RE: Private review of the Document Overview

Hi Ian!

I'm happy with your response.

I can see that the figure has just been updated, e.g. from "M'ter" to
"Manchester", etc., so that's fine now, either.

Concerning a more distinctive short-description of the Direct Semantics
(last point, in the table), we can have a discussion about this later.

Cheers,
Michael

>-----Original Message-----
>From: public-owl-wg-request@w3.org [mailto:public-owl-wg-request@w3.org]
>On Behalf Of Ian Horrocks
>Sent: Wednesday, April 15, 2009 1:12 PM
>To: OWL 1.1
>Cc: Michael Schneider
>Subject: Re: Private review of the Document Overview
>
>Thanks for the careful reading and useful comments -- I believe that
>I dealt with most of them.
>
>Ian
>
>On 10 Apr 2009, at 13:34, Michael Schneider wrote:
>
>> Hi!
>>
>> I finally found the time to have a look at the Document Overview.
>> All my points are editorial and probably simple to cope with.
>>
>> Best,
>> Michael
>>
>> * General: Consider writing all names (of syntaxes, semantics, etc)
>> with capitalized letters. It's a bit incoherent at the moment.
>
>I'm leaving this for a WG decision on what is appropriate (after LC)!
>
>>
>> * Abstract, 2nd par: The term "various" is used in two consecutive
>> sentences. Choose a synonym for one of the two occurrences.
>
>Deleted the 2nd various as it was superfluous.
>
>>
>> * Table of Contents: The "[Show Short TOC]" link seems exaggerated
>> for this short document. Consider removing it (it's actually not in
>> all our documents, e.g. not in the Direct Semantics).
>
>I would, but somebody seems to have beaten me to it.
>
>>
>> * §1 (Introduction), last par: "OWL 1 and OWL 2 are designed
>> to ...". Perhaps better something like: "As OWL 1, OWL 2 is
>> designed..." The document, after all, is a document about OWL 2.
>
>OK
>
>>
>> * §2, Figure, syntax layer: Write the full names of the syntaxes,
>> i.e. avoid "M'ter. syntax" or "func. syntax". If necessary, use
>> three lines: "Manchester/Syntax/document".
>>
>> * §2, Figure, syntax layer: "turtle" is written "Turtle" (capital
>> "T") later in the text. So should be in the figure, either.
>>
>> * §2, Figure, semantics layer: "RDF Based Semantics" in Semantics
>> layer: add the "-" between "RDF" and "Based".
>
>I agree with all three of the above comments, but I don't control the
>figure. Fine if Sandro/Ivan can fix it, but I don't believe that any
>are critical for LC -- but let's not forget to fix them later.
>
>>
>> * §2.2 (Syntaxes), 2nd par: "an XML serialization". Dangerous after
>> this LC phase! It should become clear that this is a specific XML
>> serialization that closely reflects the structure of OWL
>> constructs, and that it is clearly distinguished from RDF/XML.
>
>I'm not sure if this wouldn't add to any danger: questions might
>arise as to why other syntaxes *don't* closely reflect the structure
>of the constructs, and the "clearly distinguished" part sounds like
>"protesting to much".
>
>>
>> * §2.2 (Syntaxes), 2nd par: "and a more readable syntax used in
>> several ontology editing tools [OWL 2 Manchester Syntax]".
>> ** Please explicitly say the name of that syntax! Currently, one
>> only sees it from the citation mark, and the format of citation
>> marks is possibly going to change in the future.
>> ** Put the citation mark directly behind the name of the syntax
>> then. Currently, it looks as if the citation mark refers to
>> "ontology editing tools".
>
>OK
>
>>
>> * §2.2 (Syntaxes), 2nd par: "the functional-style syntax can also
>> be used for serialization, although its main purpose is specifying
>> the structure of the language". Can we say this about its purpose?
>> I thought the structure of the language is primarily specified by
>> the UML diagrams, although the functional syntax closely
>> corresponds to the diagrams. Maybe, it's better to say something
>> like that the "main purpose is to serve as an abstract syntax for
>> the language"? This would also make sense in this context.
>
>Maybe. I'll think about it.
>
>>
>> * §2.3 (Semantics), 1st par: "to answer queries about, e.g., class
>> consistency, subsumption and instance retrieval." Somethings wrong
>> with this sentence, I think: one doesn't answer queries /about/
>> instance retrieval?
>
>Restructured.
>
>>
>> * §2.3 (Semantics), 3rd par:
>> ** 1st sentence: The citation mark "[OWL 2 RDF-Based Semantics]"
>> should be placed directly behind "RDF-Based Semantics".
>
>OK
>
>> ** 1st sentence: The citation mark should be replaced by "[RDF
>> Semantics]".
>> ** 1st sentence: "compatible with _the_ RDF Semantics" (missing
>> "the").
>> ** end of par: That citation mark there can be dropped, since there
>> is already one at the beginning of the par.
>
>OK to all.
>
>>
>> * §2.3 (Semantics), 4th par: "The correspondence theorem in Section
>> 7.3 of the RDF-Based Semantics Document". It's Section 7.2 now:
>> <http://www.w3.org/2007/OWL/wiki/RDF-
>> Based_Semantics#Correspondence_Theorem>
>
>OK
>
>>
>> * §2.4 (Profiles). For OWL 2 QL, nothing is said about its
>> computational complexity, while there are such assertions for the
>> other two profiles ("polinomial time ..."). Do we want to say
>> something for QL, either?
>
>Possible -- we could say "... enables conjunctive queries to be
>answered in LogSpace (AC^0) using standard relational database
>technology". Given the sensitivity surrounding descriptions of
>profiles I didn't add it at the moment.
>
>>
>> * §4 (Roadmap), 2nd par: "and two alternative concrete syntaxes
>> (XML and Manchester)." Should be "OWL/XML" instead of "XML".
>
>OK
>
>>
>> * §4, table: The Direct Semantics is characterized by "defines the
>> meaning of OWL 2 ontologies in terms of a model-theoretic
>> semantics." This is exactly true for the RDF-Based Semantics,
>> either. The latter is more specifically characterized by "an
>> extension of the RDF Semantics". For the Direct Semantics, I
>> suggest to say something like "compatible with the SROIQ
>> description logic", because Section 2.3 already characterizes the
>> Direct Semantics in just this way.
>
>If anything, I would prefer to say something like "a standard first
>order model-theoretic semantics. But I fear that this might be
>offensive to some. No names, no pack drill.
>
>Ian

--
Dipl.-Inform. Michael Schneider
Research Scientist, Dept. Information Process Engineering (IPE)
Tel  : +49-721-9654-726
Fax  : +49-721-9654-727
Email: michael.schneider@fzi.de
WWW  : http://www.fzi.de/michael.schneider
=======================================================================
FZI Forschungszentrum Informatik an der Universität Karlsruhe
Haid-und-Neu-Str. 10-14, D-76131 Karlsruhe
Tel.: +49-721-9654-0, Fax: +49-721-9654-959
Stiftung des bürgerlichen Rechts, Az 14-0563.1, RP Karlsruhe
Vorstand: Prof. Dr.-Ing. Rüdiger Dillmann, Dipl. Wi.-Ing. Michael Flor,
Prof. Dr. Dr. h.c. Wolffried Stucky, Prof. Dr. Rudi Studer
Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus
=======================================================================