[sc34wg3] CTM: Semicolons

[Xuân Baldauf]

Jaeho Lee wrote:
> Hello,
>   
Hi. :-)
> I strongly support the idea of putting semicolons at the end of every
> "statement" (define if needed).
>
>   
>> Hi all,
>>
>> This spring in Oslo we introduced semicolons to delimit statements
>> within topic blocks [1] (for no technical reason but to enhance
>> readability for non-SVO languages).
>>     
>
> I think there were other technical reasons mentioned in Oslo more than just
> readability for non-SVO languages.
>   
Can you please elaborate on this point more in-depth? More specifically:

    * /What specificity of East Asian languages prevents readability of
      CTM documents without semicolons which is re-gained with semicolons?/

Or even more specifically: Given the fact that most CTM documents will
have one statement per line (because this is already custom in present
CTM documents as well as other languages to have one statement per line
and presumably because it allows faster visual scanning), the new-line
character actually serves as a de-facto semicolon-character. We could
even go that far and make the new-line character serve as a de-standard
semicolon-character. So, given this:

    * /What specificity of East Asian languages prevents readability of
      CTM documents with new-lines after every statement without
      semicolons which is re-gained with semicolons before the new-lines?/

I've been asking every person I could get hold of at the TMRA 2008 to
explain to me what the impact is for non-SVO languages (or East Asian
languages specifically) regarding semicolons. However, nobody could
explain it to me. I also asked Naito-san, and he also could not explain
it to me. He also thinks that in his opinion, semicolons are not needed
for Japanese speakers. This strikes me, as one of the most modern
programming languages (in terms of compactness and readability), namely
Ruby ( http://en.wikipedia.org/wiki/Ruby_(programming_language) ), is
invented by a Japanese native speaker, and this language avoids
semicolons where it can. So, apparently, there is no consensus about
non-SVO native speakers or just East Asian languages' speakers about the
necessity of semicolons. It is clear that CTM has a SVO-centric bias and
thus maybe a Western bias (and I'm happy to help fixing that). However,
it is unclear to me why semicolons could help improve readability for
East Asian languages' speakers. That's why I'm asking you.


I'll get to your other points as soon as I get your reply. :-)

ciao,
Xuân.


> One of them is easiness for the parser, and more importantly easiness for
> the generator of CTM statements.
> My many application programs will generate CTM statements routinely, and
> handling optional component is always headache.
> Furthermore readability and ugliness are so subjective issues. I see the
> beauty of consistent when there are always semicolons at the end of every
> statement.
>
>   


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.isotopicmaps.org/pipermail/sc34wg3/attachments/20081107/83b63d29/attachment.htm