Symbol Font correct on IE, but gibberish on FireFox

[I.N. Galidakis]

I thought I should try Mozilla FireFox on my pages to see how they render and
found a surprise:

My math Symbol font characters render as gibberish on FireFox, while they render
correctly on IE. Here's an example:

http://ioannis.virtualcomposer2000.com/math/Naturals.html

Does anyone know of any reasonable cure for this? I am using the Symbol font
according to the instructions found here:

http://www.tedmontgomery.com/tutorial/SYMBchrc.html

Many thanks,

 

[Jukka K. Korpela]

I thought I should try Mozilla FireFox on my pages to see how they
render and found a surprise:

Good for you.

My math Symbol font characters render as gibberish on FireFox, while
they render correctly on IE.

In such situations, it is usually Firefox that gets things right and IE
misbehaves.

Here's an example:

http://ioannis.virtualcomposer2000.com/math/Naturals.html

If you set IE to ignore fonts suggested on web pages, IE will display the
correctly, e.g. the character Æ (capital letter AE) as Æ and not as an empty
set symbol, etc. This is not what you want, but it is the correct rendering
of the document you have created.

Does anyone know of any reasonable cure for this?

Use correct methods for including special characters in HTML, such as entity
references or UTF-8 encoding. If your document contains a lot of special
characters, it's probably best to use a Unicode-enabled text editor such as
BabelPad and UTF-8. See e.g.
http://www.alanwood.net/unicode/utilities_editors.html

Note that this approach also lets you use proper quotation marks and
apostrophes instead of the crude ASCII characters " and ', the correct
n-ary summation symbol (instead of Greek letter sigma), etc.

I am using the
Symbol font according to the instructions found here:

http://www.tedmontgomery.com/tutorial/SYMBchrc.html

It's worse than nonsense and bogus, since it appears to "work" when you use
faulty browsers, just enough to tempt you into using this horrendous
trickery.

The late Alan J. Flavell wrote words of wisdom about this issue, too:
http://www.alanflavell.org.uk/charset/fontface-harmful.html

 

[I.N. Galidakis]

Good for you.


In such situations, it is usually Firefox that gets things right and IE
misbehaves.


If you set IE to ignore fonts suggested on web pages, IE will display the
correctly, e.g. the character Æ (capital letter AE) as Æ and not as an empty
set symbol, etc. This is not what you want, but it is the correct rendering
of the document you have created.


Use correct methods for including special characters in HTML, such as entity
references or UTF-8 encoding. If your document contains a lot of special
characters, it's probably best to use a Unicode-enabled text editor such as
BabelPad and UTF-8. See e.g.
http://www.alanwood.net/unicode/utilities_editors.html

Note that this approach also lets you use proper quotation marks and
apostrophes instead of the crude ASCII characters " and ', the correct
n-ary summation symbol (instead of Greek letter sigma), etc.


It's worse than nonsense and bogus, since it appears to "work" when you use
faulty browsers, just enough to tempt you into using this horrendous
trickery.

Many thanks for all the illuminating info. I had no idea the issue was so
convoluted. Unfortuately at this point I cannot convert any of my webpages to
UTF-8, so I will contend with just putting a warning on my math pages for non-IE
users.

The late Alan J. Flavell wrote words of wisdom about this issue, too:
http://www.alanflavell.org.uk/charset/fontface-harmful.html

When did his status change to "late"? I seem to remember his name on this forum
a while back.

 

[Jukka K. Korpela]

Unfortuately at this point I cannot convert any of
my webpages to UTF-8, so I will contend with just putting a warning
on my math pages for non-IE users.

It's not just non-IE users. Anyone with IE set to ignore font settings on
web pages (common among people with reduced eyesight), or with additional
software plugged into IE for converting content to speech or Braille, or
with a system with Symbol font removed (it's a lousy font really) will see
the real characters on your page, as opposite to their "fontistic"
misrepresentations.

I realize that converting to UTF-8 can be laborious (since you would have to
replace characters with font declarations into the characters they are meant
to stand for), but it could in fact be automated, and you could
alternatively use character references.

 

[Raymond SCHMIT]

It's not just non-IE users. Anyone with IE set to ignore font settings on
web pages (common among people with reduced eyesight), or with additional
software plugged into IE for converting content to speech or Braille, or
with a system with Symbol font removed (it's a lousy font really) will see
the real characters on your page, as opposite to their "fontistic"
misrepresentations.

I realize that converting to UTF-8 can be laborious (since you would have to
replace characters with font declarations into the characters they are meant
to stand for), but it could in fact be automated, and you could
alternatively use character references.

What he can do is:
Display your beautifull page in your IE brower then print it to a
pdf-printer. then upload the resulted pdf file and let a link pointing
to it.So *everybody* will be able to look at your work and print it if
needed.

 

[I.N. Galidakis]

What he can do is:
Display your beautifull page in your IE brower then print it to a
pdf-printer. then upload the resulted pdf file and let a link pointing
to it.So *everybody* will be able to look at your work and print it if
needed.

Thanks. What I did was to trash all the symbol characters and emulate all
symbols using regular html in all my math pages.

Again, thanks to all who responded.

 

[Jukka K. Korpela]

Display your beautifull page in your IE brower then print it to a
pdf-printer. then upload the resulted pdf file and let a link pointing
to it.So *everybody* will be able to look at your work and print it if
needed.

No, PDF format is inherently much less accessible than HTML. It is often
much slower to open a PDF file, it contains fixed font faces and sizes (as
opposite to HTML, where font settings can always be overridden by the user)
and fixed layout, and it may seriously confuse a screen reader (if you don't
take care when creating it, and most people don't). And if you create a
simple PDF document the way you describe, you lose all links.

For some data, such as math documents containing really rare (in fonts)
characters, a PDF file, if served in addition to an HTML document (and not
instead of it), may be an improvement.

 

[Raymond SCHMIT]

No, PDF format is inherently much less accessible than HTML. It is often
much slower to open a PDF file, it contains fixed font faces and sizes (as
opposite to HTML, where font settings can always be overridden by the user)
and fixed layout, and it may seriously confuse a screen reader (if you don't
take care when creating it, and most people don't). And if you create a
simple PDF document the way you describe, you lose all links.

For some data, such as math documents containing really rare (in fonts)
characters, a PDF file, if served in addition to an HTML document (and not
instead of it), may be an improvement.

Oups ! .... i forgot to mention this.
The pdf file is an extra for those who want print or view what they
can't because not surfing with IE.

 

[GTalbot]

I.N. Galidakis,

1- Conversion to utf-8 can be automated.

2- Before judging Firefox prematurely, you may want to test it:

MathML Test Suite
http://www.w3.org/Math/testsuite/

MathML Torture Test
http://www.mozilla.org/projects/mathml/demo/texvsmml.xhtml

MathML Tester
http://www.mozilla.org/projects/mathml/demo/tester.html

MathML project at mozilla.org:
http://www.mozilla.org/projects/mathml/

3- I have a test case where Internet Explorer 7 and Internet Explorer
8 beta 2 can NOT render greek letters with Symbol font while Firefox
3.0.3, Opera 9.62, Safari 3.1.2, Konqueror 4.1.1, Hv3 TKHTML alpha
16, Amaya 10.0.1, Seamonkey 2.0a2pre all can:

http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/greek-letters-with-symbol-font.html

4- PDF is definitely not recommendable for many reasons (mentioned by
Jukka Korpela).

Regards, Gérard