Search the FAQ Archives

3 - A - B - C - D - E - F - G - H - I - J - K - L - M
N - O - P - Q - R - S - T - U - V - W - X - Y - Z
faqs.org - Internet FAQ Archives

Nordic FAQ - 1 of 7 - INTRODUCTION
Section - 1.8 What are Nordic graphemes?

( Single Page )
[ Usenet FAQs | Web FAQs | Documents | RFC Index | Sex offenders ]


Top Document: Nordic FAQ - 1 of 7 - INTRODUCTION
Previous Document: 1.7 Which are the related newsgroups?
Next Document: 1.9 About measures and figures
See reader questions & answers on this topic! - Help others by sharing your knowledge
   (by Tor Slettnes)
   
   Nordic graphemes can in this context be described as "graphical
   representations of the letters that exist in the various Nordic
   alphabets, beyond those that exist in the English alphabet". Each of
   the Nordic written languages uses some additional letters compared to
   English. These are, in order of appearance in the alphabets:
   
    Letter:        Languages used:      Pronounced like:  character:
    ________________________________________________________________

    a acute        is                   'ou' in "loud"          
    eth            is                   'th' in "there"         
    e acute        is (dk, no, se, fi)  'ea' in "yeah"          
    i acute        is                   'e'  in "he"            
    o acute        is                   'o'  in "home"          
    u acute        is                   'ou' in "you"           
    y acute        is                   'e'  in "he"            
    thorn          is                   'th' in "thumb"         
    ae             is                   'i'  in "hi"            
                   dk, no               'a'  in "bad"           
    o-slash        dk, no               'i'  in "bird"          
    a-ring         dk, no, se (fi)      'o'  in "bored"         
    a diaeresis    se, fi               'a'  in "bad"           
    o diaeresis    se, fi, is           'i'  in "bird"          
    u diaeresis    (se, fi, dk, no)     'ue' in french "rue"    

   A set of parentheses around the country code indicates that the letter
   is rarely used in the corresponding language, typically only for loan
   words or names originating from another language. Other accents, such
   as ^ (circumflex) and accent grave are now and then used in foreign
   names and words in all Nordic languages.
   
   In Denmark and Norway the alphabet is ordered:
a b c d e f g h i j k l m n o p q r s t u v w x y z   

   For Finland and Sweden the order is:
a b c d e f g h i j k l m n o p q r s t u v w x y z   

   If your curiosity isn't satisfied by the pronounciation guide above,
   there are more extensive comments in the various language sections of
   this faq.
   
   
   
  1.8.1 How are these represented in Usenet postings and E-mail?
  
   The "mother" of all modern character sets for computers is the
   original ASCII character set, now renamed to US-ASCII. (ASCII =
   "American Standard Code for Information Interchange"). This is a 7-bit
   set containing the characters needed to write American English without
   accents or special letters, and little more. No "foreign" letters are
   included.
   
   Various standards exist for representing extra characters, some of
   which are: Digraph, LaTeX, ISO-646, ISO-8859-1, and the IBM codepages
   437, 850, and 865. All of these sets, except the IBM codepages, are
   usually considered acceptable on soc.culture.nordic, e-mail, and the
   internet in general.
   
   Digraphs are two-character combinations used for simplicity, and are
   often the most universally understood notation on soc .culture
   .nordic. However, when using these to non-Nordics, one should be
   careful to explain that these are digraphs, not two separate
   characters. Also, some information may get lost by using digraphs,
   since a filtering program will not be able to determine whether it is
   really a digraph or two separate characters.
   
   LaTeX notation comes from the typesetting program by the same name,
   where a sequence starting with '\' may be substituted with a given
   character. For instance, the a-ring is written as "\aa" or "{\aa}" in
   LaTeX.
   
   ISO-646 (really ISO-646-NO and ISO-646-SE) are 7-bit sets similar to
   US-ASCII, but with national characters substituted in place of the
   following characters: {, |, }, [, \, ]. This is the oldest one of the
   "true representation" standards mentioned here; it was used in e.g.
   the Nordic versions of the CP/M operating system, prior to MS-DOS.
   Today, it is mostly used in Sweden and Finland (although the ordering
   of the letters, for the sake of compability with the Danish /Norwegian
   /German equivalents, are not correct in these languages).
   
   ISO-8859-1, also called ISO Latin-1, is the first of several 8-bit
   character sets described in International Standards Organization's
   document 8859. (ISO is the maintainer of the meter, the kilogram,
   etcetera.) This sets include all characters needed for all West
   European languages, leave Smi and Esperanto. Latin-1 is a superset of
   US-ASCII, hence all ASCII characters maintain their original position
   in this set. Rather than trying to accomodate positioning in any
   spesific language, the letters in ISO-8859-1 are ordered according to
   the alphabetical position of their US-ASCII lookalikes. Latin-1 is
   supported through modern standardizations like MIME (RFC 1521).
   
   The IBM codepages 437, 850, 861 and 865 are used on Personal Computers
   in "text" mode, and is also the default set on many MS-Windows 
   communication programs. Out of the Big Blue, they were created to
   provide text-based PC programs with a means to create low-cost
   graphics, and the addition of extra characters came as a nice side
   effect. (Certain Nordic characters were not represented in the
   original codepage 437, with the consequence that in Iceland, Denmark
   and Norway, computers would occasionally be sold with cp 861 or 865 in
   the hardware. Today, alternative codepages can be downloaded to the
   video card via software). The Danish /Norwegian character o-slash is
   not represented in cp 437, and in 850 /861 /865 it is positioned with
   the dangerous code 155 (9B hex) -- "Upper Escape". Certain terminal
   types will interpret this code as the initial character of a escape
   command, and may e.g. clear the screen depending on the next letter.
   Further, it is incompatible with the established 8-bit standard
   Latin-1, and should be avoided.
   
   The various notations of the Nordic graphemes follow:
    Letter   Digraph   LaTeX ISO-646   ISO-8859-1
                                                   HTML          Octal  Char
    _________________________________  _____________________________________
                                                
    a acute       A'    \'{A}     -    alt-0193   Á Á  \301   
                  a'    \'{a}     -    alt-0225   á á  \341   
    eth           TH              -    alt-0208   Ð Ð     \320   
                  th              -    alt-0240   ð ð     \360   
    e acute       E'    \'{E}     -    alt-0201   É É  \311   
                  e'    \'{e}     -    alt-0233   é é  \351   
    i acute       I'    \'{I}     -    alt-0205   Í Í  \315   
                  i'    \'{i}     -    alt-0237   í í  \355   
    o acute       O'    \'{O}     -    alt-0211   Ó Ó  \323   
                  o'    \'{o}     -    alt-0243   ó ó  \363   
    u acute       U'    \'{U}     -    alt-0218   Ú Ú  \332   
                  u'    \'{u}     -    alt-0250   ú ú  \372   
    y acute       Y'    \'{Y}     -    alt-0221   Ý Ý  \335   
                  y'    \'{y}     -    alt-0253   ý ý  \375   
    thorn         TH              -    alt-0222   Þ &THORN ;  \336   
                  th              -    alt-0254   þ þ   \376   

    u diaeresis   U"    \"{U}    ^     alt-0220   Ü Ü    \334   
                  u"    \"{u}    ~     alt-0252   ü ü    \374   
    ae            AE    {\AE}    [     alt-0198   Æ Æ   \306   
                  ae    {\ae}    {     alt-0230   æ æ   \346   
    o-slash       OE    {\OE}    \     alt-0216   Ø Ø  \330   
                  oe    {\oe}    |     alt-0248   ø ø  \370   
    a-ring        AA    {\AA}    ]     alt-0197   Å Å   \305   
                  aa    {\aa}    }     alt-0229   å å   \345   
    a diaeresis   A"    \"{A}    [     alt-0196   Ä Ä    \304   
                  a"    \"{a}    {     alt-0228   ä ä    \344   
    o diaeresis   O"    \"{O}    \     alt-0214   Ö Ö    \326   
                  o"    \"{o}    |     alt-0246   ö ö    \366   

   The ISO-646 charsets for Denmark/Norway and Finland/Sweden are in
   practice obsolete, and there never existed one for Icelandic, but you
   may run into older 7-bits text files using them. It is to be noted
   that  is not represented in iso-646-NO for Denmark/Norway.
   
   
   
  1.8.2 Pros and cons of the different representations
  
   If you have been a reader of this group for a while, you may have
   noticed that discussion about characters and their representations
   occasionally accounts for quite a bit of bandwidth. It often does not
   take more than a question about the issue from a new reader, or
   someone posting an article with an IBM character set, to get a new
   thread going on the issue. Some want to keep 7-bit ISO-646 (be aware
   that they may call it "true ASCII", although strictly speaking, is
   not), since 7-bit codes will always get though with any setup; others
   want ISO-Latin-1 since it is more universal; and yet others promote
   digraphs as the greatest common denominator between the two.
   
   Some pros and cons for each set:
    Character set:    Advantages:             Disadvantages:
    __________________________________________________________________

    Digraphs          * Requires 7-bit only   * Ambiguous
                                                ("oe" or "o-slash"?)
                                              * Non-optimal compromise

    LaTeX             * Non-ambiguous 7-bit   * Made for typesetting;
                        representation.         somewhat cryptic for
                                                regular text.
                                              * Non-optimal compromise

    ISO-646-SE,       * Only 7-bit "true"     * Different standards
    ISO-646-DK          representation.         for each language
    <[\]{|}>          * No data loss even     * Getting harder to
                        with old hardware/      find font support
                        software/setup.         (Dying out).
                                              * Shadows the brace,
                                                sqare bracket, pipe,
                                                and backslash chars.

    ISO Latin 1       * Utilizes all 8 bits   * Requires 8-bit clean
    (ISO-8859-1)        in a byte; yet avoids   connection; older
    <..>  dangerous codes.        systems may cause
                      * Universal for all       data loss.
                        Western European      * May require some
                        languages.              setup.
                      * Supported by ISO and  * In case of stripping,
                        MIME; true subset of    becomes "FXEDVfxedv";
                        Unicode.                difficult to read.

    IBM CodePages     * Uses all 256 codes;   * Uses all 256 codes;
    Machintosh set      more characters         incl. dangerous ones.
    <Unacceptable>    * Often used in PC      * Incompatible with
                        environments such as    the "de-facto" 8-bit
                        BBS'es.                 standard ISO-8859-1

    __________________________________________________________________

   
   
  1.8.3 How do I set up support for 7-bit ISO-646 representation?
  ({|}, [\])
  
   The ISO-646 sets are still supported via varoius fonts and translation
   filters. Possible measures to set up support for them are:
     * For the "terminal" program shipped with Windows 3.x, simply select
       "Denmark/Norway", "Sweden" or "Finland" from the Translations item
       in the "Terminal Preferences" dialogue box.
     * For MS-Kermit, use the command "set term charcter-set language",
       where "language" is one of "Finnish", "Swedish", or "Norwegian".
     * For other DOS and Windows communication programs, visit its local
       translation tables and insert appropriate translations for '[',
       '\', ']', '{', '|', '}'.
     * For Unix based news readers, either find a ISO-646 font, or pipe
       your newsreader through one of the following commands (Provided
       the font you use is ISO-8859-1):
       
        
                Denmark/Norway: tr '\\]{|}' '\330\305\346\370\345'
                Sweden/Finland: tr '\\]{|}' '\326\305\344\366\345'
                
       For instance, in your .cshrc file, insert the following line:
       
        
                alias rn "rn | tr '\\]{|}' '\330\305\346\370\345'"
                
   The character '[' should not be translated, because it is used in ANSI
   escape sequences.
   
   Note that if you use this kind of translation, you will no longer see
   any of the characters '\]{|}'; in most cases this outweighs the
   benefits from seeing the national letters.
   
  1.8.4 How do I set up support for 8-bit ISO-8859-1 representation? (,
  )
  
   The ISO-8859-1 (Latin 1) set is currently the most common character
   representation standard on soc.culture.nordic, and is also quite
   frequent in e.g. soc.culture.german, personal e-mail etc. However, on
   many systems, the ability to view these characters is not provided as
   "default", so you may need to configure some things on your own.
   
     * If you are reading news through a modem, you need to make sure
       that your modem connection is 8 data bits. (The most common
       parameters are "8N1" - 8 data bits, no parity bits, and one stop
       bit).
     * For DOS text mode communication programs, you need a ISO->IBM
       translation table. Tables for Telemate, Telix and Procomm Plus can
       be found in the file "xlate.zip", available at various FTP sites.
     * For MS Windows  communication programs, select an ANSI or
       ISO-Latin-1 font. For MS-Kermit, use "set term char latin". For
       Procomm Plus for Windows, select vt220 or vt320 emulation. Be sure
       that bit 8 is not stripped.
     * For MS Windows  you can also generate 8-bit characters globally
       by choosing "US-International" keyboard layout via the
       "International" dialogue box in the Control Panel. For instance,
       '' (a diaeresis) is generated by pressing "a, i.e. double quote
       followed by lowercase a.
       A note to Windows programmers: Let the underlying keyboard
       drivers, run-time libararies etc. take care of keyboard input.
       Only be sure that the 8th bit is not stripped/masked away.
     * If your newsreader is UNIX-based, insert the following command in
       your .login or .profile file:
       
        
                stty -istrip pass8 
                
     * If your modem connection is 7 bits (and cannot be changed to 8
       bits), you can have ISO-Latin-1 characters translated to "[\]{|}"
       before they are sent over the modem. Pipe your reader through the
       "tr" command, similar to above:
       
        
                tr '\306\330\305\304\326\346\370\345\344\366'
                '[\\][\\{|}{|' 
                
     * If you use the "emacs" editor, version 19.x, and have a
       ISO-Latin-1 display font, insert the following line in your .emacs
       file:
       
        
                (standard-display-european t) 
                
       Also, if you have a keyboard with international characters that
       you want to be able to use directly, or if you in another way are
       able to generate 8-bit codes directly from your keyboard, insert
       the following line:

        (set-input-mode (car (current-input-mode))
                        (nth 1 (current-input-mode))
                        0)
   Note that in cases where the Meta key is represented by setting the
       8th (high) bit, (ie. if you are not using X-windows), this line
       will disable the Meta key, so you will subsequently have to use
       "ESC x" to generate "M-x".
       Otherwise, insert the following line:
       
        
                (load-library "iso-insert") 
                
       A new keymap, 8859-1, has now been assigned to the key sequence
       "C-x 8". You can assign this to another sequence, e.g. C-t, by
       inserting:
       
        
                (global-set-key "\C-t" 8859-1-map) 
                
       Some strokes from this map:
        C-x 8 d   gives  (eth)
        C-x 8 t   gives  (thorn)
        C-x 8 a e gives  (ae)
        C-x 8 / o gives  (o-slash)
        C-x 8 a a gives  (a-ring)
        C-x 8 " a gives  (a diaeresis)
        C-x 8 " o gives  (o diaeresis)
        C-x 8 ' a gives  (a acute)
        C-x 8 ' i gives  (i acute)
       
   
   
  1.8.5 References
  
   For an index to other literature on internationalization, try:
   <http://www.vlsivie.tuwien.ac.at/mike/i18n.html>
   
   I am: Tor Slettnes.
   

[ the sections above are available at the www-page
  http://www.lysator.liu.se/nordic/scn/faq18.html ]

   
   



User Contributions:

Comment about this article, ask questions, or add new information about this topic:

CAPTCHA




Top Document: Nordic FAQ - 1 of 7 - INTRODUCTION
Previous Document: 1.7 Which are the related newsgroups?
Next Document: 1.9 About measures and figures

Single Page

[ Usenet FAQs | Web FAQs | Documents | RFC Index ]

Send corrections/additions to the FAQ Maintainer:
jmo@lysator.liu.se (SCN Faq-maintainer)





Last Update March 27 2014 @ 02:11 PM