If you know special local kerning cases, post them here please

the "Periodcentered" thread got me thinking: are there special cases in world orthographies, like kerning the periodcentrered exactly between two l's, that are specific to a country's orthography but should be paid special attention to when developing a professional font? There are perhaps similar cases in other orthographies that non-local designers are totally oblivious to, and it would be good to hear more about these cases. :)


  • Kent LewKent Lew Posts: 747
    Kerning periodcentered between two l’s in default kerning could backfire in a non-Catalan context where the periodcentered is being used to mark syllabification or some similar purpose.

    In most font editors, managing this as a language-specific kern is not easily handled with native tools either (as discussed in that other thread).

    You could try to limit the possibility of undesirable results within default kerning by contextually kerning the entire sequence /l/periodcentered/l as a triplet. But such contextual or triplet kerns are also not easily managed with the tools in most font editors (even though the spec allows for such things).

    So, I think most type designers these days tend to use the periodcentered.locl_CAT strategy to leverage a language-specific GSUB rather than a GPOS approach.

    This solution is nicely outlined in a Glyphs tutorial on the subject of Catalan ela geminada. (But the basic solution is easily implemented in other font editors as well, though you have to adjust for the Glyphs-specific naming used in the tutorial.)

  • Nick ShinnNick Shinn Posts: 1,279
    c_h and c_k in German. ;)
  • Finnish has loads of odd Vy, Ky and Kv combos, like Vyötärönympärys, Kyläyhdistys or Kvanttimekaniikka. And maybe less critical for most kerning there is also uncommon double consonants like yy and kk, like Syyskukka.
    Not a kern, and maybe a question of taste, but also the .fi domain for all Finnish websites sometimes looks over-important when the default ligature snaps in at the URL end, like www.foobar.fi - I just think it is something not many designers consider with regards to the fi ligature.

    Also any Hawaian text is a good test not to miss in regard to single quote spacing.

    Curious to hear about more examples from your corner of the world :)

  • John HudsonJohn Hudson Posts: 1,398
    Are you asking about potential kerning pairs that occur in individual languages that don't occur in e.g. English, as in Johannes' Finnish examples? Or are you asking about combinations for which special kerning is appropriate for an individual language but not appropriate for others? Nick's observation re. traditional German treatment of ch and ck would be an example of the latter: a treatment that differs from how the same letter sequences would be handled in other languages.
  • Ray LarabieRay Larabie Posts: 757
    The L’ in French where little to no kerning is preferred but in English kerning is preferred. L’orange vs. HAL’S.
  • Alex VisiAlex Visi Posts: 42
    c_h and c_k in German. ;)

  • John HudsonJohn Hudson Posts: 1,398
    ch and ck are digraphs in German (i.e. pairs of letters that represent an individual phoneme, pronunciation of which in German is determined by the preceding vowel). In blackletter orthography, these digraphs were regularly represented by touching letters, as shown in this example from Klingspor's Tiemann-Fraktur:

    These combinations would typically be implemented as ligatures. When post-war Germany largely switched to antikva (roman) types, the ch and ck digraphs were sometimes represented by closely spaced letters or as ligatures. My impression is that this is seldom done any more, and I think Nick's suggestion of making this a special case for German kerning isn't completely serious.

    See also this earlier Typedrawers discussion.
  • Nick ShinnNick Shinn Posts: 1,279
    edited July 2
    In the immediate post-war (II) era, when German foundries issued a lot more “antiqua” styles (moving away from Fraktur with its negative associations), many would have a c_k and c_h “logotype” (to use the Linotype term) for these combinations, which are extremely common in German; clearly, it helped the efficiency of compositors, despite being an archaic concept inherited from fraktur, and not really suitable to roman type. But as metal declined, so did the practice.

    I’ve offered this quaint kerning as a Stylistic Set option on a couple of typefaces—along with lowered diereses on Ä, Ö and Ü—that have a kind of mid-century German antiqua vibe—which happened under the influence of Hermann Zapf, I suspect, as I’ve always admired his work immensely, especially Palatino.

    So yes, as indicated by the winky-face, my tongue firmly in cheek!
Sign In or Register to comment.