I'm not sure whether I can still send to thie J-3 list. We'll see...
Apparently not, probably because I unsubscribed from recieving it. (Just
too much there for me no longer being involved in the day-to-day work;
no I don't want to get every darned message emailed to me.)

Well, I'll post it here in the hopes that someone will pick it up.

This is far from new, but I find myself tripping over it now in the
process of trying to explain the standard.

Why does the BNF term specification-stmt exist? Its only use in the BNF
is to collect together a subset of the declaration constructs. One might
suspect that it is intended to implicitly define the term "specification
statement", which is used in multiple places. However, if so, that
definition does not correctly correspond with the usage. It appears to
me that the term "specification statement" is more often used to mean
any statement in a specification part. That's what I'm telling people it
"must" mean, in spite of the bnf.

For example, we have things such as (the first sentence of 12.4.3.1 in
08-007r2).

 "The interface for an internal, external, module, or dummy procedure is
specified by a FUNCTION, SUBROUTINE, or ENTRY statement and by
specification statements for the dummy arguments and the result of a
function."

Insomuch as the BNF shows that a type-declaration-stmt is not a
specification-stmt, does this mean that type declaration statements
don't specify things about the interface? (That was a rhetorical
question; of course, they do.)

If one considers the above quote to be more in the nature of informative
blather instead of normative material, let's examine a constraint, which
is hard to dismiss that way. C1251 of 09-007r2.

 "If RESULT appears, the function-name shall not appear in any
specification statement in the
scoping unit of the function subprogram."

Again, apparently a type-declaration-stmt is ok? I doubt it.

-- 
Richard Maine                    | Good judgement comes from experience;
email: last name at domain . net | experience comes from bad judgement.
domain: summertriangle           |  -- Mark Twain

Richard Maine wrote:
> I'm not sure whether I can still send to thie J-3 list. We'll see...
> Apparently not, probably because I unsubscribed from recieving it.
> (Just too much there for me no longer being involved in the
> day-to-day work; no I don't want to get every darned message emailed
> to me.)

I forwarded it.  I have continued to receive the list in spit of
not reading it most of the time.  I have piles of unread j3
mail.  But it's on my alternate e-mail account, so I usually
don't even see it.

-- 
J. Giles

"I conclude that there are two ways of constructing a software
design: One way is to make it so simple that there are obviously
no deficiencies and the other way is to make it so complicated
that there are no obvious deficiencies."   --  C. A. R. Hoare

"Simplicity is prerequisite for reliability"  -- E. W. Dijkstra