Community archive

Ask something
Discussion Locked

First revised thread display mock-up

0 helpful
16 replies

by Jay Garmon

Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

This link will take you to sMoRTy71's first mock-up of our revised top-level thread display.

http://img.com.com/i/tr/member_sig_meter_1108.gif

The major new feature here is the rating meter. We've been developing a community member "quarterback rating tool" for lack of a better term. It tracks certain key data points in community, weights them, a derives a score. The rating would use a 30-day rolling track of these stats, so being on top wouldn't mean you can rest on your laurels to stay on top.

This is our first stab a creating a more useful member rating system than TechPoints. We may as well get into that debate now.

--Jay

16 replies

by Jay Garmon

The tool is still pretty rough, but here's a sample of some notable scores I ran last week.

Maxwell Edison: 188
EdLockett: 80
tksolway@shaw.ca: 9
ozi Eagle 50
TheChas 233
CG IT 139
sgt_schultz 318
GuruOfDos 0
HAL9000 229
Oz Media 533
BFilmFan 691
James Linn 100
markusfrei@gmx.net 541
statykserver 363
willcomp 277
ReWrite 232
Garion11 114
zlito@aol.com 53

The system values Tech Q&A activity over discussion activity, which is why some of our heaviest posters may have scored low, and some quiet Q&A helpers scored high.

--Jay

by oz_media

Hey Jay, I know you explained how the participation level works, but Max scored WAY below me and he has been offering his help here for years longer than I with a points score way above my own, Colin (Hal9000) is a hardware GURU in comparisson to myself (I usually email Colin my own problems when stuck) and I know he is pretty active in Q&A.

Are you sure the results are reasonable? I scored in the top three and only hold an MCNE and know nothing in comparisson to most of these guys. While I'm somewhat flattered at the results, I still must be honest enough to question them.

OM

by willcomp

Kind of a "what have you done for us lately" rating system. The merit is that it rewards those most "effective" over a given time frame.

Hopefully, these ratings over whatever period you choose will not reward question askers more than question answerers which now seems to be the case.

And I certainly have no idea how I and some others came out on top of TheChas.

Dalton

by hal 9000

As almost every question that I place an answer to has already been visited by Chas and he has left a perfectly acceptable answer, which from my reading sounds correct.

It hardly makes it worth the effort in answering anything particularly when you see The Chas comments rejected for what appears no good reason and quite often no reason at all given.

Col

by bfilmfan

A lot of times I search the database for issues and see what some of the rest of you folks suggested to resolve issues. 99% of the time, I don't have to search any further.

About the only time I am in first is when it is an Active Directory freaky question. I am beginning to think that should be my nickname here "Freaky AD Question Guy."

by thechas

The mock up page looks great.
I love the better navigation buttons.

With the improved graphics and images, how is the page load time for our members on dial-up?

As to the ratings, I see a problem if your criteria is activity. One could theoretically get high scores from posting gibberish.

For starters, you should throw out activity in the "off-topic" threads when ranking individual activity. That way, the activity rating would be of more value to those looking to judge the merit of a posted opinion.

Compare Max and myself. Even though my rating is higher because of my activity level in the T Q&A section, I would trust an answer from Max over my own. Also, Max has a much better writing style than I have.

My point is that it would take a serious amount of data mining and number crunching to make any ranking system have true meaning and merit.

Sure, it was a real "kick" to break into the top 5 T Q&A points earned.
Still, I know that I got there mainly from persistence and spending a lot more time on the site than I really should.

I guess I would like to know what the goals are for your ranking system.

Chas

by smorty71

While the algorithm still needs to be tweaked (as most have suggested), the goal behind the meter is qualification of our members.

While it shouldn't be the only metric that readers of the boards take into consideration, we think that (when we get it right) this will be used along with other data points like the "member since" date, the "Interests" and "Technology Supported" sections of the profile page, etc. to let readers know more about the person posting.

One of the things we have heard from members who read the boards a lot (but don't post) is that they feel like they don't know enough to post.

We're hoping that things like "Interests," "Technology Supported" and some sort of activity/status meter will help those non-posting members see that not everyone who posts in our boards is a guru and that there input is valuable to the community.

by maxwell edison

.
(duplicate post - but I should have put it here in the first place.)

Okay, it seems that we're on a Q&A tangent; there have been a couple of discussions started recently about it; and, I suppose, I've been spending a little more time in the Q&A, and I'm noticing some things.

Rating and recognizing the members answering the questions, however they're rated, is great (and open to further discussion). But I've thought for a long time that you should also rate the people ASKING the questions - the acceptance rate, either by total answers accepted/rejected or questions with any answers accepted.

If, for example, a person has received 100 answere to various questions, and only one (or none) were accepted, that person would score very low, only 1 (or zero) percent. On the other hand, if that same person would have accepted all of them, he would be rated at 100 percent. So somehow rate the person asking the question.

Using ebay as an example, both the buyers and the sellers have ratings as well as buyer/seller comments that can be reviewed. If a person asks a question that might take a little time and/or research to answer, I think we need to know if that person is prone, based on past history, to reject most or all answers.

I'll bet a dollar to dirt that the following guy will reject ALL answers provided to this question:

https://www.techrepublic.com.com/5208-6239-0.html?forumID=48&threadID=163086&start0

His previous questions (some, not all):

https://www.techrepublic.com.com/5208-6251-0.html?forumID=48&threadID=162656&start=0&tag=search

https://www.techrepublic.com.com/5208-6251-0.html?forumID=48&threadID=161578&start=0&tag=search

https://www.techrepublic.com.com/5208-6251-0.html?forumID=52&threadID=161579&start=0&tag=search

https://www.techrepublic.com.com/5208-6251-0.html?forumID=48&threadID=161575&start=0&tag=search

Out of those 4 question, 5 answers were provided, and all of them were rejected. And in my opinion, they could have - or should have - ALL been accepted. They were perfectly acceptable answers.

This guy would probably have an "answer accepted" rating of around 10 percent or less. Some people would be zero percent. Contrast that with my "answer accepted" rating, which might be 99 percent. (The only answers I've ever rejected were from a guy or two that really pissed me off.)

Moreover, you might want to rate a member according to the points awarded. Again, in my case, I'll bet I've given away over 200,000 points. And many other people are generous with their points as well.

In short, recognize a members:

1. Contribution Factor (like you do now).
2. Generosity Factor (points given away).
3. Appreciation Factor (percentage of accepted/rejected)

by smorty71

Jay and I were just discussing that very topic this morning. We are definitely going to look at the activity of the person who asked the question.

In addition to the things you mentioned, we also discussed factoring in how many questions they leave open longer than 30 days.

Thanks for the good suggestions,
sm

by hal 9000

Is as Maxwell pointed out elsewhere those who post far too many questions and then either answer them themselves or have the same person constantly answering them.

Some of these seem to only remain open for a few minutes and appear nothing more than a points gathering exercise for those involved.

Incidental how did I get a higher ratting than Max as that is obviously incorrect I suspose you only took one days figures into account to get a basic view on things for the project.

Col

by willcomp

Max,

Agree with you except for one point. Don't think questioner rating should be based on number of answers accepted. That would only prompt more inane answers in hopes of generous acceptance.

Perhaps a better method would be question closure percentage with at least one accepted answer.

Dalton

by willcomp

Just got finished checking new Q&A posts. There are 2 with point value of 5.

Need to set a minimum point value (if point system is kept) that can be increased but not decreased.

I know we answer them all anyway, but fair is fair.

Dalton

by apotheon

I think that a rating based on acceptance and rejection rates should be maintained separately from expertise and participation rating(s). The acceptance/rejection rating is really only relevant to the Q&A section, and displaying that rating makes question acceptance its own reward. After all, when you browse through the questions to see if there's anything you can answer, and you find one that you have the answer to but see that the question poster has a truly abysmal acceptance rate, you're less likely to answer (after all, you might get a rejection, which could impact your point total).

Since nobody wants to answer that person's questions, that person won't get answers. This matches behavior in the real world: It's hard to get respect when you've lost it. When some guy tanks his own acceptance rating by rejecting every answer he gets, he's ruined his credibility as a question poster, and won't get answers to his questions. It becomes very difficult at that point to get answers to his questions, then, which makes it difficult for him to recover his acceptance rating.

Perfect, it seems.

by house

That's the general impression here too.

I agree entirely with your point. I don't believe that the acceptance rate of your answers plays any part in the validity of your replies.

I think that the rejection/acceptance rate directly affects the questioning member's ability to receive proper support from us.

I've also mentioned before that there should be a third and friendlier option regarding a reply to an answer. I would like to be able to say that someone was on the right track (and posting that message below their reply) without affecting their rating or the points that I put up.

Then, again, I don't usually ask questions, so it is difficult to put myself on the other end of the spectrum without a bias for our tech ratings. I've actually closed a post with rejections to change the topic, context, and location of my question. I feel bad now because I didn't realize how this would affect the other tech's ratings.

* Welcome to TRI, I've seen you battle with the ignorant posts and I appreciate your recognition of uninformed answers - their is much bickering in here about the situation out there.

PS - they posted an "open questions" link on the respective thread. Very handy indeed.

https://www.techrepublic.com.com/5204-6239-0.html?forumID=34&all=1&start=0

house, Chris G

by bfilmfan

I absolutely think that is a great idea. I've had folks reject my answer in the forum and email me asking for assistance!

I like the idea of seeing not only how many points this person has participating, but how do they rate when they ask a question.

Well met sir! Well met indeed!

by hal 9000

Answer to the question provided by the person who asked it.

What they did to fix the problem or whatever, all too often I see questions marked as Answered but every response rated as Not Accepted. Now I can understand that these real questions may be placed on many forums but still some form of answer should be provided if only so we who are interested enough can work out how they worded the question wrongly to get a Not Accepted response to what should under normal circumstances be correct answers.

All too often I get the impression that these people not only want to eat their cake but keep it as well.

Col

This discussion is closed to new replies.