21
$\begingroup$

It seems that there is a widespread concern about the influence that in many ways the switch to the stack exchange network has on this site. I infer this from the many questions on meta on this subject, for example the recent one on the "publicist" badge.

So, borrowing my words from a comment of, I ask about

Should we consider to ask Stack Exchange

(i) to exclude questions from this site from the hot list?

(ii) to remove the association bonus of 100 reputation points that any user of any forum of stack exchange get when coming to math-overflow?

I am interested in the community's opinion both and the desirability and the feasibility of those measures, and also in any other suggestions of measures to the same effect, namely distend our relations with the rest of stack exchange.

$\endgroup$
  • 5
    $\begingroup$ I juts realize that this question of Steven Landsburg (meta.mathoverflow.net/questions/435/the-association-bonus) discusses one of the measure. I agree with everything Steven says, except for one thing: "I realize we are not currently overrun by barbarians at the gates". Well, it was true when written in JUly, but I am not sure of this anymore :-) $\endgroup$ – Joël Dec 21 '13 at 13:55
  • 4
    $\begingroup$ I have another meta-question. If there is a strong community support on any of the measure suggested above, who gets the power of decision about it ? Is it our moderator, or is it people (who?) of the stack exchange network ? $\endgroup$ – Joël Dec 21 '13 at 14:01
  • 3
    $\begingroup$ There are presently 4398 users with exactly 101 points -- see data.stackexchange.com/mathoverflow/query/155069/… . -- I would say this is quite an alarming figure ... . $\endgroup$ – Stefan Kohl Dec 21 '13 at 14:20
  • 12
    $\begingroup$ @StefanKohl users are created very easily once you have an SE account, it only takes two or three clicks. Many user create then on sites they never even post or vote, the statistic should probably exclude users that never acted on the main site at all. $\endgroup$ – Mad Scientist Dec 21 '13 at 15:00
  • 9
    $\begingroup$ MadScientist is quite right, only 665 such users have voted - data.stackexchange.com/mathoverflow/query/155101/… Note also that many of these users were created before 2013-06-24. $\endgroup$ – François G. Dorais Dec 21 '13 at 15:09
  • 3
    $\begingroup$ Yes, but 655 is still a lot. Am I the only one to have the feeling that the voting pattern (which questions get voted up, which question don't) has changed since a few months ? $\endgroup$ – Joël Dec 21 '13 at 15:28
  • 6
    $\begingroup$ How many users with 101 points posted bad questions? I would guess these are far, far fewer than bad questions posted by users with 1 point. $\endgroup$ – Gerald Edgar Dec 21 '13 at 15:38
  • 2
    $\begingroup$ @Gerald: sure, but we are not discussing bad questions/bad answers, but bad votes. $\endgroup$ – Joël Dec 21 '13 at 15:40
  • 3
    $\begingroup$ The question does not mention votes. Only the comments. $\endgroup$ – Gerald Edgar Dec 21 '13 at 15:42
  • 3
    $\begingroup$ Do upvotes for this question mean we are in favor of (i) and (ii), or that we are in favor of discussion of (i) and (ii)? $\endgroup$ – Gerald Edgar Dec 21 '13 at 15:44
  • 5
    $\begingroup$ On the heart of the matter, both bad questions and bad votes are a problem. But there is not much we can do about bad questions, in addition of what we already do (trying to prevent them by the FAQ, then chasing and closing them). On the other hands, bad votes, it seems $\endgroup$ – Joël Dec 21 '13 at 16:16
  • 5
    $\begingroup$ are caused in substantial part by users not having the mathoverflow culture, but brought here by the top questions list on stack exchange having the right to vote here because of the association bonus. On this it seems that we could do something easily. $\endgroup$ – Joël Dec 21 '13 at 16:19
  • 3
    $\begingroup$ @GilKalai it seems a site has to opt in to participate; on math.SE meta.math.stackexchange.com/questions/11706/… and a couple other sites there was a poll about it done by local mods, so I assume they were asked about it. And so it might be ours were asked too, and decided or forgot to, or maybe MO was forgotten. In any case I for one am extremely glad we do not have this. $\endgroup$ – user9072 Dec 22 '13 at 22:16
  • 3
    $\begingroup$ Thank you, quid. It is not that surprising that we see it differently: Well, I can brag about it among my colleagues and add it to my CV! $\endgroup$ – Gil Kalai Dec 22 '13 at 22:48
  • 3
    $\begingroup$ Actually, I did brag about it in a lecture on Internet-mathematics (a lot about MO) "Open Collaborative Mathematics over the Internet – Three Examples" see here gilkalai.wordpress.com/2013/08/19/… (and the first youtube.com/watch?v=P4yECgveQ2c and second youtube.com/watch?v=S215MRveokg parts of the lecture) (The hats came in the second part, I think...) $\endgroup$ – Gil Kalai Dec 22 '13 at 22:59
43
$\begingroup$

Upvote this answer if you think we should ask StackExchange to allow us to opt out of the "hot list".

(I'm adding this answer because it was not clear what an upvote to the main question would mean.)

$\endgroup$
  • 2
    $\begingroup$ Also relevant: meta.mathoverflow.net/questions/1232/… $\endgroup$ – Asaf Karagila Dec 21 '13 at 23:47
  • 5
    $\begingroup$ For the Math SE perspective on the hot questions list, see meta.math.stackexchange.com/questions/12161 and meta.math.stackexchange.com/q/11994. I think there's a lot of good evidence there that this is not something that MO should be involved in if we can avoid it. (putting this here because the comments on the main post are about more than just hot questions.) $\endgroup$ – Logan M Dec 23 '13 at 7:18
  • 3
    $\begingroup$ A slight variation of this could be: only registered MO users see MO questions on their hot-list. (Or: only MO users with ≥200 rep, or something.) This would keep the possible-benefit of nudging MO users back to the site to keep them engaged, without advertising MO questions to the whole network-wide audience. $\endgroup$ – Peter LeFanu Lumsdaine Jan 4 '14 at 17:11
22
$\begingroup$

Upvote this answer if you think we should ask StackExchange to allow us to opt out of the "association bonus".

$\endgroup$
  • 7
    $\begingroup$ Before voting on this answer, users might want to look at some of the abuses of the association bonus that I documented here: meta.mathoverflow.net/questions/435/the-association-bonus $\endgroup$ – Steven Landsburg Dec 21 '13 at 20:30
  • 10
    $\begingroup$ I agree with this in spirit, but I don't see any reason to bar users without the association bonus from meta, posting links/images, flagging, or commenting as these restrictions are mostly in place to block spammers. Hence it seems easier to me to argue that we should just raise the voting & bounty threshold above that of the association bonus than to remove the bonus entirely. $\endgroup$ – Logan M Dec 22 '13 at 10:00
  • $\begingroup$ While I agree the bonus has problems, removing it for MO would I think leave a bad taste in the mouth of SE users who have come to expect it. While we may not care about this so much (since they’re not the users we want), I can see why the SE admins might be unhappy about offending/alienating users this way. The “security through obscurity” approach of wholly or partially opting out of the hot list seems safer this way than the explicit restrictions which this suggestion or raising the voting threshold would impose on users. $\endgroup$ – Peter LeFanu Lumsdaine Jan 4 '14 at 17:17
17
$\begingroup$

What do you think about the idea of further restrictions to voting? We could require 50 reputation and at least one question or answer with positive votes. This would nullify the effects of "association bonus" on voting but it would keep its other benefits.

$\endgroup$
  • $\begingroup$ Good idea. This may be simpler to "sell" to the larger stack exchange community. $\endgroup$ – Joël Dec 21 '13 at 19:21
  • 2
    $\begingroup$ There is a general proposal to that extent on meta.SO by Mad Scientist and linked somewhere on our meta too; maybe I can find it. Added: here it is meta.stackexchange.com/questions/183109/… And for irony use your association bonus on meta.SO and upvote it if you like it! :-) (though metaSO being general there is no actual irony as there it makes sense) $\endgroup$ – user9072 Dec 21 '13 at 20:07
  • 2
    $\begingroup$ Thank you for the comment, I think that in general there are valid points in the proposal against this. But since the cross-voting affects the smaller community more, i think it may still be useful for MO. $\endgroup$ – Daniel Soltész Dec 21 '13 at 21:25
  • $\begingroup$ Good approach, since a crank can be quickly downvoted to the “no items with positive votes” state, whereas there is no procedure to lower reps for an user who acquired it through bonuses or editing, without creating anything. $\endgroup$ – Incnis Mrsi Nov 7 '14 at 11:19
  • 1
    $\begingroup$ Another feature request on meta.SE similar to the one mentioned in previous comment: by user9072 (probably quid): Prevent questions on Hot List from being upvoted by casual visitors (only rep is from association bonus) $\endgroup$ – Martin Sleziak Jan 19 '17 at 6:10

You must log in to answer this question.

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