Re: help required

From: Otto Stolz (Otto.Stolz@uni-konstanz.de)
Date: Tue May 10 2005 - 03:22:45 CDT

  • Next message: Hans Aberg: "Re: Full Unicode Computer Keyboard"

    Helo Faraz Siddiqi,

    You had asked about an e-mail problem re transmission of Urdu characters.

    You had sent a line in Urdu which was sent/transmitted as garbage.
    Now, Chris Jacobs has resent your line thusly:
    > میرا نام فراز ھے اور میں نے ایم سی ایس کراچی یونیورسٹی سے کیا ھے
    which looks like genuine Urdu to me (but then I do not know anything
    about your language; I can only tell that I am seeing Arabic characters).
    You also should see that line correctly (depending on your e-mail client,
    cf. below).

    Chris Jacobs has given this explanation:
    > Use MIME headers to indicate which charset you use.

    As I deem this a rather cryptic remark, I am going to elaborate this for you.

    Chris's message contains these header lines:
    > MIME-Version: 1.0
    > Content-Type: text/plain;
    > charset="utf-8"
    > Content-Transfer-Encoding: 8bit

    This meta-information tells the reader how the bytes transmitted are to be
    interpreted, viz. as plain text, encoded in Unicode UTF-8, and with no
    additional transport-layer encoding (8 bit data per byte transmitted).
    Any decent e-mail client program will generate these header lines for
    outgoing messages (if asked to do so), and will interpret them for
    incoming ones (thus displaying the message alright). The technical
    details are specified in <http://www.zvon.org/tmRFC/RFC2045/Output/>,
    <http://www.zvon.org/tmRFC/RFC2046/Output/chapter4.html#sub1>,
    <http://www.zvon.org/tmRFC/RFC2047/Output/>,
    and <http://www.zvon.org/tmRFC/RFC2049/Output/>.

    If you, however, use a not-so-decent e-mail client, it will leave those
    header lines off. The receiving e-mail client will then assume that the
    message is encoded in ASCII (which uses only 7 bits per byte) and the
    (8-bit) bytes you have meant to represent Arabic characters were in error
    (impossible in ASCII). Some e-mail-clients will indeed mark these as
    erroneous (example: <http://www.systems.uni-konstanz.de/EMAIL/FAQ-SMTP.php#71>),
    others will guess (and usually wrongly so) on the meaning of them. This is
    what you have experienced.

    Hence, the most important step for you is to get you a decent e-mail
    client program, such as Mozilla Thunderbird, and configure it to use a
    suitable encoding for your outgoing messages (UTF-8 will always work).
    The receiver has to use a decent e-mail client, too; and he must have
    a suitable font installed on his system, capable of displaying the
    characters used for Urdu. In your case, a particular problem is the
    writing direction: Your e-mail client must be able to enter, and
    display right-to-left text.

    Best wishes,
       Otto Stolz



    This archive was generated by hypermail 2.1.5 : Tue May 10 2005 - 11:03:27 CDT