Re: Notes on SpecGL discussions [DRAFT]

Responding to David and Sandra's comments together (thanks to both):

At 11:19 AM 7/25/02 -0400, David Marston/Cambridge/IBM wrote:

>I agree with Sandra that the WG should set priorities for the existence,
>locatability, and content of conformance clause in a consistent way.
>
>I propose that checkpoint 5.4 be dropped, and the explanatory verbiage
>be moved over to 10.1, which currently has no such verbiage.

I will do this for the next draft.  I could also put some mention of GL10 
into the verbiage of GL.5.

>[...]Once that's done, GL 5 will focus on the establishment of a
>policy, while GL 10 will focus on the documentation of it.

For example, this statement would be useful in the verbiage of GL5.


>Ck 10.1 is about documenting the policy somewhere: "all specs must
>address conformance" and thus Priority 1 is justified.
>Ck 10.2 is about documenting conformance in a particular way.

Okay.  If they were the same priority, I would suggest to merge them 
together as two requirements under the same checkpoint.  But (for now) they 
are P1 and P2.  (Coming soon ... an all-ckpt review of priorities!)

>Ck 10.3 is about locatability
>Ck 10.4 should be added to cover citing dependent specs: "all specs
>should clarify their perimeter of normativity, even where it includes
>other specs" -- priority needs to be voted on

Sandra mentioned that the CK10.3 ("make TOC entry") discussion didn't make 
sense, and I had noticed the same:  "Make normative reference to 
specifications on which the current specification depends." It is my belief 
that this was intended to be checkpoint 10.4 (but lost its ckpt markup), 
not discussion of 10.3  -- similar to David's proposed 10.4, at least in 
underlying motivation.

Thoughts?


>I hope that we are nearing closure on the GL 10 material.

I think so.  The monday (7/29) editors draft -- for wednesday (7/31) 
telecon discussion -- will have changes.  I'll put in flags (@@) to bring 
attention to them, for endorsement in telecon.

-Lofton.

Received on Friday, 26 July 2002 13:26:40 UTC