21
$\begingroup$

Since September 11, the "cite" option for posts on main has been promoted from being inside "share" to directly below the post, cf. here.

Now, since the argument for requesting its promotion was that "MathOverflow is widely cited in the literature" and "the cite link is very important to [MathOverflow]", I figured it was reasonable to discuss the desirability of the current position of "cite" on Maths.SE.

For, if we find that hardly anybody ever uses it, clarity of the user interface may be preferred over the visibility of a rarely-used option.

So, does that "cite" link serve any purpose to you, or would you prefer to move it back to its original position?

$\endgroup$
35
  • 11
    $\begingroup$ I think even on MO it is fair to say "hardly anybody ever uses it." In my mind, its purpose is rather to promote citations (as opposed to actually helping create them). $\endgroup$
    – quid
    Commented Sep 21, 2015 at 16:15
  • 4
    $\begingroup$ While I can't quite speak for the author, arxiv.org/abs/1506.03533 seems like a place where the "cite" button could have been helpful. $\endgroup$
    – Asaf Karagila Mod
    Commented Sep 21, 2015 at 16:25
  • 14
    $\begingroup$ Does that extra link really bother people? I like having it available, but I can use it wherever it is. However, its greater visibility does make it more tempting to use for those who might not otherwise cite properly. $\endgroup$
    – robjohn Mod
    Commented Sep 21, 2015 at 19:27
  • 3
    $\begingroup$ @robjohn I think I'm a crotchety old person that can't get used to change :p More seriously I don't see myself using that feature as often as edit, close or flag, the button doesn't deserve being at that position. And the bibtex file produced is subpar anyway (try putting it through biblatex to see what I mean, it's even bugged if the title contains <or >), so I would probably just take the 20s needed to create the entry myself. I wouldn't care if the change hadn't shifted all the other buttons, which is annoying. $\endgroup$ Commented Sep 21, 2015 at 20:12
  • 3
    $\begingroup$ To give an example, the title Is a differentiable function $g(x)$ a strictly increasing function iff $g'(x)>0$? (a random question on the front page) became TITLE = {Is a differentiable function $g(x)$ a strictly increasing function iff $g&#39;(x)&gt;0$?},... Which is obviously not valid... Here is the result when that issue is corrected -- pretty isn't it? It also does nothing for capital letters, and doesn't enclose math markup in braces. I don't want to sound mean, but they didn't think this through at all and just slapped a half-baked feature on. $\endgroup$ Commented Sep 21, 2015 at 20:17
  • 8
    $\begingroup$ @NajibIdrissi: that looks like something to put in a bug report, not to tuck back into the "share" link to insure it doesn't get fixed. It could be that the "cite" code was written before MathJax was even around, but certainly it is now an issue on sites that use MathJax. $\endgroup$
    – robjohn Mod
    Commented Sep 21, 2015 at 20:30
  • 9
    $\begingroup$ The extra link does bother me -- there's no end to the times the last week I've pressed "cite" instead of "edit" due to muscle memory (i.e. second link from the left to edit a post). $\endgroup$ Commented Sep 22, 2015 at 16:21
  • 3
    $\begingroup$ @robjohn The place where the cite button is at belongs to edit. It's a disgusting attempt at forcing people to use the feature. As bad as Skype's "Search with bing" replacement of "copy". $\endgroup$
    – Git Gud
    Commented Sep 25, 2015 at 9:20
  • 2
    $\begingroup$ Yes, I've clicked "cite" so many times since it moved. If it was all the way to the right, I wouldn't mind. Don't put something that is rarely used at the front of your interface - even if people used cite were used as often as feasibly possible it would still be rare. @robjohn $\endgroup$ Commented Sep 25, 2015 at 23:27
  • 1
    $\begingroup$ @ThomasAndrews: indeed, as I replied to Git Gud, I would not mind it being moved to the end of the list of links. Since this is not a concern mentioned in the question, it is not something that my previous answer and comments dealt with. $\endgroup$
    – robjohn Mod
    Commented Sep 26, 2015 at 4:07
  • 2
    $\begingroup$ @GitGud just to be clear, it was asked for on MO and SE did it with some hesitation, and in doing so also distributed over a few other sites. If it was an attempt, then at least it did not come from SE. $\endgroup$
    – quid
    Commented Sep 27, 2015 at 12:22
  • 1
    $\begingroup$ @ThomasAndrews Who gives Stack Exchange (now Stack Overflow, Inc.) money every time somebody visits this site? And what for? $\endgroup$
    – user147263
    Commented Sep 27, 2015 at 14:58
  • 1
    $\begingroup$ @ThomasAndrews Can you show me an example of a commercial ad on this site? The only ones I see are those the community itself proposed here on Meta, without anyone paying anything. $\endgroup$
    – user147263
    Commented Sep 27, 2015 at 15:08
  • 2
    $\begingroup$ @ThomasAndrews Okay, then for the reference: Math.SE has never carried commercial ads, and there are no plans to add them in foreseeable future. $\endgroup$
    – user147263
    Commented Sep 27, 2015 at 15:14
  • 1
    $\begingroup$ I suppose we can all agree that it is in the best interest of the site to have traffic. However, as the situation is not as simple as @ThomasAndrews proposed it to be, first not all traffic is the same, and second traffic it not everything, as one of the site's (and sites' in the network more generally) assets is the idea that the content is in some sense of high quality. Thus, one citation in a scholarly journal that might attract the attention of a few additional experts and can also be used to underline the crediblity of the site might be worth more than 10k views via a popular share. $\endgroup$
    – quid
    Commented Sep 27, 2015 at 18:15

2 Answers 2

20
$\begingroup$

I am a big fan of the cite button, because it reminds people that they are supposed to cite things they learned here on math.SE. Citations are the currency of academia -- if I write a good answer here, I want to be able to say five years later that this answer was cited so many times in so many papers. Indeed, I get a little annoyed when I see papers that use ideas they learned on MO or math.SE and don't cite them. I know that math.SE is more oriented towards students than research, but it is also important to teach students that they need to cite where they learned things! If you are allowing your students to use math.SE as a reference, then you should be requiring that they cite it just like any other source they use, and the cite button reminds them of this.

It seems that people are complaining because they have trained their muscle memory on other SE sites, or on this site before the button was there. The easy solution to this is to move the cite button to the far right.

I support moving the cite button to the far right, but keeping it visible on the main post.

$\endgroup$
2
  • $\begingroup$ "The easy solution to this is to move the cite button to the far right. " The easier solution seems to be for those few that really care to just do it on their end. (This aside, good post.) $\endgroup$
    – quid
    Commented Sep 30, 2015 at 18:15
  • $\begingroup$ I understand that far right means near the poster's name, not anywhere close to flag, right? $\endgroup$
    – ccorn
    Commented Oct 5, 2015 at 0:46
0
$\begingroup$

There has been some discussion in the comments, and the most popular views are:

  1. Move "cite" back into "share" or elsewhere out of view.
  2. Move "cite" to behind "flag".
  3. Keep "cite" where it is now.

I think 2. is closest to a compromise so I figure we can test that. Hopefully this is technically feasible.

If it doesn't work, 1. can be considered.

$\endgroup$
7
  • 1
    $\begingroup$ I think 2 is a good option. $\endgroup$
    – Asaf Karagila Mod
    Commented Sep 27, 2015 at 11:35
  • 3
    $\begingroup$ If we wait for a while and then do 2. it will be hilarious because then some can again complain they click "flag" and "close" instead of "edit." The point of that comment is that pretty soon most will have adapted and that's that. Moving things around will just annoy more people. Also I am not really sure the "left to right" is the most apt criterion to decide where the more or less frequently used buttons should be. Especially, the last spot of a the list seems also a privileged position; mostly i for one have the cursor rather in the center or right, so it is closest and easy to aim at. $\endgroup$
    – quid
    Commented Sep 27, 2015 at 12:30
  • 1
    $\begingroup$ @quid One difference is that the cite button is not present on meta, or other SE sites (except MO). I agree though that left-to-right isn't really the order of important-to-less-important (I use edit, flag and close at least an order of magnitude more than share, personally). $\endgroup$ Commented Sep 27, 2015 at 12:46
  • $\begingroup$ @NajibIdrissi that's a valid point. But then we have "share|edit|close|flag" on questions, and "share|edit|flag" on answers (or if one does not have close-privilege) while based on this reasoning of uniformity it should better be "share|edit|flag|close." Maybe it should be like this. But it is not, and no-one seems to have problems with this. Don't get me wrong I am not a huge fan of that cite button either, and maybe it should be placed elsewhere. But I still think moving it around now is not useful and those that really want should do it on their end (as you did, IIRC). $\endgroup$
    – quid
    Commented Sep 27, 2015 at 15:13
  • $\begingroup$ @quid Well, even if nothing changes now, I think it's important that we as a community speak up about not liking unannounced alterations of UX. Tim Post said this as a reply to Asaf's comment but apparently the SE team still does as pleases them without informing us. $\endgroup$
    – Lord_Farin
    Commented Sep 27, 2015 at 15:38
  • 1
    $\begingroup$ @Lord_Farin not quite, on MO the general site update was announced by SE. I do not think it is reasonable to have each and every small change announced and discussed at length. Right, they could have made a post what this is about, but I do not consider it as that relevant either. $\endgroup$
    – quid
    Commented Sep 27, 2015 at 15:46
  • 1
    $\begingroup$ My preference is for (1), but (2) would be a significant improvement over the present setup. $\endgroup$ Commented Sep 27, 2015 at 17:28

You must log in to answer this question.

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