W3C home > Mailing lists > Public > public-vocabs@w3.org > March 2015

Re: List of commonly used property names when used as *inverse*?

From: Martin Hepp <martin.hepp@unibw.de>
Date: Mon, 30 Mar 2015 20:35:20 +0200
Cc: Gregg Kellogg <gregg@greggkellogg.com>, Melvin Carvalho <melvincarvalho@gmail.com>, W3C Web Schemas Task Force <public-vocabs@w3.org>
Message-Id: <62CC877A-E471-41E3-9754-BDAB167E46FE@unibw.de>
To: ☮ elf Pavlik ☮ <perpetual-tripper@wwelves.org>
My two cents:

Any mechanism for supporting inverse properties *at the level of syntax* are great. Reverse properties in the vocabulary are *evil* ;-) (well, undesirable).

For a previous discussion on this, see also 

    http://www.w3.org/wiki/WebSchemas/InverseProperties

    https://lists.w3.org/Archives/Public/public-vocabs/2014Apr/0200.html


Martin

--------------------------------------------------------
martin hepp
e-business & web science research group
universitaet der bundeswehr muenchen

e-mail:  martin.hepp@unibw.de
phone:   +49-(0)89-6004-4217
fax:     +49-(0)89-6004-4620
www:     http://www.unibw.de/ebusiness/ (group)
         http://www.heppnetz.de/ (personal)
skype:   mfhepp 
twitter: mfhepp





On 30 Mar 2015, at 19:36, ☮ elf Pavlik ☮ <perpetual-tripper@wwelves.org> wrote:

> On 03/30/2015 04:53 PM, Gregg Kellogg wrote:
>> On Mar 30, 2015, at 2:53 AM, Melvin Carvalho <melvincarvalho@gmail.com> wrote:
>>> 
>>> 
>>> 
>>>> On 30 March 2015 at 10:31, ☮ elf Pavlik ☮ <perpetual-tripper@wwelves.org> wrote:
>>>> Hello,
>>>> 
>>>> In Social WG we currently only use JSON-LD as normative serialization. I
>>>> just suggested that instead of defining tons of inverse properties in
>>>> vocabulary(ies), we can simply define aliases for them in JSON-LD context.
>>>> http://www.w3.org/TR/json-ld/#reverse-properties
>>>> 
>>>> Myself I already do it on my website which uses schema.org as base and
>>>> extends it with other vocabularies. Currently I define in JSON-LD @context.
>>>> 
>>>> "attendeeIn": { "@reverse": "attendee" }
>>>> 
>>>> And as of today I start adding
>>>> 
>>>> "authorOf": { "@reverse": "author" }
>>>> "subjectOf": { "@reverse": "about" }
>>>> 
>>>> ( leaving out now issue with not using "@type": "@id" )
>>>> 
>>>> Would you see it making sense to maintained list of such recommended
>>>> names for properties when used as inverse?
>>>> 
>>>> Example of one from The PROV Ontology
>>>> http://www.w3.org/TR/2013/REC-prov-o-20130430/#inverse-names-table
>>> 
>>> I've had this issue also in the past where I have a link from one thing to the other, but then I want to start at one place and link backwords.  I didnt know you could use @reverse in JSON LD in this way, that's a neat finding.  
>>> 
>>> People I know that have experience with modeling and ontologies tell me it's generally a bad idea to create terms for both a relation and it's inverse.  While it may solve an initial pain point, what I am told it that it complicates code bases further along.  I dont know if this applies also to your aliasing use case, but I've been told in the past, to try to pick one direction and use that, it may be something to note.  
>>> 
>> That's my opinion too. Just a reminder that we did add @itemprop-reverse as an experimental feature to Microdata+RDF for just this purpose, and RDFa has @rev (thought, not RDFa Lite).
> Thanks for mentioning @itemprop-reverse in Microdata+RDF and I hope we
> all stay aware about straight forward @rev in RDFa. Still JSON-LD gets a
> lot of traction and having recommended names for inverse of properties
> could save everyone effort of coming up with some sane names.
> 
> Would something like *schema:recommendedInverseName* would make sense
> whenever property doesn't have existing schema:inverseOf ?
> 
> 
Received on Monday, 30 March 2015 18:35:57 UTC

This archive was generated by hypermail 2.3.1 : Monday, 30 March 2015 18:35:57 UTC