[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [soa-rm-ra] Proposed edits to SOA-RAF Committee Specification Public review Draft
|A few comments on the draft distributed by Peter:|
- In the abstract, can we just say we're using IEEE 1471-2000 and leave the details of 42010 for section 1.3.1? That would be cleaner.
- In the abstract, there is a strike-thru the added ANSI. Is this supposed to be deleted. In any case, it should be consistent with section 1.3.1.
- In section 1.3.1, I suggest the following be used as the first paragraph:
The SOA-RAF structures its analysis based on the concepts defined in ANSI/IEEE 1471-2000, “Systems and Software Engineering – Architecture description” [ANSI/IEEE 1471]. ANSI/IEEE 1471 was eventually approved as ISO/IEC 42010-2007 and subsequently superseded by ISO/IEC/IEEE 42010:2011 [ISO/IEC/IEEE 42010]. Although the more recent standard modifies some of the original definitions and introduces new material, the modifications and additions were not found to significantly impact the SOA-RAF analysis. As such, the SOA-RAF follows the definitions and structure of the original standard.
An architectural description conforming to ANSI/IEEE 1471-2000 and ISO/IEC 42010-2007 must include the following six (6) elements:
- In section 2.2, we note the dual designation of ANSI/IEEE 1471 and ISO/IEC 42010, but [ISO-IEC 42010] was deleted from references. Should the dual designation be noted in the references?
- The remnant remaining in line 1945 seems awkward. Would we lose anything by deleting completely?
- For line 1949, did we close on the question of whether the mediator led to simpler or a more complex infrastructure? Consider instead, "Mediated awareness provides consumers and providers with a straightforward approach to mutual visibility."
- In a later email I suggested (but not strongly) that we keep lines 1974-1977 as a way for the reader to connect with a familiar and concrete example.
- In Acknowledgements, Kevin Smith should affiliate with his company and Michael Poulin is no longer with Fidelity.
On Nov 2, 2012, at 8:37 PM, Peter F Brown wrote: