Seems to me there's a flaw in the system, not that I see what could be done about it.

Someone recently posted a question that was a duplicate of If $z_n \to z$ then $(1+z_n/n)^n \to e^z$. I saw a very nice answer, if I do say so myself, so I added a new answer to that old question.

The problem is that the new answer seems to be invisible. There was already a very nice answer with a +9 score. The duplicate question led to people looking at the old question, and in a few hours the +9 went to +11. Fine. But my new answer is buried down there in the answers with a score of 0.

Not to be whatever, but if you look at it I think you'll understand why I conjecture nobody's seen it - it's a very simple answer, with the added virtue of being correct. If people were reading it people would like it. Seems to me that nobody is ever going to bother reading it, when they see the score on the accepted answer - why read the 0's?

I swear the problem is not the rep points - I've got enough of those. I don't think anyone who reads my answer will think I'm being overly immmodest when I suggest that someone looking at that question would be better off reading my answer than not reading it.

Some sort of "don't be fooled by the 0, this answer wasn't there back when the voting was going on" doesn't seem practical. But it does seem like a flaw in the system.

  • 7
    $\begingroup$ In what way is the new answer hidden? What's the mechanism that prevents people from reading it and voting on it? $\endgroup$ Dec 9 '15 at 17:39
  • $\begingroup$ I retagged this as discussion. It is not requesting a specific feature, rather it is a call for discussion how to resolve a specific (perceived) problem. $\endgroup$
    – quid Mod
    Dec 9 '15 at 18:13
  • 2
    $\begingroup$ @NajibIdrissi No physical or software mechanism, but I don't find it surprising that a lot of people would not read all the other answers when the top answer has a score of 11, and the others 0. That puts later answers at a real disadvantage. $\endgroup$ Dec 9 '15 at 18:23
  • $\begingroup$ @quid Thanks - if I'd been aware of three of the four tags you used I would have used them instead. Probably just me being dumb again, didn't see how to find a comprehensive list. $\endgroup$ Dec 9 '15 at 18:25
  • $\begingroup$ @DanielFischer In fact I not all that humbly suggest that the fact that it jumped from +9 to +11 today is evidence - I don't think I'm just being vain in thinking that my answer is at least as "useful" as the accepted one... $\endgroup$ Dec 9 '15 at 18:27
  • 2
    $\begingroup$ In about 5 months on the site you have 434 posts and 210 votes on main. I'm kind of curious how many of those votes have gone to late answers to old (already answered) questions. $\endgroup$
    – user642796
    Dec 9 '15 at 19:03
  • $\begingroup$ @ArthurFischer This is the first time I recall doing the Right Thing by adding an answer to an old question instead of answering the duplicate. Why? $\endgroup$ Dec 9 '15 at 19:06
  • $\begingroup$ @ArthurFischer Oh. I think I misread the question. You were asking about votes I've cast? Again, I suspect none or more or less none. Why? $\endgroup$ Dec 9 '15 at 19:24
  • 2
    $\begingroup$ The point could be that if you do not vote on answers to old questions, why do you think somebody else should? For example, the two old answers explaining the the argument with the logarithm in fact can be generalized do seem to have some merit, yet somehow nobody had voted on them until now. $\endgroup$
    – quid Mod
    Dec 9 '15 at 19:51
  • $\begingroup$ @quid Thanks again. I can imagine various things the point could be. What you say is the first thing that sprang to mind. I rejected that possibility, because I didn't say people should vote on old posts! Really, look again - nowhere did I say people should vote on old posts. I pointed out what seems to me to be a flaw with the system, not a problem with people being Bad. $\endgroup$ Dec 9 '15 at 20:24
  • 4
    $\begingroup$ It's hard to imagine this post is something other than a complaint that your answer hasn't received the admiration you feel it is due. True, your immediate complaint is that your answer appears to have been ignored (it's "invisible"). But the evidence of this is limited to the then lack of upvotes generated compared to the older accepted answer in the same period. But the "meta-effect" has ensured that your answer received an upvote. So, Mission Accomplished, I guess. $\endgroup$
    – user642796
    Dec 9 '15 at 21:35
  • 1
    $\begingroup$ @ArthurFischer In other words, you find it hard to imagine that I'm not lying. Have a nice day... $\endgroup$ Dec 9 '15 at 23:15
  • 2
    $\begingroup$ "But my new answer is buried down there in the answers with a score of 0." This is not necessarily true. Maybe some users view answers sorted by recent activity and not sorted by votes. (But I guess most users use votes to sort the answers.) $\endgroup$ Dec 10 '15 at 5:54
  • 3
    $\begingroup$ And another way how people may notice a late answer is that they see question among recently active questions and they click on the timestamp. This takes them to the part of the post which was most recently changed (posted or edited). My guess is that many users do this if they see that a question was bumped. $\endgroup$ Dec 10 '15 at 5:59
  • $\begingroup$ And of course if you (or anybody else( can improve answer in some way, any edit will bump the question and improve chances for upvote. (In this specific case, you wrote "someone's theorem; Vitali?". If you can find the name of the theorem or link to some resource with that theorem, you can add it to the post. Maybe this theorem can already be found on this site. If not you could post a question asking for the proof (and the name) of this result. (And link the question to your answer. This will explain why you are asking. And - as a side effect - add more exposure to your post.) $\endgroup$ Dec 10 '15 at 6:17

There are two things that are already done along the lines you say:

  • "Late" answers by "new" users have an extra queue to give them some added visibility.

  • One can sort answers by last activity, making new ones be displayed first. (Except accepted answers are treated separately and come always on top, except if it was a self-answers, except something I might have forgotten.)

Granted this is not much, but at least the fact that it is transparent when answers were given makes it so that those in the know can note what you want them to note. To make this more explicit does not seem like that good idea to me.

If you are really and truly set on the idea to give some answers more visibility, then place a bounty on the question to that end. This may not lead to universal praise and approval, but it will give the answers added visibility.

Finally, voting on the site is also governed by many factors and sometimes perhaps not quite fair. For a reasonable point: not few users likely hesitate to upvote content they do not understand, especially if it is not yet upvoted (which is a good thing in a way) but has negative effects for more conceptual arguments using more tools.

  • $\begingroup$ Reasonable points, the last of which had not occurred to me. I'm not complaining, because you didn't actually assert that I'm a rep whore desperate enough for a few clicks to start all this for that reason, but: Once again, it's not the votes I'm complaining about, it's the related fact that the way things are set up visibility depends crucially on timing and not just mathematical worth. You say this or that doesn't seem like a good idea to you - me neither, I wasn't suggesting any particular change, in fact I stated explicitly, twice, that I had no fix to suggest. $\endgroup$ Dec 10 '15 at 0:52
  • $\begingroup$ @DavidC.Ullrich: I agree that it is a bad thing that visibility depends on the timing. It was less so when (circa 4 years ago) questions from yesterday where still on page 2 (or even on the frontpage) - those were good days. But soon thereafter your answer only needed to be two hours late for it to be "lost". Irrespective of its merits. Imagine what this means to a user who is 4+ hours away from the time zones of 90 per cent of the users... It is just unreasonable to expect this kind of fairness. Users no longer have the time to revisit old threads, and it shows in the voting. $\endgroup$ Dec 10 '15 at 6:54
  • $\begingroup$ Damn. I thought this particular wound had healed. Going away from this thread. $\endgroup$ Dec 10 '15 at 6:58
  • $\begingroup$ Regarding the last paragraph: it also happens that users upvote content they don't fully understand if it's already been upvoted and/or it's been posted by a high-rep user. It sometimes leads to completely wrong answers getting nontrivial scores. $\endgroup$ Dec 10 '15 at 13:11

You must log in to answer this question.

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