Tal Leming has announced the UFOv3 format. Inkscape shoud support it.
---------- Forwarded message ----------
Hello Everyone,
We are pleased to announce the first draft of the UFO 3 specification. The
specification contains the major new features that we have planned, but
everything is not yet final. There are a few things that we are thinking
through and there are some editorial changes needed, but we think it is ready
for initial peer-review. This revision has been in the works since the day
after the last Robothon and it has gone through many, many rounds of editing,
thinking and starting over. We'd really appreciate it if the community would
read through this to see if there are errors, things that don't make sense, etc.
UFO 3 builds on UFO 2 but it adds a number of new data structures. There are a
lot of new things, but these are some highlights:
Layers
The UFO may now contain more than one glyph set. Corresponding glyphs across
the different glyph sets represent glyph layers within the font. This layering
model is conceptually unrestricted. It can represent the common
foreground+background model that Fontographer pioneered and it can just as
easily represent something entirely different.
Glyph Images
Glyphs may now contain a background image. These images are stored within the
UFO.
Links
Links are abstract relationships between points. One potential use for links is
to identify parts of a glyph ("these points form a stem," "these points form a
serif," etc.) Our hope is that this will help with things like hinting. This is
the one big section of the specification that we are still thinking through, so
any input from hinting experts would be greatly appreciated.
Kerning
The UFO has had a very powerful kerning model from the very beginning.
Unfortunately it wasn't overly documented because we thought it was obvious how
things worked. Based on implementor feedback, we've added a large amount of
documentation and added a new bit of kerning related data to fontinfo.plist.
Terminology, Algorithms, Example Code, Etc.
Our work on WOFF has taught us a lot about writing specifications and we've
applied that experience to the UFO 3 spec. We have read every single sentence
over and over and edited nearly everything to make sure that things are clear.
We've added copious documentation about how class kerning is implemented in the
UFO. We've written lots of sample algorithms for handling various parts of the
data. Our hope is that this makes the spec much more clear. This isn't
technically a change from the previous spec, but it is a significant addition.
And More
There's more, but less exciting, stuff like WOFF support in fontinfo.plist,
guidelines, a directory for application specific data, etc.
If you are interested, please take a look at the spec here:
http://unifiedfontobject.org/versions/ufo3
Please note that this is a draft, not a final specification. We're starting our
own implementation of the spec in RoboFab and our other libraries. This could
cause revisions to the spec if we find something that is
impossible/difficult/ugly to implement. We're hoping to make the specification
final no later than early next year.
Original issue reported on code.google.com by d.crossland on 6 May 2011 at 7:59
Original issue reported on code.google.com by
d.crossland
on 6 May 2011 at 7:59