DAL sessions in Victoria

Matthew Graham mjg at cacr.caltech.edu
Wed Apr 26 08:23:24 PDT 2006


Hi,

Rob Seaman wrote:

>> Does this imply that a single ISO8601 argument can handle the  
>> selection for
>> your use-case above?
>
>
> If the particular usage might benefits from not-fully-general  
> cadences, e.g., minutes or round fractions, but not a cadence of 17.3s.
>
> Was proffering that particular use case as an example.  Such esoteric  
> searches will become more-and-more important as data pile up.  LSST  
> will offer 10^3 visits over ten years for 3 pi steradians.  There are  
> a lot of spatial and temporal baselines one might choose to select  
> out of all that.  Searching on solar system resonances is another  
> such possibility.  More generally, we could benefit from some way to  
> perform searches over physically defined quantities, not just  
> phenomenologically.

There are already datatypes in XML Schema that support both instance 
(dateTime) and duration (duration) that support ISO8601 format and allow 
arbitrary precision in sec. so you could easily have two arguments: 
startTime and duration which would cover most use cases.

    Cheers,

    Matthew



More information about the dal mailing list