Re: What happens when an ontology has data literals that are outside the range handled

On 12 Feb 2009, at 22:43, Alan Ruttenberg wrote:

> To where can I point Jos to in our spec that defines this behavior?

? We don't define the behavior I described. We say that  
implementations must support at least 16 digits. If they don't support  
more we are silent on what they should do. I described two  
possibilities.

> The reason I ask is that in some cases the XML docs say operations
> involving such literals are implementation specific.

Well we don't say what non-conforming implementation should do. If an  
implementation supports on 16 digits that doesn't license it to do  
what it wants with 17 digit numbers. It can, of course, do what it  
wants, but that behavior wouldn't be conforming.

> I contended that
> that would not be the case for OWL but didn't find a place that said
> this explicitly.

I don't know that there is. We're nowhere nearly as rigid about  
conformance as I would, ideally, like. But that's probably poltico- 
socially wise.

I think at the moment the advice I'd give users is that if an  
implementation doesn't do static and dynamic analysis to determine if  
it's overflowed and throw *at least* a warning, that they complain  
about the implementation saying, "You claim to handle only 16 digits,  
but you really handle more, just not in conformance with the spec,  
loser"

Cheers,
Bijan.

Received on Thursday, 12 February 2009 23:06:29 UTC