7

I hope that asking a question just to satisfy my curiosity isn't too self-indulgent, but does anyone know what it is about this question that stands out from the rest?

It seems to have attracted a record number of views, and votes both on the question and the answers, just since yesterday. More people have voted on this one question (seemingly just asking for the name of a fallacy), than have voted on all of the next three pages of questions put together!

I can't see anything about the question having been migrated from another site (one with more traffic perhaps), the question, and all the answers seem to have come from only yesterday, so it's not an old question bumped to the homepage. I can't think of any other explanation. I'm quite surprised to learn that even 5,000 people visited Philosophy.SE since yesterday let alone just to read this one question.

Can someone enlighten me as to what has caused this anomaly?

Update

I don't know if anyone else is actually interested in these things, but there would appear to be a new example of the same thing which refutes all the theories put forward below except @Keelan's (that it simply appeared in Hot Network Questions). This question seems to have accumulated an astonishing 11,000 views in just two days, even the second highest voted answer has more votes than the best answers to the rest of the home-page's questions. The comments seem to strongly suggest that the question itself might not even be a good fit for Philosophy.SE and the question actually has two close votes. Am I the only one who thinks that what appears to be a temporary 'hijacking' of the site might be a problem, say when looking at highly voted/viewed questions/answers as a guide to the site's scope and preferences, for example?

18
  • 1
    Probably it appeared in the top network posts lists that are shown on the right of all homepages, after getting some traffic here, which strengthens the effect. It is also very accessible, because fallacies and god & devil are things many people (think they) are knowledgeable about.
    – user2953
    Jul 20, 2017 at 7:24
  • 1
    @Keelan That's really quite a worrying trend (at least to me). I was expecting some technical reason why the question had been around for a long time but didn't appears so or something similar. If there really are thousands of people ready to swoop on this site and totally sway the apparent tone (as measured by highest voted questions), and community (as measured by reputation), just because someone mentions the Devil, I think that's something we should be concerned about.
    – user22791
    Jul 20, 2017 at 7:31
  • I hadn't looked at it that way before. Just to be clear: before a question ends up in these lists, it has to gain some traffic on this site as well. So, this phenomenon only reinforces an already existing trend. And I have not checked (would not know how, if possible) if the question indeed appeared in those lists.
    – user2953
    Jul 20, 2017 at 7:35
  • 1
    @Keelan Surely other questions recently must have ended up on top network posts? It would be really good to know whether the question generated most of it's votes before, or after then. Also linked, but only just, is a trend towards seeing the homepage quite dominated by "what is the name of this fallacy" type questions, do we really want to become mostly a vehicle for helping people win petty arguments by coming up with the Latin for their opponent's fallacy (argumentum ad verecundiam fallacy, before that becomes the next top voted question!).
    – user22791
    Jul 20, 2017 at 7:40
  • 1
    I could not find such a trend (of fallacy questions becoming more popular). This graph plots the number of questions related to fallacies (as a tag or in the title) against the total number of question. It seems that for at least 2,5 fallacy questions have made up about 5% of the total. (This does not account for the activity on those questions, but I think it is reasonable to assume that more activity would lead to more questions and this does not seem to happen.)
    – user2953
    Jul 20, 2017 at 8:13
  • Unfortunately, I don't think records are kept of the hot network posts. However, it was correct that this particular question appeared there (you can check, now, on stackexchange.com/questions?pagesize=50). The highest scoring answer was also tweeted, but I have no way to check what its score was when it was tweeted. Possibly, the hot network questions list and the twitter account share a data source, so that may indicate when the question first appeared in the list - but that is conjecture.
    – user2953
    Jul 20, 2017 at 8:14
  • @Keelan Thanks for doing the graph (I love data), actually I think it does show a trend (doing a rough regression analysis), note how the 'fallacies*20' line rises to meet and finally overtop the 'all questions' line. Not that there's anything we can do about it, I suppose, I just think it trivialises the site a bit.
    – user22791
    Jul 20, 2017 at 9:25
  • I don't do twitter, does this happen often, or is it perhaps as good a reason as any why this question seemed so suddenly popular?
    – user22791
    Jul 20, 2017 at 9:26
  • I don't see what you mean. Since the second half of 2014, the blue line is around the yellow line. It overtops it a couple of times but gets below again as well. There seems to be a biannual repetition in that, so it may be related to semesters, but it's not very strong. Currently, like in the same period in 2016 and 2015, we are in a fallacy maximum, but looking at the graph I don't see a reason why the blue line would not get below again. About twitter, I wouldn't know. This account tweets several posts on a day so it is unlikely to be the sole cause.
    – user2953
    Jul 20, 2017 at 11:46
  • @Keelan, I just did a least squares regression to a straight line (the least subtle, but simplest linear trend test) to find that there was indeed a weak linear trend of increasing ratio over time if you eliminate all the noise (of which, as you say, there is a lot). Hopefully it's just a random trend and not an indicator of anything. Unless maybe people like the name this fallacy game, in which case vive la mode!.
    – user22791
    Jul 20, 2017 at 12:09
  • This trend exists over all data, but does it also over the last 2-3 years?
    – user2953
    Jul 20, 2017 at 18:22
  • Here is another "viral" question from recent memory, What is the difference (if any) between “not true” and “false”? Fortunately, this happens very rarely. See discussions on Math Meta (their hot vote totals dwarf ours) and Meta Meta, they describe how it usually happens.
    – Conifold
    Jul 21, 2017 at 3:38
  • 1
    @Conifold Thanks for the links, they do seem to confirm that Twitter and HNNQ are probably involved, but I think everyone's missing the point of what I'm asking, it's not that this question "went viral" in the same way others have done. It's view count is double the previous highest (which was only a few days before) and nearly triple the one before that. The question and all three top answers are the highest ever scoring per day by a huge margin. It's the scale of it I don't get.
    – user22791
    Jul 21, 2017 at 6:54
  • @Keelan I see what you mean. The trend is there for the last 2 yrs but very much weaker, not statistically significant, it's levelling off but at a much higher ratio (about double) than it was before the sudden spike.
    – user22791
    Jul 21, 2017 at 6:55
  • 1
    @Not_Here Yes, that's what concerned me about this example (which maybe I should have posted as a separate Meta question rather than an edit to this one), not only the close votes, but the top voted comments outlining exactly how it is both a duplicate and not really a philosophy question in any case. With regards to the traffic, a quick glance at the homepage shows that your reservations are justified, the sub 100 view tallies common before the viral question continue unaltered after it. The 11,000 viewers simply came, looked at (and voted on) that one question and then left.
    – user22791
    Sep 6, 2017 at 11:55

2 Answers 2

Reset to default
2

Pascal's Wager is well-known among the general population (at least as well as such things are likely to be known); and this is a particularly interesting variation on the problem. It got lots of upvotes locally, indicating its quality... And went viral through the hot questions sideboard. I'm not sure what else exactly you'd like to unpack here but those are some general considerations if you're trying to think about what makes a question not only great for this site, but also of broad interest (and so likely to drive more traffic in general.)

2
  • 1
    Maybe I'm just reading this wrong, but according to this data search the only other massively up-voted question this month had 2836 views in 5 days. Higher than all the next four put together. The question I've asked about here hasn't appeared on the search, but had 5000 views in just one day, which, ignoring the other recent anomaly would be ten times higher than the previous highest ever recorded. Is that not really odd, am I missing something obvious?
    – user22791
    Jul 20, 2017 at 15:32
  • 1
    Furthermore, according to this query, this post would be the highest ever scoring for less than one day, as would it's top answer, and the next three top answers would all be in the top ten highest ever scoring posts for less than one day.
    – user22791
    Jul 20, 2017 at 15:56
2

I think there's two things which Keelan identifies well in the comments.

First, fallacies questions are popular especially with people coming from a lay conception of philosophy. There's several groups that emphasize them in purporting to have the one-true-philosophy including Christian apologists, atheist apologists, and critical thinking gurus.

Second, God vs. atheism questions also tend to attract a lot of strong opinions.

This question hits both at once and Pascal's wager itself does fall under the list of things traditionally considered in Philosophy of Religion classes, so there's three reasons to boot.

Now, my sense (perhaps near yours as to why I'm not super happy that these are our most popular questions) is that

  1. the lay version and the Pascal version of Pascal's Wager have some distance between them

  2. fallacy questions quite often suffer from a confusion about the point/value of identifying "fallacies" (especially of the informal kind), and

  3. when you make it a question that riles up convinced atheists and convinced religious apologists, you often get answers that are tangential to the philosophical concerns

Compare this with say asking a technical question about Bertrand Russell's views on sets or Aquinas's logic, and it's obvious which one is going to get more views and attention from people who aren't taking a philosophy class or doing academic philosophy.

4
  • 1
    My biggest concern is actually what this says about the site's community and I'm thinking now perhaps I should have phrased the question that way. One has to have a reputation of 100 to vote so all voters must be members of this community, more people voted on this question (and the other weird anomaly, the Socrates question) in the first few hours than voted on all of the other question that week put together. Do we really have a faction of members just waiting for questions on religious fallacies to turn up? What are they doing the rest of the time?
    – user22791
    Jul 21, 2017 at 11:49
  • Um, if you have rep on a different SE, then you get 101 when you come here, so I think that means random people can upvote.
    – virmaior
    Jul 21, 2017 at 13:05
  • 1
    I see, not ideal either way. Maybe I'm just being cantankerous, but the idea that what will become one of the site's exemplar questions was made that way by a horde of people who just came here to vote on that one question and then took no further part in the site just doesn't sit right somehow.
    – user22791
    Jul 21, 2017 at 13:12
  • 2
    @Isaacson: I can wholeheartely subscribe the points made by virmaior and I do understand your concerns. It is a matter of fact that there are no more than 9-10 active members who are truly interested in philosophy proper, as you can derive from the votes given in more technical or philosophically interesting questions. Most people see philosophy as worldview, and this is a self-enhancing fact, as this is what other people get to see per the "hot" bar. The lack of more seriously interested and scholarly members had been discussed several times, this is just a symptom.
    – Philip Klöcking Mod
    Jul 21, 2017 at 14:03

You must log in to answer this question.