Re: Use of the word "should" in OWL Rec

>>
>> In addition to the possibly inappropriate use of the word SHOULD in 
>> Section 3.1.2.1.1 owl:allValuesFrom (cf. the original posting), I hit 
>> another passage in Section 4.1.1 rdfs:subPropertyOf 
>> (http://www.w3.org/TR/owl-ref/#subPropertyOf-def):
>>
>> "[..] Formally this means that if P1 is a subproperty of P2, then the 
>> property extension of P1 (a set of pairs) SHOULD be a subset of the 
>> property extension of P2 (also a set of pairs)."
>>
>> Comparing this with the semantics in 
>> http://www.w3.org/TR/owl-absyn/direct.html#3.3, which states that 
>> ER(p1) ⊆ ER(p2), the use of SHOULD here again seems misleading.
>>
>> Could someone be so kind and comment on why SHOULD is appropriate in 
>> those sections?
>>
>> Thanks!
>> Martin
>>
> 
> 
> To me "SHOULD" in capitals would have been a mistake.
> 
> The OWL Ref is intended as an informative document, and hence should not 
> be making any use of the keywords. Hence, any use of 'should' in OWL Ref 
> should be read just as 'should' not with RFC 2119 semantics. However a 
> capitalized SHOULD is so indicative of 2119 that it was good that it was 
> not used. Your 'quote' is in fact a misquote, since what you wrote as 
> 'SHOULD' is in fact 'should'.

I don't feel like elaborating on this, but since the usage of RFC 2119 
is neither stated in my email nor in the OWL Reference I don't see my 
quotation as a misquote, I just used it to emphasize what is the most 
relevant part of the quote.

If the OWL Reference is informative (in the sense that it is not 
normative), then this should be stated somewhere, since everyone expects 
a recommendation to be normative except explicitly stated otherwise 
(e.g., with a section). Hm, the only thing I found about "informative" 
in the OWL Reference is "This document gives a systematic, compact and 
informative description of [..]". Yes, I find it quite informative, 
except for the 'should' thing.

> 
> (I don't find the wording here brilliant, but look at a dictionary 
> rather than RFC 2119)
> 

I agree, brilliant is something different.

Martin

> Jeremy
> 
> 
> 

Received on Friday, 16 July 2004 03:31:17 UTC