--- a/model/ProvenanceModel.html Wed Dec 21 16:05:42 2011 +0000
+++ b/model/ProvenanceModel.html Wed Dec 21 18:48:40 2011 +0000
@@ -2066,11 +2066,11 @@
<section id="record-complement-of">
-<h4>View Record</h4>
+<h4>Alternate and Specialization Records</h4>
<div class="note">This section is currently under revision and in flux</div>
-<p>A <dfn id="view">view record</dfn> is used to establish a relationship between two entity records, which asserts that the two records provide a different characterization of the same entity. Consider for example three entity records:
+The purpose of the record types defined in this section is to establish a relationship between two entity records, which asserts that the two records provide a different characterization of the same entity. Consider for example three entity records:
<ul>
<li><span class="name">e1</span> describing "Bob, the holder of facebook account ABC",
@@ -2080,10 +2080,10 @@
<li><span class="name">e3</span> describing "Bob, the person".
</ul>
-One may make several assertions to establish that these entity records refer to the same entity "Bob", either in different contexts, or at different levels of abstraction. For example:
+One may make several assertions to establish that these entity records refer to the same the real-world characterized thing Bob, either in different contexts, or at different levels of abstraction. For example:
<ol>
- <li> Each of <span class="name">e1</span> , <span class="name">e2</span> provide a <em>more concrete</em> characterization of Bob than <span class="name">e3</span> does
+ <li> Each of <span class="name">e1</span> , <span class="name">e2</span> provide a <em>more concrete</em> characterization of Bob than <span class="name">e3</span> does
<li> <span class="name">e1</span> and <span class="name">e2</span> provide two different characterizations of the same entity.
</ol>
@@ -2092,90 +2092,57 @@
<ul>
- <li> B is a view of A, written <span class="name">ViewOf(B,A)</span> captures the intent of assertion (1);
- <li> B is a [foobar] of A, written <span class="name">foobar(B,A)</span> captures the intent of assertion (2).
+ <li> B is a <em>specialization</em> of A, written <span class="name">specializationOf(B,A)</span> captures the intent of assertion (1);
+ <li> B is an <em>alternative characterization</em> of A, written <span class="name">alternateOf(B,A)</span> captures the intent of assertion (2).
</ul>
In order to further convey the intended meaning, the following properties are associated to these two relations.
<ul>
- <li><span class="name">ViewOf(B,A)</span> is <strong>transitive</strong>: <span class="name">ViewOf(C,B)</span> and <span class="name">ViewOf(B,A)</span> implies <span class="name">ViewOf(C,A)</span>.
-
- <li><span class="name">ViewOf(B,A)</span> is <strong>anti-symmetric</strong>: <span class="name">ViewOf(B,A)</span> implies that <span class="name">ViewOf(A,B)</span> does not hold.
+ <li><span class="name">specializationOf(B,A)</span> is <strong>transitive</strong>: <span class="name">specializationOf(C,B)</span> and <span class="name">specializationOf(B,A)</span> implies <span class="name">specializationOf(C,A)</span>.
+
+ <li><span class="name">specializationOf(B,A)</span> is <strong>anti-symmetric</strong>: <span class="name">specializationOf(B,A)</span> implies that <span class="name">specializationOf(A,B)</span> does not hold.
+ <li><span class="name">alternateOf(B,A)</span> is <strong>symmetric</strong>: <span class="name">alternateOf(B,A)</span> implies <span class="name">alternateOf(A,B)</span>.
</ul>
-
- <ul>
- <li><span class="name">foobar(B,A)</span> is <strong>transitive</strong>: <span class="name">foobar(C,B)</span> and <span class="name">foobar(B,A)</span> implies <span class="name">foobar(C,A)</span>.
-
- <li><span class="name">foobar(B,A)</span> is <strong>symmetric</strong>: <span class="name">foobar(B,A)</span> implies <span class="name">foobar(A,B)</span>.
- </ul>
-
-<h5>Case of entities with known limited validity</h5>
+<!--
+ <h5>Case of entities with known limited validity</h5>
As we know from [entity record], entity records may only be valid within certain event's interval. Let <span class="name">char(e)</span> denote the validity interval of <span class="name">e</span>.
When these intervals are known, relations <span class="name">ViewOf(e2,e1)</span> and <span class="name">foobar(e2,e1)</span> can only be asserted if
the validity interval of <span class="name">e1</span> is a sub-interval of that of <span class="name">e2</span>. This condition ensures transitivity.
-
+ -->
+<p>
+An entity record identifier can optionally be accompanied by an account identifier. When this is the case, it becomes possible to use the <span class="name">alternateOf</span> relation to link two entity record identifiers that are appear in different accounts. (In particular, the entity identifiers in two different account are allowed to be the same.). When account identifiers are not available, then the linking of entity records through <span class="name">alternateOf</span> can only take place within the scope of a single account.
+</p>
+
+
<!--
-
- The relation <span class="name">ViewOf(e2,e1)</span> itself is only meaningful if those two intervals overlap, and then only for the extent of such overlap. If the entity records are valid in disjoint intervals, the <span class="name">ViewOf</span> assertion is meaningless by definition.<br/>
-Note that, as a particular case, the validity interval of <span class="name">e1</span> may be <em>contained</em> within the validity interval of <span class="name">e1</span>.<p/>
-
-With these assumptions, the <span class="name">ViewOf</span> relation enjoys two properties:
-
-<ul>
-<li> <strong>symmetry</strong>: if <span class="name">ViewOf(e2,e1)</span> holds in some event interval, then <span class="name">ViewOf(e1,e2)</span> also holds in the same interval.
-
-<li> <strong>pseudo-transitivity</strong>: Given three entity records <span class="name">e1,e2,e3</span> <em>with overlapping validity intervals</em>: if <span class="name">ViewOf(e3,e2)</span> and <span class="name">ViewOf(e2,e1)</span> hold, then <span class="name">ViewOf(e3,e1)</span> also holds <em>for the extent of their common interval</em>. <br/>
-
- </ul>
- Note that this second property is a weak form of transitivity, which only applies on condition that the derived transitive assertion <span class="name">ViewOf(e3,e1)</span> is meaningful, that is, that the validity interval of the involved entity records do overlap.
-
- There is no guarantee that this is always the case, as the simple example in the figure below (b) illustrates. In this case, no new relations are asserted from the pseudo-transitivity rule.
-
-<div style="text-align: center;">
-<img src="viewOf.png" alt="illustration viewOf"/>
-</div>
-
-<p>
-An entity record identifier can optionally be accompanied by an account identifier. When this is the case, it becomes possible to use the <span class="name">ViewOf</span> relation to link two entity record identifiers that are appear in different accounts. (In particular, the entity record identifiers in two different account are allowed to be the same.). When account identifiers are not available, then the linking of entity records through <span class="name">ViewOf</span> can only take place within the scope of a single account.
-</p>
-
-
-<div class="note">This is a proposal to add a further relation strictViewOf, which however still subject to discussion. <p/>
-
- It may be desirable to provide a stronger connotation of the notion of "different characterization", namely to state that an entity record <span class="name">e1</span> is <em>more constrained</em>, or provides a <em>more specific</em> characterization of a entity, relative to another entity record <span class="name">e2</span>. For example, one would like to assert that the entity record that describes "Bob, the holder of facebook account ABC" refers to the same entity as the entity record "Bob, the person", but offers a more specific characterization of such entity. To accommodate this stronger semantics, we introduce a variant of the <span class="name">ViewOf(e2,e1)</span> relation, denoted by
- <span class="name">StrictViewOf(e2,e1)</span>. By definition, this relation is <strong>anti-symmetric</strong>, that is: <br/>
-if <span class="name">strictViewOf(e2,e1)</span> then it follows that <span class="name">strictViewOf(e1,e2)</span> does not hold. <br/>
-<span class="name">StrictViewOf(e2,e1)</span> also enjoys the pseudo-transitivity property, as specified above.
-</div>
-
<div class="anexample">
<div class="note">RS example to be replaced once we reach consensus on the definition</div>
</div>
-
-
-<p>In PROV-ASN, a view record's text matches the <span class="nonterminal">viewRecord</span> (resp. <span class="nonterminal">strictViewRecord</span>) production of the grammar defined in this specification document.</p>
+-->
+
+<p>In PROV-ASN, an alternate record's text matches the <span class="nonterminal">alternateRecord</span>production of the grammar defined in this specification document.</p>
<div class='grammar'>
-<span class="nonterminal">viewRecord</span> ::=
-<span class="name">viewOf</span>
+ <span class="nonterminal">alternateRecord</span> ::=
+ <span class="name">alternateOf</span>
<span class="name">(</span>
<span class="nonterminal">eIdentifier</span>
<span class="name">,</span>
<span class="nonterminal">eIdentifier</span>
<span class="name">,</span>
<span class="nonterminal">optional-attribute-values</span>
-<span class="name">)</span> <br/>
+<span class="name">)</span> <br/>
|
- <span class="name">viewOf</span>
+ <span class="name">alternateOf</span>
<span class="name">(</span>
<span class="nonterminal">eIdentifier</span>
<span class="name">,</span>
@@ -2186,36 +2153,52 @@
<span class="nonterminal">accIdentifier</span>
<span class="name">,</span>
<span class="nonterminal">optional-attribute-values</span>
-<span class="name">)</span>
+<span class="name">)</span>
</div>
--->
+<p>In PROV-ASN, a specialization record's text matches the <span class="nonterminal">specializationRecord</span>production of the grammar defined in this specification document.</p>
+
+<div class='grammar'>
+ <span class="nonterminal">specializationRecord</span> ::=
+ <span class="name">specializationOf</span>
+<span class="name">(</span>
+<span class="nonterminal">eIdentifier</span>
+<span class="name">,</span>
+<span class="nonterminal">eIdentifier</span>
+<span class="name">,</span>
+<span class="nonterminal">optional-attribute-values</span>
+<span class="name">)</span> <br/>
+|
+ <span class="name">specializationOf</span>
+<span class="name">(</span>
+<span class="nonterminal">eIdentifier</span>
+<span class="name">,</span>
+ <span class="nonterminal">accIdentifier</span>
+ <span class="name">,</span>
+<span class="nonterminal">eIdentifier</span>
+<span class="name">,</span>
+ <span class="nonterminal">accIdentifier</span>
+<span class="name">,</span>
+ <span class="nonterminal">optional-attribute-values</span>
+<span class="name">)</span> <br/>
+
+</div>
+
+
+<div class='issue'>A discussion on alternative definition of these relations has not reached a satisfactory conclusion yet. This is <a href="http://www.w3.org/2011/prov/track/issues/29">ISSUE-29</a></div>
+
+
+</section>
+</section>
+
+
<!--
-<p>A <dfn id="complementOf">complementarity record</dfn> is a relationship between two entities stated to have compatible characterization over some continuous interval between two events.</p>
-
-
-
-<p>
-The rationale for introducing this relationship is that in general, at any given time, for an entity in the world, there may be multiple ways of characterizing it, and hence multiple representations can be asserted by different asserters. In the example that follows, suppose thing "Royal Society" is represented by two asserters, each using a different set of attributes. If the asserters agree that both representations refer to "The Royal Society", the question of whether any correspondence can be established between the two representations arises naturally. This is particularly relevant when (a) the sets of attributes used by the two representations overlap partially, or (b) when one set is subsumed by the other. In both these cases, we have a situation where each of the two asserters has a partial view of "The Royal Society", and establishing a correspondence between them on the shared attributes is beneficial, as in case (a) each of the two representation <em>complements</em> the other, and in case (b) one of the two (that with the additional attributes) complements the other.</p>
-
-<p>This intuition is made more precise by considering the entities that form the representations of entities at a certain point in time.
-
-An entity record represents, by means of attribute-value pairs, a thing and its situation in the world, which remain constant over a characterization interval.
-As soon as the thing's situation changes, this marks the end of the characterization interval for the entity record representing it. The thing's novel situation is represented by an attribute with a new value, or an entirely different set of attribute-value pairs, embodied in another entity record, with a new characterization interval. Thus, if we overlap the timelines (or, more generally, the sequences of value-changing events) for the two entities, we can hope to establish correspondences amongst the entity records that represent them at various points along that events line. The figure below illustrates this intuition.</p>
-
<div style="text-align: center;">
<img src="complement-of.png" alt="illustration complementOf"/>
</div>
-<p>
-Relation <em>complement-of</em> between two entity records is intended to capture these correspondences, as follows. Suppose entity records A and B share a set P of attributes, and each of them has other attributes in addition to P. If the values assigned to each attribute in P are <em>compatible</em> between A and B, then we say that <em>A is-complement-of B</em>, and <em>B is-complement-of A</em>, in a symmetrical fashion. In the particular case where the set P of attributes of B is a strict superset of A's attributes, then we say that <em>B is-complement-of A</em>, but in this case the opposite does not hold. In this case, the relation is not symmetric. (as a special case, A and B may not share any attributes at all, and yet the asserters may still stipulate that they are representing the same thing "Royal Society". The symmetric relation may hold trivially in this case).</p>
-
-<p>The term <em>compatible</em> used above means that a mapping can be established amongst the values of attributes in P and found in the two entity expession. This generalizes to the case where attribute sets P1 and P2 of A, and B, respectively, are not identical but they can be mapped to one another. The simplest case is the identity mapping, in which A and B share attribute set P, and furthermore the values assigned to attributes in P match exactly.</p>
-
-<p>It is important to note that the relation holds only for the characterization intervals of the entity expessions involved As soon as one attribute changes value in one of them, new correspondences need to be found amongst the new entities. Thus, the relation has a validity span that can be expressed in terms of the event lines of the entity.</p>
-
@@ -2290,24 +2273,8 @@
</div>
-
-<div class='note'>It is suggested that the name 'wasComplementOf' does not capture the meaning of this relation adequately. No concrete suggestion has been made so far.
-Furthermore, there is a suggestion that an alternative relation that is transitive may also be useful.
-This is raised in the following <a href="http://lists.w3.org/Archives/Public/public-prov-wg/2011Sep/0315.html">email</a>.</div>
-
-->
-<div class='issue'>A discussion on alternative definition of wasComplementOf has not reached a satisfactory conclusion yet. This is <a href="http://www.w3.org/2011/prov/track/issues/29">ISSUE-29</a></div>
-
-
-
-<div class='pending'> Comments on ivpof in <a href="http://www.w3.org/2011/prov/track/issues/57">ISSUE-57</a>.</div>
-
-
-</section>
-</section>
-
-
<!--
<section>
<h4>Transitive Derivation Record</h4>