Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

SUO: RE: RE: Vote on Merged Ontology as a 'Base Document' for SUO WG




Dear Ian,

I said below:

> MW: I have a high level model for a 4D viewpoint, a subset of 
> which might be a suitable start point for this. The main 
> issue is that it is in EXPRESS (ISO10303-11) an 
> entity-relationship language, and not KIF, so it would need 
> translation. Further, it is mostly structural at present with 
> relatively few axioms.

Of course I translated part of an earlier version of this into KIF in the
autumn in a paper entitled "Standard Upper Ontology: A Documentation Trial".
I just checked and was disappointed to find this hasn't made the SUO Web
site, but if you do not have it I can forward you a copy. Adam gutted the
document for the KIF content at some point and sent it to the SUO list.

Regards  
      Matthew
============================================
Matthew West
Operations & Asset Management
Shell Services International
H3229, Shell Centre, London, SE1 7NA, UK.
Tel: +44 207 934 4490 Fax: 7929 
Mobile: +44 7796 336538
E-mail: Matthew.R.West@is.shell.com
http://www.shellservices.com/
============================================

> -----Original Message-----
> From: West, Matthew MR SSI-GREA-UK 
> Sent: 17 January 2001 11:09
> To: 'Ian Niles'; Schoening, James R CECOM DCSC4I;
> Standard-Upper-Ontology (E-mail)
> Subject: RE: RE: Vote on Merged Ontology as a 'Base Document' 
> for SUO WG
> 
> 
> Dear Ian,
> 
> Thanks for your responses below. Despite voting "no" I do 
> appreciate the work you have done. I'm not usre what else you 
> could have done at this stage. I mostly think the question of 
> accepting the document as a Base Document was somewhat premature.
> 
> See comments below.
> 
> Regards  
>       Matthew
> ============================================
> Matthew West
> Operations & Asset Management
> Shell Services International
> H3229, Shell Centre, London, SE1 7NA, UK.
> Tel: +44 207 934 4490 Fax: 7929 
> Mobile: +44 7796 336538
> E-mail: Matthew.R.West@is.shell.com
> http://www.shellservices.com/
> ============================================
> 
> > -----Original Message-----
> > From: Ian Niles [mailto:iniles@teknowledge.com]
> > Sent: 16 January 2001 19:40
> > To: West, Matthew MR SSI-GREA-UK; Schoening, James R CECOM DCSC4I;
> > Standard-Upper-Ontology (E-mail)
> > Subject: RE: RE: Vote on Merged Ontology as a 'Base Document' 
> > for SUO WG
> > 
> > 
> > Matthew,
> > 
> > 	See my comments below.
> > 
> > -Ian
> > 
> > > -----Original Message-----
> > > From: West, Matthew MR SSI-GREA-UK 
> > > [mailto:Matthew.R.West@is.shell.com]
> > > Sent: Tuesday, January 16, 2001 2:31 AM
> > > To: Schoening, James R CECOM DCSC4I; 
> > Standard-Upper-Ontology (E-mail)
> > > Subject: SUO: RE: Vote on Merged Ontology as a 'Base 
> > Document' for SUO
> > > WG
> > > 
> > > 
> > > 
> > > Dear James,
> > > 
> > > As one of the more active participants, I think I should vote 
> > > publicly.
> > > 
> > > I vote NO.
> > > 
> > > I do think it is suitable as a working paper.
> > > 
> > > I also feel I should give some reasons.
> > > 
> > > The overall reason is that I would not use it as a start 
> > > point for anything
> > > that I am currently doing or about to do. I have only had 
> > > time to give a
> > > brief review of the combined document, however, the following 
> > > major sticking
> > > points exist for me.
> > > 
> > > 1. n-th Domain is still in there, though I'm pleased to 
> see that its
> > > siblings seem to have been dropped. N-th domain is equivalent 
> > > to writing a
> > > different language with KIF and then using that whilst 
> > > pretending it is
> > > still KIF. Further it is unnecessary, the equivalent 
> > > statements can be made
> > > in perfectly ordinary KIF.
> > 
> > Well, 'nth-domain' is really just syntactic sugar.  All 
> > formulas involving
> > this predicate can be recast, using the template provided in 
> > the "Structural
> > Ontology" section of the merged ontology document, into 
> > uncontroversially
> > first order language.  I don't see that this is a problem.
> 
> MW: Quite, so why not do it?
> > 
> > > 
> > > 2. The ontology has a 3D rather than 4D foundation. I can't 
> > > imagine going
> > > back from 4D to 3D, except to analyse and retrieve useful ideas.
> > 
> > Even if the merged ontology does indeed have a "3D" foundation, this
> > foundation could, it seems to me, be embedded within a "4D" 
> > framework.  If
> > you disagree, could you indicate why you think this is impossible?
> 
> MW: You will have seen that Pat has raised some doubts about 
> the possibility of doing this. The problem arises if 3D and 
> 4D viewpoints are extended subsets of each other. I am more 
> hopeful, in moving from a 3D world to 4D I have not noticed 
> anything I have lost, only things gained, so I am prepared to 
> bet that a 3D viewpoint can be mapped onto a 4D viewpoint. 
> What I am prepared to bet is some effort to try to make the 
> mapping - I believe in experiment as much as analysis for 
> making progress.
> 
> MW: This could get quite messy. For instance, what I pick up 
> from Pat is that a 4D object may not be a member of a 3D 
> class, but from a 4D object and its classes, we may be able 
> to infer a 3D object and its classes/properties etc, and 
> vice-versa (within the limits of what can be represented in 
> 3D and/or 4D).
> 
> MW: I have a high level model for a 4D viewpoint, a subset of 
> which might be a suitable start point for this. The main 
> issue is that it is in EXPRESS (ISO10303-11) an 
> entity-relationship language, and not KIF, so it would need 
> translation. Further, it is mostly structural at present with 
> relatively few axioms.
> > 
> > > 
> > > 3. The foundation is John Sowa's lattice of classes, and 
> > whilst I will
> > > defend another mans right to use whatever classes he chooses, 
> > > and am happy
> > > to find these classes interesting, for me they do not 
> provide a firm
> > > foundation. The primary reason for this is their basis in Pierce's
> > > Firstness, Secondness and Thirdness, concepts which in trying 
> > > to grasp them,
> > > remind me more of jelly than something I would care to build 
> > > on - each time
> > > I have thought I had them nailed down, they moved on me.
> > 
> > I agree that Pierce's tripartite distinction is a stumbling 
> > block for its
> > broad acceptance in an engineering context (whatever 
> > philosophical merits it
> > may have).  It is for precisely this reason that I removed 
> > the associated
> > notions (e.g. 'Mediating' versus 'Relative', 'Prehension', 
> > etc.) in the
> > latest version of the merged ontology.
> 
> MW: But then you are left with a bunch of classes without a 
> foundation at all. 
> 
> MW: With something I don't understand, I try to get down into 
> subtypes and instances. So from a 4D perspective if we can 
> get down to things that are relatively concrete, these could 
> be mapped to a 4D model, you could then see the superclasses 
> as defined by their subclasses and example members, and thus 
> get a better grip on Firstness, secondness, and thirdness.
> 
> MW: Be careful though, what you are then talking about is 
> really multiple ontologies, and how they relate to each 
> other. This would enable people to use the ontology most 
> relevant to their needs, but express the same thing in 
> another ontology, when that was more appropriate. I think 
> these are important things to (try) to do, but I wouldn't 
> want to suggest it was easy.
> > 
> > 
> > > 
> > > 4. The axioms on entity and class require that all classes 
> > > are a subtype of
> > > entity. This means that no class can have classes as members. 
> > > This is very
> > > restrictive.
> > 
> > I agree that this is a problem, and it is one that Pay Hayes 
> > noted as well.
> > I suggest that we proceed by asking John Sowa to present his 
> > view on why the
> > axioms should be retained.
> 
> MW: Did I miss something? I didn't see John propose this. If 
> retained, this would be more restrictive than entity 
> relationship modelling, and it was the restrictions there 
> that lead me to look at logic based approaches in the first place.
> > 
> > > 
> > > 5. I could create a similar list of questions and concerns as 
> > > Pat has done
> > > on the details of the ontology. I do not have the time this 
> > > would take at
> > > present - I can't decide whether that is fortunate or 
> > > unfortunate, and for
> > > whom.
> > > 
> > > This is not necessarily an exhaustive list, but for me it is 
> > > sufficient. 
> > > 
> > > Regards  
> > >       Matthew
> > > ============================================
> > > Matthew West
> > > Operations & Asset Management
> > > Shell Services International
> > > H3229, Shell Centre, London, SE1 7NA, UK.
> > > Tel: +44 207 934 4490 Fax: 7929 
> > > Mobile: +44 7796 336538
> > > E-mail: Matthew.R.West@is.shell.com
> > > http://www.shellservices.com/
> > > ============================================
> > > 
> > > > -----Original Message-----
> > > > From: Schoening, James R CECOM DCSC4I
> > > > [mailto:James.Schoening@mail1.monmouth.army.mil]
> > > > Sent: 15 January 2001 15:58
> > > > To: Standard-Upper-Ontology (E-mail)
> > > > Subject: SUO: Vote on Merged Ontology as a 'Base Document' 
> > > for SUO WG
> > > > 
> > > > 
> > > > 
> > > > SUO Participants,
> > > > 
> > > > 1. Attached is the latest version of Ian Niles'  Merged 
> > Ontology.  
> > > > 
> > > > 2. This message is a call for a vote as to whether this 
> > > > document should be
> > > > approved as a 'Base Document' for IEEE P1600.1 Standard 
> > > Upper Ontology
> > > > Working Group.  Within IEEE, there is no official 
> > > definition of a Base
> > > > Document; however, the definitions below from the IEEE 
> > > > Learning Technology
> > > > Standards Committee (our original sponsor) are as good as any.
> > > > 
> > > > 3. The purpose of this vote is to determine the level of 
> > > > consensus we have
> > > > for focusing work on this document.  In other words, we need 
> > > > to decide if we
> > > > are going to go down this path or not.  Multiple base 
> > documents are
> > > > possible, so anyone else is free to submit one.  
> > > > 
> > > > 4. The primary purpose of this vote is NOT to propose 
> > > > improvements to the
> > > > document, unless they are major enough for you to vote 
> > > > against focusing work
> > > > on this document.  However, such general improvement comments 
> > > > are welcome
> > > > and will be resolved (but not until after completion of this 
> > > > Base Document
> > > > vote). 
> > > > 
> > > > 5. Since we are still in the process of establishing official 
> > > > voting rights,
> > > > I suggest we first conduct this as an unofficial straw 
> > > poll, such that
> > > > anyone can cast a vote.  After resolving all comments, if 
> > > > consensus is not
> > > > clear, we will have the option of conducting a formal vote 
> > > > (once voting
> > > > rights are established).  
> > > >   
> > > > 6. Your options for voting are YES (comments optional), 
> > NO (comments
> > > > required), or ABSTAIN (comments optional).
> > > > 
> > > > 7. Please send your votes directly to me (to reduce list 
> > > > traffic), though
> > > > you may send them to the entire list if you desire.  
> > > > 
> > > > 8. Please vote by Jan 29, 2001. 
> > > > 
> > > > 9. Please send in any questions or comments.
> > > > 
> > > > Jim Schoening
> > > > Chair, SUO WG
> > > > 
> > > > IEEE LTSC Nomenclature of Documents 
> > > > working paper: Does not automatically get agenda time. Not 
> > > necessarily
> > > > related to standards words (e.g., liaison reports, the agenda 
> > > > itself). A
> > > > working paper is any document with a document number or 
> a document
> > > > introduced at a meeting (assuming the WG accepts it as a 
> > > > working paper).
> > > > Working papers don't automatically receive agenda time. The 
> > > > WG Chair may
> > > > choose to give higher agenda priority to discussion on 
> > > working papers
> > > > received two weeks prior to the meeting. (Members may object 
> > > > to discussion
> > > > on papers with less than two weeks notice, but the objection 
> > > > is based on the
> > > > need to digest the document, not to obstruct the process; and 
> > > > the objection
> > > > may prevent the paper from get agenda time, but not 
> > > > necessarily prevent the
> > > > discussion -- be considerate both ways, the chair will rule.) 
> > > > base document: Starting point for standards words (it's clear 
> > > > the document
> > > > *can* be transformed into standards words). Has no official 
> > > > weight (not a
> > > > standard). Might not be in standards wording. Might not be 
> > > > complete with
> > > > respect to the scope and/or final standard. Might be 
> multiple base
> > > > documents. Automatically allocated agenda time since this 
> > > is the main
> > > > product of the WG. Most base documents represent substantial 
> > > > work created
> > > > outside the committee and introduced into the WG. 
> > > > working draft: The base documents (assuming more than one) 
> > > > are combined over
> > > > time. The combination document now becomes a working draft. 
> > > > The working
> > > > draft is refined and polished. The working draft is adapted 
> > > > to the style
> > > > guides. Working drafts may be available to the public. 
> > Automatically
> > > > allocated agenda time (the rest below all have agenda time). 
> > > > base document vs. working draft: There may be multiple base 
> > > > documents that
> > > > become the starting points for a standard. There is only one 
> > > > working draft
> > > > for a standard (but many revisions, each with its own 
> > > > document number). A
> > > > base document identifies the origin of the work for 
> > several reasons:
> > > > technical, rationale, intellectual property, acknowledgement. 
> > > > committee draft: The working draft is now in the style of a 
> > > > standard. The
> > > > document substatially represents the intended final standard. 
> > > > Changes may be
> > > > made. There may be several committee drafts. 
> > > > working draft vs. committee draft: Each revision of the 
> > > working draft
> > > > represents a batch of edits that the WG accepts. Typically, 
> > > > there is a new
> > > > working draft each meeting while the document is being 
> > > developed, but
> > > > working drafts may be less frequent. Committee drafts are 
> > important
> > > > milestones because they represent to the public "this is what 
> > > > the final
> > > > standard will look like". Working drafts may be available or 
> > > > not to the
> > > > public. Committe drafts are public. Working drafts may be 
> > > > used to solicit
> > > > public comments. Committee drafts are always used to 
> > solicit public
> > > > comments. 
> > > > draft standard: The document is *complete*. There may be 
> > > > minor editoral
> > > > changes, but no substantive changes are intended. Draft 
> > > > standards may be
> > > > referred to (i.e., incorporated by reference) in *other* 
> > standards. 
> > > > approved standard: The document is now an official standard. 
> > > > Summary: Committee members submit working papers. Working 
> > > > papers may be
> > > > related to committee work, technical contributions, and 
> > so on. Base
> > > > documents are working papers approved as a "base document". 
> > > > One or more base
> > > > documents may be part of a standard. When base documents are 
> > > > revised they
> > > > become a new working paper but with the automatic agenda 
> > > > item. At some point
> > > > the working papers and/or base documents are combined to 
> > > > become a working
> > > > draft. Later on, additional working papers and additional 
> > > > base documents may
> > > > be added to the working draft. Working drafts may be made 
> > > > available for
> > > > public comment. Committee drafts represent the full 
> > intention of the
> > > > standard. Committee drafts are available for public comment 
> > > > and are likely
> > > > to be balloted (comments from voting and non-voting 
> > > members). A draft
> > > > standard represents the completed standard in every way 
> > > > (wording, style,
> > > > layout, etc.) except that it is not official. An approved 
> > > > draft standard
> > > > becomes a official standard after a succesful ballot and a 
> > > > process review. 
> > > > 
> > > >  <<Merge.txt>> 
> > > > 
> > > 
> > 
>