Re: Directionality Standard

From: Behnam (
Date: Tue Dec 18 2007 - 03:42:11 CST

  • Next message: Jeroen Ruigrok van der Werven: "Re: Directionality Standard"

    Yes indeed. My mistake. I'm not much into the technical side of this
    issue. I'm a simple user.
    But clearly, a simple right click and assigning rtl directionality
    for typing a paragraph (or any action of that nature) is no guarantee
    for that paragraph to go anywhere with its directionality intact. And
    this is what I put forward for Unicode to address.
    Let me tell you what triggered this thread.
    We have a Persian site with news and discussion forum. It provides
    RSS feed. But RSS readers were displaying the content in ltr format
    and totally unusable. The person who is more technical and takes care
    of this site finally found this
    and implemented language tag 'fa' to our site and our problem with
    RSS reader in IE7 was resolved. But the problem in Safari of Mac
    remains. I suppose if Safari implements the recognition of those
    tags, the problem would be resolved with Safari also because
    otherwise it is a perfectly Unicode compliant application. But I am
    puzzled why a directionality issue should be resolved by a language
    tag in the first place, and if this is not a Unicode issue, how much
    longer and how many times, are we supposed to address these problems
    and solve them, using made-up solutions, one device at a time?
    To me, this is a Unicode issue because I type an rtl paragraph and I
    expect it to remain rtl, anywhere it goes.


    On 18-Dec-07, at 3:50 AM, Richard Ishida wrote:

    >> From:
    >> [] On Behalf Of Behnam
    >> Sent: 18 December 2007 02:34
    > ...
    >> And why do we need language tag in HTML or other formats if
    >> this property is defined within the encoded paragraph?
    > As Jony said, you need direction attributes, not language tags. Wrt
    > why those are needed in HTML:
    > [1] 'paragraphs' are defined by markup in HTML, since whitespace
    > (including newlines etc) in the source is normalized for display
    > [2] attribute values in HTML affect more than a 'paragraph' at a
    > time - a single dir="rtl" at the top of a document may be all you
    > need for the whole document, and this will affect things like
    > ordering of table columns as well as stuff internal to paragraphs.
    > [3] the bidi algorithm is often insufficient to resolve potential
    > ambiguities in embedded mixed direction text, and markup can be
    > used to set the needed directional context for the embedded text
    > (see )
    > hth
    > RI
    > ============
    > Richard Ishida
    > Internationalization Lead
    > W3C (World Wide Web Consortium)

    This archive was generated by hypermail 2.1.5 : Tue Dec 18 2007 - 03:44:30 CST