Re: [Proposal] schema:NotApplicable

This reminds me of a discussion that we had around the library data 
standard. We already had "not applicable" but then someone decided that 
we needed to add "no attempt to code" for those situations where it 
MIGHT have been applicable, but that the library in question wasn't 
going to bother even trying to give that information. I was the only one 
who argued that people who aren't even going to bother to try to give 
the information are also the ones least likely to correctly code that 
they aren't bothering to give the information.

Mainly, I think that "not applicable" is pretty meaningless (not 
applicable why?) so if someone is stateless then it should be

"nationality": "stateless"

And if someone purposely does not have a telephone number, it could be

"telephone": "none"

And if you don't want to spam the Pope with your dating service 
(although that could be because he has a wide choice of dates already), 
you could have:

"datingStatus": "not interested"

Because

"spouse": "notApplicable"

is incredibly vague. The person could be single, widowed, be secretly 
married, be in a culture where marriage does not confer "spouse-ness" or 
"spouse-ness" could simply be irrelevant to the context in question.

kc

On 9/20/14, 6:39 AM, Dan Brickley wrote:
> On 20 September 2014 14:21, ☮ elf Pavlik ☮
> <perpetual-tripper@wwelves.org> wrote:
>> Hi,
>>
>> I see various use cases where I would like to state *NotApplicable*
>>
>> {
>>    "@context": "http://schema.org",
>>    "@id": "http://wwelves.org/perpetual-tripper",
>>    "@type": "Person",
>>    "nationality": "NotApplicable",
>>    "taxID": "NotApplicable",
>>    "vatID": "NotApplicable"
>> }
>>
>> To my understanding by simply omitting those properties I live it open
>> for reasoning - "Maybe he has nationality, taxID and vatID but just
>> doesn't publish it". Creating schema:NotApplicable or something in this
>> line would provide way to make such statements *explicit*. Since we
>> don't talk about boolean values here schema:False doesn't look to me
>> like something that fits.
>>
>> Following https://github.com/rvguha/schemaorg/issues/120
>> "Explore adding link to Role into per type, per property navigation
>> boilerplate"
>> Having such NotApplicable allowed on all properties would make a lot of
>> sense to me. I could also see possibility of having one more generic
>> type for that and making NotApplicable one of its sub types.
>>
>> All feedback appreciated!
>
> Can you give an example of who/how this might get used? Omission is a
> *lot* simpler. I suspect a much of the time you're saying
> "NotApplicable" you might equally well be saying "ItsComplicated".
>
> What's the nationality of Robin Hood? of Jesus Christ? Probably more
> "Its Complicated" than "Not Applicable". What were their heights and
> weights? Probably "DontKnow". What were their tax IDs? Given all the
> bible quotes around tax, this might be DontKnow, NotApplicable or
> ItsComplicated for JC, and something similarly messy for the possibly
> fictional Robin Hood.
>
> I guess my gut reaction is:
>
> Q: What's worse than having a http://schema.org/faxNumber property
> applicable to http://schema.org/Volcano?
> A: Having every Volcano description include { "faxNumber": "NotApplicable" }
>
> Do you have particular scenarios in mind where this would be needed,
> e.g. around nationality/tax/vat?
>
> cheers,
>
> Dan
>
>

-- 
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
m: 1-510-435-8234
skype: kcoylenet/+1-510-984-3600

Received on Saturday, 20 September 2014 20:25:57 UTC