2

I have included into my ontology a specific data property which is of type xsd:DateTimeStamp as I am looking into this website which provides examples of supposedly acceptable literal values of that format http://www.datypic.com/sc/xsd11/t-xsd_dateTimeStamp.html

I copy pasted both those literal values suggested into my data property separately :

2004-04-12T13:20:00-05:00
2004-04-12T13:20:00Z

But unfortunately both the hermit and the pellet reasoner whine about an inconsistency there. Can someone explain to me why this is wrong and provide a valid literal value example that would pass the reasoners ?

I am using the OWLTime ontology and this is the 'in XSD Date-Time-Stamp' data property to be more precise, and the tool I use is protege 5.5.0

GGEv
  • 843
  • 9
  • 23
  • can you show the sample data please which fails? At least, `xsd:DateTimeStamp` would be wrong as it's case sensitive and should be `xsd:dateTimeStamp`. And your literals are just `xsd:dateTime`, for timestamp you also need the seconds, don't you? Format: `[-]CCYY-MM-DDThh:mm:ssZ | [-]CCYY-MM-DDThh:mm:ss(+|-)hh:mm ` – UninformedUser Nov 21 '19 at 09:52
  • the examples failing are those I quoted above. The Range is indeed xsd:dateTimeStamp. So which is a valid example including seconds ? just write and answer, I ll test it and if reasoners are ok I will accept it immediately – GGEv Nov 21 '19 at 09:56
  • I googled for a valid literal value of that class and could not find anything, only regular expressions mentioning what the ideal format would be like @AKSW – GGEv Nov 21 '19 at 09:57
  • ok, your second literal should be fine, I created a single axiom with the literal in Protege, then started the Pellet reasoner, Pellet does not complain. Do you have some additional axioms leading to the inconsistency? Like a functional data property axiom? Or some wrong range axiom? That's why I was asking for the sample data. – UninformedUser Nov 21 '19 at 10:09
  • I have the class Tag which is a subclass of class Time instant which in turn is a subclass of Temporal Entity, fundamental class of OWLTime, I created the individual Tag1 which is of type Tag. Then I tried to add the in XSD Date-Time-Stamp data property which domains in 'Time instant' and ranges xsd:dateTimeStamp – GGEv Nov 21 '19 at 10:15
  • is it semantically ok to add to the aforementioned data property in domains the Tag class ? – GGEv Nov 21 '19 at 10:16
  • Do you use Protege? If so, once an inconsistency is found a dialog with a questionmark is shown, which provides you explanations why the ontology is inconsistent. Can you add this here please? – UninformedUser Nov 21 '19 at 10:19
  • 1
    The error was when inserting into the box the 2004-04-12T13:20:00Z I should have selected the type to be xsd:dateTimeStamp from the drop down list – GGEv Nov 21 '19 at 10:20

1 Answers1

3

Both Data properties are correct, what was needed upon insertion in the box was to select from the drop down type list the "xsd:dateTimeStamp" then the reasoner stopped complaining. Although it is weird because in other custom occasions I did not deal with the same problem

GGEv
  • 843
  • 9
  • 23
  • 2
    It's not weird, without specifying a datatype your literal becomes a `PlainLiteral` but the range of your data property is `xsd:dateTimeStamp`. Datatypes are disjoint in OWL, thus, your `2004-04-12T13:20:00Z^^rdf:PlainLiteral` is clearly not in the value space of all `xsd:dateTimeStamp` literals - that is clearly a violation, thus, an inconsistency. – UninformedUser Nov 21 '19 at 10:26
  • 1
    What box and what drop down list are you talking about. As far as I know, Pellet and Hermit do not have a user interface, so you must specify what you are talking about. Perhaps you should also tell this in the question. – Antoine Zimmermann Nov 21 '19 at 11:31
  • added that I use protege as a tool – GGEv Nov 21 '19 at 12:04