LinuxQuestions.org
Register a domain and help support LQ
Go Back   LinuxQuestions.org > Forums > LinuxQuestions.org > LQ Suggestions & Feedback
User Name
Password
LQ Suggestions & Feedback Do you have a suggestion for this site or an idea that will make the site better? This forum is for you.
PLEASE READ THIS FORUM - Information and status updates will also be posted here.

Notices

Reply
 
Search this Thread
Old 09-05-2010, 03:48 PM   #16
xeleema
Member
 
Registered: Aug 2005
Location: D.i.t.h.o, Texas
Distribution: Slackware 13.x, rhel3/5, Solaris 8-10(sparc), HP-UX 11.x (pa-risc)
Posts: 987
Blog Entries: 4

Rep: Reputation: 248Reputation: 248Reputation: 248

Quote:
Originally Posted by XavierP View Post
To be honest, I'm a terrible person to be compared against - I've been here forever, most of my actions these days are mod actions and to cap it all, I'm not terribly fussed about my scores Be happy that you have a decent number and don't let the scores affect your time at LQ.
@XavierP
Yeah, but here's my point; Out of your +18,000 posts, even if *one* percent were seen as helpful by others, you've out-paced someone like me, and therefore should show up as a "More Reputable" member than myself.
A percentage-based addition to the Reputation system doesn't do long-standing LQ members justice, IMNSHO*.

<not_sucking_up>
I was just using you as an example, because based on your posts in the past, I'll trust you to have the experience and know what you're talking about over any advice I would give (which is normally based off of a little experience and a few tests before I open my big mouth).
</not_sucking_up>

IMNSHO = In my not-so-humble opinion.
 
Old 09-05-2010, 07:14 PM   #17
slakmagik
Senior Member
 
Registered: Feb 2003
Distribution: Slackware
Posts: 4,113

Rep: Reputation: Disabled
Quote:
Originally Posted by xeleema View Post
@XavierP
Yeah, but here's my point; Out of your +18,000 posts, even if *one* percent were seen as helpful by others, you've out-paced someone like me, and therefore should show up as a "More Reputable" member than myself.
A percentage-based addition to the Reputation system doesn't do long-standing LQ members justice, IMNSHO*.
I'm not picking on XavierP either but, if *one* percent were seen as helpful, that means he didn't help 99% of people - he's wasting a lot of people's time. Someone who bats a thousand is actually doing a better job, even if he helps fewer people. If that person helps you, you're one of the lucky few and should listen carefully. But under the current cumulative system, that person looks far less reputable. I agree that helping more people is a better thing and that an absolute percentage would perhaps be unfair to heavy-duty helpers but the pure cumulative system is just as unfair to light-duty helpers. It would be nice if the number of a person's posts or absolute count of helpful posts could weight in favor of them, but not so much as a purely cumulative system does. The system shouldn't be pure quantity, though it shouldn't be purely anti-quantity, either.

Right now it's almost pure quantity, though. If I post a thousand idiotic posts and accidentally get ten right, I'm going to have more rep than if I post nine out of nine good ones.
 
Old 09-05-2010, 08:40 PM   #18
XavierP
Moderator
 
Registered: Nov 2002
Location: Kent, England
Distribution: Lubuntu
Posts: 19,176
Blog Entries: 4

Rep: Reputation: 430Reputation: 430Reputation: 430Reputation: 430Reputation: 430
A huge number of my posts are standard mod responses or even hand crafted ones, I would expect not to be repped at all for them
 
Old 09-06-2010, 04:40 AM   #19
H_TeXMeX_H
Guru
 
Registered: Oct 2005
Location: $RANDOM
Distribution: slackware64
Posts: 12,928
Blog Entries: 2

Rep: Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269
Quote:
Originally Posted by jeremy View Post
You're using numbers from many days after I gave my example, so will get a different end result.

--jeremy
Oh, I see, then I was using the right numbers. Ok, then. Still, I am against a percentage system, the numbers show the same trend.
 
Old 09-06-2010, 07:13 AM   #20
MTK358
LQ 5k Club
 
Registered: Sep 2009
Posts: 6,443
Blog Entries: 3

Rep: Reputation: 713Reputation: 713Reputation: 713Reputation: 713Reputation: 713Reputation: 713Reputation: 713
I see that a percentage system says how helpful a person is, while the current system shows the total help.
 
Old 09-06-2010, 11:49 AM   #21
jeremy
root
 
Registered: Jun 2000
Distribution: Debian, Red Hat, Slackware, Fedora, Ubuntu
Posts: 10,390

Rep: Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626
I've added the LQ Helpful Quotient to the Statistics tab so we can further evaluate this. Note that this information was always available on that tab, we're simply doing the math and displaying it now. Additional feedback is welcome here.

--jeremy
 
Old 09-06-2010, 12:05 PM   #22
jeremy
root
 
Registered: Jun 2000
Distribution: Debian, Red Hat, Slackware, Fedora, Ubuntu
Posts: 10,390

Rep: Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626
Thinking about this further, there are actually two slightly different ways to generate the LQHQ, and I'm interested in which way gives a more accurate picture:

Code:
($number of posts that have been rated helpful / ($number of posts that have been rated helpful + $number of posts that have been rated unhelpful)) * 100
vs

Code:
($number of times any post has been rated helpful / ($number of times any post has been rated helpful + $number of times any post has been rated unhelpful)) * 100
--jeremy
 
Old 09-06-2010, 04:35 PM   #23
slakmagik
Senior Member
 
Registered: Feb 2003
Distribution: Slackware
Posts: 4,113

Rep: Reputation: Disabled
I'm not sure. I suppose just observing some posters and their possible numbers under various calculation might give the best empirical answer. I think I lean towards the second one, but I'm not sure.

There are further calculation possibilities: you could factor in the aggregate positive/negative over the number of individual posts so that people could even exceed 100% if they had 5 positives and 1 negative over 3 rated posts, giving 133%. Or you could do the positive/negative over all posts, including posts that weren't rated at all, so that 5 positives and 1 negative over 20 posts (whether rated or not) would give 20%. This looks bad but maybe it could be converted into a letter grade graded on a curve or something.

I also like MTK358's post in #8 of this thread where it just shows positives, negatives, and neutrals as a spectrum. Then it wouldn't boil down to specific numbers but you'd just get a 'lotsa green, little red' impression on the more reputable posters. But I think something like that would have to calculate only the posts made since the system came on line.

Kinda bike-shedding it here.
 
Old 09-06-2010, 06:22 PM   #24
GazL
Senior Member
 
Registered: May 2008
Posts: 3,392

Rep: Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917
When the helpfulness thing was first introduced I thought the idea behind it was to provide a mechanism to rate the content of post, not to rate the poster. Now it seems this stat is being presented as some sort of worthiness metric on our stats page. Mine is currently sitting at 94%.

The irony here is that in my case this 6 point drop was a result of taking part in the original discussion in LQ Suggestions about the helpfulness system when everyone was playing with the yes/no system to test it out. Now I never really cared about those handful of no votes sitting there on my stats page in the past, but as soon as you turn it into a percentage you're effectively branding me as "6% unhelpful" and I find that I do care about that.

This post is not about my specific situation, those negative votes will trend towards zero as things progress and I get additional positives ratings. I simply want to raise the point that unlike an open ended number like the rep system currently uses, as soon as you assign someone a percentage, you're not only saying what they are, but you're also implying what they're not. You need to be very careful about the psychological effects of this. No one like to be unappreciated and you could end up driving people away.
 
1 members found this post helpful.
Old 09-07-2010, 08:32 AM   #25
H_TeXMeX_H
Guru
 
Registered: Oct 2005
Location: $RANDOM
Distribution: slackware64
Posts: 12,928
Blog Entries: 2

Rep: Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269Reputation: 1269
Well, mine is currently 89%, and a good part of it was due to arguments I had with others. Meh, whatever, I don't really care about this statistic.
 
Old 09-07-2010, 09:42 AM   #26
jeremy
root
 
Registered: Jun 2000
Distribution: Debian, Red Hat, Slackware, Fedora, Ubuntu
Posts: 10,390

Rep: Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626
Quote:
Originally Posted by GazL View Post
When the helpfulness thing was first introduced I thought the idea behind it was to provide a mechanism to rate the content of post, not to rate the poster. Now it seems this stat is being presented as some sort of worthiness metric on our stats page. Mine is currently sitting at 94%.

The irony here is that in my case this 6 point drop was a result of taking part in the original discussion in LQ Suggestions about the helpfulness system when everyone was playing with the yes/no system to test it out. Now I never really cared about those handful of no votes sitting there on my stats page in the past, but as soon as you turn it into a percentage you're effectively branding me as "6% unhelpful" and I find that I do care about that.

This post is not about my specific situation, those negative votes will trend towards zero as things progress and I get additional positives ratings. I simply want to raise the point that unlike an open ended number like the rep system currently uses, as soon as you assign someone a percentage, you're not only saying what they are, but you're also implying what they're not. You need to be very careful about the psychological effects of this. No one like to be unappreciated and you could end up driving people away.
Thanks for the feedback. The main intention of the Helpful system is indeed to help find quality posts. Some members were interested in seeing the data from another angle, and as mentioned this data was always available (albeit in a different format). This is something we're still very much evaluating, so additional comments/suggestions are welcome. Also keep in mind that you can completely disable the "Statistics" tab via your privacy settings.

--jeremy
 
Old 09-07-2010, 02:16 PM   #27
GazL
Senior Member
 
Registered: May 2008
Posts: 3,392

Rep: Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917
I'm not that bothered with the stats being visible to others jeremy. It was just that while I never really cared about a few negative ratings when it was just a count, I was surprised to find that I did care when you put it in a percentage format. I have to admit that I care less today than I did on first seeing that stat but as I had a negative kneejerk reaction to it, I just thought it worth mentioning.

Going slightly tangential for a second: Have you ever considered extending the star rating system that exists for threads to individual posts? It seems to me that that would be a good way to achieve the goals you had for the helpfulness system and would be in a format the userbase should be very familiar and comfortable with.
 
Old 09-07-2010, 02:34 PM   #28
jeremy
root
 
Registered: Jun 2000
Distribution: Debian, Red Hat, Slackware, Fedora, Ubuntu
Posts: 10,390

Rep: Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626Reputation: 2626
We originally considered a 1-$x rating system as opposed to the current yes/no options, but went with the simpler and more widely deployed/accepted option.

--jeremy
 
Old 09-07-2010, 03:03 PM   #29
GazL
Senior Member
 
Registered: May 2008
Posts: 3,392

Rep: Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917Reputation: 917
fair enough. I'll shut up then.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
LQ Reputation System jeremy LQ Suggestions & Feedback 1086 03-27-2011 01:31 PM
Email reputation mrlinux2000 Linux - Server 4 11-10-2009 06:42 AM
LQ User Reputation System jeremy LQ Suggestions & Feedback 55 12-10-2008 08:06 AM
Reputation System cs-cam LQ Suggestions & Feedback 45 01-07-2006 10:47 AM


All times are GMT -5. The time now is 01:54 PM.

Main Menu
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
identi.ca: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration