On the ChemSE site (Is it better to use a smaller, more accurate measuring cylinder several times or a larger, less accurate one for the same volume?), the following MathJax code renders my square root without an overbar:

$$\text{uncertainty for seven combined measurements} = \sqrt{7 \times (0.1 \text{ mL})^2} = 0.026 \text{ mL}$$

Here's a screenshot:

enter image description here

However, here on ChemSE Meta, it renders fine:

$$\text{uncertainty for seven combined measurements} = \sqrt{7 \times (0.1 \text{ mL})^2} = 0.026 \text{ mL}$$

Edit: Based on orthocresol's comment, it looks like it's a rendering issue. I'm using the HTML-CSS on both sites, yet it renders fine here but not there. If I switch to Preview-HTML, it renders fine on both sites.

For other typesetting reasons, Common-HTML looks fine here but not there, while SVG is fine there but not here.

  • 1
    $\begingroup$ I don't really know why there should be a difference. It renders just fine for me. At risk of sounding terribly patronising, have you refreshed the page, and it still does that? $\endgroup$ – orthocresol Jan 5 at 2:37
  • 1
    $\begingroup$ You may also want to try changing the MathJax renderer as described at chemistry.meta.stackexchange.com/a/2854/16683 -- I personally use Common HTML although with MathJax many things are browser-specific. i.stack.imgur.com/H7Z7p.png If you've previously changed this it may explain why one site is OK and the other isn't. $\endgroup$ – orthocresol Jan 5 at 2:40
  • 1
    $\begingroup$ @orthocresol I had previously tried refresh, and it didn't correct the problem. But, upon looking more closely, I noticed the overbar appears for just the blink of an eye right after I hit refresh, and then vanishes. When you look at that page (which I linked a minute after my initial post, so you may not have seen that link when you answered), does it render properly for you? $\endgroup$ – theorist Jan 5 at 2:41
  • $\begingroup$ @orthocresol Please see the edit I added to my post. $\endgroup$ – theorist Jan 5 at 3:02
  • $\begingroup$ @theorist All renderers display the square root correctly for me (screenshot of Chromium 87 on Windows 10). Can it be a browser/system-related issue? Which ones are you using, and do you have scaling (either systemwide or in-browser) turned on? Also, does the issue persist for a MWE, say, \sqrt x? As a side note, IMO replacing "uncertainty for seven combined measurements" with something like "$u_\mathrm{c}(7)$" would make the equation look better; "text formulas" are bad, m'kay:) $\endgroup$ – andselisk Jan 5 at 4:03
  • 1
    $\begingroup$ @andselisk Could be that I'm on a Mac, and SE is optimized for Windows. I'm using MacOS 10.13.6 with Safari 13.1.2. I'm running four monitors, with four different resolutions, all of which are set to "default for display" rathern than "scaled". And I'm seeing the same rendering on all four. Regarding the formatting: I agree that your suggesiton would look better. But the I would have had to define the term before using it, while for someone who was clearly a beginning student, I thought it would be simpler to just put the text into the formula. More broadly, you and I go back and ... $\endgroup$ – theorist Jan 5 at 4:14
  • 2
    $\begingroup$ forth on this :), but I feel strongly that (perhaps thanks to work by you and others like you), the formatting standards on ChemSE are actually pretty good, and what's really needed now to improve the site is to redirect focus from form to content, since the latter is currently the site's true weakness. I cringe at the horribly incorrect answers on this site that, because they have been heavily upvoted and accepted, misinform visitors into thinking they are correct. That's where the real clean-up is needed -- to remove or edit bad answers, not to prettify good ones. $\endgroup$ – theorist Jan 5 at 4:22
  • 1
    $\begingroup$ macOS 10.15.7 here -- it renders fine on both Safari 14.0.2, Chrome 87.0.4280.88, and Firefox 84.0.1 for me. I suspect there might well be a bug here, but very hard for me or andselisk to reproduce. What we can do is to see if the SE or MathJax developers respond -- usually tagging the question with bug will (eventually) get a response. $\endgroup$ – orthocresol Jan 5 at 14:12
  • 2
    $\begingroup$ It renders fine in both locations for me in MacOS 10.14.5 with Safari 12.1.1. The HTML-CSS is the oldest and most fragile of the renderers. It does use clipping regions to trim the output to the expected size, and it may be that a round-off error is making that clip region slightly too small and the line is falling outside the clip rectangle. That would be my guess. You could try using the browser DOM inspector to turn off the clip region for that equation and see if the line shows up. $\endgroup$ – Davide Cervone Jan 5 at 20:45
  • $\begingroup$ MathJax is weird. The rendering is done on your computer and depends (most of the times) also on your specific set-up, installed fonts, etc., … . @orthocresol Wasn't there some new-ish tag method to escalate such a question? I guess we could use this here. $\endgroup$ – Martin - マーチン Jan 6 at 12:29
  • $\begingroup$ @Martin-マーチン I'm not sure if anybody has anything more insightful to say than Davide. :-) $\endgroup$ – orthocresol Jan 6 at 13:14
  • $\begingroup$ @DavideCervone Unfortunately, while I can activate the Web Inspector in the Developer tab, I don't know how to do what you suggest. $\endgroup$ – theorist Jan 7 at 4:15
  • $\begingroup$ @theorist, do you have any ideas on what to do with the upvoted and incorrect answers? I'm thinking about making a meta post on it some time. Next week, maybe. And I'm glad that there are more people who think that these are a problem (there are more beyond you and I, of course). However, we have never really done anything about it... (so far). $\endgroup$ – orthocresol Jan 13 at 12:35
  • $\begingroup$ @orthocresol I'm not sure. As you know, for both questions and answers, SE's primary evaluative mechanism is crowd-sourcing (uprooting/downvoting). But while SE has an additional mechanism for questions that don't meet site quality* standards (closing them), nothing equivalent exists for answers (*of course, offensive answers can be removed, but that's different). [I was thinking, humorously, of Twitter's 'This claim is disputed!' warning.] My approach is usually to leave a comment saying the answer is incorrect, with a brief explanation of why.... $\endgroup$ – theorist Jan 14 at 5:42
  • $\begingroup$ ....To go beyond that, I'd have to either (a) edit the answer (something I'm not comfortable doing); (b) add my own answer (which I have done); or (c) add my own answer, including in it an explicit statement noting that the other answer is incorrect (I may have done that, can't recall) $\endgroup$ – theorist Jan 14 at 5:42

You must log in to answer this question.

Browse other questions tagged .