W3C home > Mailing lists > Public > public-owl-wg@w3.org > April 2009

Re: general review of NF&R

From: Christine Golbreich <cgolbrei@gmail.com>
Date: Tue, 14 Apr 2009 20:45:43 +0200
Message-ID: <b0ed1d660904141145p699afeb7q826b677c98891077@mail.gmail.com>
To: "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>
Cc: public-owl-wg@w3.org
Hi Peter and all,

As we dealt with this review off-line, here most part of the response
for the record.

Thank you for your review and helpful contribution to slim down NF&R
and improve it.

We accepted most of the suggested changes.
Some earlier content that you removed has been restored at some places
though, mainly when it was there (and added) for LC responses (hope we
did not cut too much).
We slightly modified the visual aspect of your changes to make the
meaning and syntax of the new features appear more clearly.
As you suggested, we moved the Profiles section out of 2 Features &
Rationale into its own section 3 and put back the minor features
subsections into section 2 (as it was earlier in the FPWD). I agree
that it makes more sense because the profiles is an important new
feature of OWL 2.
This in counterpart induces a few disavantages: features numbered from
1 to 15 are those considered as the main features, they all exhibit
UCs, while minor features don't. As a consequence, there is a break
between F15 and F1 to F14, some subsections exhbit now only one item
e.g. 2.4 and 2.6.
But these are small editorial details in my view, which are minor
compared to the benefits,  and might slightly be improved later, if
really necessary.

see details inline

Christine


2009/3/31 Peter F. Patel-Schneider <pfps@research.bell-labs.com>


- Afficher le texte des messages précédents -
> Review of NF&R
>
>
> General Comments:
>
> 1: NF&R is becoming too much for too many things.  I would much prefer a
> cut-down version that concentrates on what NF&R is supposed to be - an
> overview of the new stuff in OWL 2 as related to the use cases and other
> rationale that the WG has identified.  I actually don't mind the overall
> organization of NF&R, just its length.
>
> 2: There are quite a few changes needed to fix grammatical and wording
> problems in the document and to reduce the amount of unneeded prose.  I
> have made detailed edits up to Section 2.2.2 on my paper copy of the
> document and can apply them to the document if it is so desired.

> Here is my changed version of the start of Section 2:
>
> *******************************************************
> == Features & Rationale ==
>
> OWL 2 is an update to OWL adding several new features, including
> increased expressive power for properties, extended support
> for datatypes, simple metamodeling capabilities, extended annotation
> capabilities, and keys. OWL 2 also defines several profiles &ndash;
> OWL 2 language subsets that may better meet certain performance
> requirements or may be easier to implement. The new OWL 2 features
> are presented here, organized in the following categories:
>
> # syntactic sugar to make some common statements easier to say,
> # new constructs that increase expressivity,
> # extended support for datatypes,
> # simple metamodeling capabilities,
> # extended annotation capabilities, and
> # other innovations.
>
> Each feature is described in a common pattern as follows:
> * a brief sentence explaining why the new feature was added,
> * a feature description including a informal meaning, informal syntax, and a simple example issued from Use Cases,
> * the theoretical and implementation implications of the new feature, and
> * links to related use cases.


done


> === Syntactic sugar ===
>
> OWL 2 adds syntactic sugar to make some common patterns easier to write.


done

- Afficher le texte des messages précédents -

> ==== F1: [http://www.w3.org/2007/OWL/wiki/Syntax#Disjoint_Union_of_Class_Expressions DisjointUnion]====
>
> While OWL 1 provides means to define a set of subclasses as a disjoint
> and complete covering of a superclass by using several axioms, this
> cannot be done concisely.
> <span class="nonterminal">DisjointUnion</span> defines a class as the
> union of other classes, all of which are pairwise disjoint. It is a
> shorthand for owl:disjointWith statements used in combination with
> owl:unionOf to define a complete superclass from a set of mutually
> disjoint subclasses.
> [[http://www.w3.org/2007/OWL/wiki/Syntax#Disjoint_Union_of_Class_Expressions Normative Syntax]]
> [[http://www.w3.org/2007/OWL/wiki/Direct_Semantics#Class_Expression_Axioms Direct Semantics]]
> [[http://www.w3.org/2007/OWL/wiki/RDF-Based_Semantics#Semantic_Conditions_for_Equivalence_and_Disjointness_Axioms RDF-Based Semantics]]
>
> <div class="grammar">
> <span class="nonterminal">DisjointUnion</span> <span class="name">({ A } C CE<sub>1</sub> ... CE<sub>n</sub> )</span> where <span class="name">C</span> is a class,  <span class="name">CE<sub>i</sub></span>, 1 &le; i &le; n are class expressions, and { A }  zero or more annotations.
> </div>
>
> <div class="anexample">
> * HCLS
> <div class="axioms">
> {| class="axioms"
> |-
> | class="name" | DisjointUnion(''BrainHemisphere'' ''LeftHemisphere'' ''RightHemisphere'' ) (''UC#2'')
> | A ''BrainHemisphere'' is exclusively either a ''LeftHemisphere'' or a ''RightHemisphere'' and cannot be both a ''RightHemisphere'' and a ''LeftHemisphere''.
> |-
> |''-- RDF --''
> |
> |-
> | ''BrainHemisphere'' ''owl:disjointUnionOf'' ( ''LeftHemisphere'' ''RightHemisphere'' )
> |
> |-
> |''-------''
> |
> |-
> | class="name" | DisjointUnion(''Lobe''  ''FrontalLobe''  ''ParietalLobe''  ''TemporalLobe''  ''OccipitalLobe'' ''LimbicLobe'') (''UC#1'')
> | A ''Lobe'' is exclusively either a ''FrontalLobe'' , a ''ParietalLobe'', a  ''TemporalLobe'', a ''OccipitalLobe'' or a ''LimbicLobe'' and cannot be both of them.
> |-
> | class="name" | DisjointUnion(''AmineGroup'' ''PrimaryAmineGroup'' ''SecondaryAmineGroup'' ''TertiaryAmineGroup'' ) (''UC#3'')
> |An ''AmineGroup'' is exclusively either a ''PrimaryAmineGroup'', a ''SecondaryAmineGroup'' or a ''TertiaryAmineGroup'' and cannot be both of them.
> |-
> |}
> </div>
>
> * Automotive industry
>
> <div class="axioms">
> {| class="axioms"
> |-
> | class="name" | DisjointUnion(CarDoor FrontDoor RearDoor TrunkDoor) (''UC#4'')
> | A ''CarDoor'' is exclusively either a ''FrontDoor'', a ''RearDoor'' or a''TrunkDoor'' and not both of them.
> |-
> |}
> </div>
> </div>
>
> Since <span class="name">DisjointUnion</span> is simply a shorthand for
> several ''disjointWith'' statements in combination with unionOf, it does
> not change the expressiveness, semantics, or complexity of the language.
> Implementations, however, may prefer to take special notice of
> DisjointUnion for more efficient processing.
>
> [[#Use_Case_.231_-_Brain_image_annotation_for_neurosurgery_.5BHCLS.5D|Use Case #1]] [[#Use_Case_.232_.E2.80.93_The_Foundational_Model_of_Anatomy_.5BHCLS.5D|Use Case #2]] [[#Use_Case_.233_-_Classification_of_chemical_compounds_.5BHCLS.5D|Use Case #3]] [[#Use_Case_.234_-_Querying_multiple_sources_in_an_automotive_company_.5BAutomotive.5D|Use Case #4]]'''
>
> *******************************************************
>
>
>
> Changes:
>
> 0/ Abstract
>
>  OWL 2 provides new features over the previous version of OWL.  This
>  document provides a description of these new features and other design
>  choices that went into OWL 2, along with their rationale, based on use
>  cases provided to the W3C OWL Working Group.
>
> 1/ Introduction
>
>  This document provides brief overviews of the main new features of OWL
>  2 and their rationale. These language features were determined based
>  on real applications and user and tool-developer experience, some of
>  which has been documented in the OWLED Workshop Series [...].  The
>  inclusion of the features are supported by use cases provided to the
>  W3C OWL Working Group, which are listed in an appendix.
>
>  This document also describes and motivates some of the other design
>  decisions that were made during the development of OWL 2 or
>  purposefully retained from OWL 1, particularly the various concrete
>  syntaxes for OWL 2, and the relationship of OWL 2 with RDF.
>
>  OWL 2 extends the previous OWL Web Ontology Language (OWL 1) [OWL ??]
>  and inherits the language features, design decisions, and use cases
>  for OWL 1.  This document thus forms an extension of the Use Cases and
>  Requirements that underlie OWL 1 [OWL Use Cases and Requirements].
>
> Remove the last paragraph of the introduction.  It is not needed.


Done


> 2/ Features and Rationale
>
> There is no need for the "Feature", ... tags.

This division was required earlier by group members. The different
perspectives are still there but compacted and not tagged, which gains
space.

> Changes similar to those above should be made throughout Section 2.


I did a first pass according your pattern given for F#1 and cut some
useless sentences.


> 3/ Other Design Choices and Rational
>
> This section should also be slimmed down.

TBD

> 4/ Illustrative Use Cases
>
> 4.1/
>
> This subsection only repeats information in Section 2, and should be
> removed.

I think Table 4.1 and 4.2 should not be cut, they summarize, and there
were positive feedbacks
thus moved to the Appendix

> 4.2/
>
> Why are some features given by name and others by number?  This should
> be made consistent.  The legend is also unnecessary, as the information
> is in Section 2.
>

improved the explanation above the Table

explanation is given just above Table:
each UC is associated to several features F1, F2 ... F15,  which are
given by numbers
but for each UC, one single feature and a corresponding example, has
been selected to illustrate that feature : it is given by name in bold





--
Christine



2009/3/31 Peter F. Patel-Schneider <pfps@research.bell-labs.com>:
> Review of NF&R
>
>
> General Comments:
>
> 1: NF&R is becoming too much for too many things.  I would much prefer a
> cut-down version that concentrates on what NF&R is supposed to be - an
> overview of the new stuff in OWL 2 as related to the use cases and other
> rationale that the WG has identified.  I actually don't mind the overall
> organization of NF&R, just its length.
>
> 2: There are quite a few changes needed to fix grammatical and wording
> problems in the document and to reduce the amount of unneeded prose.  I
> have made detailed edits up to Section 2.2.2 on my paper copy of the
> document and can apply them to the document if it is so desired.
>
> Here is my changed version of the start of Section 2:
>
> *******************************************************
> == Features & Rationale ==
>
> OWL 2 is an update to OWL adding several new features, including
> increased expressive power for properties, extended support
> for datatypes, simple metamodeling capabilities, extended annotation
> capabilities, and keys. OWL 2 also defines several profiles &ndash;
> OWL 2 language subsets that may better meet certain performance
> requirements or may be easier to implement. The new OWL 2 features
> are presented here, organized in the following categories:
>
> # syntactic sugar to make some common statements easier to say,
> # new constructs that increase expressivity,
> # extended support for datatypes,
> # simple metamodeling capabilities,
> # extended annotation capabilities, and
> # other innovations.
>
> Each feature is described in a common pattern as follows:
> * a brief sentence explaining why the new feature was added,
> * a feature description including a informal meaning, informal syntax, and a simple example issued from Use Cases,
> * the theoretical and implementation implications of the new feature, and
> * links to related use cases.
>
> === Syntactic sugar ===
>
> OWL 2 adds syntactic sugar to make some common patterns easier to write.
>
> ==== F1: [http://www.w3.org/2007/OWL/wiki/Syntax#Disjoint_Union_of_Class_Expressions DisjointUnion]====
>
> While OWL 1 provides means to define a set of subclasses as a disjoint
> and complete covering of a superclass by using several axioms, this
> cannot be done concisely.
> <span class="nonterminal">DisjointUnion</span> defines a class as the
> union of other classes, all of which are pairwise disjoint. It is a
> shorthand for owl:disjointWith statements used in combination with
> owl:unionOf to define a complete superclass from a set of mutually
> disjoint subclasses.
> [[http://www.w3.org/2007/OWL/wiki/Syntax#Disjoint_Union_of_Class_Expressions Normative Syntax]]
> [[http://www.w3.org/2007/OWL/wiki/Direct_Semantics#Class_Expression_Axioms Direct Semantics]]
> [[http://www.w3.org/2007/OWL/wiki/RDF-Based_Semantics#Semantic_Conditions_for_Equivalence_and_Disjointness_Axioms RDF-Based Semantics]]
>
> <div class="grammar">
> <span class="nonterminal">DisjointUnion</span> <span class="name">({ A } C CE<sub>1</sub> ... CE<sub>n</sub> )</span> where <span class="name">C</span> is a class,  <span class="name">CE<sub>i</sub></span>, 1 &le; i &le; n are class expressions, and { A }  zero or more annotations.
> </div>
>
> <div class="anexample">
> * HCLS
> <div class="axioms">
> {| class="axioms"
> |-
> | class="name" | DisjointUnion(''BrainHemisphere'' ''LeftHemisphere'' ''RightHemisphere'' ) (''UC#2'')
> | A ''BrainHemisphere'' is exclusively either a ''LeftHemisphere'' or a ''RightHemisphere'' and cannot be both a ''RightHemisphere'' and a ''LeftHemisphere''.
> |-
> |''-- RDF --''
> |
> |-
> | ''BrainHemisphere'' ''owl:disjointUnionOf'' ( ''LeftHemisphere'' ''RightHemisphere'' )
> |
> |-
> |''-------''
> |
> |-
> | class="name" | DisjointUnion(''Lobe''  ''FrontalLobe''  ''ParietalLobe''  ''TemporalLobe''  ''OccipitalLobe'' ''LimbicLobe'') (''UC#1'')
> | A ''Lobe'' is exclusively either a ''FrontalLobe'' , a ''ParietalLobe'', a  ''TemporalLobe'', a ''OccipitalLobe'' or a ''LimbicLobe'' and cannot be both of them.
> |-
> | class="name" | DisjointUnion(''AmineGroup'' ''PrimaryAmineGroup'' ''SecondaryAmineGroup'' ''TertiaryAmineGroup'' ) (''UC#3'')
> |An ''AmineGroup'' is exclusively either a ''PrimaryAmineGroup'', a ''SecondaryAmineGroup'' or a ''TertiaryAmineGroup'' and cannot be both of them.
> |-
> |}
> </div>
>
> * Automotive industry
>
> <div class="axioms">
> {| class="axioms"
> |-
> | class="name" | DisjointUnion(CarDoor FrontDoor RearDoor TrunkDoor) (''UC#4'')
> | A ''CarDoor'' is exclusively either a ''FrontDoor'', a ''RearDoor'' or a''TrunkDoor'' and not both of them.
> |-
> |}
> </div>
> </div>
>
> Since <span class="name">DisjointUnion</span> is simply a shorthand for
> several ''disjointWith'' statements in combination with unionOf, it does
> not change the expressiveness, semantics, or complexity of the language.
> Implementations, however, may prefer to take special notice of
> DisjointUnion for more efficient processing.
>
> [[#Use_Case_.231_-_Brain_image_annotation_for_neurosurgery_.5BHCLS.5D|Use Case #1]] [[#Use_Case_.232_.E2.80.93_The_Foundational_Model_of_Anatomy_.5BHCLS.5D|Use Case #2]] [[#Use_Case_.233_-_Classification_of_chemical_compounds_.5BHCLS.5D|Use Case #3]] [[#Use_Case_.234_-_Querying_multiple_sources_in_an_automotive_company_.5BAutomotive.5D|Use Case #4]]'''
>
> *******************************************************
>
>
>
> Changes:
>
> 0/ Abstract
>
>  OWL 2 provides new features over the previous version of OWL.  This
>  document provides a description of these new features and other design
>  choices that went into OWL 2, along with their rationale, based on use
>  cases provided to the W3C OWL Working Group.
>
> 1/ Introduction
>
>  This document provides brief overviews of the main new features of OWL
>  2 and their rationale. These language features were determined based
>  on real applications and user and tool-developer experience, some of
>  which has been documented in the OWLED Workshop Series [...].  The
>  inclusion of the features are supported by use cases provided to the
>  W3C OWL Working Group, which are listed in an appendix.
>
>  This document also describes and motivates some of the other design
>  decisions that were made during the development of OWL 2 or
>  purposefully retained from OWL 1, particularly the various concrete
>  syntaxes for OWL 2, and the relationship of OWL 2 with RDF.
>
>  OWL 2 extends the previous OWL Web Ontology Language (OWL 1) [OWL ??]
>  and inherits the language features, design decisions, and use cases
>  for OWL 1.  This document thus forms an extension of the Use Cases and
>  Requirements that underlie OWL 1 [OWL Use Cases and Requirements].
>
> Remove the last paragraph of the introduction.  It is not needed.
>
> 2/ Features and Rationale
>
> There is no need for the "Feature", ... tags.
>
> Changes similar to those above should be made throughout Section 2.
>
> 3/ Other Design Choices and Rational
>
> This section should also be slimmed down.
>
> 4/ Illustrative Use Cases
>
> 4.1/
>
> This subsection only repeats information in Section 2, and should be
> removed.
>
> 4.2/
>
> Why are some features given by name and others by number?  This should
> be made consistent.  The legend is also unnecessary, as the information
> is in Section 2.
>
>



-- 
Christine
Received on Tuesday, 14 April 2009 18:46:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 14 April 2009 18:46:19 GMT