DR 09-0099 - SML: No Normative References or Definitions for "MDX" and "OLAP"

[Rex Jaeschke]

So how do these new terms interact with these existing terms?

behavior, implementation-defined - Unspecified behavior where each
implementation is expected to document that behavior, which would thereby
promote predictability and reproducibility within any given implementation.
(This term is sometimes called "application-defined behavior".)

behavior, unspecified -Behavior where ISO/IEC 29500 makes no
recommendations. [Note: To add an extension, an implementer must use the
extensibility mechanisms described by ISO/IEC 29500 rather than trying to do
so by giving meaning to otherwise unspecified behavior. end note]

Rex

> -----Original Message-----
> From: Chris Rae [mailto:Chris.Rae at microsoft.com]
> Sent: Tuesday, October 05, 2010 2:07 AM
> To: e-SC34-WG4 at ecma-international.org
> Cc: francis at franciscave.com
> Subject: RE: DR 09-0099 - SML: No Normative References or Definitions
> for "MDX" and "OLAP"
> 
> We discussed this DR in Tokyo at some length - I have in my notes that
> we agreed to add definitions of "implementation-defined" and
> "implementation-specific" to the standard, then to refer MDX to the
> newly-minted "implementation-defined". I also agreed to make sure that
> we have MDX-related examples in the primer (which we do) and that
> Microsoft properly referenced their MDX documentation in their
> implementer notes (which I haven't done yet, but will do - it's not
> necessary for the resolving of this DR).
> 
> I've updated the text for this DR, and the new version is attached. The
> major difference is the addition of the two terms above to the
> definitions and the modification of the "MDX" definition - for the
> definition of the terms I borrowed heavily from ISO/IEC 9075, the SQL
> standard, which uses:
> 
> implementation-defined: Possibly differing between SQL-implementations,
> but specified by the implementor for each particular SQL-
> implementation.
> 
> implementation-dependent: Possibly differing between SQL-
> implementations, but not specified by ISO/IEC 9075, and not required to
> be specified by the implementor for any particular SQL-implementation.
> 
> They also refer to a third definition:
> 
> SQL-implementation: A processor that processes SQL-statements. A
> conforming SQL-implementation is an SQL-implementation that satisfies
> the requirements for SQL-implementations as defined in Clause 8,
> "Conformance".
> 
> I chose not to do this by instead incorporating the definition of an
> "implementation" inline.
> 
> Your thoughts,
> 
> Chris
> 
> -----Original Message-----
> From: Chris Rae [mailto:Chris.Rae at microsoft.com]
> Sent: 25 August 2010 15:12
> To: francis at franciscave.com; e-SC34-WG4 at ecma-international.org
> Subject: RE: DR 09-0099 - SML: No Normative References or Definitions
> for "MDX" and "OLAP"
> 
> Hello Francis - many thanks for taking a look at this. Your definitions
> are a lot better than mine. I think the OLAP one is pretty clear now -
> I'd agree that the MDX one is still a little vague but unfortunately I
> think that's the nature of the beast - the "MDX" expressions in IS
> 29500 are not there to be processed by the spreadsheet application
> itself, they're just to be passed on to an OLAP data provider that the
> spreadsheet application connects to via any means it chooses.
> 
> We should discuss this DR in Tokyo - I've attached a new version of my
> faux-tracked-changes document using Mr Cave's definitions.
> 
> Chris
> 
> -----Original Message-----
> From: Francis Cave [mailto:francis at franciscave.com]
> Sent: 24 August 2010 16:56
> To: e-SC34-WG4 at ecma-international.org
> Subject: RE: DR 09-0099 - SML: No Normative References or Definitions
> for "MDX" and "OLAP"
> 
> Hi Chris
> 
> I don't think this quite does the job. Here are your definitions:
> 
> -----
> 
> MDX - A multi-dimensional expression language, passed to an OLAP
> provider.
> The method of interpreting of this is defined by the server-side OLAP
> provider implementation.
> 
> OLAP - A type of online analytical processing database which uses a
> multi-dimensional data model.
> 
> -----
> 
> The second sentence of the 'MDX' definition contains a typo: the second
> "of"
> should be deleted. However, this is a trivial point.
> 
> The problem with these definitions is that they introduce further terms
> that could cause uncertainty for implementers. What is an "OLAP
> provider"? What is an "online analytical processing database"? Can
> these uncertainties be avoided by either stripping out or providing
> further explanation of potentially puzzling terms? For example:
> 
> -----
> 
> MDX - A multi-dimensional expression language, used to construct
> queries on a multi-dimensional OLAP database. The method of
> interpreting expressions in this language is defined by the OLAP
> database implementation.
> 
> OLAP - An acronym for "online analytical processing", an approach to
> data analysis; hence the name of a specific type of database, used in
> online analytical processing, which uses a multi-dimensional data
> model.
> 
> -----
> 
> These definitions are still somewhat circular, but perhaps beg fewer
> questions?
> 
> Francis
> 
> 
> 
> > -----Original Message-----
> > From: Chris Rae [mailto:Chris.Rae at microsoft.com]
> > Sent: 24 August 2010 21:39
> > To: e-SC34-WG4 at ecma-international.org
> > Subject: DR 09-0099 - SML: No Normative References or Definitions for
> > "MDX" and "OLAP"
> >
> > http://cid-
> >
> c8ba0861dc5e4adc.office.live.com/view.aspx/Public%20Documents/2009/DR-
> > 09-0099.docx
> >
> > This DR covers the use of the terms "OLAP" and "MDX" in the standard
> > without accompanying definitions. It also points out a few uses of
> > "mdx" when "MDX" was meant.
> >
> > Proposed changes are attached - essentially I'm defining both terms
> > inside "terms and definitions" and then correcting the "mdx"
> instances
> > that should have been "MDX".
> >
> > Your thoughts,
> >
> > Chris
>