OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

soa-rm-ra message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: terms - how this works


I a close to Perter's take on 'management' and 'control'. However, management in SOA ecosystem is not a usual thing and, thus, requires to be addressed explicitly, especially, for cross-ownership cases where 'control as is' is rather questionable. So far, I am not sure I have addressed this topic as needed in the release-candidate of the Management Model.

- Michael


-----Original Message-----
From: Peter F Brown <peter@peterfbrown.com>
To: mpoulin@usa.com; soa-rm-ra@lists.oasis-open.org
Sent: Sun, Feb 27, 2011 7:29 pm
Subject: RE: terms - how this works

The ‘issue’ of actor and participant is a non-issue: it is a question of badly worded definitions using the words “is a” in too loose a manner. This will be addressed in the revisions we do.
The issue you have with management and control may be similar – but frankly for me, both concepts fall into the “do we really need to define terms for these?” box. I think every average Joe gets the meaning of management and of control, I’m not sure we need to define them differently here…
 
Peter
 
From: mpoulin@usa.com [mailto:mpoulin@usa.com]
Sent: Sunday, 27 February, 2011 10:59
To: peter@peterfbrown.com; soa-rm-ra@lists.oasis-open.org
Subject: terms - how this works
 
This is the question to all:
as I mentioned before, we have several definition that loop with each other. This subject was omitted but now I have a direct problem with definition of Management and Control
 
In the proposed version of Management Model, Management is defined as a 'process of controlling'; Rex proposed to define 'management' as 'control. However, the definition of 'control' in the version of 17 Jan says: "an account of how the management and governance of the entire SOA ecosystem can be arranged"
 
First, 'control' is NOT how "governance of the entire SOA ecosystem can be arranged", but how governance may be implemented ('arranged' may be read as 'create', which would be an incorrect interpretation)
 
Second, management => control => account how management is arranged. This means, we know neither about 'control' nor about 'management' (similarly, Actor refers to Participant and Participant refers to Actor, etc.)
 
Please, anybody, explain me how this works?!
 
- Michael
 
 
-----Original Message-----
From: Peter F Brown <peter@peterfbrown.com>
To: mpoulin@usa.com; soa-rm-ra@lists.oasis-open.org
Sent: Wed, Feb 23, 2011 3:38 pm
Subject: RE: [soa-rm-ra] RE: List of terms in section 3
Michael,
some quick responses to your points.
 
Capability – I see your point. If we accept that RWE is ‘subjective’ to the particular stakeholder(s) concerned, then actual RWE should be orthogonal to what (shared state changes) a capability actually delivers. A capability nonetheless delivers a RWE, whether it is intended/asked for/desired or not. A capability is used (= intention) in order “to realize one of more RWE’s” (from the RM). Subtle difference we may need to work on: I want to be careful that we don’t contradict the RM definition. I still don’t buy into your notion that somehow a service result can be slipped into the private state of an actor without it being part of the RWE.
Permission – fair point. We were trying (too hard) from a very early stage (<2008) to work with a duality of constraints, permission and obligation. I’d be happy to not formally define either.
RWE – still disagree with you! ;-) Your approach over-complicates the whole story, with private and public RWE’s, whatever they are (I’m still not clear) and, frankly, your diagram only makes the issue more complicated and does notadvance us to a clear definition. Our approach is to separate objective and subjective PoV’s: state changes are generally objective and measurable while an RWE is in the eye of the beholder – my RWE is not necessarily your RWE (your suntan, my sunburn – same sun, same exposure – sorry, but I’m just back off holidays), experienced mileage will vary. That is why we introduced the words ‘pertinent’, ‘relevant to’, ‘experienced by’ and ‘specific stakeholders’.
Shareable and Shared State – I never liked the introduction of shareable as a defined concept. I voted for and would again vote for keeping only private and shared state, stating that shared state is the part of the ‘shareable’ or public state that is actually (and not just potentially) shared. See our mail exchange of 31 Jan.
Service – at the meeting of 19 Jan, we agreed to go with a ‘definition’ (not formally defined, but used in the text) that states that a service is ‘a realisation of business functionality accessible through defined interfaces’
 
Regards,
Peter
 
 
From: mpoulin@usa.com [mailto:mpoulin@usa.com]
Sent: Wednesday, 23 February, 2011 05:47
To: soa-rm-ra@lists.oasis-open.org
Subject: Re: [soa-rm-ra] RE: List of terms in section 3
 
Folks,
 
I have very limited access to e-mail today. So, I hope to discuss my comments on the definition in the meeting today. Here are my belate comments :
 
Also, a few days ago I have sent a message to Ken with a small diagram that illustrates my proposal on the definition of RWE (that would include all chnages in the state of the ecosystem, not only public ones). You will see how this affects other definitions below.
 
DEFINITION
Term
Line in Model
Line where's defined
Definition
Comment
DEFINITION
Capability
1080
1107
A capability is an ability to achieve a real world effect.
Not only, capability is also to sutisfy the consumer needs that may be NOT in the RWE but in returned service result (in private state). The problem here is in the definition of RWE.
DEFINITION
Mediator
835
844
A mediator is a role assumed by a participant to facilitate interaction and connectivity in the offering and use of services.
Based on the Note 1 and definition of Participant, Service Registry/Repository or ESB are not mediators. It is OK with me.
DEFINITION
Permission
925
950
A permission is a constraint that prescribes actions that an actor may (or may not) perform and/or the states the actor may (or may not) be in.
The wording is strange IMO: permission does not prescribe, it permits/alowes
DEFINITION
Real World Effect
 
1181
A real world effect is a measurable change to the shared state of pertinent entities, relevant to and experienced by specific stakeholders of an ecosystem.
Still disagree with this. A consumer and its state are the part of the state of the SOA ecosystem regardeless who can/may access it. If private state of the consumer chnages, it is the change in the overal ecosystem state. If we do not go with this interpretetation, we will end-up with this: a cosumer of the service is interested, first of all, in the returned service result and only then may or may be not interested in the RWE. Examples: 1) I order a book because I am interested in having this book and do not care that somebody has a job to deliver this book to me; 2) I order a cleaning of the street in front of my house becuase I am interested in the clean road in front of my gates and in convenience of my neighbours who pass through this spot , i.e. I am interested in the returned results AND in the RWE avalibale/accessible to other potential consumenrs.
DEFINITION
Sharable State
 
 
That part of an entity’s state that is knowable to other actors
Proposal: That part of an entity’s state that MAY BE knowable to and may be accessible to, other actors, while the entity is not necessary aware of this knowlegde and access
DEFINITION
Shared State
 
907
Shared state is that part of an entity‘s state that is knowable by, and may be accessible to, other actors
Proposal: Shared state is that part of an entity‘s state that is knowable by, and may be accessible to, other actors, which is expected by the entity.
USE INFORMALLY
Service
835
 
 
IMO, we have to find a form of expression where we can refer to the RM definition and to append 'realisation of business functionality accessible though defined interfaces' simultanously making this a sort of refined definition.
 
 
- Michael
 
-----Original Message-----
From: Bashioum, Christopher D <cbashioum@mitre.org>
To: soa-rm-ra@lists.oasis-open.org <soa-rm-ra@lists.oasis-open.org>
Cc: Laskey, Ken <klaskey@mitre.org>
Sent: Wed, Feb 23, 2011 3:08 am
Subject: [soa-rm-ra] RE: List of terms in section 3
Ken,
 
request that we add an agenda item for tomorrow’s TC: the finalization of the list of terms for definition in Section 3.  Since sending out the email below, I have not received any input from the TC on amending the list of what will be formally defined, what will be dropped, and what will be used informally.  Would like to formally vote on this tomorrow so we can close it out.
 
This is important, as Peter and I would like to start the actual writing.  We have done some editing, but need to get a solid draft of Section 3 written up and delivered to the TC.  This is a necessary step for that to occur.
 
I will try to make the TC call tomorrow, but may miss it or be late due to work constraints.  Peter is planning on making the first part of the TC call, but will have to leave early for a client call, so we would like to have this as an early agenda item if possible.
 
Thanks!
 
From: Bashioum, Christopher D
Sent: Monday, February 07, 2011 10:32 AM
To: soa-rm-ra@lists.oasis-open.org
Subject: List of terms in section 3
Importance: High
 
At our last conf call, we proposed a list of terms that Section 3 will define.   Attached is a spreadsheet that contains that list.
 
The spreadsheet identifies terms that will be formally defined (identified with “DEFINITION” in the first column), terms that will not be used (DELETE) and terms that will *not* be formally defined but will be used and will rely on the reader’s understanding of the english language (USE INFORMALLY).   I have sorted the spreadsheet accordingly.
 
For the terms that will be formally defined, an initial definition is given in the  Definition column.  For those terms that will be either deleted or used informally, some amplifying text is in the Notes 1 and Notes 2 columns.
 
So ... first item is to come to agreement on what will be formally defined in section 3.  If you think that we missed some terms, or that we have included terms that should not be formally defined, please reply as such.  Would like to come to consensus on this by next conference call.  Definitions we will debate and refine after coming to agreement on the list of terms.
 
Michael P, your input has been incorporated in this list – except that i have interpreted your “delete” as “Don’t formally define, just use as normal english language in the text”.
 
=
---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php 


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]