[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [trust-el] Groups - TrustEL Architecture v01.pdf uploaded
Nothing wrong in disagreement Peter – it’s back in fashion again..
J I’ll be interested to hear what you think of V2. Cheers Colin From: Andrew Hughes [mailto:andrewhughes3000@gmail.com]
I'll work to ensure that the right level of detail is present at each abstraction level. I'm still trying to find all of the common threads - on the calls there are more than one use case and flow being discussed intermittently, and it's hard to differentiate which discussions are about the primordial use case and the advanced
use cases. Just to note that the stick and box pictures are mostly the sequence diagram from the primordial use case shown in a different format, plus a small bit of embellishment by me. The small bit of extra is to help get at the envisioned underlying mechanisms
to ensure that the higher level abstraction is representing faithfully. andrew.
Andrew Hughes CISM CISSP +1 250.888.9474 On Fri, Jan 23, 2015 at 8:16 AM, Peter Alterman <palterman@safe-biopharma.org> wrote: Well, Colin, here's one of the few areas in which we disagree: I think it's very important for lower-altitude architectural and flow models to map back up to the high level flow we've been working from. Here's
why: the Trust Elevation TC is chartered to develop standards (in our case Committee Specifications) that cover a broad range of implementations. In going to a Fourth Deliverable (unchartered, btw) the TC is developing a template for making the previous Committee
Specifications functional, but it is only a template and other implementers out there may certainly roll their own, so long as they align their work with what we've produced. Without a clear path from a generic use case to a particular implementation of it
(there are many implementations possible when deconstructing the boxes, as Andrew has observed), we run the risk of having the Fourth Deliverable be misunderstood, that we were trying to force all Trust Elevation use cases into our one deconstructed model.
Therefore, I think it's important to show the high level use case, make clear it's just the most generic one we can imagine, and then show one particular deconstruction while also acknowledging that many others can fit under the model, and that the actors,
transactions, data flows and metadata tags are an example of how to work through implementing TE for a particular case. I do agree with the rest of your comments. Naturally. ;-) Peter
------------------------------------------------------------ Peter Alterman, Ph.D. Chief Operating Officer cell:
301-943-7452 On Thu, Jan 22, 2015 at 6:14 PM, Colin Wallis <Colin.Wallis@dia.govt.nz> wrote:
|
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]