Please restrict your participation in the future to the Free For All forum and other non-Main Topic Forums (see List of Forums - you're no longer permitted in the forums listed under "Main Topic Forums").

-- Percy
EvC Forum Administrator


From http://http://www.evcforum.net/cgi-bin/dm.cgi?action=msg&f=7&t=123&m=138#138


    
Sylas
Member (Idle past 3334 days)
Posts: 766
From: Newcastle, Australia
Joined: 11-17-2002


Message 38 of 40 (85781)
02-12-2004 12:45 PM
Reply to: Message 28 by Percy
02-11-2004 9:37 AM


Percy writes:


I might have mentioned this once or twice in passing in other threads, but one of my long term plans (but hopefully this year) is to introduce a message rating system. The ratings members assign to messages would be roled up into a rating for each member. Sliding below a certain rating level would carry some automatic restrictions.

This is a place where you can learn from the experience of TheologyWeb.

They have two such systems operationing at present.

There are Pearls (of wisdom). Every user has a certain number of pearls. You can give pearls to another person anytime for any reason; in increments of 1 to 5. This will reduce your own count of pearls, and increase their count. There is thus a natural limit on giving pearls away, since you have to have them in the first place.

Then there are Reputation points; a very recent innovation. Every user has a certain number of reputation points. You can give a boost to any other post, which is either positive or negative. The actual amount conveyed depends only on the amount of reputation you have at present. There is no change to your own reputation when you boost or detract the reputation of another; and you cannot boost posts from the same person twice without having given reputation to 30 other people in between.

The reputation system is not working as well; and there are plans afoot by the programmers to find a new system in the near future. I personally think pearls are an excellent system; simple, and with natural limits to abuse.

The objective of any such system is to allow the readership some input into an evaluation of posts of other people. Given this, I think it is a mistake to worry too much about the system being abused. Let the people have their say. I'd recommend not using cummulative ratings to impose any effects at all on how a person can use the board; that would only encourage abuse. At the most, you could have some kind of indicator of a rolled up rating level displayed next to the Avatar. You will find that is more than enough to let people place some value on it.

Keep it simple. Just let everyone give a rating from -3 to 3 to every post, if they choose. Then you can look at the rating of a post. A rating can be supplied with a one line comment indicating why a rating is applied; and a list of ratings can be viewed from a button at each post. If different posters have different levels of influence, then the list can also show the comparative influence of raters; and either a raw or scaled sum of ratings.

For the cummulative individual rating, either just give a simple sum; or else allow for a more complex formula if you want to avoid meaningless boosting of one individual without regard to content.

I have a modest proposal...

A person is likely to be contributing substantively when they make a high proportion of posts to a thread; but not too high. Let the activity level be a simple count of posts by one user which are in a thread where that user contributes less than 30% of the total input. However, let activity also have a natural decay function, with a two month half life; that makes it a measure of recent activity.

Let the impact of a poster be the logarithm of their activity level. Persons with more activity have more impact; but it gets harder and harder to get more impact as your level increases. One post is still zero impact; impact starts with two posts.

Let the kudos of a poster be the sum of all ratings received, scaled by impact, and also reduced by a simple exponential decay function, like a that of a radioactive substance; kudos has a half life of two months. A cute tie into a relevant empirical phenomenon, and an encourangement to newbies that the well known experts can't rest on their laurels. Any new poster who maintains activity and garners rating will soon overtake someone who is taking an extended break.

Note that ordinary moderation as it is applied at present gives a natural regulation to such a system. Moderators should avoid actually playing with the rating system directly; though they may be empowered to remove impact from individuals who are rating posts without any regard to content.

Note also that the effect of one poster on others is related to impact and activity; not kudos. This is deliberate; and is intended to reduce the possibility of the system being dominated by one point of view. You want to remain friendly to users from all sides; especially creationists, frankly. So have a system where any poster can hop in and start to contribute, and have an impact as a result.

Cheers -- Sylas

[This message has been edited by Sylas aka cjhs, 02-12-2004]


This message is a reply to:
 Message 28 by Percy, posted 02-11-2004 9:37 AM Percy has responded

Replies to this message:
 Message 39 by Percy, posted 02-12-2004 8:41 PM Sylas has not yet responded

  
Percy
Member
Posts: 18307
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.7


Message 39 of 40 (85918)
02-12-2004 8:41 PM
Reply to: Message 38 by Sylas
02-12-2004 12:45 PM


Sylas writes:

The reputation system is not working as well; and there are plans afoot by the programmers...

Programmers? That's programmers with an "s"? Wow!

I've had nibbles of programming help from NosyNed and Joz (I think), but nothing's come through yet. I like your ideas a lot, but couldn't tell from them if you also have a software background. But if you do and you're interested in implementing some of your ideas, let me know.

--Percy


This message is a reply to:
 Message 38 by Sylas, posted 02-12-2004 12:45 PM Sylas has not yet responded

Replies to this message:
 Message 40 by NosyNed, posted 02-12-2004 8:57 PM Percy has not yet responded

    
NosyNed
Member
Posts: 8837
From: Canada
Joined: 04-04-2003


Message 40 of 40 (85920)
02-12-2004 8:57 PM
Reply to: Message 39 by Percy
02-12-2004 8:41 PM


Nibbles but some people aren't reliable. It is still on my todo list. However, I've been roped into some work for the karate dojo (they hit harder than you do) and the possibility of paid work too. I do want to help and want to learn.

There has been a lot of other things going on with me right now too.

sorry.


This message is a reply to:
 Message 39 by Percy, posted 02-12-2004 8:41 PM Percy has not yet responded

  
Prev12
3
Newer Topic | Older Topic
Jump to:


Copyright 2001-2018 by EvC Forum, All Rights Reserved

™ Version 4.0 Beta
Innovative software from Qwixotic © 2019