- From: Phil Barker <phil.barker@pjjk.co.uk>
- Date: Mon, 19 Feb 2018 11:02:49 +0000
- To: public-eocred-schema@w3.org
- Message-ID: <5e738a56-baa4-feab-17a3-6ba03ce29175@pjjk.co.uk>
OK. Does anyone object if I keep DefinedTerm in the expectedRange? (For
that matter, does anyone object to dropping it?). Taking a bit of the
Dublin Core definition, how about:
*Name*: educationalLevel
*Definition*: The level in terms of progression through a learning,
educational or training context. Examples of educational levels include
'beginner', 'intermediate' or 'advanced', and formal sets of level
indicators such as the European Qualifications Framework.
*Expected Range*: Text, Url, DefinedTerm
I am very much of the opinion that these levels are only meaningful if
they come with definitions in the context of a set of levels, hence my
desire to keep DefinedTerm in the range. Given the choice of Text and
Url I can easily imagine getting values like cryptic values like "NQF4"
or meaningless values like "4" when there is no Url for individual terms.
Phil
On 17/02/18 10:51, Richard Wallis wrote:
> “I think text is good, I’d really like to add URL"
>
> That is no problem - it is default in Schema.org to be able to use a
> URL (if you have one) for the value of any property:
>
> /some types such as Role and URL can be used with all properties,
> and we encourage this kind of experimentation amongst data
> consumers. <http://schema.org/docs/datamodel.html#conformance>/
>
>
>
>
> Richard Wallis
> Founder, Data Liberate
> http://dataliberate.com
> Linkedin: http://www.linkedin.com/in/richardwallis
> Twitter: @rjw
>
> On 17 February 2018 at 10:20, Fritz Ray <fritley@gmail.com
> <mailto:fritley@gmail.com>> wrote:
>
> Phil,
>
> Your understanding continues to make sense to me. I didn't realize
> the credentials for each level of award were nationalized... I
> don't believe that changes anything.
>
> Nate,
>
> I could imagine educationLevels being described as:
>
> "Masters"
> "SPQF Level 5"
> Taxon/DefinedTerm/reserved URL such as
> http://ed.gov/degreeLevels/associates
> <http://ed.gov/degreeLevels/associates>
> by URLs to documents.
> <https://www2.ed.gov/about/offices/list/ous/international/usnei/us/associate.doc>
>
> That's primarily why I chose a set of things that broad. I'd agree
> it doesn't serve the use cases, but there are a lot of cases where
> these levels are defined in documents, so text strings is about
> all we're going to get.
>
> I assume that at some point the web of data people will win and
> all these word documents
> <https://www2.ed.gov/about/offices/list/ous/international/usnei/us/edlite-structure-us.html>
> and pdfs
> <http://scqf.org.uk/wp-content/uploads/2014/03/SCQF-Level-Descriptors-WEB-Aug-2015.pdf>that
> describe levels will have corresponding URLs for each level.
>
> Richard et al,
>
> I think text is good, I'd really like to add URL in order to
> represent taxons and links to descriptions of levels out on the
> web (so we're at least somewhat as capable as AlignmentObject.)
>
> Other than that, I'm happy with where the conversation has ended up.
>
> On Fri, Feb 16, 2018 at 10:33 AM, Vicki Tardif <vtardif@google.com
> <mailto:vtardif@google.com>> wrote:
>
> Based on similar experiences in other Schema.org extension
> areas, when it has become complex/difficult to gain
> consensus on a particular point, especially with an
> initial proposal:
> I suggest that we agree on a property name for this
> (these) concepts and create it with a range of Text and a
> suitable, not too specific, description.
> After some use in the real world, we can then review that
> usage and come up with enhanced propert(ies) definition,
> range, etc. as part of a further following proposal.
>
> At this current stage translating the forgoing discussions
> in this email trail into a concise description, that will
> be understandable to the Schema,org community that will
> receive, and hopefully accept, our proposals seems a
> challenge too far for this initial release.
>
>
> I agree with Richard. It may be simplest to use text and see
> where the data leads us.
>
> - Vicki
>
> On Fri, Feb 16, 2018 at 12:42 PM, Richard Wallis
> <richard.wallis@dataliberate.com
> <mailto:richard.wallis@dataliberate.com>> wrote:
>
> Based on similar experiences in other Schema.org extension
> areas, when it has become complex/difficult to gain
> consensus on a particular point, especially with an
> initial proposal:
>
> I suggest that we agree on a property name for this
> (these) concepts and create it with a range of Text and a
> suitable, not too specific, description.
>
> After some use in the real world, we can then review that
> usage and come up with enhanced propert(ies) definition,
> range, etc. as part of a further following proposal.
>
> At this current stage translating the forgoing discussions
> in this email trail into a concise description, that will
> be understandable to the Schema,org community that will
> receive, and hopefully accept, our proposals seems a
> challenge too far for this initial release.
>
> ~Richard
>
> Richard Wallis
> Founder, Data Liberate
> http://dataliberate.com
> Linkedin: http://www.linkedin.com/in/richardwallis
> <http://www.linkedin.com/in/richardwallis>
> Twitter: @rjw
>
> On 16 February 2018 at 17:39, Nate Otto <nate@ottonomy.net
> <mailto:nate@ottonomy.net>> wrote:
>
> Thanks for digging in to get more precise on level here.
>
> I like how the SCQF reasons about levels of
> accomplishment. A Credential can recognize a level of
> accomplishment, a level of performance, or both. A
> Course could be "at" a level of accomplishment in
> terms of difficulty or prerequisite knowledge &
> skills. These are good use cases to target, and if I
> think of "educationalLevel", this would be the sense
> of level that would fit best, versus "level of
> performance", even though it would be possible to
> split hairs further between the two categories I
> started with, which we could abbreviate to
> "accomplishment level recognized" and "accomplishment
> level required".
>
> This vocabulary's ability to describe level of
> accomplishment should be distinct from trying to talk
> about level of performance and not use the same
> property, in my opinion.
>
> Fritz,
> I'm a little wary of "A string, term or URL". That's
> amazingly broad to the point where it would likely
> make it very difficult to serve the comparison use cases.
>
> What feels important to me about understanding the
> level of accomplishment of a credential is its
> position relative to other credentials, learning
> opportunities, etc. I am not confident I get that
> across a range of credentials unless they all use
> specific URLs pointing to level definitions like the
> ones from the SCQF.
>
> On one hand, one string property is nice and simple,
> on the other hand, it doesn't serve comparison use
> cases well unless all the credentials you'd like to
> compare use a very specific scheme established outside
> the scope of this vocabulary known to the consumer.
>
> Maybe I changed my mind on using alignment,
> particularly because AlignmentObject already has the
> "alignmentType" property, which includes
> "educationalLevel" as an option. We could suggest
> something like this, adding a numerical levelNumber
> property and using a URL either for
> educationalFramework or targetUrl (a little wary of
> targetUrl because I would think that should represent
> a URL of the exact level that alignment is desired
> for, but maybe somebody can ease my mind on this point)
>
> {
> "@context": "http://schema.org",
> "@type": "Credential",
> "alignment": [{
> "educationalFramework":
> "http://pinballsorcerers.org/levels/2
> <http://pinballsorcerers.org/levels/2>",
> "alignmentType": "educationalLevel",
> "levelNumber": 2
> },
> {
> "educationalFramework":
> "https://ec.europa.eu/ploteus/content/descriptors-page
> <https://ec.europa.eu/ploteus/content/descriptors-page>",
> "alignmentType": "educationalLevel",
> "levelNumber": 7
> }
> ]
> }
>
> It does seems like we're not going to be able to model
> this nearly as well to serve comparison use cases with
> a bare text string. Only human eyeballs could make
> sense of the difference between
>
> "educationalLevel": "Pinball Wizard Level 1: Nub" and
> "educationalLevel": "Pinball Wizard Level 6: Ultimate
> Extra Baller"
>
> Nate
>
>
>
>
>
--
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
Received on Monday, 19 February 2018 11:04:50 UTC