Re: Unicode Stability

From: Peter Kirk (
Date: Wed Mar 02 2005 - 11:38:55 CST

  • Next message: Jony Rosenne: "RE: Unicode Stability"

    On 02/03/2005 14:56, Peter Constable wrote:

    > ...
    >>>The recently accepted proposals for Hebrew do just that, they change
    >>>meaning of existing data.
    >>That is true of any disunification.
    >No, Michael, don't take his bait. The proposals accepted for Hebrew
    >don't change the meaning of data. The only issue is that the qamats
    >qatan can't be implemented in existing or new fonts without awkwardness. ...

    This is true of QAMATS QATAN, but that is only one of the changes
    recently accepted for Hebrew, and not necessarily the one which Jony had
    in mind.

    >Jony's claim is simply incorrect: meaning of existing Hebrew data is not
    >changed by the recently-accepted Hebrew proposals. ...

    The accepted proposal for HOLAM HASER FOR VAV does change the meaning of
    existing data, from correctly spelled to a spelling error. It has the
    potential (although for Hebrew words rather than isolated fragments this
    is not realised in practice) of changing meaning by incorrectly
    resolving ambiguities, i.e. a form intended to represent meaning A but
    in fact ambiguous between meanings A and B could be incorrectly resolved
    to meaning B.

    I accept Michael's point that it is impossible to make a disunification
    without this danger by adding a new character - but it is possible if a
    variation sequence or similar is used. This is a very good argument for
    adding variation sequences rather than new characters for
    disunifications, for it preserves stability in the standard in the sense
    defined by Doug Ewell:

    >it does not change in a way that causes existing ... data to break.

    That is why Jony and I among others worked hard to find a solution to
    the Hebrew HOLAM problem without adding a new character. For example, we
    looked at solutions using variation sequences. But the one who "make[s]
    additions to Unicode all the time" strenuously opposed every solution
    which did not involve addition of a new character and forced through his
    own proposal for an addition, therefore creating the situation in which
    the meaning of existing data is changed.

    Peter Kirk (personal) (work)
    No virus found in this outgoing message.
    Checked by AVG Anti-Virus.
    Version: 7.0.308 / Virus Database: 266.5.7 - Release Date: 01/03/2005

    This archive was generated by hypermail 2.1.5 : Wed Mar 02 2005 - 11:40:14 CST