W3C home > Mailing lists > Public > public-vocabs@w3.org > June 2014

Re: Inverse properties

From: Dan Brickley <danbri@google.com>
Date: Sat, 7 Jun 2014 19:40:41 +0100
Message-ID: <CAK-qy=7Dhjp2rdZ91HPABHT8ML74xArdwkASgd3FJZu4+WuxNw@mail.gmail.com>
To: ☮ elf Pavlik ☮ <perpetual-tripper@wwelves.org>
Cc: W3C Web Schemas Task Force <public-vocabs@w3.org>, Martin Hepp <martin.hepp@ebusiness-unibw.org>, Markus Lanthaler <markus.lanthaler@gmx.net>
On 7 June 2014 10:37, ☮ elf Pavlik ☮ <perpetual-tripper@wwelves.org> wrote:
> On 04/14/2014 12:11 PM, Dan Brickley wrote:
>> On 14 April 2014 11:03, martin.hepp@ebusiness-unibw.org
>> <martin.hepp@ebusiness-unibw.org> wrote:
>>> FYI: I just created and populated a W3C wiki page for the topic:
>>>
>>> https://www.w3.org/wiki/WebSchemas/InverseProperties
>>
>>
>> Thanks Martin! That's been on my todo list. However I don't yet see
>> any content in the page ("There is currently no text in this page
>> ..."), perhaps some problem saving it?
>
> Looking at:
> http://www.w3.org/wiki/WebSchemas/InverseProperties#Linking_from_member_pages_to_an_externally_defined_association.2C_sports_team_entity.2C_or_other_group
>
> I have impression that
>
> http://schema.org/member
>
> already has inverse property
>
> http://schema.org/memberOf

Yes, indeed.

In fact there's a new release of the site in preparation (I'll send a
full msg monday) which makes 'inverse' relations navigable, as well as
sub-property / super-property links in a property hierarchy too.

See http://sdopending.appspot.com/member http://sdopending.appspot.com/memberOf
http://sdopending.appspot.com/alumni etc.

It would be very useful to collect any more pre-existing pairs of
inverse properties here.

> Maybe we should document all properties which already have inversed
> properties defined? Would it make sense to make them 'legacy' pattern
> just as properties with plural names?

No, they are sometimes useful, because of the conceptual and syntactic
overhead of doing things with @rev(erse) notation. We don't want to
add inverses for all schema.org properties but we may add a few more
here and there.

Dan

> I worry that apps consuming data
> following schema.org concepts may need to deal with both ways of stating
> the same relationship, when pair of inverse properties exist but one can
> *also* use @rev
>
> I think section *Instances of (Event) may appear as values for the
> following properties* can give a good clue about when to use inverse
> properties.
>
>
> For one case i need to use inverse of
> http://schema.org/event
>
> While http://schema.org/location seems to make sense for
> http://schema.org/Place in case of http://schema.org/Organization I may
> need to use http://schema.org/event as inverse, meaning ORGANIZER of
> this event (somehow matching one in: http://www.w3.org/TR/rdfcal/#testdr
> ), or maybe it matches as inverse better http://schema.org/attendee and
> we may need to add http://schema.org/organizer |
> http://schema.org/host similar as we have http://schema.org/performer
> I already proposed adding it in:
> http://lists.w3.org/Archives/Public/public-vocabs/2013Nov/0082.html
>
> BTW somehow relevant discussion in Hydra CG:
> http://lists.w3.org/Archives/Public/public-hydra/2014Jun/0025.html
>
> ""IME, a lot of people (outside the SemWeb community) find traversing
> links in
> the reverse direction "unnatural" and "hacky".
>
> It may also be that it is much more efficient (performant) to follow the
> forward link instead of the reverse links as there are much fewer of
> them."" -- Markus Lanthaler
>
> My apologies if I throw to many loosely related thoughts in single
> message ;)
>
>
Received on Saturday, 7 June 2014 18:41:09 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:29:42 UTC