OLDaily

Learning to Learn
Sarah Drasner, CSS-Tricks, 2019/03/08

I've been in the midst of what I've been calling 'retooling' - spending some time to reacquaint myself with the basics of our field, bring myself up to date, and yes - relearn. Not that I think everything I learned in the past is falso. Far from it. But as Sarah Drasner says in this article, "By choosing to be a developer, you are choosing to learn." Or an educator. Or a researcher. That's the job. My approach is a lot less formal than Drasner's, but it's still the same job. Learn.

Web: [Direct Link] [This Post]

Towards an intuitive high-performance consensus algorithm
Heidi Howard, Read, Write & Execute, 2019/03/08

This article looks at 'distributed consensus', that is, "is the problem of how to reach decisions in asynchronous and unreliable distributed systems." As Heidi Howard outlines, the two most widely known algorithms are Paxos (which I've covered here before) and Raft (which I thought I'd covered, but I can't find it). The difficulty with consensus algorithms is that they must not only work, they must be intuitive. We won't trust a consensus we don't understand. That's what this article is about. Howard writes, "We aim to improve performance and understandability by utilising two key weapons: generality and immutability." The new work along therse lines, 'A Generalised Solution to Distributed Consensus' (20 page PDF), "proposes an immutable and generalised approach to reaching distributed consensus over a single value." Via Colyer.

Web: [Direct Link] [This Post]

How to create a broader, fairer and smarter education system
Joysy John, JISC, 2019/03/08

The message is this: "If our world is going to be driven by artificial intelligence algorithms, yet huge sections of society are underrepresented when deciding how these are programmed, then we have a problem. This has to change. Our education system must be multi-disciplinary and based on real-life problem-solving. Students need technical and creative skills." Quite so. But the main message of this post is that all this is especially true for girls and women. "There are issues around equity here... while role models are powerful, I think it's more than that. We need to build women’s technical skills, creative skills, social and emotional skills."

Web: [Direct Link] [This Post]

Round Up: Ethics and Skepticism
Amanda Hickman, GitHub, 2019/03/08

I've been thinking about the epistemology of literature-based research recently. This results from a day when I presented results from my reading on connectivism and also reported on a 'scientific' literature search that resulted in a questionable sample of 12 papers. Now obviously there should be some selectivity in research - the list of questionable methods in this post proves that. At the same time, research isn't an election or even a poll. This is true especially is the way you collect your research sample is presented as unbiased when we know that it can't be. Something to think about.

Web: [Direct Link] [This Post]

Teaching Students to Read Metacognitively
Brooke MacKenzie, Edutopia, 2019/03/08

In a way, each one of these little posts in OLDaily is me reading metacognitively. I could imagine myself asking "Does it look right and sound right? Can I picture the story? Can I retell the story? Does my mind feel good?" What I like about this is that none of these address my memory of the story - they're all focused on what I feel and what I can do after having read the story. Brooke MacKenzie writes, "After we read the poem, I ask, 'What do you think this is about? What words in the poem make you think that? What do you picture as you read it?'"

Web: [Direct Link] [This Post]

Kids' relationships and learning with social robots
Jacqueline M. Kory Westlund, MIT Media Lab, 2019/03/08

One of the things MIT and Media Lab do really well is to get their reserachers to communicate their work in straighforward and engaging narrative. I wish more institutes (including our own) did this. Anyhow, this article describes how children forming relationships with robots demonstrates greater learning outcomes. The relationships are created by having the robot itself engage in relationship-behaviour: "interacting multiple times, changing in response to those interactions, referencing experiences shared together, being responsive, showing rapport (e.g., with mirroring and entrainment), and reciprocating behaviors." Children responsed with similar behaviours, and those that did had the higher scores. There's a lot still to unpack here, and this work is worth watching with interest. For the academic treatment of the topic, see their paper (12 page PDF).

Web: [Direct Link] [This Post]

From video game to day job: How ‘SimCity’ inspired a generation of city planners
Jessica Roy, LA Times, 2019/03/08

I've spent many happy hourse playing Sim City and Cities Skylines and so I understand both the appeal of the games and how they could inspire people to want to make city planning their career (indeed, it's one of my many 'alternate careers' I could easily have fallen into). This article is the story of such people. Of course Sim City embodied a very naive understanding of civic planning. Nonetheless, " the SimCity games are a good introduction to the field. Those people will go on to understand what the game gets right and wrong."And SimCity establishes a very fundamental literacy in systems, Sanchez said. He hopes games like his will extend that literacy to some of the finer points of city-building."

Web: [Direct Link] [This Post]