J3/01-344

To:       J3
From:     Malcolm Cohen
Subject:  /DATA responses to 01-260 and 01-320
Date:     23rd August 2001

1. Introduction
---------------
This paper contains the /DATA responses to the items deferred to it by /A.

2. Itemised responses to 01-260
-------------------------------
Item [68:38-41]: Subgroup agrees, but proposes no action unless someone can
come up with a snazzy term.

Item [74:35-47]: The assertion that note 5.15 interacts with interp 31 is
mistaken, fortunately.  In any case, interp 31 is still under consideration.

Item [76:9-10]: (a) Yes, the contrapositive is true.  This follows from the
provision of semantics for such references and definitions (e.g. in section 7)
and the lack of prohibition against them.  (b) The assertion that this
subsection interacts with interp 31 is mistaken.

Item [83:40+]: No, there is no need to say anything about deferred or assumed
type parameters.  We already have rules about things declared with such type
parameters and these apply to implicit declaration as well as explicit
specification (because they use the term "declaration").

3. Response to item [246:10] from 01-320
----------------------------------------
Subgroup agrees that there is an ambiguity and proposes that abstract interface
names shall not match intrinsic type names.

4. Edits to 01-007r2
--------------------
[245:16+] Insert constraint
"C1211a An abstract interface name shall not be the same as a keyword that
specifies an intrinsic type."