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

A note on "non" and "non-"

From: Antoine Zimmermann <antoine.zimmermann@deri.org>
Date: Wed, 22 Apr 2009 12:24:51 +0100
Message-ID: <49EEFE83.1070609@deri.org>
To: 'W3C OWL Working Group' <public-owl-wg@w3.org>
The use of prefix "non" or "non-" is very inconsistent in almost all 
documents. Both forms are often used within the same document for the 
same word (e.g., "nonnegative" and "non-negative").

I can see 4 options:
1) use the same form for all terms (i.e., always use "non-" or always 
use "non");
2) have the spelling normalized in all the documents (e.g., always 
"nonempty" and "non-deterministic");
3) have each single document use the same form consistently (e.g., 
"nonnegative" in SS&FS and "non-negative" in RDF Semantics);
4) do not care.

I personally favour 1), and I prefer the spelling with an hyphen, but I 
understand that it is common practice to remove the hyphen in American 
English (but it is not possible in some case). In any case, I would not 
object any option, even 4) at this stage.

Following is a comprehensive list of all occurrences of such terms.

Document overview
In header: "non-normative formats"

In header: "non-normative formats"
In TOC: "17.2 Nonnormative References"
Sect.1, 1st parag.: "(normative and nonnormative)"
Sect.2.1: "The { ordered,nonunique } attribute"
Sect.2.2, Tab.1: "nonterminal symbols"
  "Nonnegative integers are defined as usual"
  "nonNegativeInteger := a nonempty finite sequence"
  "languageTag := @ (U+40) followed a nonempty sequence"
Sect.4.1, Lexical Spaces: "a positive, nonzero integer"
Sect.4.7, 1st Ex.: "but nonidentical data values"
Sect.5, 2nd parag.: "associate nonlogical information"
Sect.5.8: "These axioms are nonlogical"
Sect.8.3.1: "consists of a nonnegative integer n"
Sect.8.3.2: "consists of a nonnegative integer n"
Sect.8.3.3: "consists of a nonnegative integer n"
Sect.8.4.2: "and therefore have non-UK and non-Canadian"
Sect.8.5.1: "consists of a nonnegative integer n"
Sect.8.5.2: "consists of a nonnegative integer n"
Sect.8.5.3: "consists of a nonnegative integer n"
Sect.11.2, Restriction on the Usage of Anonymous Individuals, 2nd Ex.: 
"non-tree-like way"
Sect.12, Optional parameters: "transfering non-ASCII data"
  "nonNegativeInteger := a nonempty finite"
  "languageTag := @ (U+40) followed a nonempty sequence"
Sect.17.2: "17.2 Nonnormative References"

RDF Mapping
In header: "non-normative formats"
Sect.3: "n is a nonnegative integer"

Direct Semantics
In header: "non-normative formats"
In TOC: "5.2 Nonnormative References"
  "is a nonempty set called the object domain"
  "is a nonempty set disjoint with"
Sect.5.2: "5.2 Nonnormative References"

RDF Semantics
In header: "non-normative formats"
EdNote after TOC: "Non-breakable whitespace"
Sect.4.1: "takes non-negative integers as its values"
Sect.4.2: "IR is the universe of I, i.e., a nonempty set"
Sect.5, Conventions used in this Section: "INNI: The set of all 
non-negative integers"
Sect.5.3: "to any non-trivial semantic conclusions"
Sect.5.7, 2nd parag.: "usually, a non-empty facet space"
Sect.6, 1st parag.: "it becomes non-obvious"
Sect.7.3, Proof of the Balancing Lemma, item B:, 3rd bullet: "they are 
Sect.7.3, End of the Proof of the Balancing Lemma:
  "every non-built-in IRI"
  "thus all non-built-in IRIs"
  "For every non-built-in IRI u"
  "of a non-separated vocabulary"
  "all the non-built-in names in gA"
  "For every non-built-in IRI u occurring in gA"
Sect.8.4, 1st parag.: "for any non-negative integer"
Sect.9, Weakened Semantics of Data Value Enumerations [DEV]: "now 
restricted to non-empty sets"
Sect.9, Modified Abbreviations [NOM]: "the explicit non-abbreviated form"
Sect.10, Section 2: "of non-semantic aspects"
Sect.10: "Other non-editorial changes"

In header: "non-normative formats"
In TOC: " Non-Entailment Tests"
Sect.2.3: "When using non-RDF serializations"
Sect. " Non-Entailment Tests"
Sect., 1st parag.:
  "Non-Entailment tests"
  "and a non-conclusion ontology"
Sect., 2nd parag.:
  "In all non-entailment tests"
  "all non-entailment tests"
Sect. "if they were non-entailment tests"
Sect.3.2.2: "conclusion, or non-conclusion"
Sect.3.2.4: "might be a non-entailment"
Sect.4, 1st parag.: "entailment tests and non-entailment tests"

In header: "non-normative formats"
In TOC: "8.2 Nonnormative References"
Sect.4.1: "the need for nondeterministic reasoning"
Sect.5, 2NEXPTIME: "nondeterministic"
Sect.5, NEXPTIME: "nondeterministic"
Sect.5, NP: "nondeterministic"
Sect.5, LOGSPACE: "NLOGSPACE is the non-deterministic version"
Sect.8.2: "8.2 Nonnormative References"

In header: "non-normative formats"
Sect.1.2, 2nd parag.: "easier for non-logicians to read"
Sect.4.8: "with non-negative integers"
Sect.4.8, last parag.: "with non-negative integers"

In header: "non-normative formats"
Sect., ObjectMinCardinality: "n is a non negative integer"
Sect., ObjectMaxCardinality: "n is a non negative integer"
Sect., ObjectExactCardinality: "n is a non negative integer"
Sect., DataMinCardinality: "n is a non negative integer"
Sect., DataMaxCardinality: "n is a non negative integer"
Sect., DataExactCardinality: "n is a non negative integer"
Sect.3.1.3, 3rd bullet: "existence of non named individuals"
Sect.4.1.2: "normative and non normative"

In header: "non-normative formats"
Sect.2: " "n" is a non-negative integer"
Sect.3.2, column Value v: "Nonnegative integer"

XML Serialization
In header: "non-normative formats"
In TOC: "8.2 Nonnormative References"
Sect.6, Optional parameters: "transfering non-ascii"
Sect.8.2: "8.2 Nonnormative References"

Manchester Syntax
In header: "non-normative formats"
In TOC: "6.2 Non-normative References"
Sect.2, Tab.1: "nonterminal symbols"
  "one for non-empty lists"
  "one for non-empty lists with annotations"
  "two terminals or non-terminals"
  "terminals or non-terminals if its"
  "the above non-terminals."
Sect.2.1: "languageTag := @ (U+40) followed a nonempty sequence"
Sect.4.1, 3rd parag.: "lose some non-logical information"
Sect.4.2, 2nd table: "Nonterminal"
Sect.4.3, 3rd parag.: "The non-determinism in the mapping"
Sect. 5, Optional parameters: "transfering non-ascii"
Sect.6.2: "6.2 Non-normative References"

In header: "non-normative formats"
Sect.1, 1st bullet: "Similarly, non-linear"
Sect.5, 2nd bullet: "implementation of non-linear (in)equations"
Sect.5, 4th bullet: "Prolog system with non-linear inequations"

Antoine Zimmermann
Post-doctoral researcher at:
Digital Enterprise Research Institute
National University of Ireland, Galway
IDA Business Park
Lower Dangan
Galway, Ireland
Received on Wednesday, 22 April 2009 11:25:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 22 April 2009 11:25:34 GMT