Not for the institutions, nor for students, who have no way to export their learning data (content, connections, relationships with peers, etc.) out of the system in a way that could be integrated into something outside of the institution.

Reliable: As a simple example, I'd trust Google to maintain data reliability over my institutional IT support.

And you'd also need to make the argument for why learning needs to be "private", etc.

Another example of why online learning could (should?) take place in public.

Imagine if, upon graduating, an engineer already had an online, public portfolio of work that could be shared with potential employers.

If it were to be used for audits, it's doubtful what the value of the audit outcome would be. The LMS may provide data on "time spent" logged in, for example, but who cares about that?

Another example: Most universities will have some rhetoric around a commitment to principles of lifelong learning, but at the same time they use a learning management system that seems designed to reduce the possibility of learning after students graduate.

This was meaningful because I feel that I've had a similar experience over the past few months. I found myself thinking about the open web and indieweb, being drawn to it and getting excited by it again.

Having gone through the description below, I can't help but feel that there's an awful amount of manual effort going on here. My own thinking is that task and project management as part of a daily workflow needs to be something relatively friction-less.

When there's too much structure in a system, with too many steps, I feel like I would be spending more time managing the system than doing the work.

One of my goals in my own system is to reduce the number of steps required to complete any single activity.

I used to work from the premise that I have to be able to add tasks to a system that is cross-platform (because otherwise, how could I add that one task while walking between meetings, and have it show up in my work for the day on my laptop?).

But now I've switched to a "working context", which means I'm sitting in front of my laptop. It's where I capture 99% of the work I need to do. If, on the rare occasion I need to capture a task or note, I use Joplin.

Important point to make. I think of this as "Principles before tools", which means that I start with what I want to achieve, and then find tools that help me do that more effectively.

Postgraduate students who can complete their degrees remotely. We know that our own PG students struggle significantly with the costs associated with moving to South Africa from other African countries.

It seems like good advice but it's actually quite hard to divorce yourself from what you know. See the curse of knowledge.

This is why I think that having this list of questions is a good idea; you don't have to rely solely on putting yourself in the reader's shoes.

See Tim Berners-Lee's SOLID project.

"...using computers to pass time in algorithm-driven loops..."

It is strange that every interaction we make on a social media platform is, in part, a small aggregation of yet more data about ourselves, and that each aggregation increases the shareholder value of that company by a little bit. And for what? So that we can feel a bit more connected in each of our little online worlds.

Video chat does none of this currently. What's different about looking at each other through a screen? Is it an evolutionary thing, where we just haven't evolved to communicate effectively through video? Sure, we can move information with video chat but can we connect?

Nice juxtaposition.

I like this concept. It'd be interesting to see if different apps could have a "together mode".

I really like the Basecamp approach (I forget where I heard this...could have been in one of the Rework podcasts):

Don't try to get the most out of everyone; try to get the best out of them.

If you're looking for ways to build trust in a team, I can't recommend the following books published by Basecamp:

• Rework
• Remote
• It doesn't have to be crazy at work

So that the fact that the metrics are collected over 28 days is meant to maintain privacy and trust. How?

That's weird because in this article they say that this is just a confusing mistake, and that Productivity score "was never designed to score individual users."

This is going to be awesome when it debuts in universities. I can't imagine that any academics will be concerned when a departmental chair or administrator asks you why you're not sending more emails.

So this will require that you have to have video turned on. How will they sell this to employees? "You need to turn your video on so that the algorithm can generate an accurate meeting quality score using your body language and facial expression.

Sounds perfect. Absolutely no concerns about privacy violations, etc. in this product.

Does anyone belief that they created a user interface that included individuals, by accident?

It's great that MS has reacted so quickly to the outcry around the privacy of workers.

I thought it would be super-interesting to see how academics might have responded to the idea of institutional administrators keeping tabs on the number of hours that they'd spent in meetings (via Teams), composing and reading emails (via Outlook), writing articles (via Word), and so on.

And yet these would be the same academics who do this kind of monitoring of student work.

These are really beautiful and inspiring posters.

Via @chrisaldrich

And if you can get better at doing hard things it gets harder for others to replicate what you can do. When you're good at doing things that others can't easily replicate, you increase the value that you bring to an organisation.

See Newport, C. (2016). Deep Work: Rules for Focused Success in a Distracted World (1 edition). Grand Central Publishing.

Good ideas are often immediately recognised, by many people other than yourself, that they are good.

Feynman: "The first principle is that you must not fool yourself. And you are the easiest person to fool"."

Applications provide an interface to functionality.

Good point. However, loads of companies use open source libraries to build their software. And we pay the companies, not the people who built the libraries. How do we get around this?

This is true in other areas as well. Use principles to make choices about what products to use; that way you can always swap out the products. Putting together a collection of software - instead of a monolithic approach - means you can select the tools you like working with.

I chose Evernote over other alternatives because their design was better. I chose Google Reader over other apps because their design was better. When companies change their policies or shutter services, design simply isn't relevant. But openness is always relevant.

Project-based learning approaches where student outputs that are assessed are real-world projects that simulate what they'd need to create in order to solve real problems.

We should be assessing how people use information to solve problems they care about, rather than whether or not they can recall information.

Not just in the US but increasingly seen around the world as part of the anti-intellectualism associated with the Wokeness.

This simply isn't true. We know an enormous amount about how to increase memory. And memory systems have been around for centuries.

There's something here...but not "executable" which suggests a process that runs according to a preconfigured pattern.

Maybe we just haven't sufficiently defined what we mean by "goals".

How was the development of writing not goal driven? I'm no expert but I believe that early forms of writing were developed in response to the need to record transactions. How is that not goal-driven? Of course no-one sat down and said we need more irregular verbs. But they absolutely would have identified the need to include words with more nuance than those that existed. So the process may have been open-ended (they didn't articulate objectives, deadlines, milestones, etc.) but those open-ended explorations would have been in the service of achieving objectives.

This is a form of argument I detest. Reduce the problem to an absurd form (prehistoric people setting OKRs) and then make fun of it.

But achieving goals absolutely would have been the impetus for language development. Want to survive the winter? Develop a way to share ideas. Want to collect more food? Develop a way to share ideas. What to fashion a more robust spear? Develop a way to share ideas. Of course language developed in response to achieving goals.

Not necessarily. All/most of those smart people are working within the same paradigm, which means that the possibility of building something outside the paradigm is limited. Take the online advertising that's made Google and Facebook the monsters they are. How easy do you think it's going to be for someone to shift the paradigm so that online advertising isn't what's paying for the web? The smart people want to make online advertising more effective, rather than figure out a new paradigm.

Has this been formally defined anywhere in the essay? And who gets to decide what is a "genuine" tool for thought?

It's not clear to me what makes Photoshop, for example, a tool for thought. People have been editing photos for as long as we've had photos. What does photoshop add that makes it a tool for thought?

There are open source versions of all Adobe products that are free. Yes, they're expensive in terms of the hours put into development but I'm resistant to the idea that Adobe is some kind of visionary company.

This sounds a lot like the tech product design and testing experience that was contrasted to with the lone genius / deep insight analogy presented earlier. I'm not convinced that second-by-second user experience testing is what we need.

But there's nothing here that hasn't been known for 100 years. It's not just that these aren't deep enough enough, they're not even insights.

This also isn't fair. You provide what you acknowledge is a "fanciful" story for how Arabic numerals were created (a single genius with deep insight rather than a culture working over centuries) and then make the claim that this is somehow linked to the lack of progress on thinking tools in the tech industry because our current system is more like a focus group. Is the suggestion that the authors of this piece are the geniuses with the deep insight?

This isn't really a fair comparison, though. For all we know, the invention of Hindu-Arabic numerals did go through this process. It was likely the outcome of centuries of work, with people at the edge of convention pushing boundaries about what another system might look like. There would've been discussions with all sorts of people. I'm not sure what the analogy is meant to provide, since it seems so divorced from reality.

It's not an arbitrary structure; you've literally just said that the associations between memory chunks are the structure.

This reminds me of Donald Trump: "People are saying...". Who are these people and why should we care what they are saying?

This is a straw man argument...making a claim that few people familiar with the literature would support, and then taking a position in opposition to the claim.

This is a bit confusing because it uses "elaboration" in a different way to the literature on learning and memory. Like with spaced repetition, we know that elaboration (i.e. explaining concepts recalled from memory without reference to the source material) is a great way to learn new information. See Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving Students’ Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58.

How to take smart notes, by Sonke Ahrens: https://www.amazon.com/How-Take-Smart-Notes-Nonfiction-ebook/dp/B06WVYW33Y. It includes these suggestions and loads more.

Pretty much, yes.

Why do we need an informal experiment to confirm the results of loads of formal (published in peer reviewed venues) experiments?

It's not about studying more; it's simply taking advantage of the fact that memory traces are reinforced through active recall over increasingly longer periods of time. And we've demonstrated this reliably since Ebbinghaus in the 1880s.

At the moment this is looking like more (but not necessarily better) evidence in support of spaced repetition.

What does this even mean? If this refers to the practice of active recall over increasingly longer periods then we already have this built into every spaced repetition algorithm out there.

After receiving the email, opening the browser, logging in...then it takes a few seconds. Again, this just seems like a less efficient form of spaced repetition.

How is this any different to extracting questions from the text and adding them to something like Anki? Except that in Anki I have the questions with me all the time, I don't need to be online, don't need to register anywhere, and can choose the questions that are meaningful to me, can connect it to other knowledge...it seems that this simply embeds a less useful form of spaced repetition into the text. Or am I missing something?

The medium we're working in now probably isn't the optimal format to discuss a new paradigm.

We shouldn't try to use technology to close off learning but rather to open it up.

See also the Wait but Why post about something similar to this.

Don't start your day by checking email.

Fair enough re. the HIV medication. But where is the equivalent example for being "free from hunger"? Where is the Constitutional protection for hungry citizens when politicians steal from the country?

This is currently happening in South Africa where the evictions are being carried out by the police. See here and here.

Also, this seems to now be conflating the question of evictions with poverty. Of course they're aligned but are they the same thing? Is this a bait-and-switch argument?

But is it clear that "injustice" = "human rights violation"? As Foley says in his article, bad things are bad things but they're not necessarily human rights violations.

So it's a magical system that fixes everything in education? The complexity of a system that could achieve this is mind-boggling. This is not a 10 year prediction. This is a "one day in the distant future" prediction.

One of the problems with companies developing AI for education is that they're focused on things like efficiency, rather than learning.

The problem is that ML is trained on retrospective data, sometimes gathered far removed from the possibility of being considered for the training. Do we have to retrospectively get consent for this data to be used?

This links to what Downes said about the use of AI by teachers vs the use of AI by students. These are use cases that make the teachers work easier. And that's going to be the focus of the technology i.e. aimed at teachers and schools. Who is doing the work of helping students use the technology to improve their learning?

My understanding is that the algorithm uses past performance to predict the performance this year, but that seems insane. Why would anyone want to do that? I thought an AI was actually grading each piece of work.

OK, so it's not that the algorithm didn't perform according to expectations (it was 96.4% accurate relative to teacher-awarded grades). The problem is that some students were downgraded and this would affect their university performance. But couldn't the same thing be possible with teachers who make mistakes?

When it says that the algorithm "predicts grades for each school", does that mean that it's not looking at the work of each student? Which would be weird.

Who is doing the work to make sure that the same is true of teachers? I don't mean that teachers are these things in some malevolent way...only that bias, poor judgement, and lack of fairness are built into our cognitive processes and we're not even aware of it.

It's not enough to note that a text was "striking"; we need to be able to count the number of readers who thought it was.

This is a democratic way of getting ideas out into public for further discussion.

Don't worry about wasting ink; this is about getting words onto the page. It's meant to be messy and unorganised.

What stands out to the reader?

It's about responding to the content and not the structure.

The algorithm is able to infer the task from a few examples. In cases where the algorithm is fine-tuned, it can be done using instructions in plain English.

The legacy of white supremacy is that it is built into the systems and infrastructure that govern our daily lives.

This is white privilege; these advantages.

OK, so by saying that "it takes experimentation", what does the author mean if not that it's a science-based approach? Either it means that learning how to focus is a science, or we have a problem because we don't use words according to what they mean.

I agree. But "doing the work" is not the same as "having a system for doing the work".

I disagree. You absolutely can be scientific in how you think about productivity. For example, "don't multitask" is a science-based piece of advice that should form part of a productivity system.

He's conflating the process for working with the work itself.

And finding additional material can create the illusion of progress.

He can only "know" this if someone explains it to him. Did you?

It's often not the patient who makes these appointments. They are simply told by the doctor to go to the physio. This patient may not have had any idea what he was doing there. If the doctor had told him to go see the physio who would help with the pain, you can see why the patient would be upset when you didn't give him the injection.

Instead of saying that it's not part of your duty - which may sound like you simply don't want to help him - you could say that you're not allowed to give injections. Did he think you were a doctor? Did you correct him? He may simply have been confused.

You could elaborate on this process. Is it fair? Why or why not? It relates strongly to the concept of Justice in the allocation of resources, which we discussed at length in the module.

What is this referring to?

We didn't realise that the web allowed us to "play" with teaching and learning in ways that enabled what Freire called "acting on the world". We didn't realise that we could use the web to transform, instead of merely transmit.

We didn't realise that the web allowed us to "play" with teaching and learning in ways that enabled what Freire called "acting on the world". We didn't realise that we could use the web to transform, instead of merely transmit.