Module 3 - Discussion

Unedited Google recorder transcript from audio.

Now, everything running. Okay, so I'll keep an eye on in case we get more people joining us, you never know could happen. We've had a high of three participants now in this course, during the online discussions. So it's getting better and better as we go along. So for those of you who are watching the video or listening on audio, this is the discussion session for module three of ethics analytics, duty of care.

I'm Stephen Downes. I'm joined by Sherita and Mark who have been stalwarts attendees throughout so far. And quite frankly have as much to do with authoring, this course, as I do because without their contributions, this would be a very different looking course.

So, not really sure where to start. Did you guys? So there were two tasks in this week's this week's edition I guess of the course the first and I saw a bunch of things that were added to add examples or links to the ethical issues. And actually, I found myself having some ethical issues before even this this, this particular discussion because I was a, you know, I've been doing the the slides with the audio which is taking me longer than I thought it would but but that's okay because I'm still getting good content out of it, at least, to my mind, but I was doing the section on bad actors.

And so part of what happens when I do these slides, is I look up more resources because I don't have enough already. I talk implementation and I hit on a paper, and what was it called? I forgot what it was called. Got a good memory but it doesn't last very long.

Oh garden paint list. Post just post. So this is what happens when you're not ready. It was called AI enabled future crime. It's from the journal crime, science as pretty comprehensive. So I'm putting a link to that article in today's newsletter. So you'll be able to see the link and actually so you don't have to leave.

I'll put it right in the chat here. I quite liked the article. Now a lot of the prime goes beyond scope of this course, although you know, pretty much any kind could fit into the scope of this course. So I wasn't sure, really where to draw the line. Anyhow, there it is.

In the chat by good like it. And so I thought I'd share so and I'll be doing that throughout, you know, the course, adding more posts to the newsletter. That I find interesting. And if you guys find anything that you think I should share, send me a note and you know, because you know, there are good resources and resources that aren't necessarily apparent and the newsletter.

So did you find anything of interest and adding the examples to the the issues? But I did, I found it quite interesting. I found my was going down a rabbit hole because there are so many things that can pop up, you know, even on page one. Yeah, if you're screwing so and and the last thing that I put in which is just a few minutes ago was under Eminem and the tour browser, how you know, I used for before, you know for whatever.

Yeah. But what I found interesting was, when I thought about it. I thought about good possibilities for poor tour. Yeah. And then I thought about bad possibly and you know gray area in between such as you know, you know pirating a movie and right where you can protect yourself using a number of ways.

Not that I've ever done that, you know. But so I found that really, really interesting. So I thought, okay, I'm gonna go down all of these things and get lost down a radical. It's fun. Yeah, that keeps happening to me that I couldn't turn down reading and studying this crying studies thing, really?

That sounds fascinating. Yeah, was it was pretty good. How are you marks? I'm good. Yeah, I have that same reliable problem. Yeah, but this week I was at most my time in different radical. I did add some resources but I haven't gotten to the Lincoln art. Mmm. And and that's fine because I'm sort of laying back and I read a couple of my teams first.

I think it's hard to pronounce this again. Yeah, and I'm just trying to rap. I had a round. I mean, I get a concept of the way people, but the actual been doing of it and the purpose of the, knowing of it and doing it purposefully. So that's none of my hold there.

I'm gonna another standard online college class. Yeah, the sweet side that could mean something up by that price, which is unusual, but it was a good way that it was one of those kind of openings to kind of what you want. And of course, that we're on my way.

Is that? Yeah. So how did this course compare with? That course, just what was your experience? And I know this is a much smaller course, but I mean, just generally. Yeah. So this is actually one of my areas of study is online and picture and particularly learning management systems.

I'm actually certified in three of them and this college state university in Minneapolis Minnesota users of fourth one piercings, right? Space B2L which is not the best system and then they don't support their teachers and then on my features can you she's not been probably brain and she never sees this like her a lot.

And, you know, it's so to compare a navigating your course is no more difficult than navigating earths and that doesn't mean you're easy than that, but she's in a probably the most expensive about former readings system. And so in my mind as a student, I should be able to click forward them and just work through the materials.

She posted videos on this next nation, you know, you might need to jump out for would be something, but I should be able to just click through each week. Yeah. And and so it's actually structured quite a bit like yours in that you have to keep your turning to the beginning and then branch from there and which is fine.

I mean that's standing because but from a user experience and I always look at it from the student user. Yeah and I've actually yeah to experience sports more. You've just clicked straight through the course some courses sort of break into pieces where you can get on a roll and click to the material but never through 14 weeks and that to me is you know, it's an ideal.

It's probably unbelievable. But to me that it's that's how of course should potentially. So and then wanted to add a drug course has is that Well wanted to ISIS. First of all, of course, this more agents and everyone is as equal agents. And that's, that's an ideal hiring vacation as a move something?

Yeah I should say. I mean there are certain things but as you well you know exactly who's talking about. We can name the board by people. Yeah. Better talking about this North American here. And then of course, I lost the word thought there. Oh, and your course does not hide where the closest.

And this is my biggest complaint with most online. Oh yeah, it's this one. Putting each at a certain time and that is not to never because sometimes we get behind sometimes we have an opportunity to do it and you know, life is not in the neighbor's mind and there in the case of this course.

If it's not up yet. It's because it doesn't exist yet. Yeah. But I mean but anybody can click in and see where we're in. Yeah, yeah times yeah. That's you know again because. Yeah, but at least the map is. Yeah. And in this report I'm in the most highly online voices.

They've refused to show you now. I don't know, strange. Yeah, especially if they were credited. So I mean you know how you your most checked nation and the course listing in the faculty the catalog that the faculty has to work with. How many students not the only because of leaders that means they're governments and I even yeah.

Yeah you know particularly what well I would respond to is you're saying that the person who's teaching the course you know is 10 years. Yet that person may not have had very much training on how to put a course together, and, or how to teach online, and this is very prevalent in university.

I mean, even back enough when we teach in a university, who taught you to be a teacher. Yeah, no one, no one. Yeah, so really short story. So far, one of relations of my college career, turning after the great recession of 2003, once in a shared governance meeting where the chancellor of the faculty were going out.

It was a four campus community college district. So we had a chancellor and college suppressing faculty, I believe. And the chancellor who had taught in that in previous decades. And then the laugh degree. And from that was in this confrontation with the back of the radical back that. Remember and the chances said to this member, unlike you all, I have a teaching student and I was just like, you kidding me, kindergarten teachers, haven't thinking certificate, but you know that was a revelation.

And it's it's an issue that's been raised before. By, for example, Tony Bates who, who has argued that length, that all collegeing university. Professors should have, you know, required courses and how to teach ideally as part of their PhD. I have mixed feelings about that because now somebody becomes a professor of physics, not so much because they want to teach physics, but because they want to do physics research and so, you know, it sort of goes against, you know, good pedagogy to force, somebody to learn something.

They don't want to learn at least to my mind. I know we even know it's a requirement of the job that they actually know how to do this sort of stuff. So I'm a mixed feelings on that. On the other hand as soon. Yeah I would wonder why there wouldn't be a practice room included before setting a research removed so fast.

Don't let doctors come out of their classroom and into the operating room, you know. Yeah for airline pilots out of the simulator. Yeah. And in the content and I would argue that teaching is a debris, serious. Professional stop. I'm thinking though that a professor researcher and some subject probably ought to be viewed as a resource rather than, as the designer of learning because they're not going to pick up how to do it in just a few courses.

Anyways, you know, I think the analogy might be, you know, we don't want aircraft engineers to pilot aircraft not because they haven't been trained and how to pilot aircraft but because really, that's not what they do. They design aircraft, they don't pilot them. We have separate people to pilot them and these separate people need to know something about the airplane, but they don't need to know how to build one.

How's that for an analogy? Is that work? Yeah, well it's reminds me of the it's probably apocryphal but apparently you used to have to be able to build apart to get environmental license in Russia. Again, I still don't believe that that was true, but that's that was the story in the Soviet Union that in the Soviet Union, you had to demonstrate mechanical capability before the night.

I can believe that part of it the mechanical capabilities part because they didn't want you abandoning the card in the middle of the road just because you couldn't fix it. I think what you just said, though, points to my abandoned favorite word, as you can only, you know, music so much, the importance of my patient.

Yeah. Is that and this refers to our conference, you've already meets you that if you have a learning technologist or learning designing working in tandem with a subject matter expert. Yeah. Then then you're going to get a good force. Yeah. The students are going to get the better learning.

Come the institutions going to, you have a better reputation, you know, you have a better product. Yeah. Another reputation and enrollment one. So I think we're headed in that direction except for the useless. Yeah. Something that's probably true. Yeah. What that's and yet. Now that we hired, you know, 10 times administrators, we used to be decades.

Those people should be looking at cutting cross in other ways, than reduced in faculty to find their resources. Yeah, cutting themselves out and putting those resources to increase support for that. So the holy grail to bring us back to our topic of all of this is supposed to be artificial intelligence, but it's not obvious that this is easily possible from one of the physics.

Interesting, one of the bad actor things that I was doing just before this session was misrepresentation, where you have a snake oil salesman basically coming along and saying, AI all this. You should be invested in that and, you know, I can think of some of these LMS systems that bought into some of these AI systems and made their inventors rich, but didn't really add very much capacity to their LMS.

That bothers me. So, but it's interesting. Characterization makes me think by adding AI to, I'm learning designer and a subject for you would have your infinity? Yes, this education. So particularly this is yeah. Where people are not gathering in the world more around the fire. But our media is that perhaps those three elements would be the best.

This one, I would be interested in the debate between those three elements because each of those elements in a way, has a different purpose that I did plans on how they see the tasks here, I am referring to a artificial intent, you know, intelligence as an entity that you know, it's okay, it's 2021, you can do that.

Yeah, you know if that entity has has a perspective in the lens the you know, instructional designer has a perspective and a lens, they may never have taught anybody any and that comes out when they try to teach the subject expert, how to use the technology? So I love to see the debate.

I think it would be a great triumph for us. Yeah, the debate would be really interesting. You know that the you know the the learning desire designer may say we don't need that, we don't need that. We can make this happen like this and AI may say but I can predict that X will happen and you know the teacher or where the the crop might say you have no idea where I could go with this.

No. But if it was networks interactive and let's feedback boots. I think that Trinity is really create a main way. Eugene learning. I think we extremely interesting. I'm advocate for another piece is to put the students energy. Oh yes, yeah. So then yes. Quadrupon and working. Oh yes. It's now the young holy trinity.

Yeah exactly. Yeah. And then to, you know, to be able to go that you know, and that's what they're institutional score, right? So maybe an especially, isn't primary, focus isn't mowing the lawns. And even with pool cleaning, maybe institutions primary should be the oldest space, where something like this, the assemble reflect and improve growth, you know.

So do you want the AI to be a participant in the discussion or a mediator? How before we have to? We have two once a moderator. Okay? And and one is a a Siri, what do you think about this? Yeah, like just have flash of irrelevant responses coming from Siri the temperature today is 14.

Yeah. What's the background in back? Something? Whatever it is of any ideas to improve. It's responses. Yeah. Speedy. Not that. This one. Yeah, the other thing and this is part of the reason why I'm doing everything in this course, even though it's stupid is actually making the decisions. When you have multiple people, you have multiple people having a say and that work with teams both of developers and instructors and all of that.

And here I am putting on my professor hat even though I'm not one, you know, I've got a pretty good idea where I want this to go. And most of my experience working with these teams is failing to convince them that they, it should go in that direction. And now I can imagine having to argue with an AI correction and being overruled.

Yeah, but it might have to be some rules in routines. Yeah, there are plenty of existing rules and education. So again, back to renovation and of course, as professors are played them on YouTube. And, you know, I can't think of the name of that document, that an instructor has informed me to document that was submitted for accreditation of.

That course, is a name. I can't figure out all accredited courses. Have to claim and that they conform to the intentions of the outcomes and the accreditation background of that course. And I've never did that, I never did that. Because well yes because by the time of faculty members assigned, the course that's forgotten about exist.

It's it's it's in that space between the institution and yeah, it's in that space and professors. Yeah. Again, never became about that part of the system and of course, they would use this the suggestion that there was informed to it. Yeah. But the pigeon inspiration has to be able to attend that it questions about the presentation.

I've been in some presentations about anyone, right here. It's my my experience is that I developed workhorses from scratch and I had just put in certain things like you can't have, you know, all the grades happening at the end. Okay. That's something 100%. You know, it is really little, it's a bit pieces.

You know, you provide syllabus the first time you do it. You'll always change that unit and put all of this stuff in to, you know, an advisory committee and they've rubber stamped it. I never. I mean correct. Yeah, I know. See, I learned in seconds and was made yes as a student representative.

Yeah, you know, and I had the time, you know and most we still but yeah it was and it hurts me also on American war, both Canadians. Right. And so I'm speaking about an American experience in California, right? So that's a particular accreditation agency. Yeah. So I realized that I'm guessing that if YouTube doctor Brown, because your institutions are also critical, but I'm not aware of how your graduation agencies are structured with their different or the same as are.

We have regional violations. Yeah, if we do too, I think we do to imagine the provincial, you know? Yes. And also there are a couple of now separately through since the federal government has happened. That only reason since the first questions presidency is one of the federal government was completely out of business.

But they have become involved in. So there are a couple now of national whether particularly handy for online again. It's I think it's going to be a good thing and I think it can be a better to because once you get the rules and accreditation rules too tight and then your ability to change your course or to do something totally radical like I'm not going to grade, I'm not going to, you know what I'm going to do is I'm going to offer you all an aid and people who really move forward.

Will you get the a plus you know you're not allowed to do that no more. Now what I did. So you know and accreditation would basically tell me no, you're not alone. Yeah. Yeah, that is a better pedagogy than having, you know, flash tests level. It's so much better clickers.

Oh that one it was gone everywhere and so that's in my experience. So so again that's a bad thing of good. Yeah, right. Well imagine wanted to the attention between what we're doing, which has been yeah and the potential over instruction situation which is basically how it's operating these days especially with this topic administrations and all the tasks and the course data.

And so, what we're searching for, I think is a minimal way that allows more agency, more community and yet, you know, offered to some structures so that we move from. So when I move from this class to a different from completely instructive class, the overlap is going to be very small, right, right.

And so what we're searching for here is some sort of open framework that of a discipline. Whether they're something better expert or not and anticipates, the way the next horse will go based on this is loose thing around the highest structure. And my idea in my ideal world, we wouldn't actually divide all of this by courses.

Anyways, just be the professors would be participating in this overall giant graph. And then people who are studying, that would just go from place to place on that graph. And there might be advanced happening at different times and so on. But, you know, you wouldn't be locked into a series of events or anything like that.

I don't know if that would be better, but, but that's kind of what I would prefer. But then I really fall on the no rules side of things. So, the other thing that I would say, I'll go again, this has been off topic, is that teaching young children and then high school and then adults is really, really different in terms of how the people approached, what they learned, how they learn etc Once you get into adults, who are doing this, because they have an interest and because they eat a specific skill because they need an answer for something that they're interested in.

You have a totally different student. And, and one where I think the more open way of doing things is really better because, you know, I don't learn what they want. Learn. Right? I mean and they don't learn what they don't want to no they don't be really stubborn about it.

Yeah and I still have some mandatory courses that interest you wants me to take and they're only half hour of courses but they've been sitting in the hopper for several months now. Yeah. Proving my boy. Yeah. So I I use three different or these different types of learners and I promote the last one.

I'll get you in a second but it hasn't thought on yet. But anyway, so yeah, that's developmental learning and that that's a perfect word for it. And whether it takes six years or eight years and 12 years, there's some discussion about that. I argued that to I'm old my parents are old when they academy and their parents are old women.

They have them so back. My grandparents had very related informal. They went sleeping. They weren't ignoring. They weren't uneducated. So you know their case can't be made that we have almost too much. Anyway. That's gotta go. Then there's advertising adult marketing but it's all that it's more. Taskbarient. That's more careerists.

Yeah you know that's mean it's a long literature email. It's almost on you know no you have to win that. Put also and I don't have a word for it, but it's not everybody's interested in that skill. Not everybody is interested in learning something for a career. And what happens with adults is, they're often interested in learning something just because they're interested in learning something, or they wouldn't be there.

And yeah, that you have that word. What is it? It comes from Australia. Yeah, and only probably 20 years old. So it's a brand new word photos. Yeah, each, yes, I've seen that. Yes. And that is itself corrected about. Yeah, that's different than the antonym and mechanical and that's the result learning classes and second language class and you say you can high schools and and it's a perfectly good category.

In fact, I think that the modern university has somehow slipped out of the bedigo into it but the vendor is hoping and I think that's a great port because it self-directed this the first part of that stuff and that's the part that can be open and creative and a grant.

So how would AI?

So, yeah, that's that's the question. I've been thinking of, as this conversation has continued. So how it fit into that? So let's go all the way back. So is going to ask way back when we are early in this discussion, talking about accreditation and, and all of that, imagine we had a eyes that monitored the courses and decided that they should be a credited or not accredited, depending on how they were being conducted.

What are your thoughts on that?

Who programmed the AI? It's my first question here. And who does the report? Wanted, another example, I love bringing up amongst the containers. Is the Edinburgh University? Yeah. It's an actual, okay? And they own their own building on campus. They have their own funding, their own industry, their own rules and at the university process them things happen.

Yeah, real real consequences happen. They also started the pretty here started the, what the fridge, all the frames. Okay. Yeah, that came out of the bathroom and you. So if they only got it's they programmed me, there we go. Very high. Then one purchased from the corporation by the Missouri.

Yeah, yeah. And and exactly and then missing from that conversation unfortunately the space is in the United States is to faculty. Yeah. They used to run the universities saying no longer work. No, and that's, that's true. I think increasingly around the world. Yeah, yeah. And that's you see a lot of these applications of AI in learning directed toward leading people to specific outcomes, usually outcomes related to employment objectives that are determined by.

Well, basically a mixture of people in the community who hire people and university administrations and politicians. And it seems to me that that raises the issue of whether that really is the appropriate way to use an AI and learning. I don't know if I have it down as one of my issues yet, but, but, you know, just the idea of using the AI to wreck to direct people toward a certain end in a system where really we'd like to promote autonomy of some sort, especially at the higher levels.

Yeah, it seems like we're evolving and more imports. The German model of sorting people in the careers and services. Awesome. Which used to be existing is in California, and it's almost like these ones, but we might now with six hours. Six. Yeah, comes back down to a discussion or a debate or

It would even be a discussion or debate as to what would the benefit be to any of those entities? Yeah. And and there would be some kind of tension there. They would have to be some kind of thing. There, I always say that it comes down to governments, that's Google's license inside and it would be an ongoing debate.

Hopefully annual reviews are figured, oh, and, and is the AIB annual reviews. They get, they implement theirs to. Yeah. There's what we, you know, more detailed and very long. And the students, I would hope would be the most critical. And the faculty would hopefully do the most online points.

You see what else on what's needed? What improvements? Are you and then the administrators, you know, they have to be drying above not using that, so this part of it. So it would be few minutes. So we're being interested in and I guess you'd have the color code in the school group, so we were there.

Okay. I don't know. We're going to find out, people will eventually find out how this is, but they risk here and then it actually shows up one of the issues that I was writing about talking about this week, is that, when you let the AI make decisions. Now, of course, there's always the requirement to have as they say humans in the loop to, you know, affirm the the judgment that the AI makes.

But what happens is that it becomes kind of a rubber stamp thing, and sure you have the meeting. But you generally tend to defer to the AI because, you know, it's done all about work and you probably can't hold smart. It's you just rubber stamp it and it ends up being the AI making all the decisions.

Anyways, yeah, that's a bad black box problem. Yeah, yeah, yeah, that's the black box of course this machine is smarter than me, hey, you know, ever to argue with an ATM. So perhaps that's in purpose to argue with monitoring. Yeah, I keep coming back to what you said, just a few moments ago and that had to do with power, who has the power, right?

And, That's of that. Could be a real slippery slope. Yeah, when you're dealing with, you know, aon that could be a real slippery slope. And this would AI the more with AI have more of an infinity with more pictators, right? People who ask total control, okay, I don't know, maybe that, but whatever.

So, would AI have an infinity for that? Good questions sees. Yeah, but again, the AI that we've interacted with will build by corporations. Yeah. Right through our primary school and certainly power. Yeah. Primarily one could even say that they certain of those corporations have the power because they can also very easily influence legislature.

Imagine the only tool we had on our government. See, Imagine the only two we had. Sorry Mark was the Facebook algorithm for choosing what we hear what we can say? Yes, I think that points to the multiple algorithms. Mm-hmm. And then back to power multiple centers of power, and here I go to be in America but across three or four.

Exactly, not the best way to govern, but perhaps one way to make sure powers not sunflowers. Course, the final M result of that is still made. Yeah, we're looking at, yeah, but you're not the first to live that either. In ancient Rome they had clubs that were designated by color red.

The green, I think the other one was yellow. They eventually broke down into two facts and said, I believe they were blue and green and they would have pitched battles in the streets of Rome. And they, you know, the only thing is like they weren't politically agencies particularly but but you know, so there were sort of like glorified fan clubs, but you but the thing is over time, what happened is all the, you know, all of the balance competition ended up being like perfectly even and and it all boiled, not to just two sides you didn't matter how many started with you ended up at two and they were perfectly balanced and you could never get a resolution our hours you know the distillate him world war one before new technology emerge.

We had two sides that were perfectly bound, you know, and he couldn't get a resolution, and that's what happens, you know, it's good to have checks and balances, but when they're equal over time, they progressed to a point where the you can't get past. Disagreement. The US had three has three.

So in principle to could always vote the third. That's not really how it's worked out because this three checks and balances are now become irrelevant. The major political agencies in the US are the two political parties and these three areas just become the arena in which those two political parties fight against each other.

And again, never reach a resolution. That's my observation. Anyways, and it's a problem in AI too. You have competitive like there are neural networks which are having to pools, right? And you get your input into the system, and then you allocate that system into different pools of neurons, and they each do their own thing, and then they come back together and fight it out.

Right? And in one case, and that's their, there's I have the example of a duck rabbit in one of the slides, and one case, the competitive pole, just simply swamps out all other opposition. We might think about is, you know, being similar to how the US reacted to communism, right?

No, no, you can't have that period. We'll move this way or you know, or alternatively you get a case where the AI can't make a resolution. Can't decide between doc or rabbit and can't do anything. So it's a design issue and and it's, you know, it's it's a design issue in society and it's also design issue.

I think in AI and I it's not clear to me that the methods of resolving it are good methods of resolving it. I mean might be but what would be a method of resolving it? In terms of they are us, I think having that's a good question.

I would almost say not trying to resolve it if I had to answer that. I mean, I think the problem happens when you try to force the AI into rule like behavior, pick one alternative or the other pick Republican, or Democratic duck or rabbit and all of these problems are so complex that one or the others.

Never going to be the correct answer. And sometimes one. Sorry, but species of duck or a rabbit? Yeah, that's just called getting it wrong. Yeah, we're getting on dangerous character there. Yeah, I was reading this morning about early and AI not being able to distinguish between men and women.

Yeah. Artificial skin effort and apparently that's been small. But then this brings me back to human agency. Yeah. Because I really en agency and periodity because now we have people who major recognition will not be able to sort. Yeah, it's just they have figured out a way to present themselves in a way that is not fighting here and I don't know about you are running so I'm online all the time.

Yeah, and so there's the creative response of the demand binary deformity that AI might require. Now the human like no. So it's I love the term, wicked problems. Yeah. And I have that's where this force is inside of the problem. Yeah, there will be no resolution and there will be no solution.

There will just be exploration and building this graph and see what that suggests and new iteration next year. And yeah and that's why we're building that craft. I've two things just to to bring us to a close. The first thing comes up came up in my thinking, as we talked about this, it's almost like contemporary applications of AI and learning are attempting to replicate.

What Plato proposes in the Republic where all of your attributes are identified by and play those case, the philosopher king in our case and AI. And then you were placed into the right role in society and that is your role. That is your lot in life. There's been a lot of objections to republic because you can imagine and I think similar objections come up when we try to use AI in that way in order to sort and catalog, and categorize and allocate people to their places.

Second thing I just wanted to open briefly the, the graph exercise, and this is more for the people watching the video than, for you guys. But, but so, but I think it's still be useful to you as well. So I just want to show how it works. That's like I've been waiting for.

Okay good. So it is you sorry one unit market all up like you can't this is something you can break so yeah, right. All right, so okay so let's go. You should be looking at sorry and bouncing around to be here because the the stupid sharing thing covers over my tabs.

There we go, right? Hey, you should be seeing the ethics analytics and debut, of course screen right now. Okay guys. So, I'm gonna go to ethical issues and I'm going to go to the appropriate task. So add to the graph and we'll access, I didn't have to make that extra click but to access the graph tool, click here.

And so, here it is. Ignore that. Ignore these, I need to take them off. But the way I load the graph right now, that's how it works. So this is more than I wanted to put in a single instance of the graph. But I'm just working with a default file for now, I'll change up this display a bit, but the idea here is that we're trying to we're looking at these applications that are on the left and then these ethical issues that are on the right.

And the idea is to ask yourself what applications raised what ethical issues. And so if you think for example suppose you think that dashboards raise the issue of content manipulation. It's probably don't, but let's say you did, right? So just click on dashboards and the description comes up in the right in case you need to refresh yourself and then press the alt key and then clutch your mouse and drag and then once you've reached the other point unclick and you've drawn a line that's tricky because I always want to just drag and that's what happens to me.

That's I will probably change that once I figure out how you just drag the line, but right now if you just click and drag you end up moving the things and I'll probably flip those. So you have to do all, click to move the box but just click and drag.

But right now that's how it works. So just draw your lines. So we've got dashboard to content, manipulation plagiarism detection, the fear and anxiety. I think that is a real one plagiarism detection to privacy generally. Yeah, I think that's one. So, in the future, I'll just have a few on the left hand side and if you know, on the right hand side, so give up facing this overwhelming, you know, graph of a hundred different elements on each side.

So once you're done that, you're happy with the graph as you've drawn it. Right. Click and that pops up this. Now, there is this you can add a new item, it will say, the new item in the information that it sends to the server. But I don't do anything with that.

You can also wipe the claim but then all you've done is white to clean. The real thing you want to do is click export, click export and the system will take care of the rest for you. Don't tell you. You graph was submitted successfully and it'll list things that you've added click okay and you're done, it'll take you back to the test.

So that's how it works. Now the way this is going to work hello. Now right now what it does is it's saves what you've done and adds it to the end of a file of links. So I'm just building up, but this file of links would be fed back in to the system.

So that when you click on, say, let's go back to applications of learning analytics and we can look at each of these applications. So all applications. So let's pick one plagiarism detection. I think this one we were using so right now. Oh we don't even have any articles I thought I had an article in there I guess not.

But, in addition to articles, and links, there would be another section underneath that would list, the ethical issues that were raised. And so, and it's not just me defining, what issues are raised by plagiarism detection. It's all the participants in the course who are defining that, right? And so that builds this interrelationship, so that you can explore this issue, you can explore the other factors around this issue.

And as a side, that's really why you can't. I mean, we could in this course you couldn't have it. Go from the beginning, all the way through to the end. Theoretically that's doable. And and, you know, I've done that with some of these next previous things but because we're building this as a graph, you know, it's the sort of thing where you can just wonder through the graph.

There is no one path through the graph. Although what I'd like to do is create more range of different paths so you can experience this differently or you know, maybe do something like a click capture or whatever so that you can go through the course and then the way you went through the course, becomes a path for someone else.

Something like that is well the, the grass that is created in this case between the application and the ethical issue that might be selected, once it might be selected a hundred times by different people and that gives us the possibility of each connection. Each link between things is having a weight and so the we can use that in order to highlight the issues that are being selected the most most often and and, you know, wanted to highlight parts of the course, or even just highlight the display here.

We'd say, you know, well, pick maybe the five ethical issues that got the most length here, and that allows us to focus in on those. So that's how it works, and that's what it's intended to do. It's intended to be very easy for you guys but very powerful work.

Yeah, so just for my own clarification. Yeah, the reason he said you can't break it is because you're aggravating them. He's suppliers graph, right? Yes. So probably just want to be clear about that and then the white flicker does that just clear that connections you made or does that activity entire graph bottle, clear?

The whole graph and then you'd have to reload it. Yeah, nothing. Perming is happens to white clear. Now, I think let me just check this. We'll go back to the go back to the activity. Oops.

Learning how we're doing. Is we're pulling a copy? Yeah, the module to wherever it is anyway that we're pulling back copy modifying it and submitting it to the. That's right. Yeah, this is just a copy, doesn't matter what you do to this, so let's say it's wipe, it clean.

So that's all you've done. But now if I reload the page, so I'm reloading it here and there it is. Again also, I think, let me just check, I'll draw a line here, click on the line. Oh, all you can do, there is change color. I'd love to add a deleted to that, but I don't think there is a delete at the moment.

Now, there isn't just change color and even that doesn't work very well. Okay. So that's too bad but you know it'd be nice to have an edit or delete on this. It's not a huge thing but it's a certain way doable. It's just figuring out how to write the code, but yeah.

And you ever figure out a way to zoom in and out. Now it's all on one. You know, it's that I did not. Yeah that's a really good point. I'll talk about that somewhere. I think it was on Monday, I did. Yeah you message us if he knows cuz it seems to me there is a way.

Yeah but I would be a lot better for sure. And you know is this help maybe so I think yeah no yeah I know all it does is load this intro game. Well with pain dropping put that single items, white export we color drag. Yeah I don't see anything that says changed the size.

Yeah well yeah can inside the side. Yes. Zoom. What we're looking for is a zoom. Yeah, means connect drag alt plus drag wearing or red color. White acting items. No, yeah, click this and look right. Okay, yeah so I don't see anything for scale and knowing how this works on the back end, that could be a bit difficult because it uses the the web can this element.

And so each one of these boxes is defined with an xy location and so to zoom it in and out you'd have to change all of those XY locations and resize the box. Now this computer it could do that, but writing the software if that software doesn't exist to do that writing that software would be a bit tricky.

Yeah. All right. Sure. How tricky, you know, I could I can easily figure out how to make the boxes bigger and smaller. But if there may be a way, if not, I'll just make sure all of our graphs get inside the box and then you can change the size of the web page by checking and control.

And using your most scroll which works pretty well. You're still only seeing that one part. Yeah. And then, if you scroll it too big, you know, now they're sort of in trouble, but yeah, you can't see the whole graph if the graphs bigger than the box, that's the problem.

Yeah. That's so I hope that has helped and I hope that this is an easy but interesting exercise. That was the intent anyways and something different that you don't normally do our course,

So any final words, before we wrap up the, the module ethical issues

Okay, I'll just go. I'll just go back down to wherever so there's going to be lots of rabbit holes in the next module. We're looking at ethical codes and you guys will be able to benefit by me. Having God down a rabbit hole and dug up dozens and dozens and dozens of ethical codes and having running them through an analysis.

And yeah, so that that'll be a bit of fun, I think. But we'll talk about that. Starting on Monday. All right, so I'm gonna wrap up. Only nine minutes late that's not too bad. So see you all next week and have a good weekend.