[Oslc-pm] should ems:observes be required for PerformanceMonitoringRecords

[John Arwe]

> ... and other 
> resources with ems:observes predicates and no rdf:type 
> PerformanceMonitoringRecord. 


That is absolutely encouraged.  As originally promulgated, the EMS spec 
had (+continues to have) resource definitions where ems:observes was used 
in exactly this way.  If EMS had added the equivalent constraint to what 
the excerpt above might suggest is needed, Perf Mon would have been UNable 
to re-use the ems vocabulary.

This is RDF where to the best of our ability we try to give a concept one 
identifier in a vocabulary (ems:observes means the object contains some 
kind of observed information) and *re-use* that same vocabulary in as many 
contexts as it is relevant to.  Each specification is one such context, 
and may layer on additional constraints (like occurrences) appropriate to 
the scenarios that comprise the context.  Hence you "always" have at least 
one spec for a vocabulary term (that is our minimalist practice, not a 
Requirement From Above).  It does not imply that vocabulary == 
specification/domain, however.

If you want a system where wide, serendipitous, uncoordinated vocabulary 
re-use is discouraged, we have 20 years of OO systems waiting to be used 
and lots of people happy to use them.  It's just a different world-view, 
each appropriate to some set of problems and constraints.  Don't get 
tangled up in that thinking in RDF, those constraints are not there by 
default and they generally inhibit re-use.

Best Regards, John

Voice US 845-435-9470  BluePages
Tivoli OSLC Lead - Show me the Scenario

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://open-services.net/pipermail/oslc-pm_open-services.net/attachments/20121024/ece12321/attachment-0003.html>