Re: Latin-script keyboard layout (was RE: Quick Question About Korean Input Methods)

From: Kent Karlsson (
Date: Mon Jan 04 2010 - 14:18:15 CST

  • Next message: Michael Everson: "Re: Latin-script keyboard layout (was RE: Quick Question About Korean Input Methods)"

    Den 2010-01-04 00.28, skrev "Karl Pentzlin" <>:

    > Am Sonntag, 3. Januar 2010 um 22:23 schrieb Kent Karlsson:
    >>> The revision of ISO/IEC 9995-3 ...
    >>> "4 shift states" is not compatible with the group/level concept of
    > KK> I'm surprised that the new draft standard does not have 4 levels but
    > KK> sticks with the outdated 3 levels the old version had...
    > Please do not mix "shift states" with "levels", although these
    > concepts seem interchangeable as long as no more than 3 levels are
    > to be considered.


    > "Level" roughly resemble rows on a keytop while "groups" roughly
    > resemble columns.

    I have no idea what you are trying to say here. (Maybe your keyboard
    looks different from those I have used...)

    > E.g., the new Finnish standard keyboard (SFS 5966, see:
    > ),
    > has 4 shift states (unshifted, shifted, AltGr unshifted, AltGr shifted),
    > which in ISO/IEC 9995 terms are 2 groups with 2 levels each.

    I.e. four levels. Dividing it into two groups of two levels is just
    a play with words.

    > (btw, this is something the Microsoft Keyboard Layout Creator V1.4
    > can handle, while afaik it cannot handle the applying of multiple
    > diacritics for Vietnamese characters.)
    > In fact, ISO/IEC 9995 allows, as an exception, the Level 3 characters

    If you were serious about "two groups with two levels each", then please
    remove the notion of a third level. Or just merge group 1 and group 2
    into plainly four levels. (Of course leaving the layout to national
    standards; as well as vendors.)

    > to be displayed on the right lower corner on the keytop when there is
    > only one group. This is due to historical reasons, like for historical
    > reasons there are precomposed characters in Unicode.
    > KK> And (still, though changed) the "common second group". Thanks, but no
    > KK> thanks.
    > Of course, nobody excepts everybody to like everything; nevertheless I
    > miss any argument in your statement.

    A "second group" (or third group, or whatever) with a jumble of characters
    with no mnemonic connection to what is on the (group 1) level 1 and level 2
    (for each keyboard layout) is not helpful. For rarely used characters where
    one has no mnemonic connection to the (group 1) levels 1/2 layout, then it
    is probably better to use a character picker applet of some sort, rather
    than this "common second group". Note that most keytops only get a single
    character printed on it, in some cases two characters. I don't expect that
    to change much (for aesthetic reasons, leaving open for keyboard layout
    changes, ...).

        /kent k

    > - Karl

    This archive was generated by hypermail 2.1.5 : Mon Jan 04 2010 - 14:23:46 CST