Re: [DejaVu-fonts] patch Old South Arabic

[nocturnaldreamer <noc...@gm...>]

Am 20.02.2011 19:36, schrieb Ben Laenen:
> Vertical stems in capitals: 202
> Horizontal stems in capitals: 170
>
> I guess you're confused as to why these two numbers are different? But it's a
> pretty basic rule in typeface design that in most typefaces horizontal stems
> must be thinner than vertical ones. There are very few fonts with the same
> numbers for horizontal and vertical stems.
>
> Diagonals will be something in between for obvious reasons.
> Curves will be slightly thicker as well for visual reasons (if it weren't
> thicker, it would look too thin). There are much more rules like this, like
> why the curve of O goes slightly below base line, or why the diagonals of X
> aren't really just two crossing lines, but that at least one of them is
> "broken" at the center when you look at it closely.
>
>
> As for Old South Arabic (and Old Italic actually), years ago there were
> discussions about whether these scripts should be in DejaVu in the first
> place. The answer never was fully resolved, but is probably something as: yes,
> if it matches the style of the font and it wouldn't just be a set of glyphs
> within the font that have no relationship with the other glyphs. As of now,
> due to perhaps your inexperience with font design, this is a bit like such a
> set of glyphs. So I'm wondering in fact whether it wouldn't be better to
> remove Old Italic again before the release and import it later when the issue
> is resolved?
>
> Other opinions on this?
>
> Greetings
> Ben
Hi,

in the case of Old Italic I touched up most of the glyphs. However it is 
my opinion that we should be careful with changes just for the sake of 
conformance. Old Italic and Old South Arabian precede typographical 
conventions by centuries. Forcing the latter on the former might just 
create an unwarranted anachronism.

The first goal we have is having a set of glyphs that serves the user 
base, in this case historians, archaeologists, and palaeographers. The 
second goal is internal consistency as you described it. The question 
is, how to prioritize them.

Now I'd say that the more a script is removed from our LGC sets (in 
terms of time and/or culture) the more leeway we should allow for. Case 
in point: Georgian in Serif differs in line weights by the same order of 
magnitude from Latin, plus there are differences in x- and cap-heights. 
Still the submission was accepted because visually the set meshes well 
with the rest of the font. I think that this is what we should aim for.

That being said, I think it wouldn't hurt to postpone the inclusion of 
Old South Arabian in any case.

Regards,
Gee Fung Sit