Re: Representing Unix filenames in Unicode

From: Doug Ewell (
Date: Mon Nov 28 2005 - 08:30:09 CST

  • Next message: Neil Harris: "Re: Representing Unix filenames in Unicode"

    Hans Aberg <haberg at math dot su dot se> wrote:

    >>> Most facts points to that the Unicode/10646 is a human interface,
    >>> not a computer to computer to computer interface.
    >> I wonder what on earth that means.
    > Do you mean the typo (the erroneously extra "to computer")?

    No, I mean saying that "Unicode/10646 is a human interface." It's a
    coded character set.

    > Otherwise, it just means that computers works best with handling low
    > structured binary data, whereas for humans it is the opposite,
    > usually handling thinking in terms of high level structures, not even
    > binary in nature. If computers should just communicate among
    > themselves, as in done in the low level on a filesystem, they do not
    > have any benefit of knowing that the filenames can be given human
    > interpretation, as is done with the Unicode/10646 character set.
    > Right?

    Computers are great at byte-level operations, but they also do a fine
    job of taking the bytes 0xE2 and 0x98 and 0xBA and converting them to a
    single value integer 0x263A. They're also great at using that value to
    look up an entry in a table of properties, or mapping it to a glyph in a
    font. Interpreting that glyph as a smiling face is a job best left to
    humans. This is *no different* for Unicode than for any other coded
    character set. Calling Unicode a "human interface" and not a "computer
    interface" doesn't shed any light on this for me.

    Doug Ewell
    Fullerton, California, USA

    This archive was generated by hypermail 2.1.5 : Mon Nov 28 2005 - 18:21:05 CST