LaTeX Why are MathJax images not displaying on my webpage?

[Greg Bernhardt]

It still renders very slowly in IE8 because of the conditional.

oddly the emulation doesn't break IE9 now. the conditionals are off.

 

[berkeman]

Hey, for the first time in a long time, I can see LaTeX rendered again! Woo-hoo!

Don't know if it's related to whatever you just did, but thanks. I'm running IE7 on Vista on my work PC.

 

[vela]

oddly the emulation doesn't break IE9 now. the conditionals are off.

That's why I suggested ditching the inline configuration. The combination of inline configuration and IE7 emulation was breaking MathJax on IE9. Now that the pages load the configuration from a file, IE9 works in emulation mode.

 

[Greg Bernhardt]

That's why I suggested ditching the inline configuration. The combination of inline configuration and IE7 emulation was breaking MathJax on IE9. Now that the pages load the configuration from a file, IE9 works in emulation mode.

thanks! you're so smart

 

[vela]

IE8 works pretty fast now, too. A page that was taking almost 2 minutes to render before now finishes in about 10 seconds!

 

[Borek]

I might be mistaken but I thought the scripts were on a 24h cache.

Seems like they are loaded every time - or at least I see Opera waiting for mathjax server much often than that.

[qute]Do you have any other specific I can relay to them?[/quote]

Hardly :( Sometimes I get error messages in a popup - two types of these, one is in attachment, other one is

MathJax no longer loads a default configuration file; you must specify such files explicitly. This page seems to use the older default config/MathJax.js file, and so needs to be updated. This is explained further at

http://www.mathjax.org/help/configuration

I guess that's the same problem (with downloading mathjax files) in both cases. From what I understand you were tinkering with configuration details, so could be these messages are related to different configurations.

Anything in Opera logs?

You may get more than you ask for

[2011-05-24 23:31:18] JavaScript - http://cdn.mathjax.org/mathjax/latest/jax/element/mml/jax.js

Linked script not loaded
[2011-05-24 23:31:18] JavaScript - http://cdn.mathjax.org/mathjax/latest/jax/input/TeX/jax.js

Linked script not loaded
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-webkit-box-shadow is an unknown property
Line 1:
  pace: nowrap; float: none; box-shadow: 5px 5px 15px #AAAAAA; -webkit-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-moz-box-shadow is an unknown property
Line 1:
  5px 5px 15px #AAAAAA; -webkit-box-shadow: 5px 5px 15px #AAAAAA; -moz-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-khtml-box-shadow is an unknown property
Line 1:
   5px 5px 15px #AAAAAA; -moz-box-shadow: 5px 5px 15px #AAAAAA; -khtml-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
Declaration syntax error
Line 1:
  geTransform.Microsoft.dropshadow(OffX=2, OffY=2, Color='gray', Positive='true')}
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
Declaration syntax error
Line 2:
  ng: 0; margin: 0; background-color: white; opacity: 0; filter: alpha(opacity=0)}
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-webkit-border-radius is an unknown property
Line 1:
  : nowrap; float: none; z-index: 201; border-radius: 15px; -webkit-border-radius:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-moz-border-radius is an unknown property
Line 1:
  ndex: 201; border-radius: 15px; -webkit-border-radius: 15px; -moz-border-radius:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-khtml-border-radius is an unknown property
Line 1:
  px; -webkit-border-radius: 15px; -moz-border-radius: 15px; -khtml-border-radius:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-webkit-box-shadow is an unknown property
Line 1:
  html-border-radius: 15px; box-shadow: 0px 10px 20px #808080; -webkit-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-moz-box-shadow is an unknown property
Line 1:
  x 10px 20px #808080; -webkit-box-shadow: 0px 10px 20px #808080; -moz-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-khtml-box-shadow is an unknown property
Line 1:
  px 10px 20px #808080; -moz-box-shadow: 0px 10px 20px #808080; -khtml-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
Declaration syntax error
Line 1:
  geTransform.Microsoft.dropshadow(OffX=2, OffY=2, Color='gray', Positive='true')}
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-webkit-box-shadow is an unknown property
Line 2:
  loat: none; z-index: 201; box-shadow: 0px 10px 20px #808080; -webkit-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-moz-box-shadow is an unknown property
Line 2:
  x 10px 20px #808080; -webkit-box-shadow: 0px 10px 20px #808080; -moz-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
-khtml-box-shadow is an unknown property
Line 2:
  px 10px 20px #808080; -moz-box-shadow: 0px 10px 20px #808080; -khtml-box-shadow:
  --------------------------------------------------------------------------------^
[2011-05-24 23:31:49] CSS - [PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN][PLAIN]https://www.physicsforums.com/search.php?searchid=2702324[/PLAIN] [/URL][/URL][/URL][/URL][/URL][/URL][/URL]
Inlined stylesheet
Declaration syntax error
Line 2:
  geTransform.Microsoft.dropshadow(OffX=2, OffY=2, Color='gray', Positive='true')}
  --------------------------------------------------------------------------------^

 

[Greg Bernhardt]

MathJax has been added to the mobile skin

 

[micromass]

Hmm, something annoying: I often make a post, then I edit the post. But after I edited the post, the LaTeX won't run. So I have to click refresh for the LaTeX to show up. It's no biggie, but I just wanted to let this know. (I'm on firefox, by the way)

 

[Fredrik]

Sorry, I don't know what is supposed to work and what isn't. \copyright doesn't work.

http://www.mathjax.org/docs/1.1/tex.html#supported-latex-commands

 

[Borek]

Hmm, something annoying: I often make a post, then I edit the post. But after I edited the post, the LaTeX won't run. So I have to click refresh for the LaTeX to show up. It's no biggie, but I just wanted to let this know. (I'm on firefox, by the way)

LOL, that would mean old problems with preview are still here

 

[Greg Bernhardt]

LOL, that would mean old problems with preview are still here

hmmm I can't reproduce this problem on FF 4.0.1

 

[micromass]

hmmm I can't reproduce this problem on FF 4.0.1

If you want, I can make a video about what I do exactly. But I don't know the software for filming the desktop though...

 

[vela]

Do you have an add-on installed that might be causing the problem?

 

[micromass]

Do you have an add-on installed that might be causing the problem?

The weird thing is that the same problem also occurs on IE. So I don't think add-ons will be the problem.
Also, if I edit the post, then only the LaTeX of the edited post doesn't work. All other LaTeX formulas work fine...

 

[vela]

The weird thing is that the same problem also occurs on IE. So I don't think add-ons will be the problem.
Also, if I edit the post, then only the LaTeX of the edited post doesn't work. All other LaTeX formulas work fine...

I see the same behavior with Safari. MathJax runs when the browser says it's done processing the page. Apparently, saving the edited post doesn't cause the browser to send this signal, which makes sense since if you have a script that runs when the page finishes loading, you probably don't want it to run again if you modify just a small piece of the page. So the text of the edited post doesn't get processed.

EDIT: Firefox on the Mac acts the same way as well.

 

[Greg Bernhardt]

I see the same behavior with Safari. MathJax runs when the browser says it's done processing the page. Apparently, saving the edited post doesn't cause the browser to send this signal, which makes sense since if you have a script that runs when the page finishes loading, you probably don't want it to run again if you modify just a small piece of the page. So the text of the edited post doesn't get processed.

EDIT: Firefox on the Mac acts the same way as well.

Are we talking about quick editing?

 

[Borek]

Just testing...

ax^2 + bx + c = 0

Strange - quick edit under Opera, once worked, thrice didn't (OK, I am not going to edit it once again, when it worked it must have been some other problem, as teh page reloaded, which usually doesn't happen).

I guess when the page code is changed after quick edit MathJax script is not called.

 

[Borek]

Yes, that's the thing - refreshing works, quick edit doesn't.

So when it comes to preview we are back at square one.

 

[Greg Bernhardt]

Yes, that's the thing - refreshing works, quick edit doesn't.

So when it comes to preview we are back at square one.

There has never been preview for quick edit. Quick edit save would require a refresh because quick edit is an AJAX function that does not call MathJax.

 

[Hurkyl]

LOL, that would mean old problems with preview are still here

I don't follow your logic. In fact, testing preview directly contradicts your conclusion.

 

[Borek]

There has never been preview for quick edit.

Yep, obvious, I got things mixed up

 

[Fredrik]

I've had the same problem as Micromass since we made the move to MathJax. Every time I save an edit, I only see the source code. (Previews work fine. The problem is that when I view the post after saving the changes, I only see source code). A refresh solves the problem.

This isn't nearly as annoying or confusing as the old preview problem, which required a refresh after each preview. It also required me to explain it to someone else once a week.

 

[cepheid]

I've had the same problem as Micromass since we made the move to MathJax. Every time I save an edit, I only see the source code. (Previews work fine. The problem is that when I view the post after saving the changes, I only see source code). A refresh solves the problem.

This isn't nearly as annoying or confusing as the old preview problem, which required a refresh after each preview. It also required me to explain it to someone else once a week.

I've been having this problem too, in Safari.

Unrelated note: excited to see LaTeX working on mobile site!

 

[diazona]

It's possible to manually invoke MathJax on a section of the page whose content has been changed - at least, I've seen it done on other sites. So that might be a feature to consider for the future.

 

[Fredrik]

Every time I save an edit, I only see the source code. (Previews work fine. The problem is that when I view the post after saving the changes, I only see source code). A refresh solves the problem.

OK, I just did some testing, and I don't have this problem at all. I guess I just remembered it wrong when I wrote the above. There is no problem if I preview before I save the changes. The only time I see the source code instead of an image is when I just click edit, change something, and then click save. Both FF4 and IE9 behave the same way.

 

[Borek]

There is a bug. Whatever is in noparse tags should be not parsed. That allows things like showing people how to embed youtube video: enter [noparse][/noparse] to get



But it is not working for tex nor itex tags, as - even if they are between noparse tags - they are parsed by MathJax. So

[nοparse][tеx]a=x^2[/tеx][/nοparse]

should yield

[tеx]a=x^2[/tеx]

but yields

[noparse]a=x^2[/noparse]

(and yes, I do tricks to display it properly )

 

[Greg Bernhardt]

borek is mathjax still failing for you in opera?

 

[Borek]

I think I have seen the same problem about 10 hours ago, but to be honest I don't remember. You can be sure I will complain if it will repeat often enough

 

[vela]

Any thoughts on adding new delimiters for TeX code? MathJax supports multiple delimiters, so it won't break existing posts. Plus, it's faster typing something like $$ than the current tex and itex tags.

 

[Fredrik]

Any thoughts on adding new delimiters for TeX code? MathJax supports multiple delimiters, so it won't break existing posts. Plus, it's faster typing something like $$ than the current tex and itex tags.

I think it would be awesome if something like that could be implemented.

Btw, there's some discussion about that in this thread, starting at post #48.

 

[Daiquiri]

Hello, I have a problem...
I can't see the "Sigma" icon for creating LaTex code, even in advanced text mode...
I've tried both IE and Chrome, but still no icon there. What am I doing wrong?

 

[Borek]

I can't see the "Sigma" icon for creating LaTex code, even in advanced text mode...

It is not present in all subforums. Check general math or general physics - it is there, but not here.

 

[AlephZero]

Any thoughts on adding new delimiters for TeX code? MathJax supports multiple delimiters, so it won't break existing posts. Plus, it's faster typing something like $$ than the current tex and itex tags.

We probably don't want to go too far towards "posts in LaTex". There are some sub-forums here which have very little LaTeX use, and people there probably wouldn't appreciate having to type \$ to get a US Dollar sign.

The think that really bugs my typing is the foward slash in the closing tags and the backwards slash in LaTeX. [ \ tex ] (without the spaces!) as an alternative end tag would be nice.

 

[Redbelly98]

Hey, I'm glad to see that equations are automatically indented under the new implementation -- I'm pretty sure this was not the case before. I don't know if this is built into MathJax, or is a tweak by Greg, but I am happy to see it.

(At least this is true in FireFox 4)

 

[Greg Bernhardt]

The think that really bugs my typing is the foward slash in the closing tags and the backwards slash in LaTeX. [ \ tex ] (without the spaces!) as an alternative end tag would be nice.

I am open to suggestions for additional delimiters.

Hey, I'm glad to see that equations are automatically indented under the new implementation -- I'm pretty sure this was not the case before.

It is a configuration setting in MathJax.

 

[Fredrik]

I suggest

$f$ as an alternative to [itεx]f[/itεx]

$$f$$ as an alternative to [tεx]f[/tεx]

(I couldn't figure out how Borek did his magic so I just wrote ε instead of e to let you see what I wrote).

That first one might break a few old posts, but it still think it's the best option, since it's easy to type, and also what you actually type in a LaTeX document.

 

[micromass]

I suggest

$f$ as an alternative to [itεx]f[/itεx]

$$f$$ as an alternative to [tεx]f[/tεx]

(I couldn't figure out how Borek did his magic so I just wrote ε instead of e to let you see what I wrote).

That first one might break a few old posts, but it still think it's the best option, since it's easy to type, and also what you actually type in a LaTeX document.

I second this idea. $'s instead of [ t e x ] would be a very welcome change!

 

[vela]

That first one might break a few old posts, but it still think it's the best option, since it's easy to type, and also what you actually type in a LaTeX document.

It not only breaks existing posts, but, as AlephZero noted, it makes it difficult to type a dollar sign when you want a dollar sign. Something like $$ and @@ would be better, or $$ and $$$.

 

[Fredrik]

Yes, but this is a physics and math forum, not a forum for American accountants. I'm assuming that they can write something like \$3.50 instead of $3.50, and if that doesn't work, they can always write 3.50 USD.

(I'm not saying that I would find other options intolerable, only that I still like my suggestion best).

 

[Greg Bernhardt]

Yes, but this is a physics and math forum, not a forum for American accountants. I'm assuming that they can write something like \$3.50 instead of $3.50, and if that doesn't work, they can always write 3.50 USD.

(I'm not saying that I would find other options intolerable, only that I still like my suggestion best).

I bet it would cause a lot of casual and new members a lot of grief at first. \$ is not intuitive :)

 

[dlgoff]

I bet it would cause a lot of casual and new members a lot of grief at first. \$ is not intuitive :)

Well, I hope it doesn't effect the $ sign here: https://www.physicsforums.com/payments.php"

 

[Redbelly98]

I suggest

$f$ as an alternative to [itεx]f[/itεx]

$$f$$ as an alternative to [tεx]f[/tεx]

(I couldn't figure out how Borek did his magic so I just wrote ε instead of e to let you see what I wrote).

That first one might break a few old posts, but it still think it's the best option, since it's easy to type, and also what you actually type in a LaTeX document.

Yes, but this is a physics and math forum, not a forum for American accountants. I'm assuming that they can write something like \$3.50 instead of $3.50, and if that doesn't work, they can always write 3.50 USD.

(I'm not saying that I would find other options intolerable, only that I still like my suggestion best).

I think using $ would break too many old posts. Even though we're primarily a science/math/engineering forum, the Politics and World Affairs area (for example) is and has been quite active at PF. A double $$ would be much better as a LaTeX delimiter.

 

[AlephZero]

As I see it, a key design feature of the current interface (whether or not it was planned that way or "just growed") is that if you don't even know LaTeX exists, you can type text and get what you expect.

Moving away from that principle on any "general access" web forum, is a BIG strategic decision to take IMO. That certainly rules out $ as a delimeter, and probably $$ as well.

BLEH! I just tried a search to see how many posts already used $$, and the answer was

1. Sorry - no matches. Please try some different terms.
The following words are either very common, too long, or too short and were not included in your search : $$

 

[Fredrik]

I can certainly see your point and to some extent I agree with it. Not so much that I have changed my mind, but enough to start thinking about other options. How about ££math££ for inline, and $$math$$ for displaystyle?

Not sure where these symbols are on keyboards in other countries. On a Swedish keyboard, the pound sign is Alt Gr+3 and the dollar sign is Alt Gr+4, so both are easy to type.

Or how about not introducing new delimiters and instead configuring hotkeys that type the delimiters automatically. For example, right now, if I press Alt Gr+i, the computer types [noparse] and puts the cursor between and , and if I highlight a word before I press Alt Gr+i, the and the are inserted before and after that word. Is it possible to e.g. configure F8 and F9 to do that with itex and tex tags?[/noparse]

 

[Redbelly98]

U.S. keyboards do not have £, at least that is true for PC's. $ is <shift>4

Fredrik, I'm not sure what this Gr key is. We normally use Shift, Ctrl, or Alt to change the character or function of other keys.

 

[vela]

MathJax supports multiple pairs of delimiters, so it doesn't need to be one size fits all. You could have one set that's easy to type on US keyboards, and another set that's easy to type on non-US keyboards.

 

[Greg Bernhardt]

What about a double pound sign? Is that ever used in anything?

fyi, if by this weekend we haven't encountered any big problems with the MathJax setup, I will turn the old system off (images are still being generated in case we switch back)

 

[Borek]

Opera still stumbles now and then.

That's not to say you shouldn't switch, you just asked earlier...

 

[Fredrik]

Fredrik, I'm not sure what this Gr key is. We normally use Shift, Ctrl, or Alt to change the character or function of other keys.

Alt Gr is the key to the right of the space bar on my keyboard.

MathJax supports multiple pairs of delimiters, so it doesn't need to be one size fits all. You could have one set that's easy to type on US keyboards, and another set that's easy to type on non-US keyboards.

Good point.

 

[Fredrik]

I've been looking at the keyboard layout page at Wikipedia. It seems to me that almost everyone would find it easy to type the delimiters if we choose

££math££, $math$, §§math§§ and ··math·· for inline (i.e. itex)
$$math$$ and ¢¢math¢¢ for displaystyle (i.e. tex)

I would say that the §,· and ¢ characters aren't really necessary. They are just a minor improvement for Germans, Spaniards, and French Canadians respectively.