Next in thread → Next in month →

RE: [conformance] Minutes of 15 Feb 2002, Telconference

From
Lofton Henderson
Date
2002-02-21T17:21:00+00:00
ID
Thread
RE: [conformance] Minutes of 15 Feb 2002, Telconference
MHonArc v2.5.2 -->


















conformance message






[Date Prev]
 | [Thread Prev]
 | [Thread Next]
 | [Date Next]

--

[Date Index]
 | [Thread Index]
 | [Elist Home]








Subject: RE: [conformance] Minutes of 15 Feb 2002, Telconference




From: Lofton Henderson <>
To: 
Date: Thu, 21 Feb 2002 15:19:07 -0700







At 06:53 PM 2/20/02 -0500, Lynne Rosenthal
wrote:

[...]
Has anyone yet recorded a definition, exactly what we mean by
"deprecated feature".  Is it still legal but
discouraged?  Does it pertain only to MUST features?  Or also
to SHOULD features and MAY features?  (Can you deprecate a MUST NOT
assertion?)  
[Lynne Rosenthal] No. this wasn't specifically discussed. 
However, a definition is needed and will be provided.  the
pertinence to MUST, SHOULD, MAY is a conformance impliciation and not
necessarily part of the definition. 

That's probably true.  Until we make the definition, I wasn't sure
whether or not it might have some dependency on the level of normative
language.



[...]
I think it would be useful to make a precise definition of
"backward compatibility", particularly if it is going to be
used in an issue statement.  Having missed the teleconference, I'm
not sure what is being discussed.  Here are a couple of
possibilities that come to mind:



[Lynne Rosenthal] Again, agree.  A definition of what is meant
by backward compatiability would be good.  I'm not sure how precise
it will be considering there may be different flavors depending on the
type of specification. 

Maybe the Glossary isn't the right place for such detail (or maybe it
is).  But if it does turn out that different flavors need to be
addressed separately in order to be precise, then we should certainly do
that.  Perhaps in an "extended definition"
somewhere.  I have suspected for some time that a lot of
argumentation happens because the participants have different flavors in
mind as they argue their viewpoints, and their base premises (implicit
flavors) are never clarified or made explicit. 


As you further noted, this is a live topic on the W3C QA WG issues list,
and I think we have an opportunity for mutual progress and benefit
here.


-Lofton.
Next in thread → Next in month →