Re: EOCred: Identifying subtypes of credential

I have also heard Richard Wallis talking about library metadata saying
something along the lines of if you want to share information about books
between libraries use the appropriate library standard
{MARC,RDA,METS,MODS...}, if you want to share that information with the
rest of the world use schema.org (specifically bib.schema.org).

I would add “*.. preferably do both*”.

You cannot openly share data about a credential, or anything, and control
who consumes that data.  If a consumer is expert in the field they will
want/need the specificity of a domain specific vocabulary. Whereas in the
wider world a link to an authoritative description and some high-level
detail is all they want.  Also in the wider world consumers (eg. Search
engine crawlers) will only be looking for *Things* described using generic
vocabularies they understand — Schema.org.

In general the more consumers, and other data providers, that link to a
description the more authoritative it becomes, and the more likely it
becomes for visitors to view that description in *all* the detail used to
describe it.

The objective of our discussions, to enable description of credentials etc
using Schema.org, is to make their existence, identification(s),
authoritative descriptions, and details more generally available as
structured data on the web.

Some loss of specificity in description, as understood by experts in this
domain, is probably inevitable and should be expected.  However that should
be understood in the context that Schema.org data should be shared to
supplement the visibility (by ease of general consumption) of descriptions,
not to replace current and emerging detailed domain specific
descriptions/vocabularies.

The trick to this kind of process is to hopefully do it (defining
Schema.org needs) whilst not requiring major reengineering of domain
specific data models or, having to convince the Schema.org community to
consider new modelling patterns.

~Richard.



Richard Wallis
Founder, Data Liberate
http://dataliberate.com
Linkedin: http://www.linkedin.com/in/richardwallis
Twitter: @rjw

On 24 January 2018 at 10:18, Phil Barker <phil.barker@pjjk.co.uk> wrote:

> Hello Kerri, everyone.
> On 23/01/18 16:45, Kerri Lemoie wrote:
>
> Hi All,
>
> One gap in understanding that I have is the correlation between the schema
> and the CTDL and also, what are the consequences if they diverge?
>
> That's a good question, one which occurs a lot in connection with
> schema.org. You'll know schema.org's basic mission is about a standard
> way of sharing structured data on web pages with big search engine
> providers like Google (although it's also useful for other things, and it's
> important to understand that schema.org is not a Google product). CTDL
> was developed for the Credential Engine, to meet more specific use cases
> about educational and occupational credentials and going beyond search.
> Schema.org aims web-scale reach, the credential engine works closely with
> partners. So the Credential Engine can ask for quite tight conformance with
> CTDL based on an agreed understanding of a common aim, but schema.org
> isn't really able to. schema.org does have the advantage of breadth and
> reach: it can be used to describe books, videos, events, people, and it is
> used globally. So courses and educational resources from places that are
> not connected with the Credential Engine are being described with
> schema.org.  My hope is that describing Educational and Occupational
> Credentials in schema.org in a way that is compatible with CTDL (and
> other relevant specs) will allow links between the Credential Engine and
> these other places.
>
> This isn't unique to Credentials. I have heard Dan Brickley say something
> along the lines of whatever you domain, there will be a domain specific
> vocabulary that works better for that domain than schema.org. The
> strength of schema.org is in providing a simple way of linking between
> these domains. I have also heard Richard Wallis talking about library
> metadata saying something along the lines of if you want to share
> information about books between libraries use the appropriate library
> standard {MARC,RDA,METS,MODS...}, if you want to share that information
> with the rest of the world use schema.org (specifically bib.schema.org).
>
> The consequences of EOCreds in schema.org and CTDL diverging depends on
> the degree of divergence, but they include loss of specificity in the data,
> and increased difficulty in publishing CTDL data in schema.org. The
> former of those is often an unavoidable consequence of going from a
> specification for a partnership to one with global reach. It is one of the
> aims of schema.org to avoid the second of those consequences, even at the
> expense of making the data more difficult to handle for consumers.
>
> While both the schema and the CTDL serve the purpose of finding
> credentials it would seem that both are also useful referentially for
> issued credentials. For instance, I am proposing to Credentials Community
> Group that a property be added to assertions to reference credentials in
> the registry.
>
> That sounds great.
>
>
> I do think that the CredentialType and the subclasses in the CTDL are
> tremendously useful properties that not only make it possible to seek
> opportunities related to credential types button also compare verifiable
> credentials post issuance. For instance, being able to compare efficacy of
> online courses with badges or a series of courses to degrees.
>
> Again, I agree, with the caveat that here in Scotland a similar use case
> would be to compare academic and occupational careers after HND, HNC,
> Highers, Advanced Highers and A-levels, so in that case CTDL wouldn't work.
>
> With that in mind, (although curious about the divergence question), I
> definitely think we minimally need a CredentialType and while I agree that
> it would be onerous and take too long to compile an exhaustive list now, I
> prefer option three (DefinedTerm) with possibly an adaptable reference list
> to encourage consistency.
>
> Thanks,
>
> Thank you. I hope I answered your questions and didn't just rehash things
> you has already considered.
>
> Phil
>
> --
>
> Phil Barker <http://people.pjjk.net/phil>. http://people.pjjk.net/phil
> PJJK Limited <https://www.pjjk.co.uk>: technology to enhance learning;
> information systems for education.
> CETIS LLP: a cooperative consultancy for innovation in education
> technology.
>
> PJJK Limited is registered in Scotland as a private limited company,
> number SC569282.
> CETIS is a co-operative limited liability partnership, registered in
> England number OC399090
>