N276 - ISO/IEC JTC1 SC22 WG17


Minutes WG17 meeting 2017-08-07 via WebEX

Ulrich Neumerkel,
ISO/IEC JTC1 SC22 WG17 (Prolog) Convener

0.1 Annex
A Mailadress list

The meeting was held via WebEx, Monday, August 7th 17:00-19:30 UTC

1. Greeting and introduction of participants, administrativa

Mats Carlsson     (Sweden),  (left early)
Michael Covington (US),      (left early)
Klaus Daessler    (Germany), (left early)
Stefan Kral       (Austria), commitee member (joined late)
Per Mildner       (Sweden)
Ulrich Neumerkel  (Austria), convenor
Joachim Schimpf   (UK),      commitee member
Markus Triska     (Austria), commitee member

2. (N274:1) Agenda

Since at least one participant had to leave early, it was unanimously
agreed that the order of items of the second agenda N274 were changed
as below.  JS raised the issue that the meeting had been announced
only two weeks in advance, wondering about the consequences.

After a vote on an unclear question, JS proposed:

Resolution N275:R1

| The WG thanks UN for organizing this first online meeting of the
| WG. The meeting was, however, called at short notice and there was
| insufficient time to prepare and submit material. The advance
| notification of 2 weeks falls far short of the 6 weeks required by
| ISO, and also the 12 weeks that JTC1 demands for advance distribution
| of the agenda.
|
| The meeting should therefore be considered irregular, and resolutions
| on agenda items only taken if no meeting participant objects on the
| grounds of this insufficent notice.

Yes: 3, No: 5.

JS noted that this vote was problematic, because members who could not
participate due to the short notice were not being asked.  UN
clarified that all votes are considered straw-votes anyway.

(SK shows up as WebEx participant, no audio)

3. (N274:2) Passing of Alain Colmerauer

4. (N274:7) Convener's reports to ISO SC22, 2016 and 2017 this meeting.

JS reviewed the convener's report of 2016 and criticized misleading
statements that went beyond positive spin.  JS wants to see reports at
least two weeks before they are submitted to SC22.  UN agrees to
circulate future reports at least two weeks before sending them to
SC22, but will finally deliver the report to SC22 without any formal
consent.

5. (N274:8) State of the WG and future, endorsement of convenor
for the next period 2018-2020.

(MiC left before the first question round)

After a discussion about the short time span, UN and JS propose for
convenorship presenting their intentions:

JS intends as convenor to finish current work and then
disband group.

UN intends to continue work and does not want to disband the group.

Votes: UN:2,  JS:1, abstain: 1.  (Both candidates did not vote)

6. (N274:4) DCG - editorship of overdue project
After Jonathan Hodgson resigned, the editorship is vacant.
Discussion of further action including nomination of editor.

JS explained (see N275:4) that after 11 years the project was clearly
overdue and qualified for cancellation according to ISO directives.
He also called attention to the fact that the intended outcome
of the project was a Technical Report, which has no normative power.
JS further expressed the view that the long gestation period of the
project had caused vendors to wait for the results, preventing them to
address incompatibilities earlier.

KD disputed this, saying that, on the contrary, vendors had used the
existing drafts to eliminate problems.  UN gave as example
steadfastness issues and asked JS to name further unknown issues.

One view (KD) was that the quality of DCGs has been improved over the
years.  The other view (JS) was that the WDTR was too ambitious.  JS
suggested there were two valid views, DGC-as-a-language and
DCG-as-syntactic- sugar, and that the latter might be easier to agree
on, possibly even without the need for ISO involvement.  UN remarked
that WG17's original DCG effort was just a translator but that this
effort failed and was replaced by the current approach.

(At 20:00 SK joined audiowise)

Resolution N275:R4

| WG17 cancels the work item ISO/IEC DTR 13211-3, which is overdue and
| would anyway not have resulted in a normative document. WG17 instead
| appeals to the community of Prolog vendors to agree rapidly on a
| common portable DCG functionality, perhaps using recent TR drafts as
| guidance. WG17 will review the situation after a year, and if
| unsatisfactory, consider a new project.

Yes: 4, no: 4.  (Included is a vote by chat/mail from MiC)

KD stated that he would try to solicit the cooperation of external
experts to break the deadlock.  JS stated that he would work on a
simplified solution, but had lost hope in the current TR's approach
and did not want to put future effort into it.

7. (N274:5) Amendment to part II, modules

N251 is the first "draft amendment to 13211-2 Modules to allow a
common subset of SICStus, YAP, and SWI Prolog implementations to be
conforming" Res. A1, 2013, Istanbul.  Discussion of further action.

N251 is actually a preparation for an NWIP or amendment.
Inconclusive discussion about 13211-2:2000. There is a need for
having a module standard, but currently it is unclear how to
proceed.  WG17 consensually agreed to defer further action.

KD reminded in this context that IF/Prolog implements 13211-2.

8. (N274:6) Structure and content of WG-related web resources
   Official documents (N-documents) now have to be uploaded on
   iso.org/jtc1sc22wg17.  Problem of outdated comments on the Web.

JS presents his concerns detailed in N275:8.  UN remarks he has
updated his personal pages to meet JS's concerns with respect to ISO
and WG17 and refers back to N250:6b Minutes Istanbul for inclusion
criteria.  JS confirmed that his concerns regarding ISO and
conformance assessments had been addressed.  The discussion then
continued about the precise relation to WG17 of UN's personal site and
comparison pages which are under the sole control of UN.  The
subsequent discussion was inconclusive.

(MaC, KD left during the discussion.)

JS agreed to let the matter rest, and no action was raised.

9. (N274:3) Corrigenda
   After successful publication of Cor.3 in July, the Draft for further
   corrigenda has been updated and uploaded as N273.
   Discussion of further action.

UN underlines that N273 contains many old items that most probably
should be removed.  An item-wise discussion was deferred.
JS elaborated on N275:2 and proposed:

Resolution N275:R2

| In future, the WG shall make sure that corrigenda conform to the rules
| and do not introduce new features. New features shall only be
| introduced in a rule-conforming manner, either via specific amendment
| projects, or via a revised standard. Such projects will have to go
| through the normal approval process, making sure they are sufficiently
| resourced by P-members.

Yes: 2, No: 3.

10. Dresden Minutes were approved.

11. The next annual meeting will be begin of June 2018.

The meeting adjourned at 21:33 CEST, 19:33 UTC.

Annex A Mailadress List

Mats Carlsson     matsc@sics.se
Michael Covington mc@UGA.EDU
Klaus Daessler    klaus.daessler@mathint.com
Stefan Kral       skral.sk@gmail.com
Per Mildner       per.mildner@ri.se
Ulrich Neumerkel  ulrich@complang.tuwien.ac.at
Joachim Schimpf   jschimpf@coninfer.com
Markus Triska     markus.triska@GMX.AT