[sc34wg3] CTM for non-SVO languages

Lars Heuer heuer at semagia.com
Mon Feb 4 11:54:08 EST 2008

Hi all,

From my understanding of Prof. Lee's explanations, I believe that the
Kyoto proposals try to solve homemade problems; problems which do not
exist in the current CTM draft. Maybe I may be mistaken here, but let
me explain my reasons for the conclusion:

1. Kyoto removes a delimiter for type/value pairs.

   The current draft uses a colon as type/value delimiter and if I
   remember correctly, in Leipzig 2006 was a request from a non-SVO
   participant to *keep* a type/value delimiter because it would be
   easier to read.

   To make CTM readable again (especially for non-SVO languages),
   Kyoto proposes a semicolon to mark the end of a statement.
   Otherwise the type/value entities would just 'hang around' without
   an indication which one is used as type and which is meant as

2. Kyoto reintroduces template invocations without parenthesis.

             - "John Lennon";
             born-in Liverpool;
             died-in New-York;
             homepage <http://lennon.com/>;

   If I understood Prof. Lee correctly, this may cause confusion,
   especially, if someone uses nouns:

             - "John Lennon";
             Birth-Place Liverpool;
             Place-of-Death New-York;
             homepage <http://lennon.com/>;

   Again, the semicolon is *needed* to parse (either from an human or
   a machine) the statements. That was previously not necessary, since
   the current draft uses parenthesis to make template invocations
   distinct from occurrences and names. And the assignment of a value
   to a type is made clear through a colon as type/value delimiter.

           - "John Lennon"                # Name
           Birth-Place(Liverpool)         # Template invocation
           Place-of-Death(New-York)       # Template invocation
           homepage: <http://lennon.com/> # Occurrence

Again, this is analysis is not meant to find pro/against semicolon
arguments but to understand the problem for non-SVO languages better.
Maybe some non-SVO language writer can comment on this?

Best regards,

More information about the sc34wg3 mailing list