Keith and everybody,

I'm a little unclear on the status of id/href use in SOAP.  The spec
is short on details, and I don't know the other specs it references
well enough to know where to find the answer.

Is the use of id/href a MUST on the SOAP 1.1 conformance, meaning
that clients and servers must be able to accept it if used?  I know
that it is not a MUST to use it, you could just duplicate data, but
id/href is preferable.  I'm just not sure after reading the spec that
implementations MUST accept it.

Is the use of id/href a MUST somewhere outside of SOAP, such as
schemas?

Thanks,
Chris


--- In soapbuilders@[...]., keithba@[...].. wrote:
>  Dave, et al:
> 
>  I've tried to review the BDG, to see what I could about
>  interoperability and/or SOAP 1.1 compliance. And I have to echo
Sam's
>  words below: I'm confused.
> 
>  I have one concrete thing I *know* breaks SOAP 1.1 and .NET:
ID/HREF.
>  We send them, and it doesn't look like they are required.
> 
>  But outside of that, I am really confused as to what a BDG
compliant
>  app must accept. In my mind, you have to accept *everything* that
is
>  a MUST in SOAP 1.1, including Section 5, if you want full interop.
> 
>  Does this paragraph: "Where this specification is not in compliance
>  with the SOAP 1.1 specification, or the XML 1.0 specification, as
of
>  4/2/01, this specification will change" mean that?
> 
>  I think your best bet is to call this document a quickstart, as Sam
>  suggests below. Then, beef up the paragraph I quoted to say
something
>  like the paragraph above it. At that point (IMHO) you would have a
>  most excellent document that will be of tremendous help to many,
many
>  people.
> 
>  I hope this helps.
> 
>  Thanks,
>  Keith
> 
>  PS Sorry for the delays (haven't been feeling well+too much work
>  since Wed. of last week.)
> 
> 
>  --- In soapbuilders@[...]., "Sam Ruby" <rubys@[...]..> wrote:
>  > <section serious="no" note="be careful to note smileys">
>  >
>  >    Oh, sure, send a note saying that you will be up all night
after
>  I go to
>  >    bed (I'm on the east coast)...  ;-)
>  >
>  >    Ask people to tell you what to write, and then when they tell
>  you what
>  >    to write, tell them you will think about it, but never get
back
>  to them.
>  >    ;-)
>  >
>  >    Don't say that you haven't finished thinking about it yet -
you
>  are the
>  >    one saying that it should be ready to freeze at 9 am.  ;-)
>  >
>  > </section>
>  >
>  > The attached e-mail contained a very specific request.  Dave, I'm
>  still
>  > curious as to what you think.
>  >
>  > I appear to not have been successful to date in finding the words
>  that
>  > convey my concern.  I'll continue trying, sometimes borrowing
words
>  from
>  > others, sometimes trying different tactics.
>  >
>  > New approach:
>  >
>  >    Question: What do you call a spec that permits some of the
MUST
>  NOTs
>  >      that another spec does not?
>  >    Answer: A superset.
>  >
>  >    Question: What do you call a spec that does not require all of
>  the MUSTs
>  >      that another spec does?
>  >    Answer: A superset.
>  >
>  >    Question: What do you call a spec that starts with another
spec
>  and adds
>  >      additional MUSTs?
>  >    Answer: A subset
>  >
>  >    Question: What do you call a spec that starts with another
spec
>  and adds
>  >      additional MUST NOTs?
>  >    Answer: A subset
>  >
>  >    Question: Given these answers, is the current BDG a superset
or a
>  >      subset?
>  >    Answer: Both.  Neither.  A dialect.  A different protocol.
I'm
>  so
>  >      confused!
>  >
>  > A quick start would say something like:
>  >
>  >     "if you see something that looks like a header - look for a
>  >    mustunderstand attribute and if you find it - simply fail the
>  request!
>  >    At some point in the future, you might want to add in support
for
>  >    headers, but for now you don't so simply reject the request."
>  >
>  > Header support could be in a second chapter of the quick start.
As
>  could a
>  > number of other aspects of SOAP.  Don't banish them forever from
>  the "BDG
>  > subset" of SOAP, simply don't require them in the first cautious
>  steps
>  > towards SOAP interop.
>  >
>  > Again, I would like to see this document describe an on-ramp not a
>  > destination.
>  >
>  > Keep up the good work!
>  >
>  > - Sam Ruby
>  >
>  > ---------------------- Forwarded by Sam Ruby/Raleigh/IBM on
>  04/02/2001
>  > 09:26 AM ---------------------------
>  >
>  > Sam Ruby/Raleigh/IBM@[...].. on 04/01/2001 03:49:20 PM
>  >
>  > Please respond to soapbuilders@[...]..
>  >
>  > To:   soapbuilders@[...]..
>  > cc:
>  > Subject:  Re: [soapbuilders] RFC: Top of BDG
>  >
>  >
>  >
>  > Dave,
>  >
>  > My two cents: much improved.
>  >
>  > It is probably way too late for this suggestion, but I will give
it
>  anyway.
>  > Consider renaming the document to "A Quick Start to Soap 1.1".
>  >
>  > There are four occurrences of the word subset in this document.
>  The two in
>  > the first paragraph can evaporate.  The one in the scalars
section
>  can be
>  > changed to "examples include".  The one in the goals section
should
>  be
>  > replaced with something along the lines of "an accessible
>  description that
>  > people could put to use immediately and productively".
>  >
>  > There is no need to label this as establishing a subset.  Nobody
>  expects a
>  > quick start to be exhaustive.  And the wisdom of subsetting is
>  > controversial.  I, for one, believe that a subset is a good
>  starting point
>  > but not a good stopping point.  You may decide never to go beyond
>  the
>  > subset you have defined over the past few days, but if you do,
the
>  market
>  > will decide whether or not your stopping is a good idea.  And, if
>  in the
>  > future, you do reconsider, hopefully there will be a number of
open
>  > implementations available to you to build upon.
>  >
>  > So, my advice is to avoid the issue.  It is a distraction.
>  >
>  > Is it your goal to provide an on-ramp to SOAP 1.1, or is it your
>  goal to
>  > divide the SOAP community?  Think about that for a moment.
>  >
>  > - Sam Ruby
>  >
>  > P.S.  What got me concerned was the survey that you recently
ran.
>  I didn't
>  > like any of the options.  Despite (or perhaps because of?) the
>  results, you
>  > apparently have decided to made it your business to make the
first
>  option
>  > on your survey a reality.
>  >
>  >
>  >
>  > To unsubscribe from this group, send an email to:
>  > soapbuilders-unsubscribe@[...]..
>  >
>  >
>  >
>  > Your use of Yahoo! Groups is subject to
>  http://docs.yahoo.com/info/terms/