- Peter Haase, Boris Motik, Bernardo Cuenca Grau, Steve Battle, Jeremy Carroll, Peter Patel-Schneider, Uli Sattler, Michael Smith, Bijan Parsia, Giorgos Stoilos, Pascal Hitzler, Markus Krötzsch, Vit Novacek, Ivan Herman, Sandro Hawke, Deborah McGuinness, Evan Wallace, Rinke Hoekstra, Ian Horrocks, Ratnesh Sahay, Alan Ruttenberg
- Achille Fukoue, Elisa Kendall, Jim Hendler, Joanne Luciano, Vipul Kashyap, Zhe Wu
- Observers/Guests (Full-Time and Part-Time)
- Alan Rector, Peihong Ke, Robert Stevens, Sean Bechhofer, Sebastian Brandt, Thomas Schneider, Dimitri Tsarkov, Matthew Horridge, Carsten Lutz
- Anne Cregan, Fabian Neuhaus, Fabien Gandon, Martin Dzbor, Tommie Meyer
- Ian Horrocks, Alan Ruttenberg
Session 1 (Day 1, 9:00 - 10:45)
The text of this section is automatically included from F2F1 Minutes Session 1. Edits should be done there.
Welcome, Logistics, Introductions
Bernardo Cuenca Grau
Carsten Lutz, Dresden (guest, joining)
Thomas Schneider (guest)
Sebastian Brandt (guest)
Matthew Horridge (guest)
Sean Bechhofer (guest from Manchester)
Robert Stevens (guest from Manchester)
Ian Horrocks (University of Manchester [sic])
OWL 1.0 Implementation Experience
Speaker - Matthew Horridge - slides
Matthew Horridge: implementing OWL DL experience
Matthew Horridge: problems - RDF and imports
Matthew Horridge: internal API is known as the OWL API - based on OWL abstract syntax
Matthew Horridge: use of OWL API means that different concrete syntaxes can be used
Matthew Horridge: problems with abstract syntax - distinguishing between, e.g., data and object properties
Matthew Horridge: effort required - RDF parser is vast majority of effort, everything else is much easier
Matthew Horridge: similar situation for OWL 1.1 API
Ivan Herman: what is the "RDF parser"
Matthew Horridge: RDF parser is just triples to internal API, not dealing with RDF/XML
Matthew Horridge: RDF mapping - want to be fast, small, and streaming
Matthew Horridge: streaming was too hard, so the parser was not streaming
Matthew Horridge: in new parser - parser is streaming, but still takes resources
Matthew Horridge: OWL XML is very verbose - causes problems
Jeremy Carroll: what is the size increase
Matthew Horridge: not sure - 3 to 5 times
Matthew Horridge: triples to OWL API was problematic - inversing a non-deterministic mapping
Matthew Horridge: OWL 1.1 thus has two mappings
Matthew Horridge: e.g., subclass (see slides)
Matthew Horridge: other problem - n-ary constructs go to n or n*n triples
Matthew Horridge: failures of round tripping cause problems
Matthew Horridge: missing type triples make ontologies officially non-parsable
Matthew Horridge: in many cases there is a fix, but sometimes the fix is not local (may require looking at imported ontologies)
Jeremy Carroll: declaration is good style - RDF graphs are unordered - so declarations can be non-local
Michael Smith: searching for declarations require two passes - which can be expensive
Alan Ruttenberg: why not do typing "as seen"
Boris Motik: this requires deferring processing, and is hard
Bijan Parsia: in any case, there is a lot of extra work to make the RDF parsing go through
Jeremy Carroll: there are implementations that do good jobs
Bijan Parsia: no - there are bugs
Jeremy Carroll: but you do get benefits - use of RDF
Bijan Parsia: but there is a cost
Jeremy Carroll: multiple vocab is an attempt to fix this?
Bijan Parsia: yes
Alan Ruttenberg: compatibility means that there is no way out
Ian Horrocks: if we make a better way, then the old versions will die out (eventually)
Matthew Horridge: imports issues
Matthew Horridge: if //...foo.... imports //...bar... what does it mean?
Matthew Horridge: name of an ontology or a location of an ontology
Matthew Horridge: what if the name and the location don't match
Matthew Horridge: imports on OWL DL is controlled by OWL S&AS 3.4
Matthew Horridge: OWL reference says imports is by location
Matthew Horridge: OWL guide says something confusing
Matthew Horridge: want some direct and normative statement
Matthew Horridge: solution was by name (essentially)
Matthew Horridge: what is the name of an ontology?
Matthew Horridge: guide is confusing
Matthew Horridge: test cases for OWL 1.0 were very useful - we need them for 1.1
Matthew Horridge: OWL 1.1 SS diagrams were useful
Jeremy Carroll: we need some time to talk about testing
Bijan Parsia: yes - infrastructure is needed
Jeremy Carroll: probably only need a short amount of time
Alan Rector (guest from Manchester)
Ratnesh Sahay (DERI Galway)
History (including OWLED)
Speaker - Bijan Parsia (no slides)
Bijan Parsia: OWLED - started in 2005 to let people interested in OWL design and use together
Bijan Parsia: there was about 1.5 years of experiece in OWL - there were complaints (particularly QCRs and datatypes)
Bijan Parsia: idea for a "bug fix" update to OWL - things that are relatively easy and wanted
Bijan Parsia: particularly wanted by users
Bijan Parsia: workshop was (largely) to discover what this new version would be
Bijan Parsia: workshop was adjacent to ISWC 2005 - about 60 participants
Bijan Parsia: desiderata for changes:
Bijan Parsia: 1/ requested by major users
Bijan Parsia: 2/ have effective reasoning methods
Bijan Parsia: 3/ will be implemented
Bijan Parsia: (alternatively commitment from users, well understood, committment from implementers)
Bijan Parsia: other goals: quiet whining, promote apps, improve spec, move forward, path for extensions, reduce species confusion (particularly DL/Lite)
Bijan Parsia: example - move OWL-S to OWL DL
Bijan Parsia: tool feature - coercion to OWL DL in Pellet
Jeremy Carroll: tools *should* do this
Bijan Parsia: *mostly* get the right thing
Bijan Parsia: there are still things that people want to do - e.g., lists
Bijan Parsia: I implemented shadow lists -- RDF Lists with a parallel vocabulary, just a different namespace. Pellet can do this silently.
Jeremy Carroll: can rdf:list be fixed?
Bijan Parsia: issues of modelling lists (eg, breaking them) affecting syntax
Alan Ruttenberg: are lists necessary?
Bijan Parsia: yes
Peter Patel-Schneider: Lists are only in RDF because OWL-WG demanded them.
Peter Patel-Schneider: We needed them for the OWL syntax
Deborah McGuinness: consensus on what do to?
Bijan Parsia: no, just consensus on desiderata
Bijan Parsia: OWL 1.1 design is driven by the three main desiderata
Carsten Lutz: are all three needed?
Bijan Parsia: not in all cases, but in most cases
Stability vs Missing Features
Jeremy Carroll: some groups want stability
Bijan Parsia: main desiderata do lean towards stability
Ivan Herman: there is still a long process to use OWL, so change is bad
Jeremy Carroll: HP didn't participate in OWLED due to financial desires
Alan Rector: missing features in OWL have hindered uptake
Bijan Parsia: currently missing feature is keys
Ivan Herman: different markets - some want more features, some want no change
Peter Patel-Schneider: why?
Bijan Parsia: why should people who are only taking taking part of OWL care about OWL being extended?
Sandro Hawke: This is "OWL Pixie Dust". People want some of the OWL Magic, without really knowing what OWL is or does for them......
Alan Rector: standards all change so why is new OWL a problem?
Alan Rector: there are issues with backward compatibility
Alan Rector: Standards grow, with backward compatibility. And sometimes there are mistakes that need to be fixed.
Ian Horrocks: maintenance is needed
Deborah McGuinness: users want transition path and backwards compatibility
Ivan Herman: we need to take care of stability concerns
Bijan Parsia: model of development - do lots of work outside W3C, then quick recommendation, repeat roughly yearly
Ivan Herman: stability - vague uneasiness (mostly)
Alan Ruttenberg: what can we do to help?
Ivan Herman: nothing, really
Sandro Hawke: if the perception is that OWL 1.0 is broken then that is even worse than the perception that things are unstable because we're working on 1.1
Alan Rector: for me OWL 1.0 is not usable - so I need OWL 1.1
Bijan Parsia: look for people who have real blockage and try to help them
Effect of OWLED
Bijan Parsia: Bijan: OWLED attendance and submissions have been growing
Bijan Parsia: many participants felt that OWLED gave them a voice
Bijan Parsia: OWLED experience has been positive
Bijan Parsia: testimonial from Kent Spackman (SNOMED person)
Session 2 (Day 1, 11:15 - 13:00)
The text of this section is automatically included from F2F1 Minutes Session 2. Edits should be done there.
Structural Specification and Functional-Style Syntax design review
Slides for this session: Media:motik-f2f1.pdf
(Scribe changed to Jeremy Carroll)
Boris presents slides - OWL 1.1 Design Decisions
General Design Principles (slide 3)
- extend expressivity
- Bring spec closer to tools. Some features of OWL 1.0 have not been implemented correctly.
- make spec cleaner and clearer
Structural Specification: slides 4 and 5
Every OWL API wants to provide "what are the classes in this ontology", but what does that mean for OWL?
Discussion of mention; use; definition? It is very difficult to decide between these.
Jeremy Carroll: why does this matter?
Bijan Parsia: if we can improve interop on this, we should ...
Alan Ruttenberg: if these are design criteria they should be exposed
Boris Motik: there are some explicit answers in the new spec
Boris Motik: W3C should care, because these things are implicit
Peter Patel-Schneider: I still haven't heard a useful answer for tool designers
Alan Ruttenberg: two people are looking at some ontology in two different lists - "please look at class FooBar ..."
Peter Patel-Schneider: but there's lots of example where two different UIs are difficult to interop
Jeremy Carroll: let's have tests
Bijan Parsia: serialization tests would be good -
Bijan Parsia: users care abotu serialization
Bijan Parsia: OWL API
Ratnesh Sahay: a java program on two different dev environments, my program should behave similarly
Ratnesh Sahay: we care more about behaviour of program, than behaviour of tool
Ian Horrocks: the set of classes in an ontology should be well-defined
Michael Smith: for explanation and debugging it is useful to have a mapping from entity or axiom to ontology
Jeremy Carroll: isn't OWL 1.0 clear?
Jeremy Carroll: requires xx rdf:type owl:Class
Boris Motik: e.g. imports or inferred triples
Alan Ruttenberg: what is the underlying design model?
Alan Ruttenberg: If the question is "What classes are mentioned in this ontology?" then we're fine. It's not clear to me that any other question is relevant/important. What motivates other questions?
Boris Motik: we want to design OWL 1.1 as an object model
Matthew Horridge: imports was too vague
Sebastian Brandt: many industrial users like object models; descriptions of triples are much less accessible
someone (maybe Alan): Object oriented modeling of OWL, cuts both ways: A lot of teaching OWL is unteaching object oriented thinking.
expressivity enhancements uncontroversial
Metamodelling (slide 7)
metamodelling needed also in OWL DL
e.g. an OWL-S type example
punning is a possible solution,
applications want syntactic level, and don't want consequences
Peter Patel-Schneider: which reasoners could support Hilog semantics, after minor changes
bijan/boris (guest): easy to modify pellet
Bijan Parsia: easy cases would be easy ...
Jeremy Carroll: huge exlamation on first bullet
Bijan Parsia: OWL Semantics 1.0 is clear, OWL DL name, OWL Full location
Alan Ruttenberg: caching is a tool's issue
Alan Ruttenberg: caching does not break the spec
Bijan Parsia: some implementations change name when ontologies move
Alan Ruttenberg: if I moved ontology from http:... to file:... then I can't import it, and then spec is broken
Alan Ruttenberg: disagrees with first bullet
we agree that we don't agree, but we're not clear what we don't agree on
Bijan Parsia: we have session on annotations
Sebastian (guest): annotations on axioms are useful
Peter Patel-Schneider: all OWL DL reasoners are based on nonnormative docs
(sorry scribe missed a bit)
Bijan Parsia: it would be better if the implementors were working more closely from normative docs
Ian Horrocks: there is no claim that sean's nonnormative doc and normative spec say same thing
discussion on pellet and bnodes --
Alan Ruttenberg: pellet departs from spec
Bijan Parsia: we (pellet team) made choices
(Scribe changed to Pascal Hitzler)
Alan Ruttenberg: interaction of typing with RDF really a problem?
Alan Ruttenberg: is the problem in the language or in the documentation of it?
Bijan Parsia: pellet does some repairs silently. spec could go in a similar direction
Jeremy Carroll: questions on slide 15 answered on OWL 1.0 spec
Boris Motik: some may be, but spec might need fixing or made more explicit
Ian Horrocks: more clear spec desirable
Bijan Parsia: agrees about unclear parts in the spec
review of and resolution to publish Syntax, Semantics, and Mapping-to-RDF documents as First Public Working Drafts
next session: discussion and decisions on publication schedule and first public working draft
chaired by Ian Horrocks
collecting issues: (1) shortname, (2) document titles, (3) SOTD, (4) WIKI extraction, (5) attribution etc.
Ivan Herman: (1) should include namespace
Ian Horrocks: need to decide if namespace is an issue
Bijan Parsia: need to do editorial cleanup (part of (4))
Bijan Parsia: deadlines need to be watched
(1) short name (+ namespace)
suggestions: owlwot, alan: owltoo
Alan Ruttenberg: calling it OWL may overload and thus be difficult
... something neutral to version name?
Sandro Hawke: no problem with same names
Alan Ruttenberg: might be confusing
Sandro Hawke: using same name is only a problem if exactly the same document name is used
Evan Wallace: so why not call it OWL1.1?
Bijan Parsia: OWL1.1 is one possibility
Sandro Hawke: you want a URL which is the link to the latest version of the spec
Evan Wallace: you need a name which redirects to the short name?
Peter Patel-Schneider: eventually pointers to owl1.1 docs might go away ...
... but that's independent of the document names
Ian Horrocks: we can't use "owl-semantics" right now.
Ivan Herman: because that's the working draft
Bijan Parsia: OWLWOT, OWLTOO looks strange
... proposes OWL1.1 or OWL11 or OWL-11
Ian Horrocks: what about OWLTOO
Sandro Hawke: different names suggest different levels of compatibility
Bijan Parsia: OWL1.1 (with any kind of minor changes)
sugestions OWL 1-1, OWL11, OWL-1-1
Ivan Herman: decision has to be formally recorded according to charter
Ian Horrocks: should discuss point (2) at the same time
... what is going to be named in document title?
Alan Ruttenberg: procedural question: can we resolve this here? what about absent people?
Bijan Parsia: explains: may not resolve things which have not been on the agenda
Peter Patel-Schneider: may be arguable
Ivan Herman: was the issue of document titles on the agenda?
some discussion about which things that can be resolved in the F2F
Alan Ruttenberg: supports bijan that we should decide things, and people can appeal to chairs to reopen the issue
Sandro Hawke: in this case: does it need to be decided right now?
Deborah McGuinness: against owl2/owltoo
ian makes straw poll 1.1 against 2 (clear positive outcome for 1.1)
Ian Horrocks: let's decide for 1.1
proposed and resolved: it's going to be 1.1 (in some form)
PROPOSED: Our publications will refer to this work as "OWL 1.1" (not OWL 2.0, etc)
Ian Horrocks: asks for objections, abstantions on that. none recorded
RESOLVED: Our publications will refer to this work as "OWL 1.1" (not OWL 2.0, etc)
Ivan Herman: other specs seem to use similars to OWL11
Sandro Hawke: would like to postpone this and find out some background
Ian Horrocks: straw poll: do we want "owl11" ?
PROPOSED: To ask for shortname "owl11-[whatever]"
Ian Horrocks: what about namespace?
Ivan Herman: tough one
Peter Patel-Schneider: proposes brief discussion about it
Peter Patel-Schneider: should reuse the namespace
several people second reusing the namespace
Alan Ruttenberg: I know now that I don't know whrether to resuse ns
straw poll on this: tendency for reusing namespace, but not uncontroversial
Alan Ruttenberg: if owl constructs change semantics then it may be difficult to reuse name space
Ulrike Sattler: wasn't the idea not to change any of the constructs already present?
Ivan Herman: are we sure this won't happen?
Ian Horrocks: we don't have to decide on this right now
Bijan Parsia: new constructs should have new names
... we will add new things into to the namespace
... expanding vocabulary is considered difficult by some people
Ian Horrocks: summary: tendency for reusing, but issue can be postponed
Peter Patel-Schneider: need to be careful on first working draft that it doesn't cause confusion in terms of namespace
bijan takes action to take care of this
ACTION: Bijan to put alert box in all the documents about the status of the namespace
Alan Ruttenberg: say it "owl" subject to change, not "owl11" subject to change.
Boris Motik: old names are still in old namespace (current document)
Bijan Parsia: suggests to leave the two namespaces as they are right now
Bijan Parsia: let's not make owl 1.1 implementors change anything right now.
Bijan Parsia: there are owl 1.1 ontologies on the web right now.
Alan Ruttenberg: straw poll: leave as is with warnings (agreed)
boris has action to do the changes (add warnings) in the docs
ivan about doc titles: suggests owl11 DL
Ivan Herman: functional syntax doc is DL-only, so that should be in the title?
Ian Horrocks: functional syntax is not entirely irrelevant outside DL
Alan Ruttenberg: needs to be decided later
Ian Horrocks: see 1.0 docs on abstract syntax
Bijan Parsia: in some way structural syntax specifies OWL Full
Ivan Herman: but there are statements which cannot be expressed in it
... should not forget that this is an issue
Ian Horrocks: action on this?
Ivan Herman: need a list of editors first?
moving on to point (5): attribution etc.
Alan Ruttenberg: proposes for current draft that attributions should be as they are
... next draft if substantive changes, attributions should be reevaluated
Bijan Parsia: question is if chairs want to assign editors. bijan suggests chairs do that
Alan Ruttenberg: would like to not do that right now
Peter Patel-Schneider: somebody needs to put more work into it soon ... credit should be given
Ivan Herman: seconds alan: currently mentioned people stay editors for the current version
some more discussion on editors for current version
Bijan Parsia: wants editors assigned now
alan proposes current authors are editors for the current version
Ian Horrocks: should now decide whose job it will be
Ian Horrocks: proposal that boris, peter and bijan work on syntax (they would agree)
Ian Horrocks: bernardo, boris for semantics document? would agree as well
... bernardo, boris would also do the mapping document
... are we happy if they do it?
... straw poll on this: no objections
was agreed that attributions will stay the same in current version of the documents as they are stated right now
Ivan Herman: doesn't it need to be called editor?
Sandro Hawke: I doubt it
Ian Horrocks: If it has to change from author to editor, then that can be chairs decision
moving to point (4): editorial cleanup in the wiki plus wiki extraction
issues from working drafts will stay
Alan Ruttenberg: useful comments should be left but scripted away
moving to point (2): resolution to publish Syntax, Semantics, and Mapping-to-RDF documents as First Public Working Drafts
PROPOSED: Publish Syntax, Semantics, and Mapping-to-RDF documents as First Public Working Drafts
Ian Horrocks: straw poll - no objections
PROPOSED: Publish Syntax, Semantics, and Mapping-to-RDF documents (as on the wiki right now) as First Public Working Drafts
Ian Horrocks: formal vote on this (reminder: only one vote per member): W3C abstains, no objections
RESOLVED: Publish Syntax, Semantics, and Mapping-to-RDF documents (as on the wiki right now) as First Public Working Drafts
moving to (3): SOTD
brief agreement that this shall not be discussed now
Ian Horrocks: lunch break now
Bijan Parsia: thanks to sean for taking care of organisation ...
Session 3 (Day 1, 14:00 - 15:45)
The text of this section is automatically included from F2F1 Minutes Session 3. Edits should be done there.
(Scribe changed to Markus Krötzsch)
Slides for this session: Media:pfps-f2f1.pdf (go to page 5)
Peter Patel-Schneider repeats imports definitions from OWL DL, OWL Full
Peter Patel-Schneider: differences relate to whether ontology names or locations are assumed in import statements
Bijan Parsia: not all ontologies have names, right?
Peter Patel-Schneider: yes, only importable ontologies need a name.
Peter Patel-Schneider: this appears to be a bug
Boris Motik: (1) there should be only one kind of import, not three
... (2) it should be possible to reconstruct the location from whic some statement was imported from, e.g. for editing
Bijan Parsia: the name and location can be different, the question is how to deal with it. This seems to be agreed on.
Peter Patel-Schneider: OWL1.1 imports are based on ontology names only
... this is completely different from OWL1.0, where the name must be the location.
... We do not have XML inclusions (a mechanism working with location only).
... Summing up there are two different designs: name and location based importing.
Peter Patel-Schneider: various questions arise
... (1) should every ontology (be forced to) have a name?
... (2) should name and location be the same (i.e. should the name always indicate the location)?
... (3) should imports be by name or by location?
Jeremy Carroll: this seems to be a general web architechture issue.
Bijan Parsia: in general URIs are not locations, but there might be (multiple) hints for actually finding the document.
Peter Patel-Schneider: versioning is another problem
Alan Ruttenberg: this goes beyond the importing issue
Peter Patel-Schneider: every importable ontology needs some location, but it need not be on the web
Boris Motik: a typical use case is that two ontologies (one importing the other) reside in some file repository and then are moved to the web. How do you support this?
... locations change over time
... this is not just a caching issue
... What they are doing in XML Schema may be a good solution.
... When importing an ontology, I do not care where it lives. It might even have many copies.
Alan Ruttenberg: I suggest that names and locations might be different, but importing one ontologies from some location should also make this location a name for the ontology.
Bijan Parsia: I do not understand the proposal
Alan Ruttenberg: every importable ontology has a name which is also a location, but it is possible that the same ontologies have different names in the sameAs-sense.
... importing may lead to the inference (?) that two names refer to the same ontology.
General agreement that further clarification is needed.
Alan Ruttenberg: every name should be a location, they are linked together.
... just if a name does not match its location, then this alternative name should be deduced.
Bernardo Cuenca Grau: when you have an ontology name occuring in documents in different locations, how do you know they are the same?
Boris Motik: well, it is just *the*, say, Wine ontology
Bernardo Cuenca Grau: but there could be versions
... e.g. if someone adds axioms
Boris Motik: this is not specified, but a similar mechanism is found in including classes in Java. Java uses names but the application environment must resolve the locations.
Jeremy Carroll: there are two cases: creating an ontology and publishing it, and the reverse, downloading and caching an ontology from the web.
... we should concentrate on the web/caching aspect, not on the publishing aspect.
Evan Wallace: Many people in ISO want to use URNs as a name, and these are not locations.
Alan Ruttenberg: I would also say that an ontology name is a URI, not always a URL, but to import it, you need a location which would then become a synonym.
Achille Fokoue: the inclusion mechanism of XML Schema (XML Schema import not XMLinclude) is a good solution.
... names should not be tied to locations, but further sources should be used to resolve names.
Alan Ruttenberg: is this consistent with my proposal for having many locations for some ontology?
Achille Fokoue: yes, I would like some default mechanism that can be overwritten to specify alternative locations
Boris Motik: I have two points.
... (1) how many ontologies are really on the Web?
... (2) we should not specify in detail what tools are supposed to do when looking for ontologies
... It would have been easier to leave tools some freedom for determining ontology locations, e.g. similar to CLASSPATH in Java
Ian Horrocks: Re (1) appliations may still refer to the web, but ontologies might stilll be local to some server
Boris Motik: but aren't there also relevant uses of ontologies without any Web?
Bijan Parsia: it is not out of scope to consider ontologies that are not on the web
... I am disagreeing with Alan.
... I often created local copies of ontologies to modify them, while keeping internal names.
... It shoud not happen that those modified copies then are deduced to be the same as the original one.
... I do not see what Alan's proposal buys us.
Ian Horrocks: summing up, the problem could be that multiple (versions of) ontologies have the same name in their header, and those should not be considered the same.
Alan Ruttenberg: this would only happen if both were imported.
... Considers, e.g. having three variants of one ontology:
... B, B', B''
... depending on what you import you may get either
... only if you import two at one time, these would be merged.
Bijan Parsia: but didn't you say that locations and names then get equated (sameAs) on import. Why would this be good?
Ian Horrocks: these details should be discussed here, and the discussion must probably be taken offline at some point.
... including clearly written-up proposals.
Alan Ruttenberg: responding to Jeremy saying that we should leave this to the caching mechanism. The reason that I would like to have "location punning" on names is that I would like to use different tools at one time.
Sebastian Brandt (guest): I would like to partially agree with Boris: ontologies are often used offline to make money, but they still are developed online.
Matthew Horridge (guest): when developing tools, we only found it reasonalbe to treat ontology URIs as names. Protege uses a lookup table to map onologies to local files.
ACTION: Bijan to extend the wiki with information on imports and restructuring it if needed (with Sebastian)
ACTION: Alan Ruttenberg to write up his proposal on dealing with imports
Slides for this session: Media:pfps-f2f1.pdf (pp. 5-7)
Peter Patel-Schneider: basic ideas of rich annotations
... allow arbitrary syntax as annotations, including annotations
... annotations separated into "spaces" and some spaces may indicate that tools must understand the respective annotations (for extensions)
(Peter Patel-Schneider presents syntax slide)
Peter Patel-Schneider: keywords mayIgnore and mustUnderstand describe whether or not annotations are essential for semantics
... yes, annotations with "mustUnderstand" may change the semantics, also of existing constructs
... Each annotation belongs to some "space", given as part of the annotation syntax.
... There is a "default space" for annotations without explicit space annotations.
Bijan Parsia: the term "annotation" is ambiguous. In OWL1.0 it was something given to an annotationProperty. In OWL1.1 it can be any piece of syntax.
Peter Patel-Schneider: Annotations may even exist without relating to any OWL object.
Alan Ruttenberg: do the axioms of the containing ontology also belong to each annotation space?
Bijan Parsia: no, unless one would import it explicitly.
Boris Motik: I do not understand the idea of "annotation spaces"
Peter Patel-Schneider: this is because some annotations are semantic extensions, that should be keeped separate from other annotations.
Sebastian Brandt (guest): I have another use-case: I have some annotations that are just user documentations, some that contain "code" that is used by the application, and even some that are generated automatically by my applications.
Jeremy Carroll: we should have a worked example that illustrates this
Bijan Parsia: the Pronto extension to OWL provides some example
ACTION: Bijan to improve examples for rich annotations.
Boris Motik: it would also be useful if someone could explain in detail how to use this mechanism, starting from ontology creation up to external reuse.
Bijan Parsia: I can do that after coffee
Alan Ruttenberg: is there going to be an RDF serialisation to this?
Bijan Parsia: yes
Alan Ruttenberg: do annotations then distribute over differnt files?
Bijan Parsia: no, we can use reification to add extra annotation-space information
... but there are many possibilities
Peter Patel-Schneider: I think the idea of annotation spaces changing the semantics of OWL is what is most controversial
Bijan Parsia: semantic annotation spaces need to have a spec, e.g. to include RIF rules into OWL documents.
... this spec then defines the intended semantics.
... The annotation space has a URI that may specify this semantics.
Peter Patel-Schneider: annotations usually have no semantics, exceptions being the mustUnderstand annotations that must be taken into account by tools in an adequate way.
Ian Horrocks: we did only talk about rich annotations, but not about the other OWL1.1 extensions to the OWL1.0 mechanism.
... this should also be discussed.
Peter Patel-Schneider: we can do that in the compatibility session tomorrow.
Session 4 (Day 1, 16:15 - 18:00)
The text of this section is automatically included from F2F1 Minutes Session 4. Edits should be done there.
(Scribe changed to Deborah McGuinness)
The Working group would like to extend their thanks to the Knowledge Web project for the dinner we are about to enjoy.
Bijan's presentation of an annotation example
Slides sent to the list later
A few notes follow: The example uses one annotation space. One annotation blob includes who is responsible for the annotation (in this case self)
and sally checked it (thus showing structured annotations with reference to self)
namespaces at top
ontology header next
ontology uri, followed by comment (which expands to an rdfs comment)
followed by 2 namespace declarations
going down to womanaged50Plus
womanAged50Plus is a subclass of WomanWithBRCInShortTErm with certainty between .027 and .041
would be ignored if reasoner did not understand pronto
certainty between .027 and .041
Question from Joanne - can this be used to embed sbml in the document? Alan Ruttenberg: annotation and must understand would allow you to embed sbml into the document
Boris Motik: how should one encode constraints into an ontology?
this could be a way
bijan notes ontoclean and pronto are working
Alan Rector: we need rich linguistic representations for some applications
Alan Rector: we need rich structure
Alan Rector: a lot of information is provenance knowledge, its structure and how it was put together which is important for passing around between tools
Alan Rector: this is also a way of providing one type of representation and projecting it into an alternative representation
Alan Rector: want to take a model in owl and export it in another syntax
Bijan Parsia: part of this "must understand" is a retrofit
Sandro Hawke: must understand may be a reasonable option but there may be other options as well
what if someone wanted to put in rif rules?
one way you could interact with this is to make applications aware of the spaces
Ian Horrocks: question to sandro - what are possibly better engineering solutions?
Sandro Hawke: ummm
Sandro Hawke: extend the syntax
Alan Ruttenberg: this may lead to many different extensions ... maybe this is "too easy"
Jeremy Carroll: generally amused by mustunderstand that may be ignored
Jeremy Carroll: concerning named graphs... what about serialization
Jeff Pan: mustunderstand is a nice idea to allow users to specify intended semantics...
Alan Ruttenberg: is it the case that a mustunderstand on anything, then reasoners may not understand the semantics
Alan Ruttenberg: lets consider motivations
1: one wants to use editing tools
Sandro Hawke: suggest people give feedback to bijan on the general scheme
Ivan Herman: aestetic comment - word annotation is more what alan rector was describing
Ivan Herman: mustunderstand is not an annotation
Sebastian (guest): any tool that finds something strange could just say I do not know what to do with it... he likes the structure
Boris Motik: useful to group certain types of properties into annotation or extension.... not sure that this should be in the ontology.
Boris Motik: this may be worthy of putting in a separate document
about to move on to user facing documents (and hoping for test discussion to come)
Ian Horrocks: rich annotation mechanism but without notion of mustunderstand
most people agree a good idea
(jeremy objecting and, steve battelle abstained
that was for the general mechanism without semantics
now general idea with flagging semantic intention
good - 13
count good increased to 14 includes joanne
discussion about what the vote was...
some kind of decorating mechanism of the existing syntax that would indicate a semantic change
PROPOSED: decoration of existing syntax as a way to make a syntactic change
in favor 14 (including the 3 remote participants)
against - pfps, sandro, alanrut - 3 total
abstain - hp
we could consider adding a swrl extension syntax to this proposal
Alan Ruttenberg: this may be out of scope
feedback that it is reasonable to continue developing this
(Scribe changed to Giorgos Stoilos)
User Facing Documents
Slides for this session: Media:wallace-f2f1.pdf
Evan wallace is presenting the status of User Facing Documents
...documents that will help users into owl 1.1
...like guide, overview, reference
Evan Wallace: work mainly volunteered
...should these documents be produced as part of the spec?
...what syntax to use for the examples, different users have different preferences...DL syntax
...abstract syntax, etc
Evan Wallace: e.g. some docs use the DL side, while others use, like the Reference use a Semantic Web side (meaning RDF)
no much progress has been done
preference in producing an overview doc that is between OWL 1.1 overview and OWL overview
Bijan Parsia: is working in a way to choose to preview an example in the syntax of your choice
Deborah McGuinness: looked in OWL 1.1 overview
Jeremy Carroll: a question is how much of the old docs we will use or start new ones
Jeremy Carroll: do we want to extend the owl1.0 docs?
Bijan Parsia: oposes to extending the owl 1.0 docs. Finds them confusing
...don't thinks that good "tutorials" could be written within the WG
Bijan Parsia: if people want to go forward then he proposes something like RDF Primer
Ian Horrocks: thinks that an entry doc is very important and usefull
Bijan Parsia: what about the homepage?
...contains motivation, intro, etc
Michael Smith: homepage could also be improved to serve as a better intro doc
Alan Ruttenberg: people are not learning OWL from the docs. Use them as refs rather than intros
Bijan Parsia: exaplain some problems related to w3c for updating the docs
Vipul Kashyap: agrees with alan rut
Bernardo Cuenca Grau: agrees with bijan. Why does educational material go to rec?
Uli Sattler: what about updating old docs with links to new matterial
Deborah McGuinness: don't see how this could be done
Sandro Hawke: explains a way
Alan Ruttenberg: asks for a clarification on sandros example
Evan Wallace: a plain language doc that could be used by non-experts is very good
Bijan Parsia: I am not highly against rectrack docs but more in favour for notes
Ian Horrocks: put the rectrack non-rectrack aside and think about the docs
Ian Horrocks: it seems from today that overall docs are not very good
Ian Horrocks: come again to the question whether starting fresh docs or extending old ones
Ivan Herman: has found guide docs of various groups very helpfull, like XML schema
Ivan Herman: if these docs have the blessing of the group then this is much better
....producing similar docs for the community would be good
Alan Ruttenberg: do we agree that we need a reference doc
Deborah McGuinness: because there are problems with existing ones does not mean we have to start with new ones
Bijan Parsia: proposes a fresh "primer" doc as a replacement of overview and guide
PROPOSED: To meet our charter deliverable of covering the intent of 'Overview' and 'Guide', we'll publish a new 'Primer' (written largely from scratch).
James Hendler: asks about where the specification of OWL Full would go. Does not see it in OWL 1.1. functional syntax
Jeremy Carroll: agrees with jim. It would be helpfull to have a doc which explains to non-experts features of owl full
Bijan Parsia: functional syntax must not change since it is the formal specification
PROPOSED: To meet our charter deliverable of covering the intent of 'Overview' and 'Guide', we'll publish a new 'Primer' (written largely from scratch).
Peter Patel-Schneider: Aparently we should have 3 proposals: i) old docs, ii) fresh docs iii) prime proposal (bijan's)
PROPOSED: Refresh the old documents (otherwise, start mostly from scratch).
straw poll: YES=refresh docs NO=start new ones.
Ian Horrocks: lets go doc by doc
straw poll on =overview=
PROPOSED: 'Overview' requirement from charter to be met by cleaned up and expanded-as-needed version of 2004 OWL Overview
votes for no in room - 9
Ian Horrocks: people from the call are confused on what YES/NO choices meant. New poll is performed
count on room: 3
total for yes: 7
votes for no=12 (11 + vipul)
PROPOSED: 'Requirements' requirement from charter to be met by cleaned up and expanded-as-needed version of 2004 OWL publication
next doc to poll =requirements=
voting for yes:
count on the room: 2
voting for no
room (guest): 12
PROPOSED: 'Formal Specification' requirement from charter to be met by cleaned up and expanded-as-needed version of 2004 OWL publication
next doc =Reference=
Starting vote from YES
PROPOSED: 'Descriptive Specification requirement from charter to be met by cleaned up and expanded-as-needed version of 2004 OWL publication
count yes (refresh) in the room: 4
voting for no
count in room for no: 13
PROPOSED: 'GUIDE requirement from charter to be met by cleaned up and expanded-as-needed version of 2004 OWL publication
voting for yes:
room for yes: 2
voting for no
room for no: 13
Ian Horrocks: maybe not have a poll on evan's suggestion (above) since it is currently not clear
Jeremy Carroll: explains some ways for doing tests
...every feature in the spec must have a test
Jeremy Carroll: in owl test was a normative doc
Ivan Herman: not in sparql
Ivan Herman: don't need to decide now about normative or non
Alan Ruttenberg: put test cases on wiki
Ian Horrocks: what will happen with existing test, will they be on the wiki
ACTION: Sandro to develop scripts to extract test cases from wiki, coordinating with Bijan, Jeremey, AlanRut.
Session 5 (Day 2, 09:00 - 10:45)
The text of this section is automatically included from F2F1 Minutes Session 5. Edits should be done there.
(Scribe changed to Jeff Pan)
Slides for this session: Media:sattler-f2f1.pdf
Uli is presenting
OWL DL does not support user defined datatypes
Uli Sattler: users want to represent intervals
... and comparisons
... in OWL DL no inverse functional datatype properties
... not to mention composite keys (not even OWL Full supports this)
Boris Motik: we might want to keep the unit mapping out of TBox
Jeremy Carroll: second
Sebastian Brandt: there are real world examples
... that we need datatype mapping in the TBox
Bijan Parsia: both needed
Uli Sattler: we have examples of seeing class subsumption checking based on datatype constraints
Carsten Lutz: it is difficult to choose one standard set, e.g. covering integers, rational, +, *, ...
Uli Sattler: as many as possible
Jeremy Carroll: each simple example is easy
... but have concern on having all of them, which makes it hard
Sebastian Brandt: combining DL and data values are important and useful, there are many tasks that you could not solve if you treat datatypes externally
Alan Ruttenberg: do we want to detect the problematic cases?
Jeff Pan: there are existing works on datatype groups, a mechanism is already there, see http://www.websemanticsjournal.org/papers/20050811/document.pdf
... even in OWL DL, freely combinations are not possible, e.g. transitive properties are not allowed to used in number restrictions
Jeremy Carroll: what happen if data in the user databases having both integers, rationals + and * ...
Boris Motik: we need some datatype profile
Carsten Lutz: second boris point
Jeff Pan: two points: 1) profiling is a good idea, there have been work there such as datatype groups
... and we could provide a list of feasible datatype groups
2) if users have integers, rationals + and *, we could simply have type promotion, promoting integers into rational, and it is still decidable, see http://www.w3.org/TR/swbp-xsch-datatypes/ and http://www.w3.org/TR/xpath20/#promotion
Alan Ruttenberg: maybe we could have a stroll poll on this
Bijan Parsia: we all agree that some sort of datatypes are needed, no matter in OWL or RIF
... many of our cases cannot be addressed by RIF
Jeremy Carroll: transitive issue is different
Jeff Pan: besides Racer, an extension of FaCT (FaCT-DG) also supports n-ary and datatype groups
Uli Sattler: we could have some more general proposal, rather than specific ones
Bijan Parsia: we don't have to require all our implementors to implement everything, so we should be flexible somehow
Uli Sattler: the 4th point: easy keys
Markus Krötzsch: in foaf people use b-nodes rather than individuals, so the easy key might not solve the foat problem
stall poll 1: many +1, no -1, four 0
straw poll 2: (all) +1
straw poll 3 (about 2-b): many +1, two (conditional) -1, six 0
straw poll 4 (n-ary datatype): twelve +1, six -1, five 0
straw poll 5(easy key): 22 +1, one -1
Boris Motik: one profile proposal: a set of default profiles and allowing users to have arbitrary profiles
Another go, boris' profile: a fixed set of profile and also allowing people to define their owl profiles
alanr's proposal: a fixed set of profile
straw poll on profiling on datatype: eighteen +1, four 0
scribers should clean up yesterday's minutes by next telecon
(by IanH and no objections)
Session 6 (Day 2, 11:15 - 13:00)
The text of this section is automatically included from F2F1 Minutes Session 6. Edits should be done there.
OWL DL and OWL Full
Slides for this session: Media:pfps-f2f1.pdf (pp. 1-3)
Here's a brief description of how model theoretic semantics works
Peter Patel-Schneider: OWL DL has a fairly straightforward take on this
Peter Patel-Schneider: OWL Full and RDF take a slightly weird take on this
Peter Patel-Schneider: wherein properties and classes live in the world with real objects
Peter Patel-Schneider: Here are the differences between OWL DL and Full semantics
See "Two Model Theories" slide
Peter Patel-Schneider: things like rdf:type and owl:Class are not in the world in DL but are in Full
Alan Ruttenberg: In OWL DL Universe what is the status of Ontologies?
Peter Patel-Schneider: There is a separate place for them because of annotations
Peter Patel-Schneider: This description is about the specification and not practice
Bijan Parsia: The things in the OWL Full universe are in there with a theory
Peter Patel-Schneider: None of this matters in some sense
Peter Patel-Schneider: What matters is the behavior which results
Peter Patel-Schneider: ...such as entailments
Peter Patel-Schneider: Differences: It's possible to make assertions about the OWL vocabulary that change their interpretation
Slides for the rest of this session: Media:carroll-f2f1.pdf
Jeremy to take over presenting
Alan Ruttenberg: question about the intention of compatibility to be entailments of DL and Full be identical
Jeremy Carroll: for me the whole point is to get compatibility with RDF
Jeremy Carroll: A goal is "least surprise" for users of RDF when using OWL
Jeremy Carroll: OWL annotations are intended to behave as RDF annotations
Alan Ruttenberg: There are implications for RDF annotations that users may not be aware of
Alan Ruttenberg: I want to make a distinction between usage and consequences of the semantics
Bijan Parsia: I don't understand what you mean by RDF triple-by-triple semantics
Jeremy Carroll: In the OWL 1.0 semantics there are correspondence theorems between OWL Full and DL Semantics
Jeremy listed Issues related to the FULL and DL differences
such as 63, 76, 81, 69, 72, 55, 73
Jeremy Carroll: Do we want to allow semantic subsetting for fragments
Bijan Parsia: If we are going to support OWL Full do we need to support the full RDF umbrella
Bijan Parsia: described in Jeremy presentation
Jeremy Carroll: The semantic of RDF reification are essentially none
Bijan Parsia: There exists somewhere in the known universe a Statement that includes: S, P, O
Jeremy Carroll: There is no clear statement in the specs for how reification can work interoperably from system to system
Bijan Parsia: In the OWL full situation you have to interpret the reification syntax somehow
More discussion about how this can be done
Jeremy Carroll: Punning
In some peoples mind the web arch specifies that a URI corresponds to a single meaning
Punning is weaker than OWL Full because it violates this principle
Jeremy Carroll: this seems to cause user confusion
In my view, the mapping rules were the hardest part of the OWL Rec
Jeremy Carroll: The drivers behind the mapping rules in OWL 1.1 are different
Jeremy Carroll: ...and this will lead to considerable change and probably
Boris Motik is speaking
Jeremy Carroll: ...issues later on.
Boris Motik: In my opinion many of these problems are the result of
Boris Motik: ...shoe-horning everything in the same universe.
Point for discussion later:
Boris Motik: if we came up with an OWL Full that has a clean model theoritic framework
...then we could fix this.
Alan Ruttenberg: This would be a smaller OWL Full?
Bijan Parsia: Punning was intended to meet the goals of Full at least quarter way
Peter Patel-Schneider: The dogma in this case is the same syntax extension of RDF
Bernardo Cuenca Grau: The people who like OWL Full should really come up
...with features for OWL Full that they like and use
...Then we could do some research.
Ian Horrocks: The point I wanted to make was how much of this proposed
...work will be part of the work of this WG?
Boris Motik: Cleaning this up would be a huge accomplishment for this group.
Alan Ruttenberg: To my mind, it's not clear that cleaning up OWL Full is desirable to
... the Full/RDF community.
Jeremy Carroll: Dropping the comprehension principles seems like the
... smallest change that would be of value.
Alan Ruttenberg: Is this in scope for our group? Strictly speaking I don't think so.
Ian Horrocks: This kind of work just isn't in scope.
Bijan Parsia: Form an OWLED task force to look at this.
Alan Ruttenberg: We need to have a discussion about what compatibility means.
Alan Ruttenberg: If we allow OWL Full semantics changing that will affect backwards compatibility
Discussion of semantic fragments
Alan Ruttenberg: we have a delta now in the sublanguage entailments
Bijan Parsia: finding some delta that makes sense that makes the languages
...as close as possible would be a good thing.
Ian Horrocks: If we are comfortable with this semantic subsetting then
... we should be happy with the Full - DL differences
Ian Horrocks: One slight difference in Jeremy's proposal would be allowing
... more syntactic freedom but actually reducing the entailments
... by removing the comprehension principles for e.g.
Jeremy Carroll: HP might be happy with such a result if it is consistent
... with some broader framework.
Jeremy Carroll: There are easy bits in the OWL 1.1 language.
... getting those bits working are a bounded and achievable task.
Bijan Parsia: My experience is that users are concerned about not
... being able to process large numbers of RDF graphs
...with DL reasoners.
... Features like punning improves this situation.
Ian Horrocks: I wonder how hard it would really be to extend the status quo
... with some acceptable differences.
Ian Horrocks: This is a strawman for something that we could do.
Jeremy Carroll: I'd need to take this proposal back to HP before commenting on it.
Bijan Parsia: I would like us to keep the political and the user requirements seperate
ACTION: jeremy describe how punning and cardinality play poorly with each other
Ian Horrocks: I'd like to see suggestions for concrete ways of moving forward to address these problems
Jeremy Carroll: Why don't we start with Qualified Cardinality Description?
Peter Patel-Schneider: Someone made comment that Qualified Cardinality Descriptions leads to non-monitonicity
... and I remember finding it believable
ACTION: pfps inform the WG on absurdity of QCR / OWL Full
ACTION: jeremy attempt Wiki sketch of QCR semantics OWL Full
Alan Ruttenberg: on concrete actions...
... we have a set of options
Jeremy Carroll: a suggestion that Jeremy concentrate on OWL Full
... Semantics and drop out of User Facing Documents
Alan Ruttenberg: Any other specific proposals?
Ian Horrocks: Let's try and extend where are now and see where we end up.
Alan Ruttenberg: When do we evaluate when this approach is failing
... so that we can try another approach?
... I want to have some ideas about where we would go if this
... doesn't work.
Jeremy Carroll: We have had two variants proposed today.
... Sacrifice backwards compatibility and work towards 1.1
Bijan Parsia: We are spending a lot of time on this.
... I would like to know how much interest in this WG
... with Full compatibility.
How many people want to use OWL Full for 1.1?
Jeremy rephrase: When using 1.1 do you want to use Full semantics?
5 yes in room 15 no
q2 Are you a potential customer for Bijan's description of patch-up rules?
many in favor
Session 7 (Day 2, 14:00 - 15:45)
The text of this section is automatically included from F2F1 Minutes Session 7. Edits should be done there.
RIF and OWL WG Collaboration
Alan Ruttenberg: there is a proposal to have a joint OWL & RIF task force
Alan Ruttenberg: ... peter is there. is there anyone else?
Alan Ruttenberg: ...uli is a second.
Sandro Hawke: I may sort of be on it for both
Bijan Parsia: I am liason to RIF and will continue to be
Fragments - OWL Prime
Slides for this session: Media:zhe-f2f1.pdf
Slide: Oracle 10gR2 RDF
Some technical difficulties continued w.r.t slide presentation
Zhe Wu: many ways to insert data.
Zhe Wu: ...in 10r2 we also support some inferencing and rules. we use forward chaining approach
Zhe Wu: ...also query using a SPARQL-like syntax
Zhe Wu: ... this was all in 2005
Zhe Wu: this year new release with new features. faster loading, owl reasoning with proof generation
Zhe Wu: ... overhauled performance w.r.t. load and query
Zhe Wu: ... just recently added Jena / Oracle adapter
Zhe Wu: ... joint with HP
Zhe Wu: subset of owl is supported
Uli Sattler: i'm curious about what scalable and efficient means
Zhe Wu: i will show some numbers later
Zhe Wu: ... re: what is supported - forward chaining rules implementation for fast query answer
Zhe Wu: ... conclusion in ISWC 2006 paper was that existing reasoners had problems with large ABox data
Slide 7: OWL subsets supported
Zhe Wu: rdfs++ added as a "minimal" extension to RDFS
Zhe Wu: ...owl prime, what is now proposed as rdfs 3.0
Slide: semantics characterized by entailment rules
Zhe Wu: owl prime has ~50 rules
Slide: Applications of partial dl semantics
Slide: support semantics beyond owl prime
Jeremy Carroll: question about example being supported directly in the future
Zhe Wu: exactly
Achille Fokoue: question about updates to abox
Zhe Wu: i'll get to that later
Slide 13: Advanced options
Alan Ruttenberg: question about time, can we focus on questions now
Zhe Wu: ok, i'll quickly browse remaining slides, then go to questions
Slide: implementation in rules
Zhe Wu: I want to stress that we did not handle one property at a time
Zhe Wu: I'll jump to query answering slide
Zhe Wu: ...that's all I wanted to cover, open for questions
Post Presentation Q & A
Ian Horrocks: the tractable fragments doc describes fragments with known database mapping. wondering why you didn't choose one of those
Zhe Wu: we started by asking existing customers what they needed. most told us they just needed simple extension into owl from rdf
Zhe Wu: ... pretty much the approach was driven by customers and need to implement efficiently
Ian Horrocks: but, customers said you needed something small (rdf + a bit) which is exactly what the fragments are. instead you chose a large fragment and implemented incompletely
Zhe Wu: so far, for those other fragments we have not found a complete rule set (except PD*)
Uli Sattler: I want to echo ian and point out that you don't allow intersection, but a clever user would have it
Uli Sattler: ...and to be complete complexity becomes a problem
Alan Ruttenberg: they're not trying to be complete
Boris Motik: echo ian, observes that fragments exist which can be implemented with a set of complete rules
Bernardo Cuenca Grau: i'm worried about soundness and worried about what "sound and complete" means here. I don't understand the semantics
Bernardo Cuenca Grau: ...b/c you haven't implemented the OWL semantics, you've chosen some of the OWL DL vocabulary
Zhe Wu: we do care about completeness, but don't consider it critical
Zhe Wu: ... completeness is evaluated w.r.t. query answering for some benchmarks, etc.
Jeremy Carroll: what I hear from customers echos Zhe's comments.
Jeremy Carroll: ...I note that much of the questioning is hostile
Alan Ruttenberg: I agree
Jeremy Carroll: that may be b/c much of the questioning is coming from members with different user groups
Ian Horrocks: it wasn't intended to be hostile. I was trying to understand whether Oracle would be interested in more well understood and explainable fragments
Ian Horrocks: ...e.g., dl-lite which can be implemented in a database system, and also in a rule system
Discussion of PD* soundness and completeness in a rule based implementation, which scribe didn't capture
Ian Horrocks: the problem with PD* is that it doesn't implement a subset of OWL, it implements PD*
Jeremy Carroll: it depends on what you mean by fragment of OWL
Alan Ruttenberg: I hear interest in co-ordinating on database fragments with Oracle
Bijan Parsia: to standardize a fragment, we need a well defined specification that we can all understand
IRC aside on specifications, definitions, and implementations
The following IRC conversation happened in parallel to the in room verbal conversation and some other IRC exchanges. It continued until Ivan the point in the minutes where Ivan requests an end to side conversations (just before Semantic Subsets)
Fragments: (Tractable) Fragments and other Fragment Proposals
Bernardo Cuenca Grau: motivation of owl-lite was easier owl. b/c owl dl and full are rich and complex.
Bernardo Cuenca Grau: ...problem is owl-lite is broken b/c it doesn't address interactions between constructors
Bernardo Cuenca Grau: most features held out of owl-lite can be recovered through "back doors"
Bernardo Cuenca Grau: existing document includes fragments which
Bernardo Cuenca Grau: .... are well understood, documented, etc.
Bernardo Cuenca Grau: we don't expect users to go over recent literature on tractable fragments, so wanted a single document
Bernardo Cuenca Grau: most of the languages I will describe are "families" of languages, we decided to keep 1 from each
Bernardo Cuenca Grau: 1st is EL family
Bernardo Cuenca Grau: ...used in bio-medical already
Bernardo Cuenca Grau: stress that these fragments are not academic exercises, there are direct applications to existing ontologies
Bernardo Cuenca Grau: 2nd is DL-Lite family
Bernardo Cuenca Grau: ... designed for large number of instances in database technology
Bernardo Cuenca Grau: approach is similar to what zhe described, do work in tbox, then pass to database system for query answering
Carsten Lutz: reiterate bernardo, but contrast with zhe's approach. dl-lite change the ontology to use database technology, not change the database technology
Alan Ruttenberg: another difference is in oracle you can query for classes, in dl-lite only instances
Bernardo Cuenca Grau: you can do tbox reasoning, but designed for abox answering.
Bernardo Cuenca Grau: I picked the particular dl-lite language b/c it is between rdfs schema and owl dl
Bernardo Cuenca Grau: next is Horn-SHIQ
Bernardo Cuenca Grau: ...can reason without disjunctions
Bernardo Cuenca Grau: ...and low complexity for query answering
Bernardo Cuenca Grau: other fragments dlp as a bridge to rules
Bernardo Cuenca Grau: ... but it may be more "hacky" that horn-shiq
Bernardo Cuenca Grau: questions for wg
Bernardo Cuenca Grau: ....1 do we fix owl lite
Bernardo Cuenca Grau: ....2 does that mean select one of these fragments
Bernardo Cuenca Grau: ....3 or do we present a menu of fragments?
Bernardo Cuenca Grau: not in slides - do we want semantic subsets of owl full?
Bernardo Cuenca Grau: ....e.g., owl full versions of these fragments? do we care about complexity of the full fragments? about compatibility?
Ivan Herman: request to drop side conversations
Ivan Herman: ... and focus
Alan Ruttenberg: little time, can we start with semantic subset of owl full?
Ian Horrocks: semantic subset means no change to syntax, but sanction smaller set of conclusions
Jeremy Carroll: example is pd*, which specifies what semantic rules are thrown away
Peter Patel-Schneider: pd* throws away *parts* of rules
Ian Horrocks: this is picky
Alan Ruttenberg: how comfortable are people with this type of fragment
Alan Ruttenberg: ... does anyone want to say this is a lousy idea.
Peter Patel-Schneider: yes, its lousy b/c you can be arbitrarily picky
Ian Horrocks: its a lousy idea b/c it blows away the idea of interoperability
Bijan Parsia: qualm that methodological design principles are "unclear"
Bijan Parsia: ...guidance for making decisions seem more arbitrary, a dangerous rat-hole
Bijan Parsia: ... would rather people say they are incomplete than building incompleteness into fragments
Jeremy Carroll: in response to ian, any semantic subsetting would need to be clear that it is a subset of spec and an explicit, agreed semantic subset
Jeremy Carroll: ...e.g., oracle and hp would agree on semantic subset and interop on at-least the semantic subset
Alan Ruttenberg: if we call this fragment or conformance level, it seems useful
Alan Ruttenberg: ...that baseline entailments are necessary, but additional entailments may be ok
Bijan Parsia: if we shift from language fragments to reasoner conformance I'm more comfortable
Sandro Hawke: Bijan, "Reasoner Conformance" might be a more useful notion here than "Language Fragments".
Sandro Hawke: ...I have examples of people specifying this at a tool level.
Jeremy Carroll: i'd be happy with such a rewording. i don't see it as notable
Alan Ruttenberg: does such a distinction help others
some affirmation to alan in room
Zhe Wu: ?
Alan Ruttenberg: he said it would be useful to say we support same entailments
Ian Horrocks: more comfortable defining conformance that fragments
Bernardo Cuenca Grau: users are comfortable with incomplete reasoning. swoop offering rdfs reasoner as a choice is an example of this
Bernardo Cuenca Grau: ...more comfortable with that than trying to specify semantic subsets
Jeff Pan: i agree with bernardo and others.
Jeff Pan: ... implementation does not specify fragment.
Boris Motik: i just looked at pd* , this seems like definition. I think it is a useful fragment if evaluated a certain way.
Ian Horrocks: i didn't say pd* was bad, that we'd be standardizing an implementation. it was a reaction to jeremey's comments on what hp and oracle might do
Sandro Hawke: owl is unique to me b/c it doesn't specify what the tools do, people read into that. specifying the tools would be useful. as a customer I expect that and would like it
Bijan Parsia: justifying discomfort - seems likely that over time fragments specified in such a way are likely to move
Bernardo Cuenca Grau: on sandro's comment - we should specify reasoning services
Bernardo Cuenca Grau: ... it's not in the spec for OWL DL. for fragments the services descriptions would be useful
Sandro Hawke: i don't know what the terms are, the market decides
Ian Horrocks: its difficult to imagine semantic subsets not drifting apart
Ian Horrocks: ... it has been a success for owl that interoperability is so good, considering
Jeremy Carroll: responding to standardizing tools - yes. there is value to user if they know different tools perform the same
Jeremy Carroll: ... this wg could provide appropriate conformance levels where vendors and user community come together
Jeremy Carroll: ... clear that motivations from academic community are useful, but they aren't the only motivations
Alan Ruttenberg: no one is saying market is unimportant
Uli Sattler: clarification on user needs?
Jeremy Carroll: users need some sort of specification, but don't need to know behavior is exact
Alan Ruttenberg: I want to poll for consensus on how to procede
Sandro Hawke: i don't understand
Alan Ruttenberg: I want to know if people think these fragments are useful
Alan Ruttenberg: ... defined as a minimum set of entailments
Bijan Parsia: reasoners can conform to the language to different degrees
Alan Ruttenberg: we should aim for something specified like pd*
Ian Horrocks: declarative...
Alan Ruttenberg: yes, declarative
Uli Sattler: we would later know e.g., what it would mean for a reasoner to conform to particular level?
Alan Ruttenberg: yes.
Jeff Pan: what does conformance level mean? is it in terms of benchmark?
Uli Sattler: provides example
Carsten Lutz: degrees of incompleteness?
Alan Ruttenberg: degree of completeness
Alan Ruttenberg: ...fragments are syntactic fragments
Alan Ruttenberg: ...conformance levels are distinct
Jeff Pan: there might be difference between alan's and uli's suggestions
Alan Ruttenberg: distinction is unimportant now
Alan Ruttenberg: reads Q1 as above
Session 8 (Day 2, 16:15 - 18:00)
The text of this section is automatically included from F2F1 Minutes Session 8. Edits should be done there.
(Scribe changed to Uli Sattler)
Organizational Things and Scribing
Jeremy Carroll: has resigned from UFDT, but wants to cancel next monday?
Alan Ruttenberg: will arrange next UFDT
ACTION: on AlanR to arrange next UFDT meeting
Sandro Hawke: has seen 7 sessions' minutes, currently 57 pages and asks how to read to accept them and asks the scribes, when cleaning them up, to add sub headers: syntax is "===" for sub headers
ACTION: AlanR to arrange next UFDT meeting
Michael Smith: asks whether to serialize shuffled subdiscussions
Sandro Hawke: yes, please disentangle
Ian Horrocks: asks what to do with parallel discussions, esp. on the IRC
Sandro Hawke: keep them if they are relevant
Sandro Hawke: scribes finish cleaning up this wednesday
Jeremy Carroll: wants to see actions & resolutions in the minutes
Bijan Parsia: subgroups affected by discussions at F2F should update their documents with pointers to minutes
Jeremy Carroll: suggests to minimize effort on minutes
Alan Ruttenberg: asks for subjects for discussions
Bijan Parsia: non-OWL full issues with RDF mapping
Alan Ruttenberg: agrees with Bijan, mentions reification
Bijan Parsia: axioms annotation asserted versus reified
Alan Ruttenberg: wants to see both
Ian Horrocks: we already agreed that we should explore both assertions & reifications
Sandro Hawke: what about b-nodes and reification?
RDF Mapping Issues and Minutes (mixed discussion)
Bijan Parsia: can we discuss now some RDF mapping issues?
Michael Smith: has added such an issue wrt declarations
Bijan Parsia: you can't specify a signature for an ontology without using the elements of that signature in an axiom or a declaration (which requires owl11 terms)
Matthew Horridge: reports on user complaints regarding declarations
Joanne Luciano: asks for a summary sections of minutes and wants to discuss evaluation issues
Peter Patel-Schneider: disagrees with JLucianos suggestions: scribes should never paraphrase
Alan Ruttenberg: suggests to have summaries outside minutes
Ian Horrocks: suggests to post summaries on the mailinglist
Sandro Hawke: add links to presentations in minutes
Ian Horrocks: suggests clean up/mark up other material as well
Sandro Hawke: mentions that chairs could, if they wanted, blog meetings
Jeremy Carroll: doesn't want to do them
Alan Ruttenberg: hasn't seen a lot about evaluation
Jeremy Carroll: wants to give 2 examples reg. OWL Full compatibility: (1) we have an OWL11 document with reified annotions, we save and modify it....
Matthew Horridge: do we discuss punning or declaredAs?
Jeremy Carroll: is worried about (starts reading out from the text above)
Jeremy Carroll: there are various rules like this one, and they are the wrong ones
Bijan Parsia: understands why: if we have r subproperty of s, and then I add a composition, then this addition would lead to a different kind of serialization
Bijan Parsia: suggests that using different syntax for SubPropertyOf would solve this issue
Bijan Parsia: this is different from round tripping
Jeremy Carroll: suggests to have some form of switch that safes an ontology in OWL11, then we shouldn't expect it to be saved in an OWL10 format unless I require this explicitly
Matthew Horridge: asks whether the spec shouldn't specify this behaviour
Michael Smith: asks whether Jeremy wants tools to save ontologies only in OWL10 if explicitly asked to do so
Jeremy Carroll: observes a subtle relationship between the 2 OWL syntaxes
Ian Horrocks: comes back to AlanR' spoint, and points out that it would introduce nasty non-determinism wrt serialisation and that our n-ary disjointness axioms would cause trouble
Alan Ruttenberg: regards it as a bug to have these 2 possibilities for reading/serializing n-ary disjointness
Bijan Parsia: sees an issue with the mapping, we need to decide what to do with it: deal with it or not. And it would be nice to be clear on our decision in the spec and to have test cases.
Alan Ruttenberg: declarations fall into similar league
Matthew Horridge: disagree - we can throw them in/out
Alan Ruttenberg: asks whether we like declarations
Matthew Horridge: mentions that we can have both, declarations and roundtripping, but with a different mapping
Matthew Horridge: mentions discussions on the mailinglist
Boris Motik: explains that there are 2 readings of declarations. Declarations, possibly using rdf:type can and should be used for linting/simple syntactic check
Bijan Parsia: adds that we can also throw out some "used terms"
Boris Motik: wants to distinguish declaredAs from type. Since there is no notion of typing of RDF, things become problematic, especially with imports, e.g, do we need to re-declare when importing? In the owl10, there was no difference between "class" and "declaration"
Alan Ruttenberg: there wasn't even a notion of an ontology containing an axiom
Bijan Parsia: there is something about documents and ontologies (how to get one from the other)
Jeremy Carroll: suggests to use lateral thinking to solve this: use a new way of imports, namely one where we put import statements at the top of our ontologies and then all declarations will be there!
Boris Motik: seems to agree that this will help tools - if I knew what the type of things are, I could use streaming mode
Bijan Parsia: if they come late, they can still be useful (eg to find typos), but they are most useful at the top
Jeremy Carroll: suggests that we can do this via searching & process imports first
Boris Motik: asks whether typed vocabulary will be obsolote - if yes, we can re-use it
Boris Motik: we can merge the notion of typing and declarations, but cleanly
Jeremy Carroll: wouldn't it make a difference wrt model theory
Boris Motik: no, it's all syntax
Matthew Horridge: we need orphaned entities rather than declarations
Michael Smith: paraphrases that we want to be clear whether rdf:type is a declaration or something else?
Boris Motik: can we add a class to an ontology without adding an axiom? Declarations are a way to mention an entity outside any axiom.
Alan Ruttenberg: asks whether in OWL11, can we have X owl:class Class?
Bijan Parsia: yes, it's in OWL Full, but it disappears in the OWL DL mapping and in the XML syntax
Sandro Hawke: wants to add next F2F meeting to agenda
Bijan Parsia: wants to see from Boris examples explicating differences and consequences of both solutions
Alan Ruttenberg: and we need to check our claims re. what appears/disappears in mappings
Alan Ruttenberg: wants to see backwards compatibility on the agenda
Peter Patel-Schneider: F2F2 will be on April 3 and 4, in the Washington DC area, venue to be determined. OWLED might be in the area, but perhaps not. One possibility is to make use of NIST, but access is restricted
Evan Wallace: access to NIST is a bit tricky, but only first time
Peter Patel-Schneider: downtown DC or near to NIST are possible, too.
Peter Patel-Schneider: considers the possibility to move 1 day earlier to make AlanRector happier
Joanne Luciano: has mentioned MITRE
Peter Patel-Schneider: says that access at MITRE is even more difficult than at NIST. This time of year, the DC area will be extremely busy and thus we need to book Hotels early
Bijan Parsia: offers to make use of C&P rooms
Peter Patel-Schneider: reinforces the need to book hotels early
Peter Patel-Schneider: will come up with proposal together with Kendall Clark
ACTION: ppatelsc to tell us by 2 weeks where F2F2 will be
Alan Ruttenberg: wants to talk about backwards compatibility
Ian Horrocks: closes, thanks Sean Bechhofer for hosting