Welcome to Codidact Meta!

Codidact Meta is the meta-discussion site for the Codidact community network and the Codidact software. Whether you have bug reports or feature requests, support questions or rule discussions that touch the whole network – this is the site for you.

Comments on Answer sort order unstable - is it just me and my browser?

Post

Answer sort order unstable - is it just me and my browser? [duplicate]

−0

Closed as duplicate by elgonzo‭ on Jun 25, 2021 at 14:13

This question has been addressed elsewhere. See: Randomize order of answers with same score

This question was closed; new answers can no longer be added. Users with the reopen privilege may vote to reopen this question if it has been improved or closed incorrectly.

The question "Threaded comments are here!" currently has 5 answers.

I noticed that the "Score" sort order (the default sort order) does not appear to work reliably. This behavior also occurs when the "Score" sort order is explicitly specified in the URL, like https://meta.codidact.com/posts/282342?sort=score.

Often, Lundin's answer (current score +7/-1) is shown first, followed by hkotsubo's answer (current score +4/0). This is alright and makes sense, as of time of writing Lundin's answer has a better vote "ratio".

But sometimes when revisiting or refreshing the page, hkotsubo's answer appears first, followed by Lundin's answer. (large screenshot) The vote count and the score of the answers have not changed, though.

I also noticed Peter Taylor's answer (score currently +4/-1) sometimes swapping places with FoggyFinder's answer (score currently +2/0). (large screenshot) while the score of the two answers remains unchanged. The occurence of this is independent of Lundin's and hkotsubo's answers swapping places.

Worth mentioning is also the observation that during my trials, only the 1st and 2nd answer swapped places, or the 3rd and 4th answer swapped places. It never happened to me that 2nd and 3rd answer swapped places, and the 5th answer with the lowest score always was shown as last answer, too.

UPDATE: Based on hkotsubo's comment, i realized the 1st and 2nd answer having both the same calculated score, and the 3rd and 4th answer also having the same score. Thus, it's reasonable to narrow down the problem scenario to groups of answers with same score in which the intermittent change of order takes place.

Only the "Score" sort mode seems to be affected. I could not observe similar ordering issues when using the "Active" (age) sort order.

bug

posted over 3 years ago

CC BY-SA 4.0

3y ago

elgonzo‭

1 2 8 52

Raw

Markdown

History

is a duplicate

This question has been asked before and has already been answered. It should be marked as a duplicate.

Please enter the URL of the proposed duplicate in the details field below.

not constructive

This question cannot be answered in a way that is helpful to anyone. It's not possible to learn something from possible answers, except for the solution for the specific problem of the asker.

1 comment thread

Maybe status-by-design? (8 comments)

elgonzo‭ wrote over 3 years ago · edited over 3 years ago

copy link

hkotsubo‭, i don't think so. Regardless of how scores are calculated/weighed, it should not result in the observed behavior. Note that the scores don't change, just the order sometimes changes between page reloads. However, the intermitting ordering issue indeed happens only between answers with the same score. The age of the answers might be a contributing factor, but it's hard to tell what exactly is at the root of the issue i observe...

FoggyFinder‭ wrote over 3 years ago · edited over 3 years ago

copy link

I agree that's exactly what hkotsubo‭ is described. Take a look at the score formula:

with z = 2: z ^ 2 = 4, (z ^ 2) / 2 = 2 we get

(upVotes + 2) / (allVotes + 4)

We can easily calculate scores of each answer:

4|0 = 0.75 = (4 + 2)/(4 + 4) = 6 / 8 = 3 / 4

7|1 = 0.75 = (7 + 2)/(8 + 4) = 9 / 12 = 3 / 4

are grouped together as well as

2|0 = 0.66(7) = (2 + 2)/(2 + 4) = 4 / 6 = 2 / 3

4|1 = 0.66(7) = (4 + 2)/(5 + 4) = 6 / 9 = 2 / 3

hkotsubo‭ wrote over 3 years ago · edited over 3 years ago

copy link

@elgonzo Considering the formula described in Help Center, and the fact that we don't know the z value for this specific site, there are 2 possibilities: 1- z has a value that makes those answers have the same Wilson Score (and, according to this, their order is randomized) - 2- the Wilson Scores are not the same, and it's a bug. But I guess only mods or staff can tell us

elgonzo‭ wrote over 3 years ago · edited over 3 years ago

copy link

Okay, i got you. Didn't expect that +7/-1 results in the same score as +4/0 (one downvote worth 3 upvotes looks way too punishing to me). Didn't expect that the order of answers with same score is intentionally randomized either (it makes sense to spread the limelight evenly across answers of same/similar quality, though). Well, i guess my Q can then be closed as a duplicate of the second question/answer hkotsubo mentioned in their comment. Unfortunately, i can't upvote hkotsubo's comment :-(

FoggyFinder‭ wrote over 3 years ago · edited over 3 years ago

copy link

Well, i guess my Q can then be closed as a duplicate of the second question/answer hkotsubo mentioned in their comment.

I disagree. I count that this Q is a totally valid one and isn't a duplicate. I think it would be better to await confirmation from mods that z currently is equal to the default value and then hkotsubo‭ can transform comments to the answer.

elgonzo‭ wrote over 3 years ago · edited over 3 years ago

copy link

FoggyFinder‭, my Q is a duplicate. My confusion was stemming from not being aware of the As affected by the swapping having the same score, and that randomizing the order of As with same score is intentional. The score of an A is shown as a tooltip when hovering with the mouse pointer over the votes of an A, so while it is not "in your face", it is not secret information either. And the randomized order of As with same score is what the Q i marked my Q as a duplicate of talks about. (1/2)

elgonzo‭ wrote over 3 years ago · edited over 3 years ago

copy link

(2/2) The randomization/changing of the order of the As is specifically what my Q asked about, so how is it not a duplicate? Whether the scoring algorithm is correct or using wrong parameters is unrelated to my Q and nothing my Q asks about, and should imo be discussed/asked as its own question addressing the scoring methodology specifically.

Communities

Comments on Answer sort order unstable - is it just me and my browser?

Answer sort order unstable - is it just me and my browser? [duplicate]

1 comment thread