[topicmapmail] TMQL Draft Posted - Comments Requested - Notice of Plans for FCD Ballot

[Patrick Durusau]

Greetings!

Thanks for the comments!

I am sure they will be appreciated by the editors as well as the 
eventual users of TMQL.

Hope you are having a great day!

Patrick

Xuân Baldauf wrote:
> Patrick Durusau wrote:
>> Greetings!
>>
>> The latest draft of TMQL has been posted!
>>
>> For your convenience: http://www.jtc1sc34.org/repository/0887.pdf
>>
>> Please review this draft throughly and post both editorial as well as 
>> technical comments.
> I have read the draft up to page 10 so far and I have some mostly 
> editorial IMHO-comments:
>
>    1. Unfortunately, the document is not searchable (searching simple
>       words using Adobe Reader yields no result), while it should be.
>       There is also no possibility to do text-retaining copy&paste on
>       some text parts of the document. (It would be nice if the HTML
>       source, http://topicmaps.bond.edu.au/junk/tmql.html being the last
>    2. Is it by design that, while Unicode is listed in the normative
>       references, nothing is said about what actually is the character
>       set of the query language? (What is the set of available
>       characters? Is it this defined by ASCII, ISO 8859-1, UCS-2,
>       Unicode?) This question, for example, is of relevance to
>       production [12], which excludes characters belonging to
>       character class "[#x00-#x20]" to be contained by an IRI. This
>       definition is commented by "IRIs are strings not containing
>       [...] as well as no non-printable character.". However
>       "non-printable" does not seem to be defined for Unicode.
>       Although, there are characters in the interval from codepoint
>       U+0000 up to, but not including codepoint U+0020 belonging to
>       the "Other, Control" Unicode General Category, however,
>       characters in the interval from codepoint U+007F up to, but not
>       including codepoint U+00A0 also belong to this Unicode General
>       Category. Now, are these characters between U+007F and U+00A0
>       considered as "non-printable character" as well?
>    3. Is it by design that nothing is said about the mapping of
>       characters of the query language to bytes? (E.g. if queries are
>       stored in files or transmitted over a network, should software,
>       without further metadata, be able to assume a particular
>       encoding, like UTF-8?)
>    4. On page 4, "xsd::string" should be "xsd:string".
>    5. On page 4, said production [12] seems to contain a regular
>       expression, but this apparent regular expression is not enclosed
>       in slashes.
>       Also, this regular expression makes use of apparent character
>       references like "#x00". This way of referencing characters does
>       not seem to be defined in normative reference RegExp ("The Open
>       Group Base Specifications Issue 6"). Also, the usual semantics
>       of a character class expression is inclusive-inclusive, meaning
>       that character class "[#x00-#x20]" represents 33 characters,
>       including the space character. While this character is
>       graphically represented, indeed, as space, it is typically not
>       considered to belong to the class of "non-printable" characters.
>       Thus, disallowing space to be a part of an IRI should be made
>       more explicit.
>    6. On page 6, "within the it" makes no sense.
>    7. On page 6 regarding "they cannot be names as these cannot
>       contain URLs": why not?
>    8. On page 8, sed -e "s/occurrence a topic/occurrences of a topic/"
>
>>
>> I would like for an FCD ballot on TMQL to issue following the Kyoto 
>> meeting of WG 3 in December, 2007.
>>
>> Note that this means we need to make every effort to make sure the 
>> text agreed upon in Kyoto is the one we want as the final text.
>>
>> Thanks to both Robert and Lars for what looks like a very complete 
>> TMQL draft!
>>
>> I hope everyone is at the start of a great week!
>>
>> Patrick
>>
> ciao,
> Xuân.
>

-- 
Patrick Durusau
patrick at durusau.net
Chair, V1 - US TAG to JTC 1/SC 34
Acting Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)
Co-Editor, OpenDocument Format (OASIS, ISO/IEC 26300)