OASIS Web Services Calendar (WS-Calendar) TC

We have identical “Interval structures” in two places in Availability. By Interval Structure, I mean dtStart and Duration and Time Zone.     Notice we have a Time Range in the overall VAvailability typem as well as an Availability Interval within each Availability component.   1)       Should these be collapsed into a common Type? 2)       If so, should they share the same name 3)       Should this observation be shared back with Cyrus (I think not, as it is less jarring in RELAX)     “The single biggest problem in communication is the illusion that it has taken place.” – George Bernard Shaw. Toby Considine TC9, Inc. Toby.Considine@gmail.com Phone: (919)619-2104 http://www.tc9.com    Chair, OASIS OBIX Technical Committee Chair, OASIS WS-Calendar Technical Committee Editor, OASIS Energy Market Information Exchange (EMIX) Editor, OASIS Energy Interoperation blog: http://www.NewDaedalus.com      

Toby - If I have your comment correctly, there are two different intervals in Availability. This is because there are different constraints in the respective uses per https://www.ietf.org/id/draft-ietf-calext-availability-00.pdf . The structure is the same only if you ignore cardinality and the semantics that drive cardinality. Looking at the cardinality for the attributes/elements: AvailabilityIntervalType:  ALL of the attributes are optional 0..1 - used for expressing availability within the TimeRange. How about never - is never OK for you? is a valid AvailabilityInterval. TimeRangeType:  dtStart is mandatory; duration and tZ optional - the scope of the availability (a bounding interval )  From this day forth is a valid TimeRange. So these are for different purposes and have different semantics. This might have been modeled as a simple IntervalType with differing constraints, but the constraints (a) don't show in the UML diagrams clearly and (b) they have different semantic purposes.  Thanks! bill -- William Cox   Email: wtcox@CoxSoftwareArchitects.com   Web: http://www.CoxSoftwareArchitects.com   +1 862 485 3696 mobile On 7/22/15 3:32 PM, Toby Considine wrote: We have identical “Interval structures” in two places in Availability. By Interval Structure, I mean dtStart and Duration and Time Zone.     Notice we have a Time Range in the overall VAvailability typem as well as an Availability Interval within each Availability component.   1)       Should these be collapsed into a common Type? 2)       If so, should they share the same name 3)       Should this observation be shared back with Cyrus (I think not, as it is less jarring in RELAX)     “The single biggest problem in communication is the illusion that it has taken place.” – George Bernard Shaw. Toby Considine TC9, Inc. Toby.Considine@gmail.com Phone: (919)619-2104 http://www.tc9.com    Chair, OASIS OBIX Technical Committee Chair, OASIS WS-Calendar Technical Committee Editor, OASIS Energy Market Information Exchange (EMIX) Editor, OASIS Energy Interoperation blog: http://www.NewDaedalus.com      

Toby - In your diagram, the type should be AvailableIntervalType not AvailabilityIntervalType . Cardinality in that type should be duration optional [0..1]; the others seem OK. bill -- William Cox   Email: wtcox@CoxSoftwareArchitects.com   Web: http://www.CoxSoftwareArchitects.com   +1 862 485 3696 mobile On 7/31/15 10:33 AM, William Cox wrote: Toby - If I have your comment correctly, there are two different intervals in Availability. This is because there are different constraints in the respective uses per https://www.ietf.org/id/draft-ietf-calext-availability-00.pdf . The structure is the same only if you ignore cardinality and the semantics that drive cardinality. Looking at the cardinality for the attributes/elements: AvailabilityIntervalType:  ALL of the attributes are optional 0..1 - used for expressing availability within the TimeRange. How about never - is never OK for you? is a valid AvailabilityInterval. TimeRangeType:  dtStart is mandatory; duration and tZ optional - the scope of the availability (a bounding interval )  From this day forth is a valid TimeRange. So these are for different purposes and have different semantics. This might have been modeled as a simple IntervalType with differing constraints, but the constraints (a) don't show in the UML diagrams clearly and (b) they have different semantic purposes.  Thanks! bill -- William Cox   Email: wtcox@CoxSoftwareArchitects.com   Web: http://www.CoxSoftwareArchitects.com   +1 862 485 3696 mobile On 7/22/15 3:32 PM, Toby Considine wrote: We have identical “Interval structures” in two places in Availability. By Interval Structure, I mean dtStart and Duration and Time Zone.     Notice we have a Time Range in the overall VAvailability typem as well as an Availability Interval within each Availability component.   1)       Should these be collapsed into a common Type? 2)       If so, should they share the same name 3)       Should this observation be shared back with Cyrus (I think not, as it is less jarring in RELAX)     “The single biggest problem in communication is the illusion that it has taken place.” – George Bernard Shaw. Toby Considine TC9, Inc. Toby.Considine@gmail.com Phone: (919)619-2104 http://www.tc9.com    Chair, OASIS OBIX Technical Committee Chair, OASIS WS-Calendar Technical Committee Editor, OASIS Energy Market Information Exchange (EMIX) Editor, OASIS Energy Interoperation blog: http://www.NewDaedalus.com