Re: ICS for Spec Guidelines

At 10:14 AM 12/4/02 -0500, Karl Dubost wrote:
>[...]
>>Since Karl is the 2nd person to assert that SpecGL does not have an ICS, 
>>it raises a question about the checkpoint (which I asked earlier):  in 
>>order to satisfy it, you MUST publish an ICS.  MUST it be labelled as an 
>>ICS?  Or SHOULD it be labelled as an ICS?  Or ...? I.e., SpecGL's ICS is 
>>labelled as a "Checklist".  Does SpecGL pass or fail?
>
>So we should explain what's an ICS. I tried to find Implementation 
>Conformance Statement in the TR space and I didn't find it. Except if it's 
>absolutely necessary we can use another wording.
>(It seems to be my QA-checklist.html in a sense)

I agree that we should state in SpecGL (and in the Checklist) that the 
spec-checklist is an ICS for SpecGL.


>>If the answer is "fail", then it would seem that an additional normative 
>>requirement needs to be added to the "to fulfill" section of the checkpoint.
>
>Yes :)

Perhaps Lynne can add this to the remaining SpecGL discussion issues -- 
"must have an ICS [and must call it an ICS?]".

...unless everyone agrees that the [...] part above should become part of 
the fulfillment criteria of the checkpoint.

-Lofton.

Received on Wednesday, 4 December 2002 11:33:43 UTC