RE: [office] ODF 1.2 Single-Level Conformance and Floor << Ceiling Already

["Dennis E. Hamilton" <dennis.hamilton@acm.org>]

Rob,

I recalled the floor=ceiling debate as the kind of thing that has plagued
standards bodies from olden times, yet I completely misremembered the
dimensions of the debate.  I noticed that as I was sending off the note
about MIMETYPES and how <office:body> needs to align.

It happens that the floor for ODF is far below what must be accepted from a
(strictly-)compliant document.  That is because the requirement for
processing does not insist on any minimum feature set whatsoever (ODF
1.0/1.1 non-normative Appendix D presents that as an feature from the
perspective of the founding fathers).  

The following example is ridiculous (and Bart Hanssens will raise his
eyebrows), but it points out the most (or least, depending on perspective)
that any ODF processor has to deal with.  Acceptance of all sorts of
additional content must be acceptable, but support for it at the processing
or semantic level is not.  This is a far bigger deal for interoperability
than whether or not foreign elements are allowed, with or without some
proviso that the reduction to a conformant document be benign.  Keeping in
mind that the world is a mostly-practical, often-realistic, place, and no
one could get a way with this, here is all that the letter of the ODF law
requires.

I chose 1.0/1.1 because those specifications are already definite about this
and I wanted to use the minimal number of required attributes.  Don't blink
or you'll miss it:

	-	-	-	-	-	-	-

<!-- Minimum Required ODF 1.0/1.1 Text Document
     There is one for each of the main document
     types.  There is a package equivalent that
     has the same office:body in an office:document-
     content element in content.xml and has the MIMETYPE 
     in the special first-item of the package.
     For a text template, just change the mimetype.
     -->

<o:document 
      xmlns:o="urn:oasis:names:tc:opendocument:xmlns:office:1.0"
      o:mimetype="application/vnd.oasis.opendocument.text"
      >
   <o:body>
      <o:text />
   </o:body>
      <!-- Singing: I ain't got no body, no body have I ... -->
</o:document> 

	-	-	-	-	-	-	-

[It is just like me to have more commentary than code, as I'm sure is no
surprise.]

I grant that, as far as a grammar for markup goes, allowing these cases is
very convenient for definition.  It is also useful to do incremental,
test-driven development of a processor from progressive test documents like
this.  I expect to see many more of these under such conditions.

However, there is not much help in Appendix D about what would be essential
to support.  Noticing that, maybe I should give up fretting about ceiling1 =
ceiling2, although I do think that foreign elements and attributes have an
useful place, whether there is a conformance level with regard to their
benign use or not.

 - Dennis 

-----Original Message-----
From: Dennis E. Hamilton [mailto:dennis.hamilton@acm.org] 
http://lists.oasis-open.org/archives/office/200901/msg00142.html
Sent: Monday, January 19, 2009 13:30
To: 'robert_weir@us.ibm.com'
Cc: 'office@lists.oasis-open.org'
Subject: RE: [office] ODF 1.2 Single-Level Conformance and Law of Unintended
Consequences

One more thing about floor=ceiling.  

[ ... ]

It may be that loosely-conformant is the wrong term, but neither conformant
nor loosely-conformant, at the moment, determine whether interoperability
will be easy or hard.  What I like about loosely-conformant is that it
provides that there be a conformant document in there, given certain
adjustments.  (There are also some edge cases that one might worry about,
where dropping an attribute means there is no attribute in a way that
impairs the document, table cell formulas being an interesting case.  This
condition only works well for ODF 1.2 because of its completeness with
OpenFormula.  This is an argument for recognizing those cases as
loosely-conformant, but it is not my ox that gets gored if only the
OpenFormula case qualifies in a conformant spreadsheet document.)

 - Dennis

-----Original Message-----
From: Dennis E. Hamilton [mailto:dennis.hamilton@acm.org] 
http://lists.oasis-open.org/archives/office/200901/msg00136.html
Sent: Monday, January 19, 2009 10:45
To: robert_weir@us.ibm.com
Cc: office@lists.oasis-open.org
Subject: RE: [office] ODF 1.2 Single-Level Conformance and Law of Unintended
Consequences

This also reminds me of another debate, called floor=ceiling.  

Floor=Ceiling was debated in the early days of initial COBOL standardization
efforts and it continued for a while.  I can't recall which side of the
debate Committee Chair Howard Bromberg held onto and if he flipped at any
time.  

Generally, producers of implementations did not look kindly at floor=ceiling
and user communities (but not all of them) and especially standards sheriffs
of various persuasions wanted floor=ceiling.  Of course, COBOL was
modularized and there were definite implementation-specific provisions
(COMPUTATIONAL-1, COMPUTATIONAL-2, ... and similar aspects coming to mind),
so I am not sure how much it was felt that floor=ceiling was achieved, in
the end.   

[ ... ]