Streams namespaces

[William Cox <wtcox@CoxSoftwareArchitects.com>]

Calling out the namespace question separately (from line 72 below in detailed comments):

It's clear that ...energyinterop is the wrong namespace.

Should this be in the WS-Calendar/xCal URN space? In the OASIS namespace? Both?

Resolving URNs continues to be a black art; resolving URIs in URL form (http scheme) is easier, and makes it easier to find the namespace description in OASIS' administrative approach.

So it needs to change.

To what?

Thanks!

bill
--

William Cox
Email: wtcox@CoxSoftwareArchitects.com
Web: http://www.CoxSoftwareArchitects.com
+1 862 485 3696 mobile
+1 908 277 3460 fax

On 5/16/13 4:17 PM, William Cox wrote:

Toby --

So the real question is what in the PIM is the base class for any of the "new" classes. Is this a PSM? Should I look at the XSD?

Comments referenced to wd03 pdf:

• The use of StreamBaseType is not clear to me. How does this relate to PIM and WS-Calendar?
• Copyright should be 2012-2013...but TCAdmin will do that as needed.
• Bookmark not defined in the TOC
• No brackets around line 2
• line 4 : "will benefit" should be "benefit"?
• Not a biggie, but I don't hyphenate "time series"
• line 72 Why is namespace ...ns/energyinterop ? shouldn't it be in ws-calendar? or one of the URNs? see strm on line 78.
  
• line 78 Is WSDL important?
• line 109 - mixing quote of WS-Calendar? WS-Calendar as quoted by PIM? I think they're the same. Should the reference on line 115 be to WS-Calendar or PIM?
• line 116 is this section normative?
• line 128 caption uses "WS-Calendar Partition" but it's not in the definitions in the previous section. It's in WS-Calendar at 202 and PIM at 84. I assume that since it's key to line 161, that it should be defined in the tables not just in an example caption. So insert definition in table 2-1.
• Line 114 table does not say non-normative but line 143 does.
• 214 "Market Signal Payload"? - change to "Payload" assuming properly defined. Yes it is in table 3-1.
  
• 215 is the order required to reflect temporal order, e.g. sorting by start times? If a partition that seems implicit but could be explicit.
• line 109: I think the model needs to be different; in PIM the "related-to" is not "RelatedToPropType", it's RelationLink[Type]. Is that a suggestion to change the PIM type to RelationLinkType in parallel with GluonType?  Good suggestion.
• Not sure how the ArrayOf types reflect in PIM - it's more an XSD concept as I recall. We need to talk.
• StreamPayloadBase is not in the document text; viz. figure at 199.
• l203 seems a carryover from WS-Calendar Streams, rather than PIM-based. What is "conformant communications"? Sounds like a carryover from WS-Calendar as WS message content?
• 210 MAY NOT could probably be MUST NOT?  or is the intent MAY but SHOULDN"T? In which case SHOULD NOT is appropriate. MAY NOT is not in RFC2119.
• PIM DOESN'T HAVE UIDs - PUT IN NON-EXISTENT JIRA COMPONENT
  
• PIM UIDs (if included) may simplify section 3.x
• 257 didn't we have confidence, or did we put it into the payload?
• I'd like a nice compact example of observational data in l226ff. Possible non-normative appendix?
• l247 what is the "service Context"? is that line 173? Is it defined?
  

In general seems much clearer and cleaner.

Thanks!

bill

--
William Cox
Email: wtcox@CoxSoftwareArchitects.com
Web: http://www.CoxSoftwareArchitects.com
+1 862 485 3696 mobile
+1 908 277 3460 fax