I am no longer interested in math.stackexchange.com, and would prefer not to log in so that my "last login" reading reflects this fact.

However, if I log into other stackexchange sites, and then accidentally visit math.stackexchange.com, it will automatically log me in.

How do I change that behavior? (preferably for math.stackexchange.com only)

  • 12
    $\begingroup$ Sorry to see you go. $\endgroup$ – Asaf Karagila Feb 6 '15 at 10:11
  • 10
    $\begingroup$ How do you change the behavior of accidentally visiting math.stackexchange.com? Years of psychoanalysis, I'm afraid. $\endgroup$ – Gerry Myerson Feb 6 '15 at 11:39
  • 1
    $\begingroup$ Some browsers now have addons (Auto private in firefox, Ghost Incognito in chrome) which allow one to automatically enter private browsing/incognito mode when visiting specific sites/domains. I have never try any of them but it might be useful for you. If you really want something to stop you from visiting math.SE, the easiest way is to ask your anti-virus/firewall program to block it! $\endgroup$ – achille hui Feb 6 '15 at 13:20
  • $\begingroup$ This seems, to some extent, relevant: How can I disassociate an account? $\endgroup$ – Martin Sleziak Feb 6 '15 at 14:03
  • 1
    $\begingroup$ IIRC there were some users who were banned from this site on their own request. I wonder, whether a ban would work for what you want. $\endgroup$ – Martin Sleziak Feb 6 '15 at 14:04
  • $\begingroup$ Off-topic: as you know the reasons for high-reputation users leaving or losing interest are often a matter of considerable interest and relevance in later debates. In case you do not mind, could you briefly indicate your reason(s). $\endgroup$ – quid Feb 6 '15 at 14:33
  • 9
    $\begingroup$ @quid Hurkyl's profile states "I believe we have lost the fight to keep math.stackexchange.com from the path of becoming a homework mill." That may be the reason. $\endgroup$ – epimorphic Feb 6 '15 at 14:43
  • $\begingroup$ @epimorphic thanks. I could and should have thought of checking there myself. $\endgroup$ – quid Feb 6 '15 at 14:49
  • 4
    $\begingroup$ Thus is the fate of math.se if people still encourage homework questions instead of downvoting them immediately. $\endgroup$ – Mister Benjamin Dover Feb 6 '15 at 15:02
  • $\begingroup$ To comment on the off-topic comments. I'm very surprised with the reason why Hurkyl is leaving. Since I've been a member, the MSE is at the apogee of the anti-PSQ trend. $\endgroup$ – Git Gud Feb 7 '15 at 13:12
  • 3
    $\begingroup$ @Git: The will to fight seems strong, but the anti-anti-PSQ trend has been picking up. But even putting that aside, the reality on the ground (with my perspective possibly skewed by combat fatigue) feels like PSQs have too much momentum to be stopped. Of course, last time I gave up on MSE was because I thought the anti-PSQ movement was once again going to run out of steam against the pro-PSQ disguised as maintaining the status quo in the face of a lack of consensus, and I was pleasantly surprised when I checked in a few months later. That may happen again, but I'm much less hopeful. $\endgroup$ – Hurkyl Feb 7 '15 at 18:40
  • 2
    $\begingroup$ @Hurkyl: Even without the enablers (do they even have significant effect?), the forces producing PSQs and their answers would still be as crushing as they are. And I don't have any illusions about beating them back either. But I think we can be a dam in the river of questions, not stopping the flow altogether, but regulating it as well as we can. By adding value to good questions by giving them good answers, we're still doing good work here. $\endgroup$ – rschwieb Feb 8 '15 at 0:07
  • 1
    $\begingroup$ Did you change your mind? (At the moment last seen on main only 22h ago.) As it might not be clear, the point of writing this is so that you might change the accepted answer (since it does not work). $\endgroup$ – quid Mar 2 '15 at 21:57
  • 1
    $\begingroup$ @quid: The accepted answer works most of the time. I had hoped I could figure out what the difference is on those rare occasions where the software logs me in again, but I can't tell anything. I had been putting off unaccepting the answer until I had a chance to tell if I had been doing something different to make it log me in those few times; since I have nothing, I suppose by this point now is the time to do so. $\endgroup$ – Hurkyl Mar 2 '15 at 22:32

The request to make it possible to disable auto-login on per-site level has been declined by SE. The answer there contains a suggestion (item 2) for a client-based solution, which would take some work to implement. Also, there is an excellent chance that it will stop working when SE rolls out the new login workflow: see New year, new experiment: Login and Signup UI.

If you prefer to keep using your accounts on other SE sites, but not on this one, the surest way is to request the deletion of your account on this site.

  • 1
    $\begingroup$ This should be the accepted answer. ;D (I actually think it is the most informative answer, but the comment is merely posted in a playful spirit given the context.) $\endgroup$ – quid Feb 7 '15 at 23:04

Hit the community dropdown and then the inline logout link.

Clicking Log Out will clear our cookies and log you out of Mathematics Stack Exchange on all devices.

If you're on a shared computer, remember to log out of your Open ID provider (Facebook, Google, Stack Exchange, etc.) as well.

Presumably if both of these steps happen, one can safely visit the site without logging in. (Maybe the last step is just commonsense advice about working on shared computers.)

  • 3
    $\begingroup$ The second step is important whether or not the computer is shared. If one remains logged in the OpenId provider, one will be automatically logged in to the site when visiting again; this is the point of the question. $\endgroup$ – user147263 Feb 6 '15 at 12:25
  • 1
    $\begingroup$ @FamousBlueRaincoat That's a little surprising login is so aggressive :) So if you ever are logged into your OpenID provider, and then accidentally visit the site again, you get sucked in? $\endgroup$ – rschwieb Feb 6 '15 at 13:13
  • 4
    $\begingroup$ Yes, this is how it works. Described here $\endgroup$ – user147263 Feb 6 '15 at 13:16
  • $\begingroup$ Logging out in this fashion (I don't think I ever noticed that link) seems to be doing what I want, so I'll mark this as accepted unless this fails me. $\endgroup$ – Hurkyl Feb 7 '15 at 22:56
  • $\begingroup$ ... actually, it seems to work most of the time, but every now and then fails. I haven't figured out why. :( $\endgroup$ – Hurkyl Mar 2 '15 at 22:33

This is an idea that might work [added: I decided to test it myself, it does not work]; I am not sure it actually works, yet you might want to try it.

  • Create, say, a new gmail address or other new OpenID (that you do not use for anything).

  • Add this as login option to your math.SE account.

  • Remove the old login, that you also use for other things, from the math.SE account.

  • Log out from math.SE and from the new OpenID.

I believe, but could be wrong, that the logins do not propagate over the accounts [added: actually they do.]

  • $\begingroup$ So, you got +2 for a suggestion that does not work. Why the upvotes, again? $\endgroup$ – user147263 Feb 7 '15 at 20:08
  • 1
    $\begingroup$ Surely you known, sometimes people publish papers showing that some approach on a problem does not work. Why the surprise here? :-) More seriously, I thought about deleting it once I checked it does not work. But then somebody else might have the same (not working) idea so I thought it might be of some limited use to have the information at hand, which brings us back to the start of my comment. Knowing that something does not work can also be useful. $\endgroup$ – quid Feb 7 '15 at 20:18

You must log in to answer this question.

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