Re: property/class ambiguity in languages with no letter case

Thanks Bernard,

On *this* occasion I'm trying not to get into the debate about whether 
it is or is not good practice to use foo->Foo although I agree fully 
that it would certainly be a good thing to discuss in this task force 
leading to a BP doc. Here I just want to know what to tell Shuji (the 
Japanese chap who is doing the translation).

Where the English original is foo/Foo, can this be interpreted in 
Japanese as 'has foo' / Foo. (I think it can but I'm covering my 
backside :-) )

Phil.

On 11/02/2014 09:34, Bernard Vatant wrote:
> Hi Phil
>
> I've always been uneasy with those classes and properties names and URIs
> with just an initial case difference.
> Not only for the translation in languages with no capitalization, but also
> to avoid systems not case-sensitive to be confused.
> I strongly stick to having different names and URIs whatever the syntactic
> trick used (has, is, whatever). Coming out with a best practice
> recommendation on this would be a good task for W3C vocabularies task
> force, BTW.
>
> My 0.02
>
> Bernard
>
>
>
> 2014-02-10 21:12 GMT+01:00 Phil Archer <phila@w3.org>:
>
>> In the last few months I've been encouraging people to provide translated
>> labels, definitions and usage notes for vocabularies hosted in w3.org/nsspace [1]. The latest one being worked on is a translation of ORG [2] into
>> Japanese, but this has thrown up a problem. ORG uses property names
>> beginning with lower case letters, (foo) to link to classes named
>> identically except that they begin with a capital letter (Foo)*.
>>
>> In languages with upper and lower case letters this is not a problem, but
>> what about those that don't, like Japanese?
>>
>> Other schemas tend to use verbs as properties and nouns as class names, so
>> we might have hasFoo linking to Foo. I am not trying to re-open the debate
>> about which is preferable, merely to ask:
>>
>> Where a vocabulary uses foo and Foo as property and class names
>> respectively, to the extent that it might help translation into languages
>> without upper and lower case letters, do you agree that we can help the
>> translator by suggesting he/she treats the property name 'foo' as 'has foo?'
>>
>> Phil.
>>
>> * the case that came up is role and Role but I'm trying to generalise.
>>
>> [1] http://www.w3.org/blog/data/2014/01/06/vocabularies-at-w3c/
>> [2] http://www.w3.org/ns/org.ttl
>> --
>>
>>
>> Phil Archer
>> W3C Data Activity Lead
>> http://www.w3.org/2013/data/
>>
>> http://philarcher.org
>> +44 (0)7887 767755
>> @philarcher1
>>
>>
>
>

-- 


Phil Archer
W3C Data Activity Lead
http://www.w3.org/2013/data/

http://philarcher.org
+44 (0)7887 767755
@philarcher1

Received on Tuesday, 11 February 2014 09:47:43 UTC