- From: Wes Turner <wes.turner@gmail.com>
- Date: Mon, 20 Jun 2016 15:39:25 -0500
- To: Dan Brickley <danbri@google.com>
- Cc: Phil Barker <phil.barker@hw.ac.uk>, public-schema-course-extend@w3.org
- Message-ID: <CACfEFw81z-ZnEmpajj5-CZmKs-r-jtR7acTTnJD9U7oWKmTBdA@mail.gmail.com>
Fair enough.
What's likely going to occur here, then, is something like with
https://health-lifesci.schema.org/MedicalStudy
- MedicalStudy exists because experts took the time to develop an excellent
domain-specific vocabulary
- ScientificStudy does not yet exist
- ScientificStudy is a proper superclass of MedicalStudy
- domains and ranges will then need to be widened to accomodate
https://health-lifesci.schema.org/MedicalTrialDesign <-
ScientificTrialDesign (or ScientificStudyDesign, StudyDesign)
- MedicalTrialDesign > RandomizedTrial
- StudyDesign > MedicalTrialDesign > RandomizedTrial
- StudyDesign > RandomizedStudy
So,
while I share the concerns about scope creep
( https://github.com/schemaorg/schemaorg/issues/1127 )
I hesitate to make this overly domain specific because
we want it now
when very simple abstract
Credential and CredentialInstance (or CredentialAssertion)
superclasses now
could save much re-work
and schema confusion later.
C: Thing > CreativeWork > Credential
- P: issuer (d: {Credential, CredentialInstance}, r: {Person, Organization,
CredentialInstance})
C: CredentialAssertion (was CredentialInstance)
- P: recipient ( https://schema.org/recipient )
- P: credential (d: {CredentialAssertion,}, r: {Credential})
This effort is part of a broader scope;
without a base class,
class expansions wont work consistently.
On Mon, Jun 20, 2016 at 2:21 PM, Dan Brickley <danbri@google.com> wrote:
> On 20 June 2016 at 19:51, Phil Barker <phil.barker@hw.ac.uk> wrote:
> >
> > In the long run you may be right, Wes. However, my proposal is that we
> don't
> > need to do credentials in order to describe Course. We can, and should,
> > leave that to people who know more about credentials, as a separate
> effort.
> > We don't need to wait for that, we can add in what works for now, and
> > improve later.
>
> +1
>
> > Phil
> >
> > On 20/06/16 18:49, Wes Turner wrote:
> >>
> >> P: educationalCredentialAwarded
> >> d: Course
> >> r: {Text, Thing}
> >>
> >> Should this be
> >>
> >> P: credentialAwarded
> >> d:
> >> r: {Credential, CredentialInstance}
> >>
> >> P: educationalCredentialAwarded (subPropertyOf) credentialAwarded
> >> d: Course
> >> r: {EducationalCredential, CredentialInstance}
> >>
> >> ...
> >>
> >> Credential > EducationalCredential > Degree
> >> Credential > EducationalCredential > Degree > BachelorDegree
> >> Credential > EducationalCredential > Certificate
> >> Credential > EducationalCredential > Certification
> >> Credential > EducationalCredential > Badge > OpenBadge
> >> Credential > EducationalCredential > Badge > cert-schema (OpenBadge +
> >> blockchain)
> >>
> >> *Are there instances where e.g. Certificate, Certification, and Badge*
> >> *are not EducationalCredentials*
> >> *(just regular Credentials)?*
> >>
> >> So,
> >> EducationalCredential may be more useful as an annotation class?
> >>
> >> CTI models this as credType, which requires a WHERE _ IN query to list
> >> things of this type (because there's no expansions of class lineage
> without
> >> Enumerated classes which are a subClassOf e.g. Credential or
> >> EducationalCredential or Degree or Certificate):
> >> *
> https://github.com/schemaorg/schemaorg/issues/195#issuecomment-222379559
> >>
> >> Scope justification:
> >> -
> https://github.com/schemaorg/schemaorg/issues/195#issuecomment-223663304
> >>
> >> On Mon, Jun 20, 2016 at 8:05 AM, Phil Barker <phil.barker@hw.ac.uk
> >> <mailto:phil.barker@hw.ac.uk>> wrote:
> >>
> >> On 20/06/16 12:26, Dan Brickley wrote:
> >>
> >> On 17 June 2016 at 17:32, Steve Midgley
> >> <steve@learningtapestry.com
> >> <mailto:steve@learningtapestry.com>> wrote:
> >>
> >> I agree that credential has many meanings across
> >> industries. I agree we
> >> should try to find a term that unambiguously locates this
> >> credential as
> >> educational. So, being a frequent simpleton, I'll suggest
> >> "educationalCredential"
> >>
> >> This is an improvement.
> >>
> >> A friend I was talking to over the weekend suggested using the
> >> related
> >> term "accreditation". How does that sound to this community? To
> my
> >> ears it has all the right associations, and is much less
> >> evocative of
> >> lower-level technical notions of credential...
> >>
> >> Dan
> >>
> >> OK. As property name I am going to suggest
> >> educationalCredentialAwarded because we might in the future want
> >> EducationalCredential as a class name.
> >>
> >> So, my proposal is that in order to meet the use case that people
> >> can search for courses that offers a qualification the searcher
> >> would like to acquire we create a new property
> >>
> >> educationalCredentialAwarded domain Course, range Text or Thing.
> >> Definition: a description of the qualification, award,
> >> certificate, diploma or other educational credential awarded as a
> >> consequence of successful completion of this course.
> >>
> >> When the educational credentials/verifiable claims community have
> >> sorted out how they want to describe their domain in schema.org
> >> <http://schema.org> then I hope we will have some more specific
> >> schema type(s) that we can point to, but for now this seems to me
> >> to be good enough to solve the use case. Solving the bigger issue
> >> seems beyond the scope of this community group.
> >>
> >>
> >> Phil
> >>
> >> -- --
> >> Phil Barker @philbarker
> >> LRMI, Cetis, ICBL http://people.pjjk.net/phil
> >> Heriot-Watt University
> >>
> >> Ubuntu: http://xkcd.com/456/
> >> not so much an operating system as a learning opportunity.
> >>
> >>
> >>
> >
> >
> > --
> > --
> > Phil Barker @philbarker
> > LRMI, Cetis, ICBL http://people.pjjk.net/phil
> > Heriot-Watt University
> >
> > Ubuntu: http://xkcd.com/456/
> > not so much an operating system as a learning opportunity.
> >
> >
>
Received on Monday, 20 June 2016 20:39:56 UTC