• ## Discussion Feed

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.

1.

I don't normally use IE 8! I just thought it might be something worth fixing.

2.

Oh, absolutely. There is, of course, a limit to the amount of pain and suffering fixing the problem is worth. I have no idea how hard it would be in this case. But if nothing else, it is good to be aware that these problems exist.

3.
I want to confirm David Dynerman's bug mentioned above in this this thread: the recent update (7.0.517.41) to the (stable) Chrome package for Ubuntu 10.04 has broken MathJax.

Does anyone know a fix?
• CommentAuthorrmfedorov
• CommentTimeNov 9th 2010

Since recently, I cannot read mathematical formulas on mathoverflow. The symbols overlap with each other in horizontal direction. Any advice?
4.

What browser do you use? You may need to switch, at least for MO. Firefox seems to work well.

5.
Double post.
6.
The post http://mathoverflow.net/questions/45730/transitive-semigroups-of-2-times-2-matrices translates in Google Chrome 7.0 to

Notice how $v$ jumped from the end of the sentence to the top of the paragraph.
7.

At the risk of sounding like a broken record: What browser, what OS?

8.
@Harald Hanche-Olsen: Sorry, the bug persists under Google Chrome 7.0 for Windows.
Seems to be fine under Mozilla/5.0.
9.

I don't see the problem at that page, but running Google Chrome 7.0.517.44 for OS X I have a similar problem at Can all n-manifolds be obtained by gluing finitely many blocks?

10.

I've opened a bug report over at the MathJax bug tracker for the issue that Andrey and I are seeing.

11.

Is there some reasonable way to make it easy to test whether a problem is showing up specifically on MO or is a general MathJax bug? It would be nice to repost problems we're having to the MathJax user forum or issue tracker, but I'd like to make sure they're honest MathJax bugs first.

12.
@Scott Morrison: Many thanks for your efforts!
13.

I sporadically get a similar problem here and on math.SE, running Chrome 7.0.544.0 on Windows 7, so I think it's a MathJax issue.

14.

It may be specific to the MathJax interaction with Chrome (on multiple platforms).

• CommentAuthorJDH
• CommentTimeNov 14th 2010

Using Windows Internet Explorer 8 under Windows Vista operating system, I get a notice of the following error on all question pages except those that have been closed. On all the same pages where I get this error, the "cite" button does not appear. It used to appear fine on all pages, but recently (within the past month or so) became broken on the non-closed pages.

Webpage error details

User Agent: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0; Trident/4.0; SLCC1; .NET CLR 2.0.50727; Media Center PC 5.0; .NET CLR 3.5.30729; .NET CLR 3.0.30729; .NET4.0C; BRI/1) Timestamp: Sun, 14 Nov 2010 18:05:34 UTC

Message: Object doesn't support this property or method Line: 41 Char: 3 Code: 0 URI: http://mathoverflow.net/custom/footer.js

Message: 'parentNode' is null or not an object Line: 35 Char: 3 Code: 0 URI: http://latex.mathoverflow.net/mathjax/trunk/mathjax/MathJax.js

Message: Object expected Line: 1 Char: 166 Code: 0 URI: http://mathoverflow.net/questions/45774/is-subamenable-the-same-as-amenable

15.
I left a comment at http://mathoverflow.net/questions/46368/the-property-of-kendall-mann-numbers and OP left a comment in reply and both of them run off infinitely to the right (at least on the software I'm using). What's happening there?
16.

Mikhail accidentally typed a very long formula; the comments are being typeset in a column wide enough to fit this formula.

17.
@Scott, thanks - can this be fixed?
18.

Not without someone deleting an reposting the comment, unfortunately; ideally Mikhail.

19.
Is it considered normal to see this overlap? The line appears in an answer to this MO question http://mathoverflow.net/questions/47611/exact-formulas-for-the-partition-function

I get this problem running Google Chrome 7.0.517.44 on Windows 7.
20.

It's not browser specific. What was needed was backquotes around the formula so the double backslash (the formula is a twoline multline environment) could do its work. This is a fairly well known interaction between the various syntaxes in use.

I added the necessary backquotes, so it looks okay now.

21.
@Harald Hanche-Olsen: thank you for fixing this.
22.

Regarding the problem discussed above (equations jumping to the top of a paragraph). The MathJax people have identified that this is a bug in WebKit (the underlying engine behind many modern browsers, including Chrome and Safari), and have reported a bug there.

• CommentAuthoralgori
• CommentTimeDec 18th 2010 edited

Since a day or two my browser does not show math formulae correctly when I open a post: basically, the formulae get squeezed horizontally so that all characters overlap. A couple of days ago everything was fine and I was wondering if there is a way to bring it back to normal. It seems that this problem hasn't been mentioned above in the previous comments.

OS Windows 7, browser Google Chrome 8.

upd: after having posted this I've reloaded a page and everything seems to be working again.

upd2: woops. After reloading again the problem reoccurred: all characters are getting squeezed together again, and reloading again does not help.

upd3: problem solved.
23.

I've got troubles with visualizing ${}pF{p-1}$ when answering http://mathoverflow.net/questions/51123/ and, obviously, the poster had got them as well (see the end of his question). The closing buck mysteriously disappears... Windows XP, Firefox. The way I've succeeded was typing ${}_p F _{\substack{p-1}}$.

24.

The usual fix of adding backticks seems to have worked ... I've changed it to $_pF_{p-1}(1)$.

25.

Thank you, Anton.

26.
This morning I am experiencing problems with MathJax scripts. This is using Firefox under MacOS, on two different machines. I get the error,
"Warning: Unresponsive script.
A script on this page may be busy, or it may have stopped responding. You can stop the script now, or you can continue to see if the script will complete.
Script: http://latex.mathoverflow.net/mathjax/MathJax.js:29"
Then I have to stop the script to proceed.
• CommentAuthoralgori
• CommentTimeJan 7th 2011

Since this morning I've been having trouble viewing MO pages with Google Chrome (OS Windows XP Home SP3). Typically a page would open but none of the LaTeX formulae is rendered. Then after a minute or two scrolling becomes impossible and reloading results in a blank page. This happens with Google Chrome only, Mozilla works just fine.
27.

I'm having the same problem as algori in Chrome (8.0.552.224) running on Windows 7. LaTeX is rendering but the pages become unresponsive after a minute or so.

28.

I am getting the same problem as Joseph O'Rourke (Firefox on OSX 10.4), but I get the "unresponsive script" error for both MathJax.js and jax.js. Even if I let the scripts continue for a long time, they don't bother to render the math.

29.

I did update MathJax last night.

If you open Chrome's javascript console (Ctrl+Shift+J I think), do you see any errors? If yes, please copy them here. Could it be that some bits of MathJax are cached longer than others? If you clear your cache, does the problem persist?

30.

Thanks, Anton. Clearing the cache seems to have done the trick.

• CommentAuthoralgori
• CommentTimeJan 7th 2011 edited

Anton -- thanks. I've cleaned the cache; the problem of unresponsive pages seems to have gone, but the latex formulae at the bottom of a page aren't rendered.

upd: everything seems to be working now.
31.

I've been getting some bunched-up LaTeX intermittently, e.g.

In the time it's taken to write this, the problem's gone away again, at least for the time being. I'm using Windows 7 64-bit, Chrome 8.0.552.237.

32.

In this question it seems that \mathfrak is not rendering Fraktur fonts. This is on Firefox 4.0b11 on Mac OS X 10.6.6.

• CommentAuthorWillieWong
• CommentTimeFeb 16th 2011

@Jose: it looks alright to me. Firefox 3.6.13 on linux. Either a regression in firefox or a settings issue, I would bet.

33.

@Willie: it looks alright to me as well on Chrome (same OS, different machine). I've had less than optimal experience with the Firefox beta and MathJax, though. I'll try to see whether it's something at my end.

34.

@José: it could be that there's a font setting in some cookie. Do you see a proper \mathfrak g at the top of this page? If yes, try deleting all your cookies from MO (or at least all the ones that look like they have to do with mathjax).

35.

@Anton: many thanks for the suggestion. I have cleared the mathjax.org cookies and the \mathfrak g at the top of your page renders fine, but still not the one in the page I mentioned ??? I'll try resetting things locally and will get back to you later.

36.

@Anton: I have now cleared all cookies having to do with mathoverflow, but the problem persists.

• CommentAuthorgrp
• CommentTimeMar 2nd 2011

Regarding a post from Jan 15 2011 of Zev Chonoles, his picture is similar to what I get when I (accidentally) change the zoom level in IE. I think the issue of different zoom levels in IE is not address by MathJax, but I don't need it to be. Workaround is to keep it at 100% magnification.

Sorry to be off topic: where is the preview for either asking or answering questions on MathOverflow? I thought my (registered) account had enough reputation to get one, but I tested it and see only a narrow grey bar below the edit/answer window (shape close to _____________________ ) . Did I do something not to deserve a preview?

Gerhard "Ask Me About System Design" Paseman, 2011.03.02
37.

@grp: Everybody should see a preview, regardless or registration or reputation. Something must of happened so that your browser is not loading the javascript responsible for producing the preview. Is this a new problem? Did you see a preview before?

• CommentAuthorgrp
• CommentTimeMar 2nd 2011

It might be a javascript issue. I will try a couple of different things. Once upon a time, long ago, I had a preview, but also I had javascript issues. I'll mess with it and let you know. Thanks for the hint, Anton.

Gerhard Paseman, 2011.03.02
38.

Do you see buttons at the top of the editor? They look something like this (but only one row of them):

• CommentAuthorgrp
• CommentTimeMar 2nd 2011

I appreciate (and will accept) your debugging help, but will take this to email.

Gerhard Paseman, 2011.03.02
39.

I thought that MathJaX wasn't enabled for the main page. I seem to be getting quite a few "busy script" error messages these days on the main page.

(Though it might be that the script is still trying to work on the page I was on before navigating to the main page; nonetheless, I feel as though I'm getting more of these than I used to.)

40.

For some reason, MathJax renders everything as PNG images for me in Firefox (both Firefox 3.6.13 and the Firefox 4 RC) on OS X (10.6.6), while both Chrome and Safari correctly render everything using web fonts. This problem is specific to MO; the demos on the MathJax web site, for instance, work fine in either version of Firefox. Could this be related to this issue?

41.

@Evan: as you can see from the comment thread to that issue, we actually updated to those fonts right away. Today (about twenty minutes ago) we upgraded to MathJax 1.1, which has some improvements in browser support. Try clearing your cache. If the problem persists, please leave a post saying so.