I love FontForge. It is what I always use for creating fonts. I tried
FontStruct once but found it highly unsatisfactory (not least because of the
geometric limitations and the licensing of the resulting fonts). I know next
to nothing about creating fonts, but for the simple things I need (letters &
diacritics, but no ligation et cetera) it works fine.

—Calculator Ftvb

P. S. I'm pretty sure is not the real
CSUR; I think it is an extension of the main one (at, as is explained at the page. I'm not sure about that though so
feel free to correct me if I'm wrong.

On 26 June 2011 00:06, Rebecca Bettencourt <[log in to unmask]> wrote:

> On Sat, Jun 25, 2011 at 7:30 PM, yuri <[log in to unmask]> wrote:
> > Hi Conlangers,
> >
> > I'm new to the list. I started creating my own conlang about 15 years
> > ago but it got put on the backburner. I'm resuming the project.
> Hi Yuri!
> > I'm interested in what software people recommend for creating a font
> > for my script.
> >
> > I have installed fontforge and inkscape, with the idea of creating the
> > glyphs in inkscape and importing the SVG for each glyph into
> > fontforge.
> For the price, you can't beat FontForge. It can do everything you
> need, though it gets a bit technical. A tip if you're going to be
> importing from SVG files: put a rectangle of a constant height around
> each glyph so they remain the same size when imported, then remove the
> rectangle after importing.
> The only other font software I've used is something called ScanFont,
> but only just to trace outlines from a scanned image; it's pretty
> terrible at actually producing a valid font. :P
> > Also, I understand the unicode address space includes areas for
> > private use. Can anyone tell me more about that?
> Yes, code points E000 to F8FF and F0000 and above are set aside as
> private use. That means Unicode itself won't ever define what those
> code points mean; whoever is exchanging text using those code points
> has to come up with their own agreement as to what those code points
> mean.
> The conlang community has their own agreement about what private use
> code points mean. It's called the ConScript Unicode Registry. More
> information is here:
> If you'd like some Unicode code space, find some unassigned space in
> that list, put together a proposal including a list of all the
> character names and the code points you'd like to assign them, and
> then I can add it.
> > One required feature is the ability to compose any combination of base
> > letter and diacritics.
> > I'd like to avoid having to have a font full of precomposed characters
> > because the diacritics above each base letter is the vowel, and the
> > diacritic below the base letter indicates voiced/unvoiced and
> > aspirated/unaspirated, so the combinations could run into the
> > thousands.
> >
> > The base letters themselves are all consonants.
> If the base letters are all the same width, this is easy. The
> diacritics are just zero-width characters with the glyph to the left
> of the origin. If they're all different widths, though, you'll need to
> do a lot more work, but it can be done. Something about GPOS tables or
> something; I've actually never done it before.
> -- Rebecca Bettencourt