RE: mustUnderstand reformulation

Henrik Frystyk Nielsen writes:

>> Sounds good! - a few comments:
>> 
Good, thank you!

>> * I would use "Blocks tagged with a SOAP.." rather than "Elements 
tagged
>> with a SOAP..." as the attribute is context sensitive
>> 
Whatever is consistent with the rest of the specification is fine with me.

>> * There might be a few places where "block" replaces the earlier 
"entry"
>> term
>> 
Likewise.

>> * I would replace "In general, processors SHOULD NOT..." with
>> "Processors SHOULD NOT..." as it always is a SHOULD NOT.
>> 
I considered that, but left the "in general" in recognition of the fact
that we specifically discuss later how extensions can indeed be
created that require the generation of such faults.  I agree this
could use at least some more consideration, and suggest that it be
left to the editorial team.

>> * Maybe also consider removing the last sentence as it is slightly
>> redundant and format the last paragraph as a note (italics?) in the 
spec
>> so that it is clear that it is a suggestion (not that I think it is a
>> bad one :)
>> 
Again, I suggest that all of this the left to the editorial team to
take a crack at, with the rest of us having a chance to comment in the
unlikely case that we are unhappy with the results. 

>> * Replace 'labeled mustUnderstand="1"' with 'tagged with a
>> mustUnderstand attribute with the value of "1"' as it seems slightly
>> more consistent with the first paragraph and less syntax specific. Btw,
>> I like using "true" instead of "1".
>> 
I think we have already seen a proposal to do this in infoset terms,
which probably supersedes both my original formulation and the one
that you are proposing.

>> Thanks!
>> 

Thank you!

------------------------------------------------------------------------
Noah Mendelsohn                                    Voice: 1-617-693-4036
Lotus Development Corp.                            Fax: 1-617-693-8676
One Rogers Street
Cambridge, MA 02142
------------------------------------------------------------------------

Received on Friday, 10 August 2001 22:49:12 UTC