[J3] Consideration of Paper 18-242 at Meeting 218 (Was: 18-242)
Van Snyder
van.snyder at jpl.nasa.gov
Sat Dec 1 19:29:48 EST 2018
On Sat, 2018-12-01 at 18:53 -0500, Steve Lionel via J3 wrote:
> 18-242 was marked "no action" because JOR did not consider it worth
> pursuing. The reason that ac-i-do-variable and data-i-do-variable had
> typespecs added was that, as construct entities, there was no
> reasonable way to give them types that might be different from the
> variable in the enclosing scope. In the case of loop-control and
> io-implied-do, these are NOT construct entities and can already be
> declared as needed. 18-242 proposes adding syntax that has the
> side-effect of turning them into construct-entities, which is a major
> and potentially confusing side-effect. JOR did not think this was a
> good idea.
ac_i_do_variable and data_i_do_variable are statement entities.
The index variables in DO CONCURRENT are construct entities.
io-implied-do would be a statement entity. There was no possibility of
turning it into a construct entity.
A loop-control declared in the statement would be a construct entity.
That's the point! It's not an accidental defect.
Students still marvel at the inconsistency. It's just an excuse for
computer-science professors to refuse to teach Fortran, and lecture
their students about the defects in FORTRAN 66.
> Subgroups act as filters on proposals. Certainly any declined proposal
> can be reintroduced, but one had better offer a convincing argument as
> to why it should be reconsidered. In my opinion, one of the reasons
> F2018 took as long as it did was that certain proposals kept being
> reintroduced, despite expressed lack of interest by the majority of
> J3.
In 2004, for development of f08, every proposal was discussed in
plenary. There was no subgroup filter.
f18 took as long as it did because we had two TS's that should have had
independent schedules. That's the point of a TS. Instead, we took full
committee time for them, without adding any more meetings to the
calendar. And we spent some time on the WG23 TS, without adding more
meetings to the calendar.
And a subgroup opinion isn't "lack of interest by the majority of J3."
It's "lack of interest by about a third of J3." And maybe less because
we don't know if the subgroup opinion was unanimous. When subgroup says
"no action," how many others read a paper?
> I have not studied all the various rules Craig Dedo refers to, but I
> don't think it is productive to keep trying out an idea that has been
> rejected for cause. I do agree that it would have been useful to have
> each "no action" accompanied by recording of reasoning, and I'd
> encourage that to happen in the future.
Just announcing a reason and a decision is inadequate. Announcing an
opinion is fine. But if it seems that subgroup didn't understand a
proposal, there ought to be an opportunity for more explanation and
discussion.
> Steve
>
>
> On 12/1/2018 6:28 PM, Craig T. Dedo, CAPM via J3 wrote:
>
> > Hello J3 Members:
> >
> > This reply addresses Van Snyder’s questions in the last
> > paragraph of his message: “Can we consider 18-242 at meeting 218?
> > Not re-consider it, because (in my opinion), it was clearly not
> > fully understood at meeting 217. If so, does it need to be
> > re-submitted to the server as a 218 paper?”
> >
> >
> >
> > Short Answer: Yes, PL22.3, aka J3, can consider paper
> > 18-242 at meeting 218.
> >
> >
> >
> > Details: Here are the details in case you are
> > interested.
> >
> >
> >
> > PL22.3 is governed by the SD-2, which in effect is its
> > bylaws, and by the current edition (11th edition) of Robert’s Rules
> > of Order Newly Revised (RRONR). Unless PL22.3 has adopted special
> > rules of order that allow a subgroup to kill a paper or the bylaws,
> > i.e., SD-2, authorize such action, the paper is “live”, i.e., a
> > candidate for action until PL22.3 takes action on it in plenary.
> > Such plenary action could be to approve it, kill it, postpone it to
> > the next meeting, refer it to a standing subgroup, create a special
> > subgroup to study the paper, or something similar.
> >
> >
> >
> > Paper 18-242 was administratively referred to JOR and
> > not acted on. JOR did not report the paper back to plenary so paper
> > 18-242 is still in the jurisdiction of JOR. At Meeting 218 PL22.3
> > could discharge the paper from JOR and act on it. Or, PL22.3 could
> > instruct JOR to study the paper and report back with a
> > recommendation and even impose a deadline for reporting back with
> > its recommendation.
> >
> >
> >
> > Each week-long meeting is a session within the framework
> > in RRONR. Untransacted business automatically transfers over from
> > one session to the next. This means that paper 18-242 automatically
> > transfers over to Meeting 218.
> >
> >
> >
> > Thus, Van, you can either have PL22.3 take action on
> > 18-242 as submitted at Meeting 217 or else re-submit your paper with
> > a new paper number for consideration at Meeting 218. Either course
> > of action is within the existing rules. I’m not sure which one
> > would be more acceptable from a human relations point of view. You
> > will need to find out from your fellow committee members.
> >
> >
> >
> > Hope this helps.
> >
> >
> >
> > Please feel free to contact me at any time with any
> > questions or concerns that you may have. I am looking forward to
> > hearing from you soon.
> >
> >
> >
> > Sincerely,
> >
> > Craig T. Dedo, CAPM
> >
> > 17130 Burleigh Place
> >
> > PO Box 423 Mobile Phone: (414)
> > 412-5869
> >
> > Brookfield, WI 53008-0423 E-mail: <craig at ctdedo.com>
> >
> > USA
> >
> > LinkedIn: http://www.linkedin.com/in/craigdedo
> >
> > -----Original Message-----
> > From: J3 [mailto:j3-bounces at mailman.j3-fortran.org] On Behalf Of Van
> > Snyder via J3
> > Sent: Friday, November 30, 2018 04:03
> > To: j3
> > Cc: Van Snyder
> > Subject: [J3] 18-242
> >
> >
> >
> > The minutes for meeting 217 report that 18-242 was dismissed by JOR
> >
> >
> >
> > Seemingly syntax-sugar change would have semantic effect.
> >
> > No value seen.
> >
> >
> >
> > My notes record only an announcement "no action" on Monday AM.
> >
> >
> >
> > JOR did not ask me for clarification or amplification. There was no
> >
> > discussion during plenary.
> >
> >
> >
> > Am I the only one who has colleagues who ask why (quoting 18-242):
> >
> >
> >
> > In <ac-implied-do-control>, the <ac-do-variable> can be
> > preceded
> >
> > by <integer-type-spec> ::.
> >
> >
> >
> > In <data-implied-do>, the <data-i-do-variable> can be
> > preceded
> >
> > by
> >
> > <integer-type-spec> ::.
> >
> >
> >
> > In <concurrent-header>, the <concurrent-control-list> can be
> >
> > preceded by <integer-type-spec> ::.
> >
> >
> >
> > This is not possible in <loop-control> or <io-implied-do>.
> >
> >
> >
> > We clearly decided there was value for the <ac-implied-do-control>,
> >
> > <data-implied-do>, and <concurrent-header> cases.
> >
> >
> >
> > In those cases, there is semantic effect that was originally
> > intended,
> >
> > not "discovered" by an interp.
> >
> >
> >
> > Why is there no value for the <loop-control> or <io-implied-do>
> > case?
> >
> >
> >
> > My colleagues see value for it, and marvel at the irregularity.
> >
> >
> >
> > Indeed, in Ichbiah's response to requirements for the language that
> >
> > became Ada, the ONLY form of what we call a DO construct declared a
> > loop
> >
> > index that had a scope of the construct. This is NOT A NEW IDEA!
> >
> >
> >
> > This sort of gratuitous and unnecessary irregularity mystifies
> > students,
> >
> > and invites computer-science professors to refuse to teach Fortran.
> >
> >
> >
> > Can we consider 18-242 at meeting 218? Not re-consider it, because
> > (in
> >
> > my opinion), it was clearly not fully understood at meeting 217. If
> > so,
> >
> > does it need to be re-submitted to the server as a 218 paper?
> >
> >
> >
> >
> >
> >
> >
> >
>
>
More information about the J3
mailing list