RE: [office-formula] ABS != Abs?

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

Funny!

Right, the first note (in 2.3(1.1)) fails to take the Absolute value and
incorrectly delivers a negative number.  The result should be "3", not "-3".

I suspect that ABS was chosen because it is easy, it only accepts a single
scalar argument, and it is trivial to present the result.  But not quite
that trivial [;<).

 - Dennis

BIGGER WORRIES

I have no idea what "implicit intersection of the argument with the
expression's evaluation position" means.  It is painful that there is no
attempt to define it until 5.3.3.  Color me incredulous.  Why are we
immortalizing this?  It depends entirely on ambient properties not in
evidence (e.g., the existence of an evaluation position, something not every
OpenFormula occurrence is going to have, it seems to me.  (I assume that the
only use case where this can't be handled by requiring the correct unique
cell reference is when the same literal formula is meant to be used with
different evaluation positions.)

A simpler way to say this in 5.3.3 is that when the a multiple-cell
reference [list] is provided and a single scalar value is required, the cell
that supplies the single scalar value is that unique cell of the
multiple-cell reference [list] that has at least one cuboid coordinate in
common with the evaluation position that applies to the formula being
evaluated.  That there be a unique cell should be a *constraint* on the
automatic conversion and then we don't have to say anything else.  [Note
that 5.3.3 does not address the cuboid case and cuboid is undefined in the
OpenFormula text although it is explained indirectly in 5.4.11.] 

It strikes me that 2.3 need not discuss automatic conversions but only what
matters when the operands are of the expected kind.  When operands do not
directly satisfy the form required, the discussion of implicit conversions
should determine whether there is resolution or there is an error.

I also notice that 3.8 treatment of ReferenceList (no space intentional) is
rather different than the 4.9 Reference List treatment, but the prose uses
"reference list" indiscriminately.  (It looks like 4.9 should have
ReferenceList as its title, and 3.8 should have Reference List.  There are
more confusions of this sort.

-----Original Message-----
From: Patrick Durusau [mailto:patrick@durusau.net] 
Sent: Monday, February 08, 2010 06:09
To: office-formula@lists.oasis-open.org
Subject: [office-formula] ABS != Abs?

Greetings!

I making another run at 2.3 Non-Scalar Evaluation in an attempt to state 
it without the use of examples or at least to make it clear even with 
examples and I noticed that the first numbered paragraph uses ABS in a 
different sense than we define at 5.16.2.

In some of the reference productions it appears as "Abs."

Just wanted to check before I invest too much time in thinking about how 
to disentangle the two uses.

Hope everyone is at the start of a great week!

Patrick

-- 
Patrick Durusau
patrick@durusau.net
Chair, V1 - US TAG to JTC 1/SC 34
Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)


---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php