Re: [soa-rm-ra] A thought for S3

[Francis McCabe <fmccabe@gmail.com>]

There seems to be some misunderestimating here ...

I am primarily talking about collecting existing text together, not writing a whole bunch of stuff.

A PhD in logic is always useful :), but I do not think that it is needed to follow what is going on. As I said last week, there are no symbols in the RAF. There is no strong mathematical rigor in our work. I do not suggest changing that, primarily because this is NOT a PhD thesis.

On the other hand, if you need to understand WHY we have some aspects of our architecture then you probably do need a triple Phd: in Logic, computer science and human psychology. This does not just apply to S# but the entire effort.

As for the RM, I was less than happy with what happened there; and I would argue that we are currently paying the price for those shortcuts.

Having said that, I am not completely convinced of the merit of a separate model for semantics. Semantics is pervasive. I propose it as a way of collecting together some underlying themes that keep surfacing.

Frank


On Oct 6, 2010, at 4:52 AM, Laskey, Ken wrote:

> Frank,
> 
> I have  a couple of concerns:
> 
> - Any material needs to be adequately explained.  Any reorganizing of S3 and additional material -- at least, connective material for shuffled pieces --  takes time and delays finishing.
> 
> - We spent an hour last week on a couple definitions as you explained how these fit with the foundations of Logic.  Most of our audience is unlikely to have that background and will not have the benefit of personal explanations.  Again, this points to the need for a significant level of explanation and the time to write it.
> 
> - Material that is written for S3 needs to be reviewed.  As you get into the technical beyond the background of many of us, I'm concerned we don't have the resources for adequate internal review.  More than that, this group will be the leading evangelists of this work and we have to be able to answer questions from our audience and ourselves be able to provide additional explanation.  I'm certainly concerned about my ability to do that and I assume that applies to others.
> 
> - A PhD in Logic requires a different level of ability to read and understand material than an undergraduate course.  The PhD material is certainly more accurate and more complete but the undergraduate text is sufficient for its purpose.  We need to identify the level for which this is being written and do a good job of writing to and being consistent with that level.
> 
> - We need to work Management.
> 
> Frank, we closed out the RM by having several hours of discussion, realizing there was no way to adequately get that discussion into the RM and we reduced several pages to several paragraphs.  Use of and reference to the RM by others would indicate we weren't perfect but did a pretty good job.  We need that for the RA.
> 
> Ken
> 
> ---------------------------------------------------------------------------
> Dr. Kenneth Laskey
> MITRE Corporation, M/S H305              phone: 703-983-7934
> 7515 Colshire Drive                             fax: 703-983-1379
> McLean VA 22102-7508
> ________________________________________
> From: Francis McCabe [fmccabe@gmail.com]
> Sent: Tuesday, October 05, 2010 1:23 PM
> To: soa-rm-ra@lists.oasis-open.org RA
> Cc: Danny Thornton
> Subject: [soa-rm-ra] A thought for S3
> 
> I have been ruminating a little on the issue with assertions, policies, goals, etc. etc. in S3.
> 
> I think that it may be worth factoring out the semantics aspects of the ecosystem view into a separate model. Thus, you would have three models in S3:
> 
> The Social Structure Model
> 
> Acting in a SOA Ecosystem Model
> 
> Semantics in a SOA Ecosystem Model
> 
> The last model would bring together the semantic aspects of the ecosystem:
> 
> assertions = proposition + 'force'
> 
> fact = proposition + belief
> 
> policies = proposition + enforcement
> 
> goal = proposition + desire
> 
> purpose = proposition + use/role
> 
> proposition = syntax + vocabulary + decision procedure
> 
> In addition, there would have to be a big message in the beginning of S3 (in bold flashing letters) that a linear exposition is not possible, hence the profligate use of hyperlinks.
> 
> I think that semantics is actually pretty important in SOA. What this model does is highlight the different ways in which semantics plays a role in the ecosystem. It also shows that it is not 'one thing' but a related set of concepts.
> 
> Comments?
> ---------------------------------------------------------------------
> 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
>