CVS difference for ais/ai-00254.txt

Differences between 1.12 and version 1.13
Log of other versions for file ais/ai-00254.txt

--- ais/ai-00254.txt	2004/09/10 00:43:32	1.12
+++ ais/ai-00254.txt	2004/09/14 01:25:56	1.13
@@ -186,7 +186,7 @@
   and the expression P'Access occurs within the body of that generic
   unit or within the body of a generic unit declared within
   the declarative region of the generic, then the ultimate ancestor
-  of S shall be either a non-formal type declared within the generic unit or
+  of S shall be either a nonformal type declared within the generic unit or
   an anonymous access type of an access parameter.
 
 Replace 4.9.1(2) by
@@ -960,7 +960,7 @@
 and the expression P'Access occurs within the body of that generic
 unit or within the body of a generic unit declared within
 the declarative region of the generic, then the ultimate ancestor
-of @i<S> shall be either a non-formal type declared within the generic
+of @i<S> shall be either a nonformal type declared within the generic
 unit or an anonymous access type of a access parameter.>
 
 !corrigendum 4.9.1(2)
@@ -1083,7 +1083,7 @@
 !reference RM9X-3.10.2(16);4.0
 !reference AARM-12.3(12.p,12.q);4.0
 !reference LSN-1042 on Accessibility Checks in Generics
-!from Bob Duff $Date: 2004/09/10 00:43:32 $ $Revision: 1.12 $
+!from Bob Duff $Date: 2004/09/14 01:25:56 $ $Revision: 1.13 $
 !discussion
 
 Two issues related to access types and the accessibility rules came
@@ -1496,7 +1496,7 @@
 !topic LSN on Accessibility Checks in Generics
 !key LSN-1042 on Accessibility Checks in Generics
 !reference MS-12;4.6
-!from Bob Duff $Date: 2004/09/10 00:43:32 $ $Revision: 1.12 $
+!from Bob Duff $Date: 2004/09/14 01:25:56 $ $Revision: 1.13 $
 !discussion
 
 This Language Study Note discusses accessibility checks, and their
@@ -3550,4 +3550,249 @@
 OK. I'm put my horns back in the drawer. :-)
 
 ****************************************************************
+
+From: Gary Dismukes
+Sent: Friday, September 10, 2004  12:47 AM
+
+> Humm. There are a lot of uses of "non-something" in the RM:
+>    non-inherited (C.6(9))
+>    non-binary (3.5.4(27.1/1))
+>    non-others (3.8.1(15))
+>    non-intrinsic (3.10(13), 3.10.2(1))
+>    non-protected (3.11(10)), etc.
+
+IMO those are wrong too because they don't follow contemporary American
+English spelling conventions, but I'm not going to push the point :)
+
+> and since the wording comes from the original approved wording, I think it
+> is correct.
+
+The existing uses of that word in the RM are spelled without the hyphen,
+so you should change it for consistency if no other reason.
+
+****************************************************************
+
+From: Randy Brukardt
+Sent: Friday, September 10, 2004  12:52 PM
+
+There are no uses of "non-formal" or "non formal" in the RM (or AARM, for
+that matter). So the only precedent is the other uses of "non-something";
+indeed, there are no uses of "non " in the AARM, while there are hundreds
+with the hyphen. So it is clear that to be consistent with the existing
+standard, we need the hyphen.
+
+****************************************************************
+
+From: Gary Dismukes
+Sent: Friday, September 10, 2004  12:52 PM
+
+Ah, I see the confusion.  When I said "spelled without a hyphen", I meant
+to remove the hyphen, not replace it with a space. :)  "Non" is indeed not
+an English word and shouldn't appear.  There are two cases of "nonformal"
+in the RM (TC1), in a note in 12.5 and in J.4(2).  There are also lots of
+other cases of "non" words without a hyphen (nondiscriminant, nonderived,
+nondispatching to name a few).  There are also various cases spelled with
+a hyphen (though not nearly as many as without, though the AARM tends to
+use them a lot).  Anyway, I'm only asking for consistency in the spelling
+of a given word throughout the RM, which I assume you agree with. :-)
+
+****************************************************************
+
+From: Randy Brukardt
+Sent: Friday, September 10, 2004  5:24 PM
+
+"nonformal" seems to be a nonword (pun intended) as well. I hadn't considered
+inventing new words as an option for the Standard. [Yes, "nonword" isn't a
+word, either.] And I'm surprised that it would bother someone now when no one
+complained about it during the editorial review of the AI (I just went back and
+verified that).
+
+I'd rather not use it at all, but I don't have the energy to rewrite all of
+this.
+
+****************************************************************
+
+From: Robert A. Duff
+Sent: Friday, September 10, 2004  1:24 PM
+
+> There are no uses of "non-formal" or "non formal" in the RM (or AARM, for
+                                        ^^^^^^^^^^
+> that matter). So the only precedent is the other uses of "non-something";
+> indeed, there are no uses of "non " in the AARM, while there are hundreds
+> with the hyphen. So it is clear that to be consistent with the existing
+> standard, we need the hyphen.
+
+There are uses of "nonformal" (without the space!).
+A quick Emacs hack shows the following 44 words that start with
+"non" with no dash.  This is from the text version of the RM,
+the 2000 version with COR 1 (not the AARM).
+
+index_non_blank
+leading_nonseparate
+max_nonbinary_modulus
+non_currency_picture_string
+non_end_of_line_character
+non_quotation_mark_graphic_character
+nonabstract
+nonactive
+nonatomic
+nonbinary
+nonblocking
+nonboolean
+noncontiguous
+noncontrolled
+nondefault
+nonderived
+nondiscriminant
+nondispatching
+nondistributed
+none
+nonempty
+nonenvironment
+nonexistent
+nonformal
+nongeneric
+nongraphic
+noninherited
+noninstance
+nonlibrary
+nonlimited
+nonlimited
+nonnegative
+nonnull
+nonoverlapping
+nonpathological
+nonportable
+nonpreemptible
+nonpreempting
+nonprivate
+nonstandard
+nonstatic
+nonterminal
+nonzero
+trailing_nonseparate
+
+****************************************************************
+
+From: Robert A. Duff
+Sent: Friday, September 10, 2004  1:35 PM
+
+...and the following 33 starting with "non-":
+
+non-abstract
+non-access
+non-ada
+non-array
+non-binary
+non-breaking
+non-controlling
+non-derived
+non-distributed
+non-edited
+non-fixed
+non-generic
+non-governmental
+non-held
+non-inherited
+non-intrinsic
+non-negative
+non-normative
+non-others
+non-overloading
+non-overridable
+non-preelaborable
+non-preelaborated
+non-primitive
+non-private
+non-protected
+non-remote
+non-reserved
+non-standard
+non-terminated
+non-trivial
+non-void
+non-zero
+
+The following are spelled both ways at different places in the RM:
+
+nonabstract    /   non-abstract
+nonbinary      /   non-binary
+nonderived     /   non-derived
+nondistributed /   non-distributed
+nongeneric     /   non-generic
+noninherited   /   non-inherited
+nonnegative    /   non-negative
+nonprivate     /   non-private
+nonstandard    /   non-standard
+nonzero        /   non-zero
+
+So what were you saying about consistency, Randy?
+
+****************************************************************
+
+From: Tucker Taft
+Sent: Friday, September 10, 2004  5:00 PM
+
+My personal rule is to use "non-" when the combination
+is rare in "normal" English writing, and use
+"non..." when it more common.  For example,
+"nonzero" and "nonstandard" are fairly common
+words, whereas "non-abstract" and "non-access" are not.
+
+For what that's worth...
+
+****************************************************************
+
+From: Randy Brukardt
+Sent: Friday, September 10, 2004  5:35 PM
+
+Looks like the Standard to me - there's rarely any consistency. I'd much prefer
+not to clutter the Standard with all of these non-existent words. Since you
+must have a lot of free time on your hands to spend it writing search tools
+:-), perhaps you should try to rewrite some of these sentences to avoid
+nonwords (or non-words, if you prefer).
+
+****************************************************************
+
+From: Robert Dewar
+Sent: Saturday, September 11, 2004  11:10 AM
+
+> "nonformal" seems to be a nonword (pun intended) as well. I hadn't
+> considered inventing new words as an option for the Standard. [Yes,
+> "nonword" isn't a word, either.] And I'm surprised that it would bother
+> someone now when no one complained about it during the editorial review of
+> the AI (I just went back and verified that).
+
+nonformal is not a word
+non-formal is not a word (hyphenated words cannot be just made up in english,
+they have to be recognized and established, this is not German, where you can
+make up new words)
+non is not a word
+
+So none of these is correct strictly, but if you do feel an urge to invent,
+it would seem clear that nonformal is the preferred form. We don't write
+non-sense or non-entity or non-descript in english :-) The standard is not
+really in english, since many words have defined semantics that do not
+correspond to the normal dictionary usage anyway. So given that the standard
+is in a kind of specialized formal english, I think it's fine to invent new
+words where they are useful.
+
+****************************************************************
+
+From: Robert A. Duff
+Sent: Saturday, September 11, 2004  11:36 AM
+
+But "informal" is a word.  Unfortunately, it doesn't mean what we
+mean by "nonformal".
+
+Similarly, we say "nonlimited type" rather than "unlimited type".
+Nonlimited types are not, in fact, unlimited.
+
+I was going to say, "I put a dash after non if I'm making up a word".
+But that's not quite true.  I think the truth is that I used to put dash
+after non most of the time, but during the Ada9X review process,
+somebody told me to quit doing that.
+
+****************************************************************
+
 

Questions? Ask the ACAA Technical Agent