[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: syntaxes-10 notes


Hallvard,

Hallvard B Furuseth wrote: 
Syntaxes-10 has this new text in

3.3.13. Generalized Time 


 The above ABNF allows character strings which do not represent valid
 dates (in the Gregorian calendar) and/or valid times (e.g., February
 31, 1994).  Such character strings SHOULD be considered invalid for
 this syntax.



I just realized that dates predating the Gregorian calendar are
not Gregorian, so the above can be read as implying pre-Gregorian
dates SHOULD be considered invalid.  Was that intentional?


Yes. I expect GeneralizedTime was only ever intended to represent modern dates
in protocols, rather than to suit the needs of historians and archeoastronomers.
It seems rather odd to me to use a format with embedded time zone information
to represent the dates and times of events occurring before the introduction,
in the nineteenth century, of standardized time and time zones.

> And if 
not, is it worth worrying about?  If so it might be simplest to
just drop "(in the Gregorian calendar)" after all.


4.1. General Considerations 


 Servers MAY implement the wordMatch and keywordMatch matching rules,
 but SHOULD implement the other matching rules in Section 4.2.


I would s/but/and/. 

Okay.



Appendix B. Changes from RFC 2252 


The appendix should mention that uniqueMemberMatch now differs
significantly from the X.500 variant.


I'll add a suitable note to Appendix B.

Regards,
Steven