RE: SUO: RE: Charter vs. Consensus
Eric,
At 11:11 AM 6/27/2003 -0400, Eric Peterson wrote:
>Thanks for your thoughts, Adam.
>
>Please see below.
>
> > -----Original Message-----
> > From: Adam Pease [mailto:adampease@earthlink.net]
> > Sent: Friday, June 27, 2003 10:34 AM
> > To: Eric Peterson; Jon Awbrey
> > Cc: jim.s3@juno.com; standard-upper-ontology@ieee.org
> > Subject: RE: SUO: RE: Charter vs. Consensus
> >
> > Eric,
> > I would guess that because Dave and Murray funded projects like
>HPKB
> > and
> > RKF which use first (and higher) order logic, that they see an
>advantage
> > in
> > that degree of expressiveness. I know that Jim Hendler (the DAML PM)
>has
> > always said that "you can do a lot with a little semantics" and
>therefore
> > I
> > think it's fair to paraphrase him in saying that more than DL isn't
> > needed.
>
>[ELP] Are we allowed to admit to seeing some validity in someone else's
>points in this group;^) This is really noble, Adam, but you shouldn't be
>so out of step with the rest of us;^) You should ignore the valid
>points, clip them out of your replies and divert attention away from the
>valid points by niggling on nits. Or your could blithely, confidently,
>and dismissively state facts not in evidence and act like you're exempt
>from needing to supply such evidence ;^)
:-) I like to think that if I'm patient enough in being nice, that
everyone else will follow suit.
> I think I'm more conceptually aligned with Murray and Dave on
> > this
> > issue.
> > But I think we're off getting topic here, conjecturing about what
> > different funders may want. To get back on topic, do you prefer
>OpenCyc
> > to SUMO because it lacks rules? Do you see that as an advantage?
>
>[ELP] I do see the advantage of DL being a much more attainable and much
>more immediately useful as a *first release*. The semantic web is
>crying out for some solid sharable content description, and as you
>pointed out DL is all they can use from us right now. But under that
>rationale, we could trim SUMO down to DL.
Ah, but we have. You can get a DAML version of SUMO at
<http://ontology.teknowledge.com>.
>So as to your first question, no. I prefer OpenCyc because it has a
>track record and I have used it productively for six years.
Hmm. That's a valid approach for you as an individual - without compelling
incentive for change, stick with what works for you. But it's not a valid
argument for a group. Why vote against SUMO, if your main reason for not
using it is simply that you've been using something else?
Adam
>Cheers,
>
>-Eric
>
> >
> > Adam
> >
> > At 09:10 AM 6/27/2003 -0400, Eric Peterson wrote:
> > >Hi Adam;
> > >
> > >In my conversations with Dave Gunning and Murray Burke (the DARPA
>HPKB
> > >program managers) they never let on to a lack of enthusiasm for or
> > >confidence in the DL-subset of logic or of Cyc. But you know them
> > >better than I do. Did they ever express such feelings in your
>hearing?
> > >
> > >Does one invest in the career of an athlete with an unsound skeleton
> > >because of the athlete's musculature? The US army would even reject
> > >someone with an unsound skeleton.
> > >
> > >FWIW,
> > >
> > >-Eric
> > >
> > > > -----Original Message-----
> > > > From: Adam Pease [mailto:adampease@earthlink.net]
> > > > Sent: Thursday, June 26, 2003 10:52 AM
> > > > To: Eric Peterson; Jon Awbrey
> > > > Cc: jim.s3@juno.com; standard-upper-ontology@ieee.org
> > > > Subject: RE: SUO: RE: Charter vs. Consensus
> > > >
> > > > Eric,
> > > > I'm not claiming that OpenCyc isn't a subset, or isn't
>consistent.
> > >I
> > > > am
> > > > questioning your line of reasoning that financial support for the
>full
> > > > product constitutes an endorsement of the subset.
> > > >
> > > > Adam
> > > >
> > > > At 08:15 AM 6/26/2003 -0400, Eric Peterson wrote:
> > > >
> > > > >Thanks for your thoughts, Adam.
> > > > >
> > > > >I worked with one of the former Cyclists that excised OpenCyc
>from
> > > > >full-Cyc. She never gave me the least indication that OpenCyc
>was
> > > > >anything but a subset.
> > > > >
> > > > >Are you aware of any reasons that we should suspect that the
>subset
> > > > >lacks the robustness and soundness of the whole? Can you point
>us to
> > > > >any full-Cyc axioms that would support such a suspicion?
> > > > >
> > > > >
> > > > >Best,
> > > > >
> > > > >-Eric
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Adam Pease [mailto:adampease@earthlink.net]
> > > > > > Sent: Wednesday, June 25, 2003 6:53 PM
> > > > > > To: Eric Peterson; Jon Awbrey
> > > > > > Cc: jim.s3@juno.com; standard-upper-ontology@ieee.org
> > > > > > Subject: RE: SUO: RE: Charter vs. Consensus
> > > > > >
> > > > > > Eric,
> > > > > > I agree. The fact that Cyc has been funded for a long time
> > > > >certainly
> > > > > > gives it some credibility and is one data point in any
>decision
> > > > >process
> > > > > > about which ontology to choose. One might say though that the
> > >funding
> > > > >has
> > > > > > been for a product which includes non-DL axioms, so whether
>the
> > > > > > consideration of funding history supports OpenCyc or not would
> > >still
> > > > >be in
> > > > > > question.
> > > > > >
> > > > > > Adam
> > > > > >
> > > > > > At 04:43 PM 6/25/2003 -0400, Eric Peterson wrote:
> > > > > >
> > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: Adam Pease [mailto:adampease@earthlink.net]
> > > > > > > > Sent: Wednesday, June 25, 2003 3:33 PM
> > > > > > > > To: Eric Peterson; Jon Awbrey
> > > > > > > > Cc: jim.s3@juno.com; standard-upper-ontology@ieee.org
> > > > > > > > Subject: Re: SUO: RE: Charter vs. Consensus
> > > > > > > >
> > > > > > > > Eric,
> > > > > > > > No one is "censuring" anyone else, but I believe that
> > >axioms
> > > > >are
> > > > > > >needed
> > > > > > > > to meet the charter. You don't. I don't know if you're
> > >asserting
> > > > > > >that
> > > > > > > > they're not beneficial though so that could still be a
>reason
> > >for
> > > > > > >choosing
> > > > > > > > one "charter compliant" offering over another.
> > > > > > >
> > > > > > >[ELP] I'm participating in this group on my customer's nickel
>and
> > >am
> > > > > > >trying to influence this group in a direction that will
>support
> > >my
> > > > > > >customer's needs. My customer needs a common model for
> > >integrating
> > > > > > >scads of disparate DB's. This is my agenda.
> > > > > > >
> > > > > > >I learned logic and Prolog nearly two decades ago and
>continue to
> > > > >revel
> > > > > > >in the miracle of writing specifications that are also code.
> > >I've
> > > > >even
> > > > > > >contemplated celebrating Frege's birthday and I glory in what
>FOL
> > >is
> > > > > > >going to do for us in the coming years. But two decades
>later,
> > >the
> > > > > > >marketplace still isn't ready to embrace anything close to
>full
> > >FOL.
> > > > > > >
> > > > > > >I claim, therefore, that a DL-based upper ontology has by far
>the
> > > > >best
> > > > > > >claim on being our initial version-one deliverable. If this
>is
> > > > >true,
> > > > > > >DL-ness is no detriment to being considered as a possible
> > > > >starting-point
> > > > > > >for our work.
> > > > > > >
> > > > > > >If not true, choosing an upper ontology with a *long* track
> > >record in
> > > > > > >DARPA's marketplace sounds to me like the best foundation on
> > >which we
> > > > > > >may add non-DL axioms.
> > > > > > >
> > > > > > >Your small group has doing something really amazing in
>getting
> > >the
> > > > > > >notice and some degree of blessing from the ontological
> > >community.
> > > > >But
> > > > > > >you haven't paid your dues yet. When DARPA places their
>loving
> > >arms
> > > > > > >around SUMO for a decade or so, let's talk more about why you
> > >don't
> > > > > > >think OpenCyc would be a good starting point for the SUO's
>work.
> > > > > > >
> > > > > > >FWIW,
> > > > > > >
> > > > > > >-Eric
> > > > > > >
> > > > > > > > I do agree with you that we should have a charter and
> > >follow
> > > > >it. I
> > > > > > > > take
> > > > > > > > much of the controversy in this group to have at its root
>that
> > > > > > >different
> > > > > > > > groups of people have different goals. Rather than
>impeding
> > >each
> > > > > > >other, I
> > > > > > > > think it would be much better to codify charters that
>express
> > >the
> > > > > > >goals of
> > > > > > > > a small number of groups and then have this group split
>up,
> > > > >amicably.
> > > > > > > >
> > > > > > > > Adam
> > > > > > > >
> > > > > > > > At 12:57 PM 6/25/2003 -0400, Eric Peterson wrote:
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > > > -----Original Message-----
> > > > > > > > > > From: Jon Awbrey [mailto:jawbrey@oakland.edu]
> > > > > > > > > > Sent: Wednesday, June 25, 2003 11:01 AM
> > > > > > > > > > To: Eric Peterson
> > > > > > > > > > Cc: jim.s3@juno.com; standard-upper-ontology@ieee.org
> > > > > > > > > > Subject: Re: Charter vs. Consensus
> > > > > > > > > >
> > > > > > > > > > o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
> > > > > > > > > >
> > > > > > > > > > Eric,
> > > > > > > > > >
> > > > > > > > > > This brings up yet another facet of the objective of
> > >achieving
> > > > > > > > >consensus,
> > > > > > > > > > as it begins from a state of disconsensus, namely,
>that
> > >many
> > > > >of
> > > > > > >the
> > > > > > > > >words
> > > > > > > > > > and phrases of the charter, or scope and purpose
>document,
> > > > >have
> > > > > > >just
> > > > > > > > >as
> > > > > > > > > > many different ways of being read from different
> > >perspectives
> > > > >in
> > > > > > >this
> > > > > > > > > > (dis-)community as any of the other words and phrases
>that
> > >we
> > > > >have
> > > > > > > > > > to discuss in the content-oriented and method-oriented
> > > > >subtasks.
> > > > > > > > >
> > > > > > > > >[ELP] Language can be tightened down. Clarity can
>certainly
> > >be
> > > > >an
> > > > > > > > >iterative process sometimes.
> > > > > > > > >
> > > > > > > > > > And though referring to the prior compact certainly
>helps
> > >to
> > > > > > > > > > reduce uncertainty from time to time, it cannot, in
>the
> > > > > > > > > > absence of consensual interpretations, achiving which
> > > > > > > > > > is a part of the original problem, reduce it all that
> > > > > > > > > > much, nor ever absolutely, when it comes to that.
> > > > > > > > >
> > > > > > > > >[ELP] Aren't you arguing against all codified law and
> > >governance?
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > > All subcommunities, some of the time, and some
> > > > > > > > > > subcommunities, all of the time, would like
> > > > > > > > > > to appoint themselves the sole-sufficient
> > > > > > > > > > interpreters of the charter, but there
> > > > > > > > > > is no reason to expect that they will
> > > > > > > > > > happen in a free and open society.
> > > > > > > > >
> > > > > > > > >[ELP] If it is a point that really matters, we can vote
>on
> > >it.
> > > > >And
> > > > > > >all
> > > > > > > > >our votes need to be reflected somewhere.
> > > > > > > > >
> > > > > > > > >But when calling for concreteness and specificity in
> > >criticisms
> > > > >of
> > > > > > >Cyc,
> > > > > > > > >I was expecting references to more ontologically
>egregious
> > >sins
> > > > >than
> > > > > > >not
> > > > > > > > >having SUMO's way-cool axioms.
> > > > > > > > >
> > > > > > > > >With Adam, I was simply making the observation that it
>didn't
> > > > >seem
> > > > > > >fair
> > > > > > > > >to me to publicly censure Cycorp for being charter
>compliant
> > > > >under a
> > > > > > > > >reasonable interpretation of the Charter. I certainly
>wasn't
> > > > > > >attempting
> > > > > > > > >to Mirandize him prior to incarceration.
> > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > So what Jim said initially continues to be apt.
> > > > > > > > > > The question that remains is how to achieve this
> > > > > > > > > > consensus in a way that is genuine and not forced.
> > > > > > > > >
> > > > > > > > >[ELP] Jon, you seem to want to reinvent long-standing
>methods
> > >of
> > > > > > > > >committee organization. I don't have time to read your
> > >papers on
> > > > > > > > >consensus building. I prefer starting with Roberts and
> > >fixing it
> > > > > > >when
> > > > > > > > >it breaks. That is already our de facto practice.
> > > > > > > > >
> > > > > > > > >BTW, where are you getting the "F"-word from in your last
> > > > >paragraph?
> > > > > > >A
> > > > > > > > >charter is consensus and far from force.
> > > > > > > > >
> > > > > > > > >FWIW,
> > > > > > > > >
> > > > > > > > >-Eric
> > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > Jon Awbrey
> > > > > > > > > >
> > > > > > > > > > o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
> > > > > > > > > >
> > > > > > > > > > Eric Peterson wrote:
> > > > > > > > > > >
> > > > > > > > > > > Moreover,
> > > > > > > > > > >
> > > > > > > > > > > I was told that the charter was a passed motion. If
>so,
> > >it
> > > > >is
> > > > > > > > >binding
> > > > > > > > > > > and is amendable only by a 2/3 majority (RRO 10th
> > >Edition p.
> > > > >12,
> > > > > > > > >line
> > > > > > > > > > 25).
> > > > > > > > > > >
> > > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > > From: jim.s3@juno.com [mailto:jim.s3@juno.com]
> > > > > > > > > > > > Sent: Wednesday, June 25, 2003 6:54 AM
> > > > > > > > > > > > To: standard-upper-ontology@ieee.org
> > > > > > > > > > > > Subject: SUO: Charter vs. Consensus
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > All,
> > > > > > > > > > > >
> > > > > > > > > > > > Consensus is what matters. The PAR Scope and
>Purpose
> > > > >really
> > > > > > > > >doesn't
> > > > > > > > > > > > matter that much. It shouldn't prevent us from
>doing
> > > > > > >anything, or
> > > > > > > > > > > > force us to do anything else. If and when we
>finish a
> > > > > > >document
> > > > > > > > > > > > with enough consensus to pass an IEEE ballot, if
>it
> > > > >doesn't
> > > > > > > > > > > > match the Scope and Purpose, we simple amend it or
> > >submit
> > > > > > > > > > > > a new one.
> > > > > > > > > > > >
> > > > > > > > > > > > Jim
> > > > > > > > > >
> > > > > > > > > > o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o