But that looks like there can only be one poster in a thread that gets thanked. But on a technical forum like this, that just won't work, because it's not like everyone immediately posts complete solutions instantly. It's more gradual, and many posts can be helpful toward reaching the goal.

Thanks for the feedback, gnashley. I'll take a look.

This is something I put on the LQ TODO a while back, so you should see something similar here at some point. The issue that MTK358 brings up is something we need to think about a bit, which is why we haven't implemented it already.

--jeremy

MTK358, when users agree with another poster response, they earn points as well, IIRC.
Jeremy, I think this addresses your concern as well. It really helps make past threads easier to use and more valuable in the future. Cuts down on new threads which cover the 'same old ground' as existing threads.
I'm all for users being able to have pride in having a good reputation or being thanked for their participation. This system just makes it easier to tie the accuracy of a given answer to a the problem at hand.

Occassionaly the OP will choose the wrong answer, but not often. If they
do they might be punished by having points taken away -if they chose a 'wrong' answer as being the best answer, they were probably not phrasing their question correctly, anyway.

I am not an expert on Linux. There are a very very few things I know about the internals of linux. But then now and again I come across a post that has an issue I am familliar with and I try and help this person with his issue. But then again due to the lack of detailed knowledge I may not be able to really help this guy fix his issue. Now the thing here is 'reputatation' is a very strong word for me and to numerically calculate reputation is something I would not really be looking forward to. The simple 'thank you' button was good. But then again, you guys have been here a lot longer than I have and perhaps there is something that I have missed out. If given the option I would definately delete 'reputation' from being displayed in my posts. Hope I have made my position clear on the topic. BTW, repling to a post by Jeremy somehow is very intimidating so please forgive me in case I am not really eplicit.

Actually, I agree, it was a lot simpler, and I think better. Now, things are more complicated, but who knows, maybe it'll work.

to be honest , i appreciated the old "Thanks" system much more.

i think that , a user would only need the raw metrics themselves rather than an arbitrary algorithm to calculate the reputation of a member..

for me , knowing 3 basic things such as Join Date , Post Count and Thank Count would be more than enough to assess the reputation of any member !

i wouldn't need automation for that , that's my thought

cheers

entz, thanks for the feedback. What we're trying to accomplish is something more useful than a simple anecdotal mental summation. As the system matures we'd appreciate any feedback on how we can better accomplish that.

--jeremy

interesting ideal

Thanks for your dedication and all that hard work!

Nice ah

And that is the problem. From what little I have seen of a description of how the system does work or will work (and yes, I read the whole thread referenced at the top of this one), I see no sign of adequate measures against various forms of cheating, especially knocking down reputations unfairly.

That is why though I may still occasionally return here, LQ will no longer be my primary source of Linux information.

Only a very small subset of members have the ability to down rep. From there, there are actually many measures in place to ensure it's not abused. Even with those measures in place, negative rep is something we watch very closely to make sure it's not being abused.

--jeremy

Additionally, negative rep (as a percentage) is continuing to go down as the system gets more usage. It will likely settle at under 1% of usage (and < .3% of rep points) given the current trends.

--jeremy

An additional note: if you see specific cases of abuse or have specific realistic theoretical cases in mind, do let us know. We've tried to address everything possible, but any system can always be improved. Thanks again for the feedback.

--jeremy

Well, for sure I will report real cases of abuse, so far none.