On Jul 7, 2010, at 9:14 AM, Estefan, Jeff A (3100) wrote:
Frank,
Read through your updated Section 3 and, overall, I though the updates looks very solid. Couple of items of note below that we can discuss on the next call and as we begin to tackle the issues list items for this section. These line numbers are based on the soa-ra-pr-6-29-10.docx revision.
1. Line 629. Fig 3 needs to be updated to reflect two models now instead of the old three models
In fact, several of the diagrams need updating. I have tried to leave a comment on the diagrams that need to be changed.
2. Line 712. Sentence that begins “However, we do not need to delve..” is overly complex and recommend dropping.
Agreed.
3. Line 936. Proposition is in bold text but is not formally defined. Needs formal definition.
Agreed.
4. Line 1043. For consistency, change “SOA-RM” to “Reference Model”
Agreed.
5. Line 1069. For consistency, change “eco-system” to “ecosystem”
Agreed.
6. Line 1116. Should “goals” be “objectives” here since you’ve now introduced the notion of objectives and the concept of goals is not introduced until later in sub-section 3.2.2?
The distinction between goals and objectives can be hard to make. As you noted, goals seems to carry a longer-term aspect to it than objective.
So, on balance, I would say something like:
goals are states that the actor wishes to achieve
an objective is a specific state that an actor wishes a particular action to achieve
Something like: I would like enough light to read by. By flipping the switch, the lamp should be turned on.
So, yes, the diagram needs adjustment.
7. Line 1134. Similarly, should Goal in Fig 15 now be Objective?
Yes.
8. Line 1134. An “Event” does not report on “Effect” but rather Event Notifications do. However, we cannot change this to Event Notification in the figure because this would violate the Actor to Event named association of “may be aware of”. This needs resolution. One way or another, we have to fix the relationships here. More about this later.
I agree with this. As I recall, there was a certain amount of sweat in this definition.
Personally, I think an event is an occurrence that at least one actor has an interest in. Many people link events to state, but I do not think that that is right: a clock tick is an event, but is not itself a change of state.
9. Line 1141. Recommend adding “measurable” between “is a” and “real world effect”. It is usual convention that objectives are measurable. I know later you introduce goals and have goals as measurable but this is not traditional convention in which goals are typically long term statements of intent that may or may not be achieved and objectives are measurable steps toward trying to achieve goals. We’ll need some consistency here.
not opposed to this.
10. Line 1145. Again, shouldn’t “goals” be “objectives” here? Perhaps a holdover from the earlier document.
No. Based on the analysis above, an actor is still interested in its goals, which are parsed into pieces as objectives linked to actions.
11. Line 1164. Is this consistent with the RM definition of Real World Effect? On lines 1158-59 you introduce a new concept of “Effect” and here differentiate it from Real World Effect. We’ll need to discuss this on a future call.
I believe it is. It is a restatement certainly. But I tried to keep the essence of the concept.
12. Line 1169. Add “notification of” in between “by means of” and “events”
Agreed
13. Lines 1170-71. This is not a correct definition of Event. An event is a notable thing that happens. Recommend introducing a formal definition of Event earlier (before Line 1169) and changing line 1170 to “Event Notification” and keeping the definition you have on lines 1165-66 to formally define Event Notification.
Agreed. I have adjusted it to read:
An event
is an occurrence of which at least one participant
has an interest in being aware
Clumsy english but closer to the spirit
14. Line 1178. Change “an event that reports” to “a notification of an event that reports”
Agreed. Have added this definition
Event
Notification
An event
notification is the action of an actor
becoming aware of an event.
15. Line 1189. Why is the third occurrence of joint action in this sentence shown in bold blue text?
This is an experiment. It signifies a hyper-link.
16. Lines 1193-95. This definition of Choreography is different than what we just agreed on for Section 4.3.4. We need to be consistent or more precise. This could mean we change the 4.3.4 term to “process choreography” or a similar variant, in which case, that might ripple into having us change orchestration to “process orchestration” but I’m not too keen on that idea. Another topic for discussion.
We should talk:)
17. Line 1232. The English in this sentence does not make sense to me.
Suggest simply deleting the sentence.
18. Lines 1235-39. Here you use “measurable” and that should be associated with the earlier introduction of “objectives”. Again, a topic for further discussion.
Goals are measurable too. This is key.
19. Line 1454. Add, “i.e., delegates” after “automated actors”
Not opposed to this. But not all automated actors are delegates, and not all delegates are automated.
20. Lines 1516-1544. I’m not convinced on this notion of “service action”. This section will require further discussion.
Well, not wedded to this. But it seemed natural to distinguish communicative actions -- which are kind of overhead/bureaucracy -- with service -- which are what the actors are really trying to do.
21. SIDEBAR: I thought the willingness, trust, and risk discussion was very good.
22. Finally, I see you are using the convention of bold type for keywords in the RAF. This means we’ll have to do the same in the other sections; however, there are several instances even in the new Sect 3 (that I’ve noted on my hardcopy) that are not consistent in its use. Kind of a pain in the neck actually. So we should talk about a convention we can all live with on a future call.
We did discuss this a few times. The problem with capitalization is that it is pretty hidden and gets confusing when the defined concept is the first word of a sentence.
Again, overall, I thought the updated write-up looked pretty solid.
Thank you for taking the time to read it so carefully!