Re: IDN Security

From: Cary Karp (ck@nic.museum)
Date: Wed Feb 16 2005 - 04:42:57 CST

  • Next message: Werner LEMBERG: "Re: 03F3 j Greek Letter yot"

    Simon Montagu wrote:

    > GERESH seems an even more likely candidate than GERSHAYIM to
    > me, because it's not uncommon in proper names. There is a
    > tendency for the GERSHAYIM to fall out of use as abbreviations
    > become more commonly accepted as words in their own right,
    > e.g. תפו״ז, the Hebrew for orange, which is originally an
    > abbreviation for תפוח זהב but today almost always written
    > without GERSHAYIM.
    >
    > GERESH on the other hand, when used as a modifier for letters
    > like ג׳ ז׳ or צ׳, would cause a change in pronounciation if it
    > were omitted, so I would expect www.חבד.org.il to be a more
    > acceptable substitute for www.חב״ד.org.il than
    > www.זבוטינסקי.org for www.ז׳בוטינסקי.org

    The key IDN policy consideration is whether or not to permit the
    registration of characters that are listed as punctuation in the
    Unicode code charts. If one or both of GERESH and GERSHAYIM are
    clearly justifiable, they can be placed in a Hebrew inclusion
    table with equal ease.

    The interesting twist here is that a language community might
    wish to populate its IDN namespace with characters that cannot be
    entered from the basic local keyboard. This may be innocuous
    enough but, as I understand it, most users wouldn't understand
    why keyboarded single and double quotation marks were not being
    accepted as GERESH and GERSHAYIM. Is the resultant confusion a
    reasonable price to pay for the orthographic flexibility?

    My interest in this is not purely academic. I am responsible for
    establishing the IDN policies for the .museum TLD. Although
    Hebrew is not one of the languages that it fully supports,
    production policies are being developed and the registration of a
    few names containing Hebrew labels has already been accepted. The
    issue with GERSHAYIM became apparent with the acronym used by the
    Israeli National Committee of the International Council of
    Museums, http://איקו״ם.ישראל.museum. (This name is provisional
    pending the outcome of the present discussion and it will also
    resolve without the GERSHAYIM. Substituting a quotation mark
    will, however, not work -- the name server balks.)

    IDN support for a language in .museum is normally based on the
    corresponding policies adopted by a country TLD that has an
    obvious association with that language and can be expected to
    understand the detail and nuance of its orthographic practice.
    (Reconciling variant practices where there are several candidate
    ccTLDs is a concern of its own.) In the case of Hebrew, such
    guidance would likely only be forthcoming from the operator of
    .il. IDN support has, however, not yet been initiated in that domain.

    The participating museums are fully able to contribute requisite
    familiarity with Hebrew, but the situation described here is
    beset with residual uncertainty. It thus seems reasonable to
    elicit the further guidance of the members of this list. I would
    also be very interested in knowing if similar situations are
    likely to appear in other languages. (I had initially imagined
    supporting the Hebrew alphabet, save for BiDi concerns, to be a
    very straightforward matter.)

    /Cary



    This archive was generated by hypermail 2.1.5 : Wed Feb 16 2005 - 04:41:41 CST