RE: [soa-rm-ra] ancillary changes comment

[Peter F Brown <peter@peterfbrown.com>]

Hi Jeff, thanks for the effort in going through this, much appreciated,
particularly as this particular version is something of a mongrel - it shows
all the insignificant edits but, for various reasons and track change
settings, shows some others in a half-cocked manner. So kudos for making the
attempt! Some comments inline below... I've marked with ** those where I
feel we need the TC's input...

All the best,
Peter

| -----Original Message-----
| From: Estefan, Jeff A (3100) [mailto:jeffrey.a.estefan@jpl.nasa.gov]
| Sent: Tuesday, 04 January 2011 22:22
| To: soa-rm-ra@lists.oasis-open.org
| Subject: [soa-rm-ra] ancillary changes comment
| 
| Feedback on the Ancillary Changes Document (focus on cover pages, general
| issues, and Sects 1, 2, & 3 only).  I have not completed reviewing the
Main
| Changes Document.
| 
| General:
| 
| (1) Discussion about list of editors.  Current callout absolutely
unacceptable.
**[Peter] Agree - my faux pas. Intended initially with a big question mark
asking the same questions as you do. Didn't get reviewed before going out
the door. The TC needs to agree this.

| Reshuffle names based on authorship content versus alphabetical surname if
| necessary.  And what about Rex and David, both key contributors past 4+
years.
| In other words, what should be reserved for page 1 versus Acknowledgements
| section.
**[Peter] I'm looking for guidance on this. Editors are normally the editors
at the time the spec is put to bed, no more no less. All contributors are
normally considered (wrongly, IMO and for exactly the reasons you raise) as
the full members of the TC through its lifetime, whether someone has
actually done any lifting or not. Only TC members belonging to sponsor-level
OASIS member organisations get any mention in any press release or other
promotional materials.

| (2) Footer "soa-ra-cd-XX" needs to change to "soa-raf-cd-XX"
[Peter] OK, thx

| (3) First use of definition called out in bold blue text started with
Frank's Section
| 3 updates.  What does this mean for the rest of the document?
[Peter] For now, I propose removing all the blue highlighting/hyperlinking
and bolding that seems to have been applied to nearly every appearance of
keywords. We can then add coherent formatting later, e.g. bolding (first
reference(s) to a defined term) and hyperlinking (only when a term is used
before it is defined and then only on the first occurrence). For the moment,
it is all clearer without...

| (4) Adding index OK.  Adding glossary is sacrilege.
[Peter] ;-) As explained at the TC last meeting, that was an editorial
exercise to clarify definitions - we will replace that with a simple index
with references to first mention of term and reference to its formal
definition.

| (5) Need to update Acknowledgements section (e.g., Peter and Chris not
listed).
**[Peter] as point (1) above. Only one personal observation: I've done more
editing on this spec so far than I think I did on the whole RM ;-)
| 
| Section 1:
| 
| Line 156 (following). Recommend adding the figure from the white paper
like
| our counterparts at TOG.  Just because it reflects "OASIS SOA-RA" versus
"-RAF,"
| we can state the update/correction in the title caption.   The figure
helps paint a
| very important visual in terms of what is more abstract and foundational.
[Peter] Good idea. I'm guessing the copyright issue doesn't arise as OASIS
is a joint rights holder
| 
| Lines 257-259:  Drop words related to "recommendation of a UML Profile for
| SOA" as this language predated the development and release of OMG SoaML.
| We do not need to mention this and the text is out of date.  Simply need
to state
| that every attempt is made to use normative UML 2.
[Peter] OK, done
| 
| Lines 299-381:  For some reason, these Architectural Implications got
moved to
| Sect 1 and they should be at the end of Sect 3.  Incidentally, we DO want
to keep
| these Arch Implications.  They should not be dropped.
[Peter] They are back as discussed at the last TC. That seems to have been a
transient tracking failure. Normal service resumed! ;-)
| 
| Line 382:  I do not like the suggested names for the viewpoints, i.e.,
Ecosystem,
| System, and Process.  None of these naturally fits the 1471/42010
conceptual
| model.  Ecosystem, as Peter mentioned before, has a relationship of
| Stakeholders, Concerns, and System.  System, of course, is System.
Process is a
| potential viewpoint name but I don't think it fits.  Again, I recommend
the
| following (only one change from the original set):  Participating in
Service
| Oriented Architectures, Realizing Services Oriented Architectures, and
Owning
| Services Oriented Architectures.
[Peter] How would you feel about 'Participation View', 'Realisation View'
and 'Ownership View' ? I'm with you on the need for a coherent (and
relevant) name set but I like a bit more concision. Not a blocking issue for
me, as long as we have coherence...
 
 What DOES need update in this table are the
| Stakeholders addressed.  Right now it is either too general ("All
stakeholders in
| the ecosystem") or too vague ("Standards Architects, Enterprise
Architects, etc.).
| This is a topic that needs further attention from the subcommittee and I'd
like to
| hear some thoughts.
[Peter] Agree. Let's hope we're not listening to the sound of silence...
| 
| Section 2:
| 
| Line 645:  Dropping "Separation" seems suspect to me.  Right now, we are
left
| with "of Concerns".  Further, the text in the Statement section would need
| update.  I believe what is capture in the Main Changes Doc changing
| "Separation" to "Distinction" so that the principle becomes Distinction of
| Concerns is more correct; however, the concept as Boris pointed out is
more
| foreign to architects than Separation of Concerns.
[Peter] I have been inadvertently the source of a potential new problem here
- my concern (no pun intended) was that, at least with the first viewpoint,
we were decidedly *not* separating stakeholder concerns, as they are many
and varied in this VP but all rolled up together in a single view. I accept
that it may be a step too far and have noted that we need to maybe change it
back but maybe modify the text so that we understand 'separation' as
including distinguishing that different stakeholders have different concerns
even if represented in the same VP.
| 
| 
| Section 3:
| 
| Line 683:  Do not like "Ecosystem" in a view/viewpoint name.  Recommend
| Participating in Service Oriented Architectures View.  (Similar reference
| throughout.)
[Peter] OK
| 
| Lines 713-714:  Recommend dropping this figure.
[Peter] Yup, this is out already in the main changes doc.
| 
| Line 737:  Update figure to reflect new name of model(s).
[Peter] Yup, already captured, thx
| 
| Lines 755-756:  Many what seems to be missing words in the sentences of
these
| paragraphs.  For example, what comes after "together for a ." ?  Also,
don't like
| starting a sentence with I.e., T.
[Peter] Again transient tracking problem.
| 
| OK.  I'm stopping there for Section 3.  It's a mess from a Word point of
view.
| Something is hosed up.  I'll turn my attention to the Main Changes Doc and
see if
| the Word issues have been resolved.  Will comment on that revision.
[Peter] I understand. I had hoped that generating a diff between the
original 17 Nov draft and the 'non substantial' editorial changes would help
highlight the minor changes but it is very difficult to read, I agree, as
lots of transient noise is included too....
| 
| Regards...
| 
|  - Jeff
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| ---------------------------------------------------------------------
| 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