Open cygri opened 14 years ago
added xsd:boolean and do not plan to add more for the time being
Hi fadmaa, I repeat my request to allow users to choose any regular xsd datatype ;-)
Reason 1: Otherwise, one cannot use the extension to create proper GoodRelations data. With GoodRelations being the one single ontology from the Semantic Web adopted by Google, I think it makes sense to support this usage.
Reason 2: You cannot now which range a property from a given ontology defines, so not allowing a user to correctly model the value is a limitation, and, to be frank, kind of arbitrary.
Reason 3: With your constraints, the resulting data will not match properly in SPARQL queries, at least in the case of xsd:string (which is needed for content that is not human language text but other alphanumeric data). As for xsd:float - again, you prevent users from modeling the data as shown in the regular recipes.
An arbitrary breach with the RDF specification will bring problems in the long run.
So please, try to find a way that allows experts to use any valid xsd datatype while not confusing other users. Maybe a simple "experts mode" button?
Thanks!
Martin
Hi, Thanks for the input!
The discussion seems fairly convincing and having "expert mode" looks like a win-win solution.
Maybe by making all xsd types available via some sort of "advanced" option? This will make them one further click away but will keep the interface simple
I will definitely consider this in the upcoming release!
Regards, Fadi
Martin Hepp requested support for xsd:float, xsd:string, xsd:boolean.
I can see the argument for xsd:boolean, but xsd:float is a bit redundant (why would anyone not choose xsd:double instead?) and adding xsd:string is in my eyes counterproductive, one should simply use plain literals instead. Deciding between plain literal and xsd:string is just one more completely meaningless decision that users would have to make.