PLEASE PROOF: Drafts of 29500-1/-4:2016; feedback due by the end of 2016-04-29

[Rex Jaeschke]

See my replies inline. Rex



-----Original Message-----
From: caroline arms [mailto:caroline.arms at gmail.com] 
Sent: Wednesday, April 13, 2016 9:34 AM
To: Rex Jaeschke <rex at rexjaeschke.com>
Cc: SC 34 WG4 <e-SC34-WG4 at ecma-international.org>; TC45 <e-TC45 at ecma-international.org>
Subject: Re: PLEASE PROOF: Drafts of 29500-1/-4:2016; feedback due by the end of 2016-04-29

Back to item 152.

I'm remembering that there were a bunch of formula-related changes made during the period when ODF 1.2 Part 2 (Recalculated Formulas) was being reviewed by MS.  I suspect DR 14-0003 relates to things discovered then.

So I have looked at the ODF spec for the PROB function, which says.

Constraints:
? The sum of the probabilities in Probability shall equal 1.
? All values in Probability shall be > 0 and <= 1.
? COUNT(Data) = COUNT(Probability)

Looking at https://en.wikipedia.org/wiki/List_of_mathematical_symbols#Symbols_based_on_equality,
I see "<>" mentioned, but only in a note in the row for ?.  The note indicates the use of a variety of ASCII-only ways to indicate "is not equal to".

So MS should confirm that the second bullet in  18.17.7.260 is supposed to mean the same as the first ODF bullet.  Assuming that "is not equal to" is what is meant, I recommend using words or ? rather than the symbol "<>".  The audience for this standard is not limited to programmers in languages that use "<>" for this conditional.  Such a fix would be editorial, I believe.

Rex> Rejected. WML fields and SML formulas use <> to represent not-equal, not only in narrative but as part of expression syntax in the grammars.


    Hope this helps.   Caroline.