OLDaily

Adding structured data on Canada.ca pages
Canada.ca, 2021/04/14

What I like about structured data is that it basically turns every web page into a feed of sorts. And it allows the page to be aggregated and added to a linked data system, enabling services to build knowledge bases. This page describes the government of Canada use of linked data (at least aspirationally). For more, see the Google guide to linked data, or if you prefer, this introduction by Alexis Sanders.

Web: [Direct Link] [This Post]

Introducing POLLEN
Julia Krolik, Medium, Nightingale, 2021/04/14

This article takes you through the development story for a Slack application, but the real story is in the header image: the use of packed hexagons to show the structure of a Slack channel (really, it could be for any data collection at all; it just happens to be Slack content). It's an evolution of force graph illustrations that depict the organization using circles and lines. Here's the application, and I'll be watching for Javascript-CSS libraries that organize contents into packed hexagons.

Web: [Direct Link] [This Post]

Online Proctoring is Not Essential
Matt Crosslin, ElearningInside, 2021/04/14

The argument is essentially that online proctoring is not essential because "the exams or tests themselves are not essential." That, asserts Matt Crosslin, is because "quizzes, tests, exams, assignments (cannot) measure learning or skill mastery. Not directly at least." In his own classes his assignments "revolve around projects that have students creating artifacts that match closely what they would do in a real life position in our field." And if you think about it, there's no proctoring or plagiarism detection system in the workplace - and no need for one.

Web: [Direct Link] [This Post]

Ellucian launches 'Experience,' an extendable, single sign-on platform
Colin Wood, EdScoop, 2021/04/14

Single sign-on is probably one of the most in-demand online services, yet the hardest (it seems) to find. Institutions work to ensure all their services can be accessed with one login, and make use of service like EduRoam to provide access to visitors. Ellucian, "known for its enterprise resource planning and student information systems platforms," works with roughly 150 partners and feels this need acutely, which explains why it has launched Experience. "it also intends the new platform to serve as a hub around which new modules, called 'cards.'" All very nice, but real single sign-on won't exist until it is removed from the domain of application providers.

Web: [Direct Link] [This Post]

E-Literate’s Michael Feldstein Weighs in on Digital Content in Higher Education Today
e-learn, 2021/04/14

This is overall a good interview but saves its most interesting point to the very end. Here's Michael Feldstein on learning content: "I think there’s going to be more creation of university-specific content, and in those cases, being able to understand their tech toolkit will be key." While this wouldn't surprise me, especially for institutions with recognizabl brands, it would have the potential to shake up the existing content market. Feldstein also addresses what instructors need to know about instructional design (starting with learning goals, and analytics) and how instructors view the LMS (as a virtual classroom, or as a file distribution method).

Web: [Direct Link] [This Post]

Artificial Consciousness, Superintelligence and Ethics in Robotics: How to Get There?
Louis Vervoort, Alexey A. Melnikov, Munesh Chauhan, Vitaly Nikolaev, Mind and Matter, 2021/04/14

So I think it's clear that this isn't a turn I would make, though a lot of people would: "the key capacity that computers and robots should possess in order to emulate human cognition and (ethical) consciousness is the capacity to learn and apply ‘coherent  webs-of-theories’... where classic AI has been, in essence, ‘data-driven’, the greatest leap forward would be ‘theory-driven’ AI." The important concept discussed in this paper (36 page PDF) is the Conscious-through-Monitoring-through-Theories (CMT) Model, and it's used to argue that "all  truly  human-like  AI  will  somehow  have  to  be  theory-driven, rather than purely data-driven, as is now the case. A similar claim has, in essence, been made by cognitive  scientists." There's some good discussion about whether neural networks could develop these theories themselves, including a recognition that there is probably no general model for theory generation.

Web: [Direct Link] [This Post]