There are two problems:

1. You can't paste any characters beyond the Mac Roman codepage into FontLab Studio’s Metrics Window on Mac.
2. You're using the Cyrillic Schwa codepoints, but on the website they probably use the Latin Schwa characters which should be called and encoded as /uni018F (UC) and /uni0259 (LC).

Hope this helps,
Jens

.

Thanks Jens.

But does that mean /uni018F (UC) and /uni0259 (LC) should replace the existing /uni04D8 and /afii10846 glyphs or be added to the character set - I don't suppose it would do any harm to have all four?

BTW - where did you get that information? is it on a codepage, and if so, which one?

That would depend on which languages/scripts you intend to support.

If a font supports Azerbaijani in the Latin alphabet, it should include 018F and 0259. If it supports Azerbaijani in the Cyrillic alphabet it should include 04D8 and 04D9. If it supports both scripts, it should include all four. If the font doesn't otherwise contain cyrillic glyphs, including 04D8 and 04D9 would be rather pointless.

The information is all available from the Unicode Consortium.

André

That's great André - it does support Latin and Cyrillic, so all four are needed.

Thanks also for the link - very useful.

While on the subject of schwa:
Do readers simply expect to see a rotated "e", or is there any merit in applying the thematic construction principles employed elsewhere in the typeface?

And is this character really necessary: is there even the remotest possibility that an Azerbaijani will pay to license a font from a Western foundry?

This opens up a whole new can of worms, not to mention a lot more work – I can't really just add the two extra characters and hope for the best. I would really need to add the whole Latin Extended B codepage. Which I hadn't planned to do at all!

Do I just suggest using the Cyrillic keyboard instead of Latin?

And is this character really necessary: is there even the remotest possibility that an Azerbaijani will pay to license a font from a Western foundry? Ha ha, that's funny Nick – I seriously doubt it, but my client is Romanian (and they pay incredibly promptly!), while their client (I presume) is Azerbaijani.

I wouldn't think in terms of unicode blocks, but rather of *languages*. After all, the set of languages included in any given unicode block is rather arbitrary and there's no real reason to allow that to dictate your language support choices. If your client is (indirectly) Azerbaijani, you'd want to ensure you include the latin and/or cyrillic characters necessary to meet their needs. That certainly doesn't include all the characters in latin extended B or IPA extensions. Schwa (018F/0259) is the only character in these blocks needed to support Azerbaijani, and since the web page you show above is clearly in the Latin script, you'd definitely want to include this.

A cyrillic keyboard would only be appropriate for Azerbaijani written in cyrillic (which the client would likely want as well) -- in which case you'd need to to include the standard cyrillic characters along with schwa (04D8/04D9), Ғ (0492/0493), Ј (0408/0458), Ҝ (049C/049D), Ө (04E8/04E9), Ү (04AE/04EF), Һ (04BA/04BB), Ҹ (04B8/04B9).

André

Brilliant! many thanks for your kind assistance - that's probably saved me a lot of work. I have those glyphs already.

my client is Romanian (and they pay incredibly promptly!), while their client (I presume) is Azerbaijani.

:)