new use cases and scope, was Re: Proposal: starting document for what we cover.

Hi Tavis, everyone. Comments inline, below

On 04/02/2016 23:59, Developer, SleepingDog wrote:
>> On 03 Feb 2016, at 12:36, Phil Barker <phil.barker@hw.ac.uk> wrote:
>>
>> Question: looking at the use cases, scope description and definition of an educational course, does anyone want to suggest any changes, additions, deletions...
> Hi all
>
> Some questions on the appropriate way to contribute and what scope bounds our contributions.
>
> I have put a couple of new use case suggestions (career/goal and apply status) in the discussion, hopefully that is the right place.
It's good to have the details on the wiki for reference. I prefer email 
for discussion, so long as it doesn't get unmanageable.

> https://www.w3.org/community/schema-course-extend/wiki/Talk:Outline_use_cases
"some users will start with the goal that a course tends towards." i.e. 
in terms of career.
Yes, I think that is realistic scenario.  It's similar to what I had in 
mind to use case 2, find a course that offers a qualification, in that I 
was assuming that the searcher would want a qualification because it 
lead to a job. That's too narrow, so I agree we should add information 
about career progression independently of qualification.

"Some indication of whether a course offering can be applied to"
Yes, something like "Having found a course, the searcher wants to know 
how to apply" is a more general case that I think is missing.

> Is the user search experience in scope? I appreciate that we are concerned with markup, but I imagine it will help if we consider possible search interactions (not that search engines are obliged to offer anything specific). For example, some searches like for images have advanced interfaces which utilise controlled vocabularies (dropdowns). And some search results show tiles like for recipes which show common properties with variable values.
I understand what you are getting at, and agree about its importance, 
but getting agreement on controlled vocabularies for use globally is 
very difficult. Also, schema.org's support for enumerations is evolving, 
i.e. it's a slightly moving target just now[1]. If we can agree on some 
recommended values and illustrate those in examples I think that would 
be great. If we can formalise these, that would be even better. Ideally, 
other people would provide those vocabularies, and we just need to point 
to them (and there are people on this list who are doing that). 
Ultimately what we do will, I think, depend on the vocabulary. So, a 
suggested enumeration of modes of study (relating to use case 1.6 [2]) 
seems achievable, whereas a single enumeration of careers for the use 
case you describe above, not so likely.

1. see https://github.com/schemaorg/schemaorg/issues/894
2. 
https://www.w3.org/community/schema-course-extend/wiki/Outline_use_cases#use_case_1.6_refining_UC1_by_mode_of_study


> The image search is deliberately chosen, the recipe search tends to be implied. How does the latter work? If a resource (a web page) can be more than one thing (a recipe and something else like a course) how does the search results logic decide which tile to show? For example:
> https://www.google.co.uk/search?q=how+to+make+chocolate+cake
>
> could return both recipes and course pages, so is a course result going to be triggered by a “course” keyword? Because you do not need a “recipe” keyword.
pass

> Also, are we considering the nesting and element placement of markup terms, or is that not significant for schema.org? Consider a recommendation to put a course title in a h1 element which can then contain further elements like:
>
> <div itemscope itemtype ="http://schema.org/course">
>    <h1 itemprop=“title”><abbr title=“Higher National Certificate" itemprop=“qualification”>HNC</abbr> <span itemprop=“subject”>Accounting</span></h1>
>    <p>description etc.</p>
> </div>
Yes, the interplay of tree-like html and graph-like RDF is one of the 
bigger issues with RDFa as a whole. One way around it in schema.org is 
what Richard Wallis has called "islands of structured mark up" in JSON-LD.

> One reason I am asking is that these questions of markup will affect how easy it will be to generate from structured sources (like relational databases or XCRI CAP), and whether things like abbr elements to provide expansions is best practice.
An XCRI-CAP (or MLO or other) to schema course extension transform would 
be a great way of testing the proposal. I think there are a few of us 
who could work on that.

> Apologies if these points have already been made clear; I have already checked through the documentation for guidance but may have missed some.
No need to apologise, all requests for clarification are helpful

Phil


-- 
Phil Barker           @philbarker
LRMI, Cetis, ICBL     http://people.pjjk.net/phil
Heriot-Watt University

Workflow: http://www.icbl.hw.ac.uk/~philb/workflow/



----- 
We invite research leaders and ambitious early career researchers to 
join us in leading and driving research in key inter-disciplinary themes. 
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.

Heriot-Watt University is a Scottish charity
registered under charity number SC000278.

Received on Monday, 8 February 2016 15:26:46 UTC