[EversonMono] everson mono - bugs with Yiddish (Hebrew) with dagesh

Mark H. David mhd at yv.org
Tue Mar 3 01:49:49 GMT 2009

I'm sure it's not just random placement. I noticed the dots were placed perfectly,
which I thought was suspicious.  Then I looked at the rendered glyphs in Acrobat
Reader, zooming them way up, and compared them to the glyphs of the characters
I'm alleging they're from in the font using FontLab (font editor), and they matched.
Also, you can tell because the placement has to be different to get this effect
depending on the following character.  For example,


places the dot in a totally different place vs. 


In fact, the placement varies along both the x and y axes in these two cases.

Seems like a rendering bug almost for sure.  But is the renderer that exposes
this problem common to Firefox and Nisus Writer Express?

I'll get on my mac later and get some more info, and send more/better screen shots.

Meanwhile, I'm happy to report this all works fine on Firefox running on Linux,
with the following version info from Firefox:

Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv: Gecko/2009020911 Ubuntu/8.10 (intrepid) Firefox/3.0.6



----- Original message -----
From: "John Hudson" <tiro at tiro.com>
To: "Discussion of the Everson Mono font" <eversonmono at evertype.com>
Date: Mon, 02 Mar 2009 17:21:20 -0800
Subject: Re: [EversonMono] everson mono - bugs with Yiddish (Hebrew) with dagesh

Mark H. David wrote:

> OK, I think I've got a description that precisely accounts for all the 
> problems: the dagesh is combined with the base character *following* its 
> base character, and the base character is rendered unadorned (without 
> dagesh) and the following base character has the corresponding 
> char+dagesh character from FBxx (alpahabetic presentation forms Unicode 
> range) rendered in its place.

It is possible that what you are seeing is not actually combination of 
the mark with the following base, but merely a blind offset to the left, 
causing the mark to appear as if combined with the following letter. 
That could happen if a layout engine simply has no idea what to do with 
the dagesh at all, is not performing character-level substitution of 
Aphabetic Presentation Forms, is not performing OT GSUB ligation, and is 
not performing OT GPOS dynamic positioning. If this is that case, you 
will see a similar offset onto a following letter even if that letter is 
not one that normally takes dagesh.

I'm afraid I have deleted your initial mail on this subject. Could you 
provide a screenshot showing the problem?

If the font is behaving correctly in Mellel, as I believe you indicated 
previously, then I think you can be reasonably sure that the bug is not 
in the font.

John Hudson

EversonMono mailing list
EversonMono at evertype.com

More information about the EversonMono mailing list