Andreas Stötzner

You can hardly define this issue by exact figures. Common sense is your friend.
My golden rule nº 1: Do very careful spacing of all glyphs first.
Golden rule nº 2: Restrict kerning to pairs which are absolutely neccessary.
– Makes the best of your font and keeps your file size and work load handy.

Small capitals: I still believe that putting them in a seperate font is more sensible, anyway. Saves me a bunch of extra work, for the font user it is equally comfortable or even more handy. And it avoids that stupid case when it also transforms the first Capital letter of a name (which I mark in order to change ist to sc.) which is annoying, of course.

As I said earlier, features are overrated by many.

Jens Kutilek said:

…
E.g. FF Trixie, the original distressed typewriter font.

• Quality of glyph outlines = horrible
• Quality of spacing = monospaced, what do you expect?
• Quality of kerning = kerning is nonexistent

…

Well, in that case, those choices were based on a particular reasoning, which was important for that special design. And then, executed accordingly, it shall be right and that makes it good.

Hard to measure all typefaces with one rule. A brilliant text face is difficult to compare to an excellent display face.
Features and range of weights are much overrated, i.m.h.o., whereas drawing and spacing are far too often underestimated by fontists.
Idea and concept I would rather leave aside here because that you can’t measure objectively.

I would distinguish essential requirements from extras and individual aspects.

1. Essentials
• Quality of glyph outlines (drawing)
• Quality of spacing (width and sidebearings)
• Quality of kerning
• Character coverage

2. Extras
• Font family concept (Italics, weights, widths, others)
• Languages supported (apart from the usual suspects)
• Figure sets and fractions
• Ligatures
• Variant glyphs and alternate sets
• symbols and ornaments

In my opinion, a basically well crafted single font with 400 glyphs and no feature extras is worth more than a feature- and alternates-packed superfont of 2000 glyphs which are lousily drawn and poorly spaced.

The aspect of character coverage is an interesting one. It always evokes a ‘search for completeness’ but in practice this is hardly to achieve. Who is going to tell you what goes in and what does not? Nobody. Out there is no reliable common standard which tells you that. Even well-known references (e.g. the Adobe glyph sets) are not entirely reliable in that respect.

Christian Thalmann said:

… many, if not most, of the Dresden realizations out there range from suboptimal to downright ugly,

Sorry, but this is downright nonsense.