(j3.2006) another example of a forward reference

[Robert Corbett]

On 03/13/15 01:51, Malcolm Cohen wrote:
> Bob Corbett suggests:
> <<<
> I suggest that paragraph 5 of Clause 4.2 be split into two paragraphs,
> as the two sentences in the paragraph have nothing to do with each
> other.
> That is a good point, actually the same can be said of the last sentence of p4.
>
> In fact maybe the first sentences of p4 and p5 should be the new p4, and the
> second sentences thereof can be the new p5?
>
> I will include these suggestions in an editorial paper for next time.  (I think
> the wording might need a little more work though.)

I am switching my references to 15-007.

Is there a reason to keep the last sentence of paragraph 4
of Clause 4.2 [52:8-9]?  Paragraph 2 of Clause 4.4.1 presents
some of the same information.  If the part about distinguishing
multiple representations is considered important, it could be
tacked on there.  The last sentence of paragraph 5 of
Clause 4.2 could also be moved there.

The first sentences of paragraphs 4 and 5 of Clause 4.2 also
appear to be redundant.  The same information is provided in
paragraph 5 of Clause 4.5.3.1 [67:19-20].  The description
there is clearer than the description in Clause 4.2.

Although paragraph 5 of Clause 4.5.3.1 says that type
parameters may be used as primaries in specification
expressions, the definition of specification expressions in
Clause 7.1.11 does not include type parameters among the
permissible primaries.  Constant expressions and kind type
parameters have a similar issue.

Bob Corbett