Re: Run-time checking of fonts for Sinhala support

From: Roozbeh Pournader (roozbeh@htpassport.com)
Date: Mon Sep 28 2009 - 20:34:06 CDT

  • Next message: Jeff Senn: "Re: NFC/NFKC Normalization Edge Case"

    On Mon, 2009-09-21 at 02:21 +1000, Harshula wrote:
    > a) 'Operating systems shall only recognise Level 1 compliant fonts as
    > Sinhala fonts.'
    >
    > Taking into account the concerns that were raised, I've revised it to:
    >
    > b) 'Operating systems shall, to the extent of their capabilities, only
    > recognise Level 1 or above compliant fonts as Sinhala fonts.'
    >
    > Is there a clear preference for replacing "shall" with "should". i.e.
    > Changing it to a recommendation?:
    >
    > c) 'Operating systems should, to the extent of their capabilities, only
    > recognise Level 1 or above compliant fonts as Sinhala fonts.'
    >
    > Please indicate your preference. Don't expect me to extrapolate from
    > your previous comments.

    None of the option are good enough for me. I think such language would
    be forcing the hand of implementers.

    But if SLS 1134 authors insist on having such language in the standard
    in a newer version, I would suggest using "may". That would mean that
    according to the standard, if an operating system applies such a rule
    for filtering fonts, it will remain conforming to the standard.

    Roozbeh



    This archive was generated by hypermail 2.1.5 : Mon Sep 28 2009 - 20:38:47 CDT