LinuxQuestions.org - Please rethink your policy on review post counting

- LQ Suggestions & Feedback (https://www.linuxquestions.org/questions/lq-suggestions-and-feedback-7/)

- - Please rethink your policy on review post counting (https://www.linuxquestions.org/questions/lq-suggestions-and-feedback-7/please-rethink-your-policy-on-review-post-counting-301707/)

On the subject of post counting would it be possible for other people to rate your posts ?
Ie all posts get a default rating eg 2 (and all posts in General are not rated) and other users can mark them higher. After all if user X post 50 times asking for help
and user Y posts 10 times solving critical problems, it seems strange that for a newbie Y at first glance looks less knowledgable than X.
I know that this could be difficult to implement and may have other problems but I still think it would be a good idea.

I am not re-opening the issue of "General" post counting. But I consider "Review" post counts to be of importance because I consider it a small acknowledgement on the part of the community towards my contribution.

I request everybody not to go offtopic and start another discussion about the "General' posts counting to your total or whether post counts really matter or not.

That is not the issue here.

Quote:

Originally posted by davholla
On the subject of post counting would it be possible for other people to rate your posts ?
Ie all posts get a default rating eg 2 (and all posts in General are not rated) and other users can mark them higher. After all if user X post 50 times asking for help
and user Y posts 10 times solving critical problems, it seems strange that for a newbie Y at first glance looks less knowledgable than X.
I know that this could be difficult to implement and may have other problems but I still think it would be a good idea.

This is available, and unfortunately is under-used.
There's a "Thanks" button on each posts to rate them using Affero.

Quote:

Originally posted by J.W.
It's true that some folks may initially give more weight to posts from members with high post counts, but in the end, I would say that what matters most is whether a recommendation actually works or not. To put it another way, would you rather get one reply that works from someone with 50 posts, or a dozen replies from people with 5000 posts, none of which solve the problem? Granted, at least initially a high post response may appear to have more credibility, but as they say, the proof is in the pudding.

Please understand, I'm not trying to argue here. As I see it though, the reality is simply that post count != experience/knowledge, which was the point I was trying to illustrate. My apologies for not being more clear. -- J.W.

Of course - no need for apology - I understood quite clearly and was just making the point.

Quote:

Originally posted by syg00
Lose all the counts.

As stated above, they are at best misleading.
Content is the paramount consideration - readers without the experience to determine the value of a post are potentially going to be mis-led by falling back to a reliance on the post count.

Ooh ooh.. I vote for that! :)

And not to lead off topic but its still sort of on topic since were talking about what counts and what shouldn't in a number system next to one's name on this forum.

I did compare to the General section in how we don't count posts in members totals as you have to think, there is more non-technical threads and off topic discussions, but there are plenty of technical threads and even windows questions asked there, but those who ask such questions there don't get any added points to their total post count? I find it along the same lines of why we don't, err didn't count the posts compared to the reviews.

Perhaps we should make it an option to toggle our own post counts and scores off and on, cause I would certainly toggle mine off, I think its misleading and members look at you differently if they judge by the numbers you have.

OK, we're mixing a few to many topics in this thread to keep them straight I think. The topic at hand it how to handle the "Reviews" line in each post. I can see wanting to show both products and reviews, but I'm still not 100% convinced it's worth the extra query. Feedback on that is what I am interested in. If someone wants to start a thread about post count please feel free (although removing it at this point wouldn't likely happen, I'm always open to discussion).

--jeremy

if the extra query is significant, it's not worth it given the value of the information. just a total count of reviews would be fine, imho.

Yes, too wide... Starting new thread as ::
Post And Review Counting, Next Thread

Edit-P.S.:
Cute sig, Genesee [sp?], I'm grinning widely & will probably remember that one.
(Best-yet belongs to a Win4Lin-list user: "My computer said to install Windows XP or better, so I installed Linux :)" I might have a word or two wrong there, but..... :) :) ;) :)

Post And Review Counting, Next Thread

Continued/widened from thread:
LinuxQuestions.org > Forums > Website Suggestions & Feedback > Please rethink your policy on review post counting

Ok, My $0.0002-worth (Not typo!)...

Human nature requires that: "if there are counts, mine will be biggest". Sigh, but true.
(Even women have a healthy dose of this, but arguably are a little less aggressive about it, at least to start with...)

OTOH, counts *ARE* useful. They also mean more server-cpu's to extract. Sigh, in stereo.

Yes, the student should read through all answers and figure out which one(s) is(/are) best, but the simple truth is, that newbies and other less-expertise types, myself frequently included, aren't going to be able to do this very well, leading to frustration-with, avoidance-of, and ultimately discarding-of, Linux. And don't belittle the fact that a lot of things require making a backup & then restoring when the test fails, which is non-trivial-time, frustration, and drive-space of some kind. Human nature at statistical work. Truth.
Ok, I'm not *quite* a newbie, but, given the hu-monstro-jig-awesome volume occupied by Linux and its programs et al, "'taint _possible_" to know much about everything, and I'd ask y'all to admit to yourselves whether this applies to you --personally--. Thus there is only so much I'll(/we'll) do unless it is life-or-death.
So for at least some things, we need counts. SIGH, in 5.1 surround-sound.

And yes, based on the 'nature requires' that I started with, I'm afraid that the best of all *is* to have both counts -- kudo's to the path-breakers, and still credit to the trail-smoothers that follow them. sigh.

Somebody suggested to filter out and ignore reviews that are too small to have any (probable!) use. I agree, but this minimum needs to be *just!* large enough to allow a single-point note. And the submitter needs to offer at least -some- description and details. This would increase the usefulness of the resource, while not requiring that the additional submitter(s) to write a whole review, which would, humans being humans, tend to reduce submissions, thus usefulness.
I would think maybe 200--300 characters, as a shoot-in-the-dark guess... Probably needs to be higher than that...

We DO need to solve the "post-count // post-value" dilemma for this to be *most* useful for *any* counts.
Sugestion: add or switch-to total affero's granted, with or without, or maybe with *AND* without, donations.
Of course that means convincing more of us to be good citizens here, and I admit that I got splattered with that brush too. (sigh -- and i bet you're getting tired of that sound, right?...)
Whether-or-not to keep the raw count...: I vaguely suggest keeping 'em.
It would be hard to separate questions from replies, though; a lot of replies are not direct attempts to answer the question, and besides, if the questioner is asking a lot, AND remembering it, they are learning, and might thus turn into useful answerers/helpers, too.

To make all this easier, it might be possible to dB so a single query gives all these values back from a single record -- assuming that this is not already true! --, and 'paste' the values in order into the info-block being created. I'm guessing that the dB record would contain the appropriate quantity of binary entries; the bin-to-text conversion takes a few cycles, but no re-delves into the dB with associated drive-access.

Getting off small wooden toy pony now.

Comments?

rgh.

P.S.: will someone please make this message start a new thread with the subject I gave it, while I research how to do this here?
Please?
Thank you.