- From: Wes Turner <wes.turner@gmail.com>
- Date: Tue, 12 Apr 2016 11:50:04 -0500
- To: Phil Barker <phil.barker@hw.ac.uk>
- Cc: public-schema-course-extend@w3.org
- Message-ID: <CACfEFw8-BR8r=B+V-uj8vk7A6hrU+6-_b7xpk+6sCH4GD1K_bQ@mail.gmail.com>
so, in terms of the downstream user interface,
* when and how often is the capacity attribute updated?
- HTTP polling
- WebSockets, WAMP
- many users hitting refresh in a skewed peak at/around the endDate
- limiting the utility of caching
Eventually, the interface flow is:
* addtocart(course_instance_2
(url (courseCode, (name, instructor+)))
- if numberEnrolled > capacity
- waitlist? OR other CourseInstance
- <course/instance_2>
waitlistStatus FIRST_AVAILABLE ;
- other CourseInstances
- fulfillments for the same given requirement
- Credential > [Degree,] requirement
- Date/time pattern
- openingHours
- CourseInstance Events
- startDate ISO8601
- endDate ISO8601
- topically linked
- (keyword/entity ordered set intersection)
- by frequency statistic
- sensitivity to
* addtocart(this course)
(url (courseCode, (name, )))
- numberEnrolled = sum(instances[])
- capacity = sum(instances[])
- if numberEnrolled > capacity
- waitlist? OR other CourseInstance
- <course/instance_2>
waitlistStatus FIRST_AVAILABLE ;
- other CourseInstances
- fulfillments for the same given requirement
...
Two different tasks:
* Scale a massive open online course (MOOC) to n-people
- [ ] capacity unspecified / NONE / N/A
* Optimally allocate zero or more instructors, rooms, equipment given
learner availability constraints / optimization goals
* "first available [satisfying (our) required criteria]"
Received on Tuesday, 12 April 2016 16:50:32 UTC