RE: [energyinterop] Error and short-fuse comments

Emix interface is already part of the resource. There are gml elements associated with Service Location. Wouldn’t multiple occurrences  of Resource Description type accomplish the same thing? That’s related to my earlier question on why we would define gml elements in both location and resource.

Bruce Bartell
Xtensible Solutions
Mobile
:  +1.321.258.6500


    www.xtensible.net
From:
William Cox  [mailto:]
Sent:
Tuesday, November 08, 2011 5:46 PM
To:
Girish Ghatikar ; 
Cc:
Bartell, Bruce ; Joshua Phillips;  William Cox
Subject:
Re: [energyinterop] Error and short-fuse comments

After several emails, we need to go with 3. (the thread has split a few times).  See earlier messages for details. This is the minimal perturbation, and  keeps exactly the types as we've had them while working around an XML restriction.
This means that there is an emixInterface -- so location: emixInterface will be just an emixInterface.
Current cardinality is [1..1] BTW.
Bruce, does this work for you without manual resolution?
Thanks!
bill
On 11/8/11 3:42 PM,  Girish Ghatikar  wrote:
In the interest of time and ease of adoption, I prefer using GML directly.
-Rish
On Tue, Nov 8, 2011 at 12:06 PM,  Bartell, Bruce  < 
> wrote:
Same issue we ran into. Circuit location is what we are really after.

Bruce Bartell
Xtensible Solutions
Mobile
:
+1.321.258.6500



www.xtensible.net
From:
  [mailto:  ]
On Behalf Of
Joshua Phillips
Sent:
Tuesday, November 08, 2011 10:40 AM
To:
William Cox ;

Subject:
RE: [energyinterop] Error and short-fuse comments

I believe 2 or 3 will be the best solution.  I don’t believe the  IRC is using GML to define a resources location.  It’s a system location, so geographic information is not accurate enough.  A resource could be miles away from its actual connection point to the grid, but closer or in another geographic connection point to  the grid.
From:
  [mailto:  ]
On Behalf Of
William Cox
Sent:
Tuesday, November 08, 2011 9:18 AM
To:

Subject:
Re: [energyinterop] Error and short-fuse comments

Numbers refer to the *3*  routes, numbered 1 through 4 below...
I think 1 and 4 aren't candidates. They share the same problem - they focus on only the geospatial aspect of the interface. This loses the capabilities which we spent a great deal of time on, including eliminating things like service delivery point, Pnode,  Apnode, etc.
The point is that emixInterface generalizes the notion of "location". And that that generalization is important.
Of 2 and 3, 2 is the simplest workaround keeping "location", and 3 is my preference.
Why would we have a "location" for BOTH the ResourceType (inherits from EnrolleeType)

AND in the contained emixResource? So another choice might be to give up the word "location" and do 3 as it directly allows any emix interface.
Thanks!
bill

--
William Cox
Email:

Web:
http://www.CoxSoftwareArchitects.com
+1 862 485 3696  mobile
+1 908 277 3460  fax
On 11/8/11 9:33 AM, Gray, Gerald wrote:
My preference is to reference GML directly.

Good development of the pros/cons Toby, however, on balance I come  down on the side of referencing GML directly.

Referencing GML directly, if it diverges, is a single order of divergence.
Referencing GML via EMIX, if there is divergence, potentially creates  two orders of divergence.
From:
  [ mailto: ]
On Behalf Of
Toby Considine
Sent:
Tuesday, November 08, 2011 9:29 AM
To:
' Bartell, Bruce ';

Subject:
[energyinterop] Error and short-fuse comments

Ok, Bruce has zoomed in on an unexpected feature of XML. This feature  breaks the Resource location. See below for thread.

The abstract type is inherited properly. It breaks the substitution  group rules, though. To recap:

<type name =”fooType” abstract=”true”>
<element name=”foo” type=”fooType” >
<type name =”fooType” abstract=”true”>
<element name=”notFoo” type=”fooType” >

<xs:element name="fooFighter" type="emix:fooFighterType" substitutionGroup="foo"/>
<type name =”fooFighterType” >
                <xs:extension base="fooType"/>
</type>
<xs:element name="fooLover" type=" fooLoverType " substitutionGroup="foo"/>
<type name =”fooLoverType” >
                <xs:extension base="fooType"/>
</type>

fooFIghter and fooLover are substitutes for foo
fooFIghter and fooLover are NOT substitutes for notFoo

We have three routes we can go.
(1)

WE can make location a name (or reference) to an existing concrete type,
(2)

we can make location a container for an EMix Interface
(3)

We can simply stick an EMix Interface in there can get rid of the word Location.
(4)

We can lose the entire EMIX stuff and refer to the GML directly

Discussion:
1) Location is all that matters, we don’t care about any  of the other possible meaning of interface, we want a geographic location only. The counter-argument is that what matters may well be “what substation is this resource attaching to” which is covered  by the interface but not by the location

<element name="location"
type="emix:ServiceAreaType"/>

2) If we just put an Emix Interface there, no one will understand  it. The word and concept Location is important, especially since a Resource, could, in concept, be a collection of [devices] each with its own EMIX Interface.

<element name="location"
type="locationType>
<type
="locationType>
<element ref=
"emix:emixInterface"/>
</type>

3) Emix defined it. Add an annotation and be done with it. THe other  interfaces (see 2) will be embedded inside emix:resources anyway, so why worry about it.

<element name="emixInterface" type="emix:EmixInterfaceType"/>

4)

We can lose the entire EMIX stuff and refer to the GML directly. It is just confusing to reference that EMIX stuff. There would be  no issue if we went directly to the GML. Well, we would have to include the gml profile defintions stuff, and it could, in concept, diverge from the normalization of GML performed in EMIX. The down-side is that if a future version of emix defines standard  gml profiles for, say “within 50 feet of a Transmission line” or “distribution zone associated with a sub-station” we will not inherit them.

<element name="location" type="gml:AbstractFeature"/>

Please discuss and vote

tc

"A designer knows he has achieved perfection not when there is nothing  left to add, but when there is nothing left to take away."
-  Antoine de
Saint-Exupery
Toby Considine  TC9, Inc
TC Chair: oBIX & WS-Calendar
TC Editor: EMIX, EnergyInterop
U.S. National Inst. of Standards and Tech. Smart Grid Architecture Committee
Email:

Phone:
(919)619-2104
http://www.tcnine.com/
blog:
www.NewDaedalus.com
From:
Toby Considine [ mailto: ]
Sent:
Tuesday, November 08, 2011 8:48 AM
To:
' Bartell, Bruce '; '  '
Subject:
RE: [energyinterop] [OASIS Issue Tracker] Commented: (ENERGYINTEROP-474) eiEnroll Resource Not in WD25

OK, I see the symptom in xmlspy schema view.

Not sure why, though. Let me poke around further.

tc

"A designer knows he has achieved perfection not when there is nothing  left to add, but when there is nothing left to take away."
-  Antoine de
Saint-Exupery
Toby Considine  TC9, Inc
TC Chair: oBIX & WS-Calendar
TC Editor: EMIX, EnergyInterop
U.S. National Inst. of Standards and Tech. Smart Grid Architecture Committee
Email:

Phone:
(919)619-2104
http://www.tcnine.com/
blog:
www.NewDaedalus.com
From:
Considine, Toby (Campus Services IT)

Sent:
Tuesday, November 08, 2011 8:45 AM
To:
' Bartell, Bruce '; OASIS Issues Tracker;

Subject:
RE: [energyinterop] [OASIS Issue Tracker] Commented: (ENERGYINTEROP-474) eiEnroll Resource Not in WD25

Not sure what is going on there. I will wiggle the parts and see  if something squeaks.

EMixInterface is, of course, abstract. It relies either on inclusion  of gml schemas (for spatial references) or EMIX Power schemas (for node/meter/etc references). Any chance those are not being resolved properly?

tc

"
It  is the theory that decides what can be observed
."   - Albert Einstein
Toby Considine
Chair, OASIS oBIX Technical Committee
U.S. National Inst. of Standards and Tech. Smart Grid Architecture Committee
Facilities Technology Office
University  of
North Carolina
Chapel Hill ,
NC
Email:
Toby.Considine@ unc.edu
Phone:
(919)962-9073
http://www.oasis-open.org
blog:  www.NewDaedalus.com
From:
  [ mailto: ]
On Behalf Of
Bartell, Bruce
Sent:
Tuesday, November 08, 2011 8:13 AM
To:
OASIS Issues Tracker;

Subject:
RE: [energyinterop] [OASIS Issue Tracker] Commented: (ENERGYINTEROP-474) eiEnroll Resource Not in WD25

The type shows up in the latest version, but I have to resolve it manually.

Bruce Bartell
Xtensible Solutions
Mobile
:
+1.321.258.6500
        www.xtensible.net