CVS difference for ai12s/ai12-0196-1.txt

Differences between 1.8 and version 1.9
Log of other versions for file ai12s/ai12-0196-1.txt

--- ai12s/ai12-0196-1.txt	2016/12/28 04:14:29	1.8
+++ ai12s/ai12-0196-1.txt	2017/01/14 02:55:52	1.9
@@ -1235,6 +1235,83 @@
 From: Randy Brukardt
-Sent: Wednesday, December 27, 2016  8:51 PM
+Sent: Tuesday, December 27, 2016  10:21 PM
+> Right, it just changes the AARM notes, which don't require 
+> any formal action.
+It turns out that I'd already removed "reentrant" from most of those,
+replacing it with some version of "concurrent calls".
+I've now removed the word from a handful of remaining notes. That just leaves
+a single use in the Distributed Systems Annex (which has nothing to do with
+either of these AIs). So far as I can tell, that was never properly defined
+anyway, so I'm going to ignore it. If someone cares enough to want to fix it,
+submit an AI to change the wording. (That is, constructive criticism rather
+than obstructive criticism).
+As far as the AI discussion goes, the term was used in the sense of "required
+to work in concurrent calls", and many of those uses already used the phrase
+"concurrent calls". I suspect that the editorially changed AI actually is
+easier to understand than the old one.
+Check it out yourself at:
+From: Erhard Ploedereder
+Sent: Monday, January 2, 2017  6:57 PM
+So, since there seems to be some consensus forming, here are the matching
+words: [He meant this as a change on AI12-0200-1, filed here with the rest of
+the thread - Editor.]
+Two concurrent calls on any two (possibly the same) language-defined
+subprograms are said to be non-interfering if they perform as specified,
+as long as all pairs of objects (one from each call) that are either
+denoted by parameters that could be passed by reference, or are designated
+by parameters of an access type, are nonoverlapping.
+The implementation shall ensure that all concurrent calls on all
+language-defined subprograms are non-interfering.
+Scanning the RM, it's about 50/50 on hyphenation, i.e. non-interfering vs
+From: Randy Brukardt
+Sent: Tuesday, January 3, 2017  4:37 PM
+> So, since there seems to be some consensus forming, ...
+I'd hope that it takes more than two people talking to each other to form a
+There is no need to define a term here, be it "reentrant" or "non-interfering"
+-- there is no use of any technical term in either existing RM or AARM wording
+-- including AI12-0196-1 (as I pointed out last Wednesday). The discussion in
+AI12-0196-1 seemed (to me) to be improved by eliminating the terms and just
+referencing the rules directly. Replacing "concurrent call rules of Annex A"
+with "non-interference rules of Annex A" seems to be a sideways move at best
+(and for me, actually a step backwards since "concurrent" is IMHO the
+important part).
+It's always possible that some future wording would be improved by having a
+term, but I'd rather not be investing resources on a prediction of the
+future. (There'd be no problem introducing a term if and when it is needed.)
+The ARG should work on fixing bugs in the standard, not inventing ones that
+don't exist.
+So I don't believe any change is needed to the approved AI at this time.
+> Scanning the RM, it's about 50/50 on hyphenation, i.e. non-interfering 
+> vs noninterfering.
+Right, but we adopted a metarule a few years ago that new "nonwords" would not
+have hyphens, while existing ones would continue to be spelled consistently
+(not wanting to mess around removing hyphens from many but not all of hundreds
+of words in the Standard). So this would be spelled "noninterfering" if we
+defined such a term.

Questions? Ask the ACAA Technical Agent