12

The problem

As a site, we slightly diverge from the SE standards:

  • We have unique rules regarding voting, for example unreferenced questions should be downvoted to zero no matter how much we agree with their conclusions.
  • Our format allows us to tackle controversial topics in a civilized manner, but we must exercise restraint in the use of comments, especially with regards to "pseudo-answers".

On the other hands, our topics are interesting, we appear on the hot network questions list almost every day, and this brings in a lot of network users. Their voting strategies follow the network standards, e.g. "this was useful to me, +1". Also, they can't vote down at all!

This makes it impossible to have fair voting on posts on the hot questions list.

It is finally also evident that many want to offer their opinions on controversial topics, but that requires continual moderation -- we have hundreds of deleted comments.

This is not only a moderation problem, this is a problem for the whole community.

My proposal

  • Change the "comment everywhere" privilege to require 105 rep instead of 50
  • Change the "vote up" privilege to require 125 rep instead of 15 (similarly to the "vote down" privilege)

This would prevent network users from casting votes and commenting until they have done some useful action on the site, such as asking questions, edit them, or answer them. This, of course, will only apply on this site.

Another consequence is that it would be much harder to get +100 or +200 score questions and answers. This is unfortunate, but unavoidable.


Some numbers on posts with score 10 or more (or answers to questions with score 10 or more) since June 1st, with users grouped by rep (101 rep are "external" and not 101 are "internal"):

Comments

Internal comments deleted: 90 (19.4% of internal comments)
External comments deleted: 75 (30.5% of external comments)

Questions

Questions are considered "bad" if they are closed or deleted, good otherwise (remember these are posts with at least score 10). We only consider upvotes because there are basically no downvotes on questions.

Internal upvotes on bad questions: 52 (5.8% of internal upvotes)
External upvotes on bad questions: 41 (8.5% of external upvotes)

Answers

Answers are considered "bad" if they are deleted or appear on a closed or deleted question. We consider "inconsistent" an upvote on a bad answer or a downvote on a good answer:

Internal upvotes on bad answers: 133 (9.1% of internal votes)
External upvotes on bad answers: 113 (13.3% of external votes)

data here

  • 11
    Is there any chance to actually change this? I try to avoid running full speed into concrete walls, and as far as I understand the official SE position, this is that kind of wall. I don't see SE allowing any site to change the comment and upvote privileges. I've argued about the hot questions list and voting by network users without participation on the actual site in the past, pretty much for the same reasons you made this meta post, but I didn't have the impression that there is any room for change in this regard. – Mad Scientist Jul 2 '15 at 21:58
  • 1
    This is just a configuration change, not a code change. It's a matter of making a good case for the community managers like Shog and Jon Ericson. – Sklivvz Jul 2 '15 at 22:35
  • 17
    Bandwagon voting from HNQ is a problem on all sites (especially ex. Puzzling). Why does Skeptics in particular need changed privileges? Why solve a problem on one site when you could solve it on all of them (perhaps not with the exact same details, but still)? – Doorknob Jul 2 '15 at 23:12
  • 4
    @Doorknob let me express myself better: more sites might have the same problem. It's not a huge problem on most sites. Each site can pitch their own case, imo. I don't see changing the standards for all sites as currently feasible. – Sklivvz Jul 2 '15 at 23:20
  • Alternative: disable hot network questions, or tweak how they work? – Andrew Grimm Jul 3 '15 at 0:54
  • 3
    Please, yes. Just yes yes. Not even a hmm, maybe yes, just a Yes. (I'm a fan of this proposal, in case that was unclear) – Jamiec Jul 3 '15 at 8:03
  • 12
    Seems reasonable. One issue, though. People who can't comment tend to leave answers. Bad answers – Paul Jul 3 '15 at 9:53
  • 1
    @Paul true, but our users have a much better track record of dealing with those. Also losing reputation makes users learn fast what works and what doesn't. – Sklivvz Jul 3 '15 at 10:20
  • 7
    A vote on this question does NOT mean "this is useful question": instead it means "I agree/disagree with this feature request". – ChrisW Jul 3 '15 at 16:50
  • 1
    Voting for previous feature-request topics e.g. meta.skeptics.stackexchange.com/q/3228/2703 suggested that user don't know (or don't follow) that voting convention for feature requests. – ChrisW Jul 3 '15 at 19:25
  • So I think that the statistics which you just posted say that, if you eliminate all input from new/external users, then you eliminate 40% of all bad input (75/165 bad comments, 41/93 bad question upvotes, and 113/246 bad answer upvotes)? – ChrisW Jul 3 '15 at 21:37
  • 1
    @ChrisW it's probably much more than that, as for example bad comments might instigate bad internal comments, bandwagon effects, etc. which I've not investigated at all. This is just to show that the quality is objectively inferior. I don't really want to fuel the discussion with these numbers, just show I'm not making up there's an objective issue to talk about. – Sklivvz Jul 3 '15 at 21:41
  • 1
    For people concerned with the number of comments deleted, note that "obsolete" is a common reason - i.e. a request to fix a question or answer that is carried out by the OP, and therefore is no longer relevant for anyone to read. – Oddthinking Jul 6 '15 at 1:40
  • 4
    30.5% versus 19.4%, 8.5% versus 5.8%, and 13.3% versus 9.1% may be statistically significant (unless confounding factors are involved), but doesn't seem large enough to justify new rules being taken against them. – Andrew Grimm Jul 13 '15 at 12:45
  • 1
    Related: meta.stackexchange.com/q/238420/321647 – Revetahw Oct 25 '16 at 9:06
25

I strongly disapprove of this suggestion.

I only approve of the argument saying that 101 users should have the privilege to vote down.

What is meant by "guest users" is users who just signed up on Skeptics Stack Exchange and got the 100 rep. association bonus.

The problem is that the OP assumes that:

All guest users (who just signed up and got the 100 association bonus) are bad contributors and create bad comments.

This is not true. Many people who come from other sites post high-quality comments; this proposal will strip them of that right. Here are examples of useful comments from 101 rep. users on highly up-voted answers:

enter image description here

enter image description here

enter image description here

enter image description here

enter image description here

Good comments from 101 rep. members are countless. There might be a lot of off-topic comments which "require continual moderation" but there are also a lot of good comments from 101 rep. members. We shouldn't prevent 101 rep. members from commenting just because there are comments which "require continual moderation". See for example, a 101 rep. member, with comments more useful than mine:

enter image description here

Moving on:

Their voting strategies follow the network standards.

Citation needed. But it doesn't matter anyway because even if their voting strategies followed the network standards, bad answers (and even highly up-voted) are removed. Moderators constantly clean-up and delete answers that don't meet the site standards regardless of the voting.

This will also cause a loss of reputation for good contributors (who get up-votes from members of sister sites). This will decrease the excitement of writing answers and this will slow quality members from gaining access to privileges.


Summary: The proposition fails to provide concrete data showing that most comments from "guest users" are not constructive and useless. The proposition also fails to provide concrete data showing that most "guest users" have "voting strategies" different than the voting strategies of skeptics.se.

  • They still can't downvote, you didn't address that George – Sklivvz Jul 3 '15 at 17:38
  • @Sklivvz, I agree with you on that. They should be able to vote down. – George Chalhoub Jul 3 '15 at 18:37
  • 3
    Speaking of evidence, do you have any evidence that upvoted comments are "high-quality" or "useful"? I suppose Sklivvz might be suggesting that new users, who don't know the site's rules, don't know what a "good answer" and "bad answer" are: and therefore don't know how to vote; and neither do they understand how to comment (and, perhaps more importantly, how to refrain from commenting). – ChrisW Jul 3 '15 at 19:03
  • 4
    Also George, saying that I assume that external contributors are all bad is completely incorrect. Their total input to the site is negative, but singularly there are surely many great contributors. But if with them come hordes of bad ones, then we have a problem. – Sklivvz Jul 3 '15 at 20:00
  • 2
    @GeorgeChalhoub you say in the answer that there is (basically) no problem, but then you agree that we should lower the downvote privilege below 101. Which is which? As it stands, nothing will be done -- which may not be what you intended, based on your comment. – Sklivvz Jul 7 '15 at 12:55
  • @Sklivvz: I edited my answer to make it clearer. – George Chalhoub Jul 7 '15 at 13:26
  • 1
    @Sklivvz: I a;so turned my downvote to an upvote (since you mention the problem that 101 users can't downvote and added numbers). – George Chalhoub Jul 7 '15 at 13:32
0

I don't support the proposal. The numbers don't convince me that external voters are so much worse or add so much workload to moderation that they require special handling.

I volunteer to moderate if it is simply a workload issue.

We also have no idea what affect the proposed change will have on our user acquisition/growth funnel.

  • 2
    Thanks for volunteering, with your reputation there's tons you can already do. The only alternative to reducing the incoming bad votes is increasing the good ones. If you think we can undo the bad voting with ours, let's try it, but it will take dozens of users, not one. – Sklivvz Jul 4 '15 at 15:57
0

While I agree that you have now shown that the contributions from guest users are (statistically) significantly less useful than the contributions from experienced users, in my opinion you have not yet justified how they are not worth the time and moderation effort required to make their votes and comments suitable for the site.

Especially with the very high degree of moderator intervention here as compared to the average SE site (personal experience with the sites I have used), I am of the opinion that the relatively small amount of bad votes due to network standards would not significantly impact on the fact whether bad questions/answers are deleted, since mod deletions do not require the answer or question to be downvoted to begin with.

Unless the network users start posting questions and answers (which would then quickly turn them into experienced users), they would not marginally increase the moderator workload in closing/deleting questions and answers.

On the other hand, they would increase the workload involved in deleting comments, but the numbers involved seem to be relatively small (75 deleted per month, which IMO is worth the time considering that they do in fact produce many useful comments).

  • The "number of comments" point doesn't make much sense to me: yes, they are 75 comments, but concentrated in a handful of questions (less than five!). – Sklivvz Jul 7 '15 at 11:11
  • 1
    @Sklivvz How is that problematic? If the questions are concentrated in a few questions, won't it be easier to moderate these bad comments? – March Ho Jul 7 '15 at 13:20
  • It's problematic because of two reasons: firstly, these comments on controversial posts in the HNQ list are often due to people having arguments and taking sides. In this case removing them is not enough as the argument has already happened. Secondly, many of these are from users posting unreferenced answers as comments and this is a problem that is endemic on Skeptics. It is not worse on these particular posts posts, but it still a problem. – Sklivvz Jul 7 '15 at 14:12
  • @Sklivvz Is it however true that arguments are more likely to be caused by guest users than experienced users? I don't see a reason why guest users are more likely to start arguments. – March Ho Jul 7 '15 at 14:14
-4

A similar but variant proposal: can you remove the +100 'association bonus' which people get from having been users on other sites?

  • A good idea. Not easily, that would also remove a lot of other privileges which we want/need to keep (e.g. chat). The proposed change is a configuration change. – Sklivvz Jul 2 '15 at 21:48
  • Can you say whether a majority of the deleted comments were from new users, or from old users? Does it even matter if new users can't participate in chat, to talk about a hot network question, until they get 20 points (i.e. until they can vote which happens at 15 points)? Anyway though, a config change might be more feasible than a change in hard-coded behaviour. – ChrisW Jul 2 '15 at 21:54
  • 4
    Talking in chat wouldn't suffer, that's based on network rep anyway and doesn't care about your rep on the room's parent site IIRC. Flagging, though, would take a hit, which probably isn't desirable. – Undo Jul 2 '15 at 23:11
  • 2
    Downvote. You want to strip the 100 reputation people get when they visit skeptics? That will prevent good contributors and quality guests from contributing to skeptics. – George Chalhoub Jul 3 '15 at 15:16
  • 3
    @GeorgeChalhoub It would mean they must gain reputation (experience) on Skeptics, by "contributing" an upvoted question, and/or an upvoted answer, and/or some approved edits, before they're allowed to vote or comment. – ChrisW Jul 3 '15 at 16:01

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .