Guide: review - boring (this review not the guide)

Summary: Mike's spellchequer needs to have an RDF/XML mode.

I split my review into two parts, the boring bits here, and comments that
are a little more interesting; the interesting comments probably get a
message each, at least one of them touches on Pat's doc as well.

The boring comments are split into the moderately boring, and deadly boring.

I have only really reviewed the guide with an RDF/XML hat - I haven't
attempted to comment on its pedagogical value.

Moderately Boring
=================

History section - move to appendix?

[[
New information cannot it cannot negate any previous information. Facts and
entailments can only be added, never negated.
]] - typo and how about adding
"OWL <em>does</em> allow negative information to be explicitly stated."
or something like that.

[[
MinCardinality defaults to 0. There is no default value for cardinality or
maxCardinality.
]]
misleading, unnecessary, delete (there are no defaults).



Deadly Boring
=============

(global)
whitespace is significant in RDF literal values, and is not trimmed.
Hence probably not what was intended:
[[[
<owl:Class rdf:ID="WINE">
  <rdfs:subClassOf rdf:resource="#POTABLE-LIQUID"/>
  <rdfs:label xml:lang="en"> wine </rdfs:label>
  <rdfs:label xml:lang="fr"> vin </rdfs:label>
  ...
</owl:Class>
]]]

I suggest
<owl:Class rdf:ID="WINE">
  <rdfs:subClassOf rdf:resource="#POTABLE-LIQUID"/>
  <rdfs:label xml:lang="en">wine</rdfs:label>
  <rdfs:label xml:lang="fr">vin</rdfs:label>
  ...
</owl:Class>

And simply incorrect (many times)
[[[
    <owl:Restriction>
      <owl:onProperty rdf:resource="#REGION"/>
      <owl:minCardinality> 1 </owl:minCardinality>
    </owl:Restriction>
]]]
" 1 " is not an integer, you need "1".
i.e.

[[[
    <owl:Restriction>
      <owl:onProperty rdf:resource="#REGION"/>
      <owl:minCardinality>1</owl:minCardinality>
    </owl:Restriction>
]]]


Many times:
[[
rdf:parseType="collection"
]]
sp: Collection



(approx doc order)
</owl:ontology>   sp: big O
<rdf:rdf          sp: <rdf:RDF
</rdf:rdf>        sp: </rdf:RDF>

[[
The syntax ID="WINE-REGION" is
]]
I prefer ... rdf:ID=...
(ID by itself is illegal)

[[
This is the familiar ID attribute defined by XML.
]]
no it's not.
How about
[[
This is the
<a
href="http://www.w3.org/TR/rdf-syntax-grammar/#idAttr">
rdf:ID</a> attribute
which is like the familiar ID attribute defined by XML.
]]
[[
Given this namespace definition, VIN:WINERY can be used to
]]
misleading, the URI http://www.w3.org/2003/??/owl-guide#WINERY refers to the
resource. the QName VIN:WINERY can be used only on typed node, property
element and property attribute constructions. (See on, you used this
incorrectly later).
[[
SubclassOf is the fundamental
]]
sp: <code>subClassOf</code> even at the beginning of sentence?


[[
<owl:Thing rdf:about="VIN:CENTRAL-COAST-LOCATION">
   <type rdf:resource="VIN:CALIFORNIA-REGION"/>
</owl:Thing>
]]
illegal VIN is not a URI scheme name. "<type" must be namespace qualified.
What you probably mean is:

<owl:Thing
rdf:about="http://www.w3.org/2003/??/owl-guide#CENTRAL-COAST-LOCATION">
   <rdf:type
rdf:resource="http://www.w3.org/2003/??/owl-guide#CALIFORNIA-REGION"/>
</owl:Thing>

If you want to shorten those URIs you need to use XML entities.
[[[
<owl:ObjectProperty id="MADE-FROM-GRAPE">
  <rdfs:domain rdf:resource="#WINE"/>
  <rdfs:range rdf:resource="#WINE-GRAPE"/>
</owl:ObjectProperty>
]]]
sp: rdf:ID

[[[
    <owl:Restriction>
      <owl:onProperty rdf:resource="#MADE-FROM-GRAPE"/>
      <minCardinality> 1 </minCardinality>
    </owl:Restriction>
]]]
sp: owl:minCardinality x 2

[[[
TransitiveProperty
...
P(x,y) and P(y,z) iff P(x,z)
]]]
iff should read ->

[[[
     <owl:minCardinality> 2000 </owl:minCardinality>
]]]
whitespace error
[[[
<owl:sameIndividualAs about="#EVENING-STAR" />
]]]
sp: rdf:about

[[[
  <owl:differentIndividualFrom about="#PARIS-FRANCE" />
]]]
sp: rdf:about


[[[

<WINE-COLOR rdf:about="#WHITE">
  <rdf:label> White </rdf:label>
</WINE-COLOR>
]]]
I prefer "White" wine to " White " wine.


Jeremy

Received on Monday, 30 September 2002 14:42:34 UTC