Suggestion for rating threads.

[bartonski]

I would like to suggest a "thumbs up" or "thumbs down" rating system for threads, a-la digg/reddit/commandlinefu. No one seems to use the ratings in stars, I don't know why, but I think that I would be a lot more likely to bump a thread that I like up if there were simple "up" and "down" buttons.

[Hangdog42]

I guess I'm at a bit of a loss as to why we would need to rate threads. And in general, the only reason to bump a thread is if the problem isn't solved. Threads here are designed to solve a problem, and if the conversation reaches an end, there is really no reason to bring the thread up again.

[jeremy]

Thanks for the feedback. There already is a "thumbs up" functionality for individual posts and the 5 star rating for threads that you mention. At this time we have no plans to change the way rating a thread works.

--jeremy

[bartonski]

Originally Posted by Hangdog42 I guess I'm at a bit of a loss as to why we would need to rate threads. And in general, the only reason to bump a thread is if the problem isn't solved. Threads here are designed to solve a problem, and if the conversation reaches an end, there is really no reason to bring the thread up again.

Ladies and Gentleman, if you've ever wondered what the true definition of a moot point is, this would be a perfect example. A moot point is a point of law that has been decided by the court, but which remains interesting to argue. I say this because I do realize that Jeremy's decision is final.

So, Hangdog, I respectfully disagree :-). Threads on Linuxquestions.org remain useful far beyond when they end. They are a valuable resource for other people researching the same problem. There's also a lot of linux knowledge buried in some of the better threads; these would act as good training tools. I maintain that there is good information buried in Linuxquestions' databases which is not bubbling to the top because we don't have effective mechanisms for making this happen... but as I said, it's a moot point.

[onebuck]

Hi,

Originally Posted by bartonski <snip> So, Hangdog, I respectfully disagree :-). Threads on Linuxquestions.org remain useful far beyond when they end. They are a valuable resource for other people researching the same problem. There's also a lot of linux knowledge buried in some of the better threads; these would act as good training tools. I maintain that there is good information buried in Linuxquestions' databases which is not bubbling to the top because we don't have effective mechanisms for making this happen... but as I said, it's a moot point.

I'll add a simple point. If you are doing a search here on LQ with broad keywords then the refinement of the words as associated to the problem will produce some of the knowledge within those posts. The rating system will not do you that much other than when YOU associate the problem/solution. Now tags will help define the search along with your keys while searching thus producing a helpful list of threads.

The activity of the threads will cause the thread to move to the top. No activity, the thread elevation descends. If it was a moot point then why raise your question?

If we classed the threads within a forum then that may aid some.

[Hangdog42]

Threads on Linuxquestions.org remain useful far beyond when they end. They are a valuable resource for other people researching the same problem.

Actually I agree quite strongly on this. I rarely have to start a thread here due to the huge volume of existing information.

There's also a lot of linux knowledge buried in some of the better threads; these would act as good training tools.

Maybe I'm strange, but when I search here, I've never wished I had a ranking system for the threads. My opinion is that good information can be had from most threads and defining what is a "good" vs. "bad" thread is so subjective as to make the rating system kind of pointless. Also, the definition of a good thread changes over time as software progresses. What was a good thread five or six years ago might be completely pointless (or worse) today.

I maintain that there is good information buried in Linuxquestions' databases which is not bubbling to the top because we don't have effective mechanisms for making this happen.

I'm reading between the lines here, but this suggests that somehow the search functions aren't living up to expectations. My experience here suggests quite the opposite. I find that the only real modifications to the search that I need are time based (newer threads are frequently, but not always, more relevant). However, if you're seeing some deficiency, that is certainly worth discussing.

Ladies and Gentleman, if you've ever wondered what the true definition of a moot point is, this would be a perfect example. A moot point is a point of law that has been decided by the court, but which remains interesting to argue. I say this because I do realize that Jeremy's decision is final.

Actually, I don't believe this is a moot point for the sole reason that Jeremy is largely open to suggestions from the LQ community, even when they go against his prior notions. That is one of the truly outstanding features of LQ and I think a huge contribution to its success.

[bartonski]

Originally Posted by onebuck Hi, I'll add a simple point. If you are doing a search here on LQ with broad keywords then the refinement of the words as associated to the problem will produce some of the knowledge within those posts. The rating system will not do you that much other than when YOU associate the problem/solution. Now tags will help define the search along with your keys while searching thus producing a helpful list of threads.

But what if you are not doing a search? If you're browsing for interesting things to learn? Yes, LQ is a wonderful place to get questions answered, but is that the only way that it should be used? Let's say that someone posed a very insightful question, which was followed by an exceptionally Lucid answer. Two posts, Five years ago. No one thought to rate the thread with 5 stars, because, well, who does that? My point is that I'll never know about those two posts, even though I might very well like to know about that little thing about bash, or perl or whatever. Just because I wasn't looking for it doesn't mean that I don't want to learn it.

The activity of the threads will cause the thread to move to the top. ]No activity, the thread elevation descends.

A flame war has a lot of posts, but I doubt that anyone would rate it highly.

If it was a moot point then why raise your question?

a) The point wasn't moot until Jeremy weighed in. b) The definition of a moot point is that even though the point has already been decided, there is still benefit in the argument.

[bartonski]

Originally Posted by Hangdog42 Actually, I don't believe this is a moot point for the sole reason that Jeremy is largely open to suggestions from the LQ community, even when they go against his prior notions. That is one of the truly outstanding features of LQ and I think a huge contribution to its success.

Thank you for this observation. I haven't been around and active here for long enough to really see that in action.

Duly noted.

[onebuck]

Hi,

Originally Posted by bartonski But what if you are not doing a search? If you're browsing for interesting things to learn? Yes, LQ is a wonderful place to get questions answered, but is that the only way that it should be used? Let's say that someone posed a very insightful question, which was followed by an exceptionally Lucid answer. Two posts, Five years ago. No one thought to rate the thread with 5 stars, because, well, who does that? My point is that I'll never know about those two posts, even though I might very well like to know about that little thing about bash, or perl or whatever. Just because I wasn't looking for it doesn't mean that I don't want to learn it. A flame war has a lot of posts, but I doubt that anyone would rate it highly. a) The point wasn't moot until Jeremy weighed in. b) The definition of a moot point is that even though the point has already been decided, there is still benefit in the argument.

Knowledge is gained! If someone shouts in the woods the mysteries of Life or science and no one hears it then nobody gains. What about that hidden book that contains the answers to all your questions? Do you just wonder between books or define the search? My point is if you really want to learn something that is worth it's weight then you will work to find that same jewel. You can be mentored, steered or whatever you wish to call it but until YOU take the steps to open the door and go through it to gain some understanding then your goals will never be met.

Perusing may get you some information immediately but until you gain the detail for the whole then the answers will be fragmented. It seems we differ on how to attain knowledge but that does provide color to the world of knowledge.

Flame wars are just that 'wars'. Heated debates with respect along with good communication will win out for me anytime.

[bartonski]

Originally Posted by onebuck Perusing may get you some information immediately but until you gain the detail for the whole then the answers will be fragmented. It seems we differ on how to attain knowledge but that does provide color to the world of knowledge.

Ok. That may be so, but it's not the whole story. I ran across a method of incrementing serial numbers within strings in perl somewhere in the last 3 months, e.g.

filename000.txt
filename001.txt
filename002.txt
filename003.txt
filename004.txt
...

It was the most elegant solution that I've ever seen to this problem.

a) Without explicitly searching in the threads that I've posted to, how would you find that?

b) I didn't know that I was looking for that. I stumbled on it because I was replying to a newbie who had a 0-reply post of a Perl question, and I know a little Perl. It just so happened that several Perl gods converged on the thread all at once, and there were several excellent suggestions about improving the quality of the OP's perl code. If I had an 'increment' button on that thread, I definitely would have rated it up, and other Perl people would find it, probably even some time in to the future.

Education is the process of learning what you didn't know you didn't know. My example above was pure serendipity, but I'm always happy when someone else makes me more likely to bump in to interesting and useful stuff that they've happened across.

[titanium_geek]

couldn't you have used the 5 star thing?

TG

[bartonski]

Originally Posted by titanium_geek couldn't you have used the 5 star thing? TG

Yeah, I'm about to do that, but I still maintain that a cumulative rating system has the "wisdom of the masses" thing going in its favor.

[dragonslayer48dx]

Originally Posted by bartonski Yeah, I'm about to do that, but I still maintain that a cumulative rating system has the "wisdom of the masses" thing going in its favor.

I believe it might work if everyone rated threads honestly. Sadly, I doubt that would be the case- too many people base their opinion solely on their own personal interest.

Cheers

[XavierP]

And even more people will not rate a thread at all. Rating threads is so very very subjective that I cannot see any use for this at all. What is the definition of a highly rated thread? Speed of answer? Number of answers? Detailed question/detailed answer? Would we downvote a "simple question" or "simple/short answer"? What value would the rating have? - if I have a KDE issue, would I really care that the top 10 rated threads are all about Gnome?

Personally, and just my 2p, this would add to the confusion factors and not give us anything of value, especially since the primary method to find useful info should be the search function.

I downvote the idea in this thread. Or I will once the button is implemented