Which Unicode characters can you depend on?

[ehamberg]

http://www.johndcook.com/blog/2013/04/11/which-unicode-characters-can-you-depend-on/

[fafner]

I wish there were a great monospace font with extensive Unicode symbol support. DejaVu seems to support a lot of math symbols. But it's far from being perfect.

My favourite fonts Inconsolata and Adobe Source Proc support barely anything beyond latin1. So I usually end up with some ugly font mix due to fallback.

(Of course if you want all of Unicode and hate yourself there is GNU Unifont. edit: Just checked and Unifont only supports all of BMP.)

[tomse → fafner]

I used DejaVu for a math symbols app, and it covered every symbol required http://mathsymbols.net/

The main problem was Apple products, surprisingly. It's crazy the iPad and iPhone don't have better Unicode support!

[shared4you → fafner]

Try out FreeMono, thousands of glyphs. Almost entire BMP. Excellent coverage as well [1]. Looks reasonable on my Emacs [2]

[1]: http://www.gnu.org/software/freefont/coverage.html

[2]: http://i.imgur.com/gTk5H00.png

[demetris → fafner]

Maybe Droid Sans Mono or Liberation Mono? They both have an extensive range of glyphs but I am not sure about math symbols.

[demetris → demetris]

Replying to myself... For anyone interested, here are two screenshots showing the Mathematical Operators blocks of Liberation Mono and Droid Sans Mono in the current Debian Sid:

https://dl.dropboxusercontent.com/u/390716/20130412-003357-d...

https://dl.dropboxusercontent.com/u/390716/20130412-003523-d...

[fafner → demetris]

Droid Sans Mono's 0 and O look too similar. Which makes it useless as a programming font.

[derleth → fafner]

Hey, I like GNU Unifont! What's wrong with it?

http://unifoundry.com/unifont.html

Terminus is nice, too, I suppose:

http://terminus-font.sourceforge.net/

http://terminus-font.sourceforge.net/shots.html

Inconsolata is popular:

http://en.wikipedia.org/wiki/Inconsolata

http://www.levien.com/type/myfonts/inconsolata.html

[mixmastamyk]

I'm not sure the intersection is necessary. I believe that as long as a character is included in any font on the system it will get shown (in web browsers at least).

It may not be the case on mobile, but it appears to work this way on Windows and Linux.

[vorg]

Some of those characters in Cook's blog didn't render in Chrome. Perhaps Chrome should ship with a default font that renders every Unicode character!

[ctdonath]

Didn't Hofsteader have some tools for mapping characters into the essential aesthetic form of a font? Could we at least get approximations of all Unicode characters (from a simple base font) into all typefaces installed on a system? (Or have I just been subjected to too much Doc McStuffins this evening?)

[lilyball]

Droid Sans is a font for Android. How exactly is that a "common font"? I mean, there's a lot of Android users out there, but I would have to assume "common font" means fonts that are available on multiple different platforms.

[johndcook → lilyball]

Droid Sans not as common as the other fonts on the list. I threw it into the mix for a couple reasons. For one, the feedback from twitter suggested that Android might support fewer characters than other platforms. For another, I expected a lot of commonality between the other fonts, but Droid fonts were an independent effort so I thought there might not be as much overlap.

[duskwuff → lilyball]

"Common font" here probably means those fonts that your content is likely to be displayed in. Some Android applications will use Droid Sans, so it counts.

[desas → lilyball]

It is available for multiple platforms, I have it on at least one of my desktops.

[k7lim]

ಠ_ಠ