Stephen Downes
Knowledge, Learning, CommunityUnedited transcript
Today's topic is the future of learning technology, 10 key, tools, and methods.
It's going to be focused mostly on technology and mostly on technology, trends. I know that this whole bunch of other stuff to cover, but I am going to keep fairly narrowly focused. That said, the breadth of this presentation will be, unfortunately, staggering. So, I want to remind you to keep in mind.
The idea here, isn't for you to remember, everything is I talk about it. The idea here is for you to have thoughts, whatever those thoughts are questions, Ideas, etc, and maybe jot them down or or type them down. As the presentation goes on, put them into question area, put them in the chat area.
I'm not really able to see the chat right now, but it's not for me necessarily to answer these questions. These questions are the future avenues of research and development. I want to be again by talking about briefly about the future of learning that's already here. And these are the things I want to be talking about.
Although, pretty much everyone is talking about. These, these them, these days are artificial intelligence. The tools we have for analytics, the tools we have for generating new content, deep, fakes all about sort of thing the tool where using to produce the live transcription of this presentation. It's here now, and of course, there's a lots going to happen but it's not a future trying anymore.
It's here. Now the matter of verse again it's here now it's not fully developed and a lot of things that I'll talk about in this presentation. Come from the meta verse and will apply to the meta verse but things like virtual reality and augmented reality. These are technologies that are current being currently being used in a learning setting.
I've done work myself developing and trying some of these VR and AR technologies. And there are widespread pilot projects and fully implemented systems today. Blockchain Everybody's talking about blockchain again. Many of the technologies that I'm talking about today are derived from blockchain but blockchain. Something that's here. Now it exists obviously and it's beginning to improve.
We have things like non-fungible tokens or NFTs. We have things like the ethereum merge, which converted the ethereum blockchain to a new kind of energy efficient system of records etc. Again, a lot of pilot projects, it isn't fully deployed but it's not a future anymore, It's here. So these are the things that I'm not talking about.
What I am talking about is where the future will go and we can draw some lessons from the recent pandemic that we just experience, for example, the pivot to online learning that we experienced in that we're still experiencing today with. Yes, yet another zoom presentation. He was in many ways flawed, but it was flawed in ways that helped us.
Learn about how we really do want to do online learning in the future, is well, the pivot to remote learning exposed issues of equity and accessed, diversity, equity inclusion. These things came to the four. And we saw how existing educational systems and existing online systems are not serving. All populations equally.
Well, I live in rural area and it wasn't until the covid pandemic, that internet service providers got serious about providing broadband internet out here. So, all of these issues now are in the balance. As we move toward the future, we've got the existing technologies of blockchain. AI metaverse, we've got issues of pet pedagogy issues of access and more and then we have these future tools and methods.
So what I'm going to do is talk about these future tools and methods, but these questions these gaps still remain and it's not from me to answer all of these questions. I have my thoughts, so, obviously, but the questions have to come from all of you. The answers have to come from all of you.
And you know, if there's a central message to this presentation, that might be it. So what are these 10 things? Well, the first one is the web of data and this is probably the most significant transformation in knowledge and learning happening right now. We are shifting from an environment where teaching and learning is conducted as a narrative sequence of events.
Well, simple sequencing. Moving from one topic to the next. A course, as a series of sessions, we're moving to something different. Now, and what we're moving to is a course as a web of data of interconnected data. Now I have in my mind, this lovely animation for this particular slide that I would do, right?
Have one web or graph and, you know, all of the issues. And then another one, all of the concepts and another one, all of the models and they all linked together, but that would have taken me several weeks to make. But imagine that and imagine a course looking like that imagine a course looking like this diagram that you see right in front of you how do you teach that we're not storytellers anymore?
We're not narrators. We're now serving more in the role of explorers or guides, the web of data, some points to consider. First of all open data, there are numerous open data initiatives, just getting a started, they haven't really been integrated into educational applications, but it's coming very shortly. I'm thinking of things like the government of Canada open data initiatives or the announcement just yesterday from crossref.
That they've opened up millions of cross referenced citations from academic publications. There's a world of open data out there. And all of that is going to become accessible and part of our courses. Another topic to consider is the idea of data literacy. I did a big project on that this past year and you can see the link to that on the slide data literacy.
Involves understanding, how to look at comprehend, manipulate used data and we can think of it from a variety of perspectives. We can think of it from the perspective of data management, we can think of it from the perspective of data science. We can think of it from the perspective of data literacy information, literacy, things like that.
So different ways of looking at data literacy and of course, it's going to become an increasingly important topic to consider in our educational systems of the future. Similarly data ethics, how do we manage data? How do we collect data, how do we clean it? How do we augment it?
There have been a variety of reports and inquiries on the question of data ethics it. Of course, ties directly into the question of ethics in artificial intelligence and analytics, and it also ties into ethics as related to instructional design and online, learning finally designing for data. And this is what's going to impact what we are doing as educational.
Technologists. Most directly how do we organize our learning materials to take into account present and allow people to work with all of this data that is now suddenly available to them. It's different from instructional design because the data is live, it's dynamic, it comes from a variety of sources and it doesn't come with instruction built in.
So the way we present it in the way we organize, it needs to take into account the nature of the data as well as the purpose that we have in presenting the data. Number two, told you we're just zipping right through these virtualization. Now, in many respects virtualization is already here and some of the systems that I talk about like, VMware or docker have been around for a very long time, a long time in internet years, which is still sure, then the lifespan of a dog.
But we haven't actually seen virtualization come into our everyday experience. It's, you know, second nature for developers and web professionals to use. And there's a lot of it's happening behind the scenes. This whole webinar is almost certainly being run in a virtual environment, but we don't have these for personal use.
Yeah. And that's what's coming. What is virtualization? Well, you take a typical computer. Say your computer on your desktop or perhaps your phone has hardware operating systems and applications. That's the way it's structured. And what we do is we simulate the hardware with the virtualization layer and so what that does is it allows us to well, for one thing run different operating systems in different applications on the same computer.
So, depending on with my computer, I can run a Python application and decent artificial intelligence processing or I can run a container with unity or the unreal engine and do some virtual reality development etc. It also allows us to take these images once I've created an image and run it on many different hardware platforms.
So I create and application fully configured with all of its data in it once and then I can share that with 10 the files in the million people. So this image can run, not just a mic computer, but on your computer, your neighbors computer etc. So these new kinds of resources, these new virtualizations allow us to redefine what we mean by textbooks or learning objects.
In fact, it we doesn't make sense to use a textbook anymore and and just as an aside, I don't know why people still write books. Hi, honestly. Don't when they could be doing so much more instead of simply getting a sequence of text. A student can access a complex computing environment already pre-configured with activities access to data etc built right in.
So, for example, they could work with an online interactive map with real time traffic and weather data. Oh yeah, just like Google Maps but as a learning resource or they can build their own resources. Virtualization has allowed us to develop entire development, environments, specifically, tailored to allowing people to build an application without having to install all of the material that they need including text-based code etc, on their computer.
Here we have what we call a no-code application development studio and we can see what's happening here. They're just dragging and dropping elements of their interface or, you know, they can also drag and drop data sources or resources and animations, whatever. And they're building their app in their, seeing their appear in a virtual environment on the right hand side of our screen here.
Number three graph. Again, it's an old concept. It is very old concept, it's already here, but it hasn't, really hit us yet. And, and let me draw the distinction for you. If you're still using things like MS, Word documents, or powerpoint, slide, presentations, or even Excel. You haven't plugged into the graph yet.
The graph is a network of things that are connected together and there's a change of thinking that we need to have a way from standalone documents and toward a whole set of interconnected entities. And it's kind of it's a hard mindset to to achieve and and, you know, until it's actually working right on your desktop.
It's very difficult to to imagine it but we already use the graph and many ways, for example, your social graph connects you with other people. And you already think of people and populations that way. You don't think of people, one person and another person. And another person, you think of groups communities networks, etc.
You think of people as though they were a graph a hashgraph will talk about that later, connects hashes and we can think of all kinds of other graphs. I didn't create a whole bunch of slides for those a cricket graph. Connects. Crickets a narrow graph. Connects neurons. Etc. Etc.
Graphs can be of different types. For example, a graph can be directed or undirected and undirected graph. The connection can go both ways in a directed graph. The connection only goes one way. This is important time is a directed graph. It only goes one. Way facts, real. That becomes really important and a lot of network in data processing.
I won't get into those but trust me it really does crafts. Can also be cyclic or a cyclic in a cyclic graph. You can create loops in the graph but in an a cyclic graph, everything eventually flows to one point or you know everything flows. You never go back to where you begin.
So you can have a directed, a cyclic graph, where everything flows only in one direction. And the ordinal cycles in that directionality. That's called a dag and that structure is used to underly a lot of the technology that we already used. Today blockchain is one example of it but if you do coding on github that's another example of it.
But graphs are important. Not just because of the way they represent knowledge but because of the way they allow us to access knowledge not simply as it cannot collection of facts and statements but rather as patterns that we can see what you're looking at, on the right, is a graph.
You know, it's cyclic, it's undirected but still it, the properties change as the distance between the different entities in, the graph changes in the different entities in the graph change, but we perceive movement in that graph. We see phenomena that emerge from that graph. So graphs enable the possibility of pattern recognition and pattern recognition is an important component of knowledge and learning, okay?
As I said, graphs or everywhere in modern technology, we just don't see them yet. And we certainly don't think in them yet and I can speak of that from personal experience. How hard it is to go from a world where I just write text in a word document or maybe I write computer code in a text processor to a world like GitHub where all of these things are connected in a graph of got versions of got, well, they don't call the masters anymore, but the, the main branch.
We have commits, we have saves, we have blobs tags. All of that mess. That is a way of representing software, but what get help allows is well for many people to work on the same code for the code to be maintained consistently for us to be able to roll back to an earlier stage in the code for us to incorporate documentation in the code.
This kind of process now is also being used for web pages get books and a variety of other types of content. In addition to software, blockchain is a graph at its heart, you know, it's a directed a cyclic graph enabled by cryptography, but at its core, it's a graph and to understand blockchain again, chains, sort of makes us think of sequences, but really, we need to think of it as a graph, neural networks and the, which is the core of artificial intelligence, buttograph, different types of neural networks, different types of machine.
Learning, algorithms are different types of graphs with different properties. Things like activation functions for each individual entity connection weights for the connections between the entities. Etc. Graphs, allow us to produce the fourth item on my list distributed resources. This is an interesting concept, it's a hard concept but it's going to be coming increasingly important over time.
Now, we already have to a certain degree distributed resources, There are things called contents delivering networks out there now that many websites use and so when you access a website, like say the New York Times or some other prominent publisher, you're not actually retrieving a document all the way from New York.
You're retrieving a copy of that content that is been cached on a nearby server. Obviously you can see the advantage of that the content delivering that work allows us to say save on a lot of internet traffic and increase access speeds but distributed resources can be used for things like content sharing networks as well content.
Delivery networks are contracted by the content providers such as say the New York Times content sharing that works. Each node in the network is a standalone thing and the content is just past from one person to another person to another person to another person, you get the same result, many copies distributed across the network and really, the major difference between them is in the ownership and the legality.
Now, the question is, how do you find a resource on a distributed network? And there's a technology called content-based addressing or hash addressing that to use to identify particular piece of data, suppose the New York times, publishes a story and puts it on their network and then it sends out a link to that story.
Well, we could look for it by location, and that's what a lot of these systems work. There's an awful lot of overhead involved in that or we could look for it based on what the content actually is. And the way you do this is you take the original, what we call a pre-image, some text perhaps are an image, or a web page or whatever.
Run it through a hash function and out the other end pops a digest, a cryptographic hash. Now these hashes are unique for each piece of content. So the hash can be used to find the content. You simply make a request to a server on the distributed network. Do you have this hash if they have this hash, they send you the content and you can tell that they sent you the right content because you run the content.
They sent you through this same hash function to get the right result. So distributed where content protocols are in pilot phase right. Now they're not widely used their inefficient, you're slow, they're clunky, there's a lot of overhead. A lot of the tech doesn't work but they're becoming more and more important.
So you have protocols like that or the interplanetary file system and others that use content to addressing. So what happens is you make your request? You request it from the nearest node in the network. If the node has it, it sends it to you if it doesn't have it, it passes the request along to weather nodes.
Eventually it reaches the node that has it then know that has it sends it to the other node and sends it to you. And that way you're getting the content from the nearest node in the network that has it and the content itself through the network. And the more popular it is the more it will be spread through the network.
So there are many uses for distributed resources in education, I developed something, I called content addressable resources for education or care, maybe not the best choice of acronyms in the world. But the idea, of course, was to create educational resources. And instead of using things like licensing or you know, repositorys etc.
In order to make these resources widely available, I made them content addressable and put them on something like IPFS. So once they're out there, they're spread out through the network. The fact that they're out there is what makes them open and accessible. We're beginning to see other kinds of distributed resource networks, for example, digital badge and micro credential networks.
If you think about it, usually when we think about badge and micro-credential networks, we think about badge authorities or credential granting institutions. But once you have distributed resource network established, anyone can issue badges? Anyone can verify badges or micro credentials and this has profound implications for the future of education.
One thing it does, for example, is allow for credential transferability or credential portability, so that a person can amass a range of credentials from different institutions or different providers that together form. A picture of what they've learned now. Is it a degree in the traditional sense? No because everybody will have a different collection of credentials.
But having enough credentials and credentials of a certain sort is kind of equivalent to a degree or so people may argue distributed resources, mainly live on your computer or your French computers or more often. They'll live in the cloud and when they live in the cloud, the resources that we're thinking of, don't need to just be things like images, tax, books, whatever they can be these content and they can be these.
The server virtualizations entire applications that are available on these cloud services. Again this is beginning to become important. You may have seen some of the artificial intelligence generated images, you know, you give it some text to generate this image from scratch for you. Now, I tried to install one on my machine.
I have a pretty good machine here, but still, they required too much memory, but then I accessed the version of it that sitting on the cloud took me a second. And I was generating images and so cloud plus decentralization. Plus virtualization means that anybody can access these complex technical resources, on whatever platform.
I have a Chromebook, a phone, whatever. In fact, I'm using this right now on my phone to do an audio transcription. So I have a backup of their transcription. So One of the mechanisms behind all of this is called consensus in a network consensus is the problem of synchronizing data, but in Society consensus is a problem of making decisions.
And from a certain perspective, the two amount to the same thing. Now, in a democracy a lot of decisions are made through some form of voting and majority role, but there are a lot of weak points to this one week point, of course, is the education of the voters.
People might vote for silly things, you know, from your perspective. Another thing is the integrity of the voting process, you know, there's an awful lot of good will and trust required in traditional democracy and we're moving into a more and more complex society. We're trust is more and more difficult, especially when there's money on the line.
So one of the things that blockchain enthusiasts have attempted to do is to establish trustless trust in a decentralized system and they did not after thinking of alternatives to democratic consensus, a lot of the alternatives, you know, we think like, yeah, we're in a democracy, everything is democratic. But in fact, we have a lot of other kinds of consensus generating processes out.
There authoritarianism is one and we see lots of examples of that out there in the world or in our workplaces or perhaps in our classrooms hierarchy and delegation is another markets. The invisible hand of the marketplace is in theory, supposed to create consensus and practice creates market failures, inequality, poverty, shortages, infrastructure, failures, etc, rumor and innuendo fake news are ways of creating consensus and anarchy.
Which doesn't seem like it consensus generating mechanism but is kind of a junk consensus generating mechanism where everybody just generates our own consensus. What has transpired over the last 30, 40 years is the development of a host of what are called consensus algorithms? The grandfather of these is an algorithm called paxos.
They're intended to solve what's called the byzantine generals. Problem, the Byzantine generals probably is you have a collection of people who are supposed to work together, but some of those people are untrustworthy and how do you make the network work? When someone the members are doing that work are untrustworthy, can't get into the details of consensus networks here.
But you get an idea of how they function from the graphic on the screen where signals are sent from one entity to the acts and through a complex, interplay of signals, different interplays, depending on different algorithms. You read your state, where all the members of the network agree that something is the case.
You know, this ties into a concept in modern data management called single source of truth and single source the truth is the idea that there is one and only one place where a particular piece of data is recorded. And one of the things that data analysts do when they go into a company and we've done that in our sea is look at their data management system and ask, how do you determine that such and such is true and that problem is magnified in a decentralized network.
But as we come to grips with this and as we depend more and more on single sources of truth in distributing networks, our idea of community begins to change as well community is, you know, seems fairly obvious to find by membership in an hour. You can belong to different communities different.
Now networks and maybe one overall social network consisting of all of society or all society, is defined by the government, whatever membership in that network is based on agreement with the consents of protocol. If you don't agree with the protocol that's fine. You're not part of the network, but if you do agree with the protocol then, you know however, limited your involvement, you are part of the network.
These networks depend on the concept of single sources of truth and the consensus protocol defines what the members agree is true or how they actually come to that agreement. Now in blockchain we have very inefficient methods like proof of work or more efficient methods, like proof the state. There may be proof of authority or proof of ownership or simply plain old trust.
That's possible. All of these different ways of defining truth are ways of defining communities. So we have a new type of community emerging community. As consensus as opposed to say, community, as people who live in the same place or community, as people who happen to be stuck in the same classroom, the adoption of new technology, learned, new learning technologies is going to need in my view, to take into account, these new forms of community, and promote, new literacies, enabling students, to thrive in them.
Again, there are many different ways of creating community, people talk all the time about collaboration and yeah, teams collaborations unions. That's part of them. But there are many alternative kinds of community, based on alternative ways, of establishing consensus cooperatives, and networks, or again in the world of blockchain distributed autonomous organizations where consensus is defined by a computer algorithm.
Yeah, there are issues with that six. Digital identity right now. We live in. Well, password help. I'm sorry to be so colorful with my language, but that really is what it is. The only thing that's worse than that is two factor authentication where I'm can't even type my password in.
It's going to send me a message on the phone or some such thing or maybe I have a token or something. Well, soon sooner than you think I think passwords etc will become a thing of the past and identity theft will become even important. Respect a thing of the past and our attention will shift at that time, from how we prove who we are.
And think about how much attention in the educational space is based on how we prove who we are and more toward, how we define ourselves on line. So what's coming new security? Sorry, new secure identities, backed by decentralized identifiers DID. Now that's actually already a worldwide web standard. You can look at up at the URL.
There you are ready? Carry sort of like DIDs proprietary. Siled DIDs in your devices After you use your phone to make payments, for example, or if you have a chip credit card, Any of these things is a digital identifier. But when things become really interesting is, when these become one single network, one single graph so that you can use your digital identifiers.
The same way, you use your wallet today, to contain your identifier's from all manner of different providers, and you have relations with different entities. The driver's license people, the blood donation, people, the credit card people, your bank, you know, comedian auto association. I'm just thinking of all the things that are in my wallet and that's what will carry with us in the future.
What will the physical form? Be well open for question. Could be a chip in our hand. I don't know if we want that but it could be could be on our devices could be in jewelry. The others probably going to be as many varieties as there are people. Now, these secure identities backed by decentralized identifiers, create verifiable credentials and they'll be issued and verified in a distributed network all that technology that I've been talking about can and will be used for credentials and here's an example of one, they're using a public blockchain.
They don't have to use a public blockchain but they are you have the issue or like the academic institution. You have the proverb who's the person, the student, who wants to prove that they actually did pass the course. And then the verifier who takes the proof, offered by the proverb checks, it against the blockchain, where the credential was first submitted and verifies that, it's true.
Notice once the credential has been issued the proverb and verifier can work on their own without referencing back to the issue or that creates some resilience credential networks, it doesn't exist today, you know, institutions close. They go out of business credentials disappear. That's why it's so important to ensure that they don't close right now.
But in the future, where anybody can be an issue or you have chaos and us, you have permanent records of credentials. But now, these new robust kinds of identity rays. All kinds of questions, how do I identify ourselves who controls access to these credentials? What does this digital identity look like and on and on, and on, there's a whole host of questions and these are the kind of questions that we're going to have to grapple with, not just as individuals but also as educational institutions.
And you think about how we create a digital identity already out there in the advertising world is the thing called identity. Graphs profiles of you These exist. Now, the only thing that doesn't exist is our ability to create them for ourselves and and manage some ourselves. But what happens out there in the world is there are systems ingesting billions of events, all of your Facebook posts, all of your Twitter tweets, your website, your work history, stuff, you've put on LinkedIn, your credit history etc.
These are put into a machine language or machine learning model. In other words, they're putting to a big graph that graph is constructed. All of the connections are made and then these agencies look for patterns in the graph patterns that we might not even have thought of, you know, we think of demographics, we think of where, you know, what, national a nationality are you where do you live, what language do you speak?
What your gender, etc? They're looking at patterns, we can't even name and these patterns exist because they are relevant from the perspective of advertising and marketing. Well, how do we control that? How do we influence and manage our own digital identity and what role do academic institutions, play in creating a person's digital identity and helping them manage their digital identity?
We need to change the way we we look at things, right? And whoops, that was supposed to be an animated gif with them all waving but they're not waving. So I'm disappointed in this slide in this new data rich world, we are other content. All of the graphs stuff, all of the virtualization, all of the data ultimately is about us and how we interact together.
The things we need to know the things that we can sin, the things that we make where we live, why? We live the way we live, and there's a movement sometimes called Indy web sometimes called web three sometimes called sovereign identity. It has many hats and many forms where we are taking care of our own content where we are managing our own content.
Now, nobody can own all of the data about themselves entirely because all of the data about ourselves are virtually, all of the data involves some kind of connection between ourselves and some external things, it's all in a big graph. And so the question is, how much influence do we have over that graph?
We'll come back to that. I'm going to jump a bit and look at creative experiences. So much of education currently relies on independent methods, right, we tell people about things, rather than having people do things, you know, we use vectors reading, lectures videos, even this presentation, I've tried to give you more of an experience by creating the animated gifts, but, you know, imagine how poor it would be without the narration behind it.
So again, still indirect experiences in the future and indeed even today to some degree education becomes a case where instead of delivering content, the teacher models and demonstrates successful practice. And this student tries to emulate that based on what they experience. In other words, learning becomes immersive and experiential, but it's not just about doing things in a virtual world.
It's actually about creating and designing things in a virtual world or in the real world. The creative experience is, can take many forms, it can be open working. I have an example of that. I have a playlist on my YouTube channel called Stephen follows instructions where I take, you know, some set of instructions, install the software, create this application, whatever, and try to follow the instructions with hilarious results, because the instructions are very often, bad job shadowing, is another way of doing it apprenticeship mentorship.
You can think of a wide range of immersive creative experiences in the whole idea here is to get the direct experience rather than to be told about it or learn about its second hand, more and more people are doing things and these virtual and online environments, and doing an openly and sharing this with their network of friends.
We see this already in gaming, a lot and pretty much only in gaming where people are playing their game, and streaming it on Twitch, and you see the person they're doing, or running commentary. I've done a whole bunch of those with a video game called no man's sky. I didn't put on the link because unless you really into no man's sky, you probably won't like it.
But the idea here is that we have tools today, that allows us to work and do things and also share the experience twitch is what we're looking at here, open broadcasting system. Something that I use like even tools like teams or slack or discord or ways of trying to make the work experience, more open and more interactive and all this dialogue.
All of this interactivity takes the work and puts it into a context and enables learners to see it as a process rather than an artifact. I talked about recognition earlier being in this workplace environment, or in this creative environment, allows people to recognize things much, like the advertisers recognize demographics in ways that can't be articulating ways, that can't be talked about.
But in ways they can learn. So that you can see when a fish is well done. You can recognize when somebody has given you. A good reception, you can feel the passage of time. He's these are ineffable kinds of personal knowledge, as Michael Polanya. Would say leads us to recognition today.
We rely on artificial assessment, we rely on artificial instruction. We rely on artificial assessment like tests and exams and I don't need to talk about the many flaws of such a system or the misuse of some of the current technologies of. I've been talking about like artificial intelligence and surveillance and the rest in order to try to make these artificial assessments better.
We're not going to be able to make them better. Not too many significant degree. One way we can make some a bit better is in the assessment side. We used to assess them by hand. You know, it was manual labor and I know because I I've had the experience of having a foot of marking on my desk.
You know, measuring my marking in linear feet and I start coffee and one hand pain in the other starting my papers and marking my hand. Not a very consistent and certainly not a very efficient way of assessing work. And so a variety of artificial intelligence technologies have come on stream these exist today if you automatically assess long form, material like tests and essays and assignments, etc, and it's a recognition task.
We put them into groups, you know, ABC, and sometimes we can explain why the AI made the decision. It did. Sometimes, we can't imagine issue. Of course. But why are we automatically grading tests? That doesn't make sense. We need to refine these tools and we need to refine them, not for tests but for actual authentic tasks.
So we use a, I to look at how somebody is actually. Yeah. And an existing expert is actually performing a task and then use that model in the assessment of other people attempting to perform the task. This has a variety of advantages. From one thing, it reduces the potential for bias because we're not working with signs, and symbols and indirect signifiers anymore.
We're working with actual real authentic practice, Does it eliminate the bias? Well, no. And that's something that we need to keep into account, But also it allows anyone to perform an assessment. Yes. Like when you're sitting there watching figure skating, right? Yeah. The judges have their score but you have your score too.
And you know, you're all working with the same, shall we say, source of truth, the video performance of the videos of the figure skating or perhaps the actual performance. We're drawing from this person's presentation of their skill and ability and we each go away with our own assessment and you know the the scores in the arena will perform one function.
But, you know, you as a spectator, you have your favorite figure skaters the people you follow. The people. Your cheer for based on your own kind of assessment. And that can happen not just on a performance level, in structured settings like this. But for everybody all the time in actual public, performances are actual personal performance or portfolios, if people are working openly, they can be assessed openly.
Not something that scares a lot of people but it's also something that has a lot of potential and in the potential is this, you don't need credentials anymore. Actually, the credential of the future will be a job offer, a company out there has run their assessments again, who knows?
Hundreds thousands, millions of personal portfolios. The digital identity that people managing their own digital identities. Using say the indie web of presented to the world, they identify someone who according to their systems matches. The needs of the job that they have based on what they're previous employees or experts have done, and they just simply offer the job to that person.
And so, a learning institution isn't trying to create a single set of skills for everybody anymore. That wouldn't actually be very useful. They're trying to consider how to help students create their own desirable profile. That satisfies a wide range of actually undefined performance criteria. I want to use to say, you know, the when I was teaching, I'd tell the class the assessment method I was using this course is arbitrary and unfair just like real life.
Now of course I was kidding but it will be in the future arbitrary and unfair. And unless we think about how to do it, properly agency is necessary in such a system. What we learn what we do, why we learn are ultimately, is shaped by the learner. And, and the reason why it shaped by the learners, because the standards that are set by external agencies become less and less relevant.
Especially the more static. They are the less able to respond to dynamically changing environments. They are agencies is a shift in the relative standing of an individual with respect came to community institutions, governments for better or worse and it's impacted by technology. And there's a growing acceptance of this technical knowledge.
Everything from automated publishing to stock trading content, alerts etc. The ways. We institutions, etc project ourselves in the world. Individual agency is what these consensus-based decentralized. Communities are designed to augment. But how much agency should people have? How much is too much and what other types of agency are there?
How do we create and define community agency, network agency, organizational agency, and and keep that open and voluntary for the members involved? And what in this model counts as success again. We're so used to us as educational institutions defining success but priorities are going to be established by individual conditions and character and defined by a variety of elements.
And so we need to be able to allow success to be evaluated by these wider metrics. And that's a tough call for educational institutions. All of this combines to create an infrastructure and this infrastructure touches everything that we touch in our lives and what to happen in the pandemic.
And what's happening in new technological environments is that we're going through a reorganization of social infrastructure to fit these new priorities and we're seeing everything from, you know the the great resignation to quiet quitting. All of these things are signs that we're looking for different priorities and we've got tensions between things like anonymity and encryption zero, knowledge data, and things like accountability, resilience, security robustness.
You know where does the panama papers fit in this? We're addressing gaps in our social fabric. We're addressing the need for greater individual and collective capacity. And we're at the same time, trying to establish through things like single source of truth, resilience of our scientific and industrial infrastructure, including supply chains, including our news media, and a whole range of other relating things.
And finally, we can't escape the pressing and immediate concern of climate and environmental security sustainability. So all of this together is the technological and and and, and looking for that other word, but it's in the title, the trends of the future, and these form the framework for the questions that I think that we as an educational community need to be asking in the future, you know, once we've dealt with artificial intelligence them, editors and blockchain.
Stephen Downes, Casselman, Canada
stephen@downes.ca
Last Updated: Apr 25, 2024 4:25 p.m.
.