[SCL] Re: Common Logic status with respect to standards

[pat hayes]

>Pat,
>
>I am willing to do the work wherever the action is.

A CG concrete syntax for SCL would be an excellent start. Also if you 
had any comments on the proposal for a 'role' syntax, I would 
appreciate them.

>
>>John, as I keep reminding y'all, I do not own and am not responsible
>>for the CL project. That is Mike's baby. I am concerned with SCL, not
>>CL.
>
>I do not know what problems you perceived with the CL group,
>but I cannot see any difference between the CL and SCL groups
>as far as progress or lack thereof is concerned.

The CL group, in the year before I formed the SCL group, had done 
literally nothing: it had not met, not made any decisions, and not 
written anything. It had to all intents and purposes ceased to exist. 
After a considerable effort of nagging had failed to resuscitate it, 
I gave up on it and started a new group, with some overlap with the 
CL group but not the same group, either in membership or in goals. I 
think it is very important that the differences between the SCL and 
CL groups and initiatives be emphasized in any discussions. If you 
have been telling Nancy and others that SCL is simply a continuation 
of CL, then I would ask you formally to make the true situation clear 
to whoever may have been misled by this. The politics of ISO approval 
may be relevant to this matter, so it is not without importance. I am 
quite willing to explain this to anyone who may be interested, 
directly.

I refuse to be held responsible for the morbidity of CL or to have 
the SCL project blamed for CL's delay. In the approximately 10 months 
we have been working, the SCL group has designed and drafted a 
complete abstract syntax, designed an overall framework for defining 
concrete syntaxes, revised the model theory in a new way to solve a 
new problem noted by Horrocks, aligned the SCL and RDF model 
theories, discussed and almost solved the issue of how best to map 
SCL+= into SFOL with axiomatized =y, and begun an intensive 
discussion on how to construct an SCL XML syntax, none of which 
topics were even remotely approached by the CL group.  As best as I 
recall, the CL group spent its last 24 months arguing about whether 
or not the syntax should allow empty conjunctions, and failing to 
agree even on that.

>JS>>And as Nancy pointed out
>>
>PH>When did that happen, by the way?
>
>Nancy, Harry, I, and miscellaneous other people were discussing
>the glacial pace toward a standard at the ICCS meeting in
>Dresden at the end of July.

At that time, the SCL group was eight months old. I do not think that 
our progress has been glacial.

>Harry was willing to carry the
>ball in the ISO court, and Nancy arranged to get funding for
>him to do so.  I sent a note about that in August.

The message is dated 9 September, a full month after the Dresden 
meeting.  BTW, I had not realized that there was funding involved. I 
could have used some of that myself :-)

John, I have to say that I do rather find your actions somewhat 
lacking in due process.  First, it seems evident that you have been 
misrepresenting SCL as a continuation of CL. Second, you could have 
suggested *first* to this group that Harry be invited to join it, 
before arranging funding for him to serve in this role and inviting 
him to be a member on your own initiative. I have no reason to think 
he will not do an excellent job, but I would like to have been 
consulted on the need to involve him in the work and on what his role 
was likely to be, and even maybe had the opportunity to debrief him 
on some of the overall goals and technical matters involved. I would 
remind you that you are not the convener or chairman of this group, 
and although I confess that our progress has not been as rapid as I 
could have wished, I have to note that you hardly have the moral high 
ground to complain of lack of progress, in that that you have not, to 
my knowledge, so far made any contributions to the technical work of 
the group *at all*. I would be delighted if you would.

>  >Sure, but we need to ensure that it is indeed 'we' who are doing
>>this. That is, we need to speak for the group when telling people
>>what we are doing, or at least carefully distinguish our opinions
>>from the group's decisions.
>
>Great!  Let's get a compendium of the group's decisions on
>the CL web site so that we can point to it.
>
>>They were not the only ones waiting, which is why I formed the SCL
>>group out of frustration.
>
>It's the same people who are doing the work.   And putting
>the letter S in front of the acronym doesn't do anything
>to cause them to move faster.

It is not the same people, except to the extent that the only actual 
technical work done for the previous 2 years in the CL activity was 
done by Chris and me.  Now Tanel has joined us, at least.

>  >The SCL group has been in existence only for less than a year, has no
>>funds or external support, and has formally met only once. Under the
>>circumstances I think we are doing reasonably well.
>
>Those attributes are also true of the CL group, since they're
>basically the same people.

No, they are not. I will not rehearse in this public forum the 
differences in composition between the CL and SCL groups, but they 
can be researched by anyone interested. The SCL group is open to 
volunteers, and I believe I am correct in saying that none of the CL 
group who were not explicitly invited to join the SCL group have 
volunteered.

>  > ... Who ARE these 'ISO people'? Why are they not
>>mentioned by name, perhaps with their emails attached so that the
>>points can be discussed with them and maybe even an archived trail of
>>discussion laid down?
>
>Pat, you have been invited to attend their meetings and never
>bothered to show up.

Bothered? I have exerted a considerable amount of 'bother', as you 
put it, to attend the only ISO meeting to which I have been invited, 
at my own expense, only a matter of weeks after forming the SCL 
group. It is financially impossible for me to attend overseas 
meetings of the ISO.

>  You were in Santa Fe where the meetings
>were held and left on the day before you could meet them.

As you know, the meeting was rescheduled at the last minute by the 
local organizers, with no advance warning and ignoring an explicit 
appeal: so your implication that my nonattendance was the result of 
my casual indifference is a calumny.  Chris and I left at the times 
we had scheduled our exiting flights, on your advice I might add, 
some time in advance. It was not possible to reschedule a flight at 
that late stage, particularly when I was paying for the damn thing 
myself.

>Harry has been going to their meetings, so he can forward
>any emails to you.

I would be grateful for such forwarding, or indeed any insight into 
exactly what is being said, and by whom.

>Basically, their concerns boil down
>to one simple question:  (1) Where is the [S]CL document?

I doubt if there will ever be a CL document, but that is for others 
to decide. The SCL document will appear when we have the technical 
work done to our satisfaction, which will be soon, but no promises. 
I might add that this technical work is entirely new and was 
conspicuously not done either by the KIF group or the CL group; in 
fact, the problems that are being dealt with were not even considered 
by those groups.

>  >... the central semantic problems with KIF that we have at long
>>last laid to rest or avoided in the SCL model theory, matters that
>>require some delicacy and care in the use of wording; so I would
>>prefer that, as far as possible, any discussion that bears on
>>technical SCL matters is done with me or Chris personally involved,
>>at least until we get the technical matters documented unambiguously,
>>which we have not yet completely succeeded in doing.
>
>I have been well aware of those issues and the approach that
>you and Chris have been taking.  I finally came around to
>accepting the typeless semantics.

As far as I know, there has never, at any time, been a proposal for a 
typed semantics in the CL or SCL discussions, or even in KIF.

>  I have never been happy
>with seqvars and have always wanted true sequences instead.
>However, I am willing to go along with seqvars as a solution
>to a problem that I believe was a mistake in KIF (and which
>I would prefer to correct rather than perpetuate).
>
>>It seems premature to be composing an ISO-compliant document in time
>>for Dec 17th when there are many technical issues still to discuss,
>>however.
>
>I never said that we need a finished document with all the
>ISO boilerplate by Dec 17th (or 12th).

Harry did, though, and he is the editor.

>  But I believe we can
>produce something better than what's now on the CL web site.
>
>Can we schedule a telecon?  Perhaps for Nov 11?

I cannot make the 11th. Could we use a different day, later that 
week? But in any case, let us first have some suggestions for an 
agenda, or there is little point in having a telecon.

>  Between now
>and then, we should have everybody send in a short summary
>of what they plan to do for the Dec 12th document and what
>issues they feel should be addressed in the telecon.

I have already suggested a quick breakdown of the work that needs to 
be done and by whom. Murray, for one, seems to be taking up the 
challenge. I hope to have my piece done by mid-week, but other work 
is pressing.

Pat

>John


-- 
---------------------------------------------------------------------
IHMC	(850)434 8903 or (650)494 3973   home
40 South Alcaniz St.	(850)202 4416   office
Pensacola			(850)202 4440   fax
FL 32501			(850)291 0667    cell
phayes at ihmc.us       http://www.ihmc.us/users/phayes