Training
In this episode we discuss training: what it can realistically achieve, why it often fails and how people actually become good at things.
The conversation begins with Aleph’s past experience delivering analytical training, and Nick’s frustration that training often strips away the excitement of discovery. The group explores whether people really learn best through formal instruction, or whether genuine understanding comes from practice, mistakes, motivation and real-world need.
Links:
Michael Polanyi's 'Tacit knowledge' - We know more than we can tell. https://en.wikipedia.org/wiki/Tacit_knowledge
Transcript
Hello and welcome to the Cognitive Engineering Podcast brought to you by Aleph Insights and produced by me, Fraser McGruer. I'm here with Peter Coghill and Nick Hare of Aleph Insights. On this podcast we look at a wide range of topics from an analytical viewpoint.
And today we're discussing training. Nick, Training, tell us.
Speaker B:Well, you may not be aware that Aleph Insights is a company used to deliver training used to analytical methods and things. Yeah, we used to.
We actually have a sort of training course which is designed to give people an insight into more the fundamentals of analysis rather than specific techniques as such.
Speaker A:But I sense a, however coming.
Speaker C:Well, Elizabeth, this isn't an advert.
Speaker B:No, it's not. Don't get in touch.
Speaker C:We don't want to do it.
Speaker B:Get in touch and ask about it because we don't do it anymore.
And one of the reasons is that I found it a bit dispiriting because I could never transfer to people or I felt like, you know, it was very difficult to get people to, to have the same sense of kind of discovery, I suppose, that I, that I had about the things we were trying to get people to understand. So things like uncertainty, people have a, have a, have an intuitive understanding of uncertainty that is not very coherent.
And we want people to think about what uncertainty really is, fundamentally, which is about information, and that's a better way of thinking about uncertainty, for example.
But when you deliver it in the form of a training course, even though, you know, you have interesting exercises and you have interesting discussions about it, there is almost this sort of lack of motivation to see why what we're trying to tell them is something really, really interesting is, is anyway, it. Comparing it to my own experiences of being trained to do stuff, I, I get it. I.
Because, you know, I, I don't think, I don't think I've really learned things from being trained to do stuff. I think I've learned nearly all the stuff that I can do well from trying to do it, you know, from trying to.
So, for example, all of the things I'm good at in terms of, you know, say, Python programming, which I'm not claiming to be an expert on, but I'm, you know, I'm much better than I was, say, five, six years ago. And a lot of that is just having, wanting to do stuff with it, you know, and then you come across something you can't do, you work out to do it.
You, you end up being able to solve those problems and, you know, and at the end you actually have A pretty good understanding of what you're doing. But then you take that understanding that you've distilled and you go, right, I've had to solve all of these problems myself.
I've, I'm going to shortcut that. I'm going to get this training course put together which cuts out all of that.
You're not going to have to spe, you know, five years making all of these mistakes because I'm going to tell you how to not make those mistakes in the first place. And lo and behold, it doesn't mean anything. People don't get it. They don't, they haven't made those mistakes.
They don't really see why what you're saying is important. And so yeah, that's my feeling is, well, maybe training is a waste of time or is it, you know, what things can you train people to do?
What kinds of skills are amenable to training versus you know, practicing, which I guess is a form of self training where you sort of think, well, I'll tell you what, I'm going to try and I'm going to try and do this to get better at it versus experience, which is where you're getting better at it because you're doing it, not because you're trying to get better at it, but merely because you are doing it to achieve some real world effect learning on a job. Yeah, that's it.
Speaker A:So, yeah, no, great question, quick question about the people that you would be training. Presumably these cohorts, would they be lay people or would they be people?
Speaker B:No, they're generally people who are doing analysis already.
Speaker A:This is really surprising then because one would have thought that they would have had the curiosity to want to analyze better. Right. If that's what they're already doing.
Speaker B:I mean, but I think there's a sense in which once you distill something down to a set of key learning points and you go, we want you to understand this and this and this. It feels like it's a thing on a page that someone from somewhere has, that just exists and, and it hasn't got the excitement of a genuine discovery.
And I suppose think about something like Newtonian mechanics, which is really amazing, like from a world of not really having any systematic way of thinking about physics, you know, the large scale behavior of physical objects, suddenly you have like this revolution.
Newton comes along, invents a few, kind of a small number of laws that suddenly explain everything, suddenly explains planets and birds and you know, balls falling off a table and, and you know, you learn it at School and it's like oh, it's a thing, it's written down, I've got to learn it. It's come from that process of discovery is missing because you just are shortcutting straight to the end point. And.
And I feel like that is partly the issue. It does it. I mean and you can be really interested in physics but it. You are still not seeing why.
That's an exciting, interesting thing to thing to try and understand.
Speaker C:I would suggest that that's partly a fault of the way it's delivered though. I think yeah, if you just said here's, here's the Newton's formula is just learn these. That's not very good.
I think there are ways of doing training that lead you to lead you on a shortcutted path journey to those. So I remember doing physics at physics A levels. I had a great, great teacher and it was mostly experimentation driven.
So we'd each do experiments using various different equipment and we would sort of end up deriving some fundamental or ourselves from our results. So we'd go through the motions of a well trodden experiment and it would lead us to a law of motion or a law of optics or.
So we'd sort of, we wouldn't just be given the end, we'd be given a journey to go on and that would lead us to the end and that would sort of concrete our understanding of how this law was found or an understanding of the law.
Speaker A:I mean. Can I posit something here?
Speaker C:So I think, yeah, I don't think it's. I think there's not necessarily. That's necessarily a problem of the shortcutting bit. I think it's how you do the shortcutting that's the.
That makes it good or not.
Speaker A:Okay. I mean there's something to suggest.
I mean so maybe drawing together what you're saying is it a mixture of like how stuff is delivered, the training and the quality of that and the content of that. But also the fact is it doesn't matter how well something is being delivered to me on I don't know, quantum mechanics or whatever.
I'm never gonna really get that great at it. Cause part of the reason is I don't really care about it sufficiently and I'm probably not mentally equipped to do anything with that or motivated.
It's never gonna work on me. Right. So you've gotta have an interest in that area. But it's gotta be even to be a good trainee. Then it goes a little bit deeper.
Cause like you were saying you were Training analysts or trying to train analysts, but even analysts didn't, you felt, didn't quite have that curiosity?
Speaker B:Well, it was the frustration that I could never communicate why this thing was a really good, interesting way of looking at it. And I think it's interesting you mention actually, so this, what is it that makes someone a good teacher?
Because I think we're touching on these two separate things here. One is being able to explain the concepts, but the other is making you interested in it.
And I think that that kind of motivation is absolutely necessary. Right. If you, you can't train someone to do something if they're not interested in knowing how to do that thing.
Speaker C:And I think, I think our experience is slightly soured though, because not disparaging our previous colleagues too much, but a lot of people just fundamentally weren't interested in what it was to be an analyst.
They didn't see their job, they didn't see their job as being partly produce some output, which is your analysis, but also partly improve yourself so your output gets better. They saw that second part as being very much not part of their job or not part of being an analyst.
They existed in a world where analysis is reading loads of stuff, having a think and then writing about it. And that was like, that was as sort of as concrete as a method as you could get.
Speaker A:And it's quite a finite world that.
Speaker B:Right.
Speaker A:It doesn't look beyond that to the framework that you're using.
Speaker C:Whereas actually our argument was always, well, yes, that is fundamentally the mechanics of what you do.
But the way you think about it, there are ways to improve that, to patch that with techniques, tools and sort of approaches that make your output better, more robust. But yeah, but I think fundamentally it comes down to a lot of the analysts that we get aren't.
There's no incentivate, there's very little incentive for doing it better and there's very little kind of checking of like, okay, so you're out this, you, you wrote this paper six months ago. Let's just check this predictive paper six months ago, let's check how much of what you predicted came true.
And therefore, you know, we'll pay you more or less or we'll give you some prize or we'll give you some sort of incentive for doing it more like that next time doesn't exist. None of this sort of incentives exist.
Speaker A:Well, quite.
Speaker B:But, but also, I don't think, I mean my, well, my subjective impression, and I think this is backed up by the, by the research, is that extrinsic Motivation isn't a patch really on intrinsic motivation anyway. And it's like, you know, this. This kind of.
I suppose another example would be something like playing the piano, something we try and get children to do. And you think, well, you know, the thing that distinguishes people who end up being good at music from people who don't is.
Is going to be partly about being enthused by a teacher. You know, you got a good teacher who kind of encourages you to want to do it.
But actually, I mean, in my case, what motivated me was seeing someone play the piano, actually an old lady at a wedding playing Scott Joplin on the piano. And I'd already spent a few years learning the piano, but. But it was always a thing I was supposed to do.
There was an ought component to it, you know, and. And I thought, hang on, I want to be able to do that. And suddenly I had this motivation to start to start learning again.
And I sort of think, well, that extrinsic motivation, the, you know, you. Your mum to learn the piano, or even you, well, you know, you might get a prize for doing it.
None of that really works compared to, you know, wanting to be good at something.
Speaker A:Quite.
Speaker B:And so, yeah, I don't know if that's true of all types of skill,.
Speaker A:But I think it's probably true because, I mean, we probably want to move on to other stuff. Right. But before we do, I mean, as you know, I do a little bit of teaching and I teach.
Speaker B:Aren't you a professor or something?
Speaker A:I'm not quite that. Yeah, pretty much. Yeah, yeah, yeah, exactly that. But also what's interesting is I'm a little bit older than the other people with whom I work.
And my view is essentially I'm, you know, I'm a filmmaker, photographer first, and then I teach as well. Right. Which.
And that's not to say that the teaching is not as important, but how I view teaching is very much related to the fact that I practice this stuff as well. Right. So what I'm not particularly interested in is grades. I'm not particularly interested in evaluations and these kinds of things.
I recognize there's a place for it, and it's a very systemized way of teaching we have at the moment. And I also recognise that the kids need to do that stuff.
But the thing for me that it always comes back to is you've got to enjoy doing this stuff and always try and work on projects that you enjoy, because ultimately what that means is the drive to improve just comes from within.
Speaker C:Yeah, No, I think that's a useful way of framing it is go where the energy is. That's go do the things that you're intrinsically motivated to do because it will.
Speaker A:Give you that competitive advantage. I feel.
Speaker C:Right.
Speaker B:Well, unless it's something that has no economic value at all, which is the unfortunate thing.
Speaker A:I mean, competitive advantage is actually not the right phrase really. What I mean, it gives you a bit of competitive edge.
Speaker B:Yes.
I mean, I think the thing is, you know, in a sense that you're born with a natural advantage if the things you want to do happen also to be commercially viable, you know, and I think so.
I was a kid, you know, in the:Some of those people went on to become, you know, tech guys and CTOs and, you know, and, and found computer game companies and all those kinds of things. And you think, well, you know, there wasn't training courses like there are now.
And I, I would suspect there's a much higher percentage of those self taught programmers from the 80s becoming successful computer engineers than are all the countless people going on now. We have computer science degrees everywhere and we have a massive throughput of people learning programming, but they're.
Because that's where the money is, not because they're driven by this desire to make computers do stuff.
Speaker A:Exactly.
Speaker B:And so I would think the percentage of people going on to become highly successful pregnancies is much lower now in terms of the. Probably the absolute number is higher, but. Yeah. So there we are.
Speaker A:I want to move us along, but just before we do, to your point there, I was recently offered some training at my work and I rejected it, which everyone was quite surprised at. And I was offered training on.
Speaker B:Was it equality and diversity training? Nah, I'm already good at that.
Speaker A:Got it.
Speaker B:No, it was after the fire.
Speaker A:It was on. Yeah, it was, it was, you know, they said, Fraser, after the incident, you have to do this.
No, it's on a bit of software called After Effects, which is essentially a special effects thing I'm going to call it. And I said, no, I don't want to do that.
And the reason why is I had training on that years ago and I had like, like a week's worth of training and then I forgot everything. And the reason why is one, I wasn't that I'm not that interested in visual effects and two, I didn't have the chance to practice it. Okay.
And so the knowledge Just ebbed.
Speaker B:But also, you know that when you are suddenly presented with the need to include a big explosion in a film you're making, you will teach yourself how.
Speaker A:To do it, possibly, or get someone else to do it. However, something that I do know reasonably well is another software called Premiere Pro, which is essentially video editing software.
And I know that pretty well. But how do I know it? Because 1. I mean, I essentially taught it myself, all those learning of YouTube videos, etc.
But also I use it day in, day out and have done for like 10 years or so or more. So it kind of. Again, going back to your point about, you know, and. But I'm motivated to, because I really enjoy it as well.
Speaker B:Or.
Speaker A:Anyway, where are we going with this? What do you want to move on to?
Speaker C:I mean, there's lots of interesting ways you could take it.
I think it might be worth thinking briefly about what training is versus other things which are to do with knowledge development, like education and coaching. I was thinking you're picking your example up, Fraser, of these tools which you've been offered training in.
Given your level of experience, it sounds like training course is not the right sort of thing. What you need is coaching. So you're the package you're familiar with.
You don't want to go on an introductory course or even an intermediate course on it. I think what you want is somebody who can lead you through didactically to. To improve. To take you to the next level on what you do with that tool.
I either sort of giving you things to do and tips on how to improve your workflow to make you more efficient, or using more of the features and everything else in that tool to make your product better. So it's like it's not a sort of. It's not a predefined, static training course.
I think what it needs is somebody to sit on your shoulder every now and again and say, what are you working on? How do you go about doing that? Have you thought about this?
Why you could try doing it this way and give you a bit more kind of like bespoke experience transfer rather than a sort of predetermined set of outcomes?
Speaker B:Yeah, because I, and I think going back to what we were.
What we were talking about earlier, and the thing which is really undervalued or at least not considered that we always think of teaching and training as a transfer of skills.
But actually what I think we all agreed on and what the evidence suggests is that is also it's equally important to transfer motivation, that actually what you really need to do is make someone want to know this stuff like that. Without that they're just gonna ignore it or forget it.
And so yeah, I feel to me like the interesting thing about coaching is that you can take the motivation for granted. Like actually the job of a coach is to get someone who already wants to be better and help them get better.
And I sort of think actually that is a totally different kind of intervention really.
And I'm thinking like with my, I've got a piano teacher for example, and she is, it doesn't have to make me excited about playing the piano because I might, I want to know what specific things I need to do to be better because I'm at the point where I can't identify that myself. And, and that is actually a very different.
Like when she's, she also does children and you know, when she's teaching a 8 year old it's going to be a totally different thing because actually you're trying to encourage them. You should make them excited about doing it.
Probably much less focus on technique and trying to get them to be perfect at stuff and more about getting them to enjoy it.
Speaker A:Absolutely. And like there's that sort of first level, right? Prior to all of this is well, what we're doing this for, what the aims, right.
Why do we want to do this? Right. The risk of saying the obvious, right? And that needs to be established first.
Again, a bit like in the past when I first started making films for clients and they say we want to make a film, I go, brilliant, let's go make a film. Now if someone says that we want to make a film, the first question, why do you want to make a film?
Because maybe the film's not, you know, maybe you should be doing something else instead, right? So it's that, well, you know, why are we doing this? Look, but still, where are we going with this? What do we want? Where have we got to.
So far we've talked about training, what direction next?
Speaker C:Well, I naturally being me, I want to take it in a sort of systems engineering direction and say, well, what is training doing and viewing training as a, as a technology, you know, what is it doing?
And therefore once we define what it's, what it's actually doing, try to see if there are ways, if there are factors in any given training that you can tweak to maximize its efficiency for any given topic or between any two different people or between the circumstance where that, where that student is in their, in their career, all these sort of, there must be factors which Govern how good training is or how training needs to be designed for that, for that given situation. Fine, yeah.
So viewing training as a technology, I think what ultimately training is doing, I'm thinking I'm talking pure training now, not education, not coaching. Think training course where your employer books you on a training course for two weeks to go and learn how to use CAD software or something.
Its job is to try to bootstrap your understanding of that thing, the topic, to a point. Where is it, which is higher than where you are now. So if you're.
If it's an introductory training course, it's the first time you've possibly ever seen the thing, touched the thing, and you're starting to actually, you know, through that training course, hopefully there's practical elements to it.
If it's a practical thing, it's there to get you starting to go through some of the basic mechanics of using it, so you're at least at a level that you can then develop further and, or maybe even start using it productively to do a certain task.
So what you're doing there is you're transferring knowledge, or taking knowledge out of somebody's head who's experienced using it and putting it into somebody who's inexperienced and therefore getting them up to a. Bumping their, their competence up in, in the topic. I think that's a sort of a reasonably robust definition of what training is.
Now, the knowledge we're talking about sort of famously comes in two very broad kinds.
You have explicit knowledge, which is stuff that can, can and is easy enough to write down, and that will be what the learning outcomes of the training are. But then you've got this bigger, more amorphous, very difficult to define set of knowledge, which is the tacit knowledge.
This is the stuff that an experienced person has, but often lacks any kind of tidy, introspective access to, to it.
Speaker B:And they may not even know that they're doing it.
Speaker C:They may not even know they're doing it. So this is like, then this, this is stuff that you can't put in a training course, but it's actually where most of the value comes from.
But the training course acts as a sort of guardrail, handrail thing to help you sort of a tree, like the stalks and branches of a tree that you can then start to walk down and put those tacit knowledge leaves onto to develop your understanding. So in a software package kind of thing, it's like, this is how you save a file, this is how you start a project.
This is what a project is in this context, there's all this kind of scaffolding stuff that might be abstract and esoteric when you first encounter it, but when you start to practice it, you can then say, ah, okay, now I understand. What I can now sort of start painting my own experience onto my scaffold.
Speaker B:Yeah, it's almost like.
I feel like where the value of those things comes in is, is that they're telling you what you can do with something and not actually necessarily how best to do it. You know, that actually how best to do it is much harder to. To give someone without them needing to achieve an effect, you know, without.
Without them having a project that they want to produce. It's very difficult to show someone the best way of doing something because they don't know why you'd be doing it in the first place.
But what they would like to know is, well, there's this menu here, and this is where you find the, you know, the effects menu. And this is. That's what a layer is. And if you want to add a layer, you can do that over there. And it's like, okay, right.
I've got a vague idea now of where these bits are, but I'm not at the moment, you know, they're not trying to teach me what to do or why.
Speaker A:Well, also, I don't know if this is what you're getting into, but again, I can't help comparing when I sort of teaching what I do, which is that a lot of stuff. I just see my thing is a sort of a simplification process through my own experience. Right.
Because I pretty much say to all my students, especially when it's with adult students, look, everything I'm about to tell you, you can see this stuff online, you can find it. Right? Okay, there's no point. And if I'm talking about, I don't know, composition, go have a look, you'll find stuff.
Speaker B:Right.
Speaker A:There's no point me telling you about that. But what I can tell you about is, from my experience, what I think about this. And I may be right or wrong, but this is what works now.
It's kind of a transfer.
Speaker C:I think that's a key part of the transfer, that a good training course. So they've got like purely vocational training.
Like if you imagine a factory worker whose job it is to turn a handle on a machine, the training is really just turn this handle, push that button. Did it work? Yes. Repeat. It's kind of like very. It's very, very mechanical. The training.
A key output of the training is that what you're, what sounds like you're trying to do there is transfer some of your. The mortacity bits. And I would class that as taste. So this is like, I, you know, I.
Speaker A:This.
Speaker C:I will show you how I use this software to achieve this effect, which I think has value because of X, Y and Z. Now you have your own tast, which you will develop. But I think that what.
This is kind of you saying what good looks like and what bad looks like, because you. But that's, but that's sort of from. That's your taste, and you're trying to transfer that to. To. To somebody else.
And I think that's particularly important. No, don't go there. So what, what, what, what. What the training is trying to do there is.
It's trying to compress knowledge that you've got and transfer it to other people. Problem with compression is that it always risks loss. You know, you can't transfer that.
You can't take that whole thing out and patch somebody's brain. You're trying to compress it through the, through the materials and through the dialogue.
Speaker A:Exactly. But the other thing, just to say, I don't know if weakness is the word, but as you say, you know, the word taste implies something quite personal.
And so I do always caveat it, say, look, this is, for me, this is how I do it. There might be better ways to it. There probably are, but this is how I do it. Where are we going?
Because I think we want to talk about cultural stuff, but I don't know if we. I don't think. Don't. If we've got time to sort of talk about other kinds of learning. Nick, help me out.
Speaker B:Well, I don't know. I mean, we, you know, we've. I think we kind of. I know it's all felt a bit abstract really, but I mean, I think, you know, we.
We've kind of been focusing on skills on like, how to do something, which.
Speaker A:Training, the word training implies, does suggest.
Speaker B:Whereas I think education is more about knowing that. It's like knowing about things that are true.
You're trying to get people to know things, which is not necessarily the same as doing stuff, you know, but obviously there is a big link between the two that there's. It's quite difficult to, you know, to draw a sharp line between those two.
But no, I just thinking, I mean, in terms of, like, obviously the fact that we're living in a world made of podcasts and cameras and light switches and not still trying to work out how to Start a fire in a cave shows that we've very successfully passed on, collectively, a lot of skill in terms of being able to do stuff. Like, you can now start a fire using one of those.
Speaker C:Not in here.
Speaker B:No. Not a good idea. But, you know, using one of 10 different technologies you probably got at home, you could get a fire going within about 10 minutes.
You know, it's. It's. And yet you probably are much, much less skilled at starting fires than a caveman would have been. Right. So.
And not only that, but you can do all sorts of other things. You can put heating on and heat your house, or if it's more the light you're after, you can turn on a light switch.
And we can do all of these things vastly more effectively than our, you know, ancestral forebears. Right. Despite the fact that we haven't really inherited the skills.
We, you and me, the kind of end users, have not inherited any of the skills of being able to do any of that. We've somehow been able to pass on collective skills, the collective how to do stuff without actually needing to train people.
Like, we don't start by saying, okay, welcome to your new home, and before I show you how to turn the boiler on, let's. Let's work out to start a fire with a pair of sticks. We don't bother with that anymore.
Speaker C:So what you're saying is we've got. The outcomes are much more accessible than they were to the caveman without the Necessarily pass. Without the necessary.
Speaker A:Well, yes.
Speaker B:That we've somehow. We passed on the capability.
Speaker C:Yeah.
Speaker B:But we've done it through technology rather than through training. Right. And I guess, you know, we could start thinking about, well, what actually, what does that look like?
How can we embody something, a set of skills or knowledge in a technology in such a way that I can give that to you without needing to bother with training.
Speaker C:Yeah. So I like.
Speaker B:I like.
Speaker C:I like the lens that technology is a sort of embodiment of a load of cleverness.
Speaker B:Yeah.
Speaker C:Instead of having to hold that all in your head, you can have this widget that does a thing. And loads of clever people have worked out how to build a widget. You just need to push the button on the widget.
Speaker B:Yeah.
And it's almost like the technology to go right back to the sort of first thing we were talking about of like, I want to make someone else have that aha moment that I had when I discovered it, you know, and. But actually, if I can get them to replicate the effect without even needing to understand it, you know, do. Do I need to worry about that? And.
And it sort of. I don't know. Napier and the logarithms, right? Logarithms are flipping brilliant. You get taught them at school. They seem very boring.
You don't know why you're learning them.
But actually, you know, as a computational tool, to be able to do, you know, and especially because it was analog, it was little, actually, literally numbers on sticks that you kind of lined up next to each other. Logarithms are absolutely amazing. And. And part of me thinks I really want to get people to understand why they're so cool.
You play with the slide rule, you're like, why are the numbers like that?
Speaker C:Yeah, you can shift numbers into a different domain, which makes them easier to work with. Yes, yes.
Speaker B:But.
But actually, you know, if I can just give you a calculator and you can calculate stuff, then actually we've achieved all of that without me even needing to teach you any of that stuff. So, you know, that's. Is that better? Well, I mean, it's certainly more effective.
Speaker A:But maybe what you're talking about there is. We're starting to drift into this murky area between training and education, Right?
Because it seems to me that training part, potentially, maybe a definition we could have there, is the skills of how to do something, and there's a kind of a. You operating in the area of utility, maybe. Right.
And so when technology comes along that removes the need to do that, then, you know, the utility's gone. It's there already.
Speaker B:Right.
Speaker A:However, what you're saying is, when you're talking about logarithms, is that there comes a point where, okay, this stuff can be done with a calculator, whatever it might be. But, hey, guys, look at this.
Isn't this sort of wider, wonderful beauty of discovery and curiosity, which that feels to me like that's sort of getting into more broader principles of education. What do you reckon? Or am I talking rubbish?
Speaker B:Except for, you know, the Napier's bones, which were a way of embodying logarithms in effectively a way of doing. Being able to do calculation, calculation, complicated big calculations with effectively numbers on sticks. That's.
At one point in time, that was state of the art, that was the calculators. You know, you would have to learn how to use them because it was how to do something. You know, it's a way of doing calculations.
Whereas now you would be learning them as almost.
Well, not quite historical interest, because learning, Learning how they work gives you an insight into what are but you don't actually need to learn them, understand them at all. You don't need to have heard of them.
Speaker C:And it's sort of like it's a learning. They're taught as a learning aid.
By using the physical item, you're like, I can now see how this maps onto this more abstract concept and you can then start to think, oh, what other abstract things can this, this approach. So you've learned a meta skill there of like how does, what else does. What else can logs do that beyond this example that we're working on?
So yeah, there's. Yeah, this is the blended line between training and education, I think.
Speaker A:Are we about done or is there something you want to go into?
Speaker B:Well, the question is, are we actually. I mean, does technology effectively end up making us all completely stupid? Is that you know, where we're going?
I mean, part one, you could argue that the reason we bother with education is ultimately some sort of how to. It's ultimately because we think that this is going to make you more effective.
You know, knowing things, knowing facts about the world, knowing a bit of history might make you more effective, you know, being, being a modern day present.
Speaker C:I think there's value to education beyond being a functioning, profitable member of society. Well, I'm just saying it's arguably there's value in, in just enriching people for the sake of doing that. That.
Speaker B:Well, yeah, okay.
Speaker C:I mean, it's nice to know stuff.
Speaker B:Obviously, I think that. But actually, as it happens, that has been really one of the only ways we've been able to get people to be able to do other stuff.
Like, you know, write an essay, let's say, you know, which, which is good training for writing about things in general. And a lot of jobs involve writing about stuff.
So, you know, we teach you to read some history and then we teach you how to write an essay and you get the hang of that and then you become more functional and if you enjoy history, well, so much the better.
What I guess we have gradually developed technology to do to require less and less understanding of us so that we can just achieve the effect at the push of a button.
And I suppose the thought would be, well, you know, if we have LLMs that are now doing that, you know, we're kind of doing the, doing the writing the essay for you. Can we completely give up in the same way that GPS means that nobody has a sense of, of direction anymore and in a sense you don't need one.
No one learns the, the actual physical map.
You just look at the representation on your screen and you follow the instructions and it says go forward and you go forward and you don't really think about where you are. Is this the end result that we pass down immense capability, but we do it with requiring less and less understanding. Is that the.
Speaker C:Well, I mean I think not in the current paradigm.
So I think until we get the AGIs, whether or not you think that's inevitable or likely and very soon at the moment you need human taste to steer the LLM to steer your AI in the right direction so you can automate things.
So for example, experimentation and the LLM can do much of the work, but to steer the LLM into what to experiment on is the key thing that the human experience gives. So LLMs are naturally at the moment very good comparative to other things. At software engineering they can given a given a spec for a functional.
For even larger parts of software they can write a pretty good job, do a pretty good job of that, of something that will work. And the key thing though that they can't do is work out what to do in the first place, what good.
Speaker B:Looks like, what the answer is to how should this behave? What, what color should the buttons be? All of those sorts of. Yeah, yeah.
Speaker C:So they, so they could keep getting better as they are, but they, the way that they work, they won't necessarily be able to decide what is a good idea to start with.
Speaker A:Yeah, no, quite, yeah.
Speaker B:I mean are we, but I mean are we all.
We're gradually, let's say we gradually technology moves us further and further away from the practicalities, the real world things that you have to touch and understand in the processes because things become more and more like pulling a lever or pressing a button and, and the actual mechanics of it. We have taken care of all of that. Right.
And I just wonder if, you know, the future is inevitably that we will be these sorts of button pressing Eloy served by the mechanical morlocks, you know, and, and you know what happens then when it all goes wrong? You know, actually we just sit about doing nothing.
You know that we've transferred all of the knowledge of humanity and as a result nobody has to know how to do anything or learn anything.
Speaker A:So yeah, nicely done.
Speaker B:The beautiful end point of all of this.
Speaker A:Well, nicely done because that's a beautiful segue into a question I want to ask to finish this off on. Right. So the question is this could go, you could either go backwards or forwards with this.
So what I was going to say is what's a skill that you would like to Know, talking about this sort of departure from, you know, the effects of technology, it sort of moves us away from certain skills. Okay. Is there a skill that you would like to learn that is no longer available to us? So that being the case, Nick, go for it. Have you got anything?
Speaker B:Yeah. Have you heard of dancing?
Speaker A:Yes.
Speaker B:I'm just, I've. I've got, I've got such a poor quality sense of. Is it proprioception?
Like where my body is at any given time and you know, which I suppose you might think, obviously I'm fingers, no problem. I could play the piano quite well. I'm alright. We're playing a piano, my fingers know what they're doing.
Speaker A:Right.
Speaker B:But the whole, my whole entire body is just slumps, usually awkwardly into some position. And I think I would like to learn to dance properly now.
Speaker A:Is this a loss? Hold on. But this is a skill that you would like that you don't have.
Speaker B:Yeah. Not saying it's a lost art.
Speaker A:Okay.
Speaker B:But I, you know, but at the same time I sort of think, well, if I had grown up in, you know, a tribe in 10,000 BC, I would probably be dancing a lot. He's doing a lot more dancing.
Speaker A:Less piano playing.
Speaker B:Yeah, less of that. But maybe a bit of drumming. Maybe playing a bone flute.
Speaker A:Yeah.
Speaker B:But probably much more dancing. And so I feel, you know, I feel like society has removed my. The necessity to dance.
Speaker A:Will you say that? But going back to this intrinsic versus extrinsic, maybe it's better off that. Because maybe your value in the tribe.
Speaker C:Or maybe society is better off because Nick's not dancing.
Speaker B:Yeah, undoubtedly true.
Speaker A:Because if you were meant to be a great dancer, you'd already be.
Speaker B:I'd already be the Nijinsky of North London.
Speaker A:Yeah. And. But you know, on this sort of little flight of fancy.
Speaker C:I don't know, I think, you know, if you, if we were in a tribe, I think people would specialize and there'd still be people who like dancing being a bigger cultural thing still doesn't make them want to do it anymore. I think people were specialized in a big people who are sort of naturally good at dancing. They like doing it.
Speaker B:And then there's the bone flute.
Speaker C:And then there's a bone flute. Yeah. I think you'd be a bone flute drummer guy and you'd be like. People would try and get you to dance. You'd be like, oh no, I hate, I hate dancing.
But I'm quite good at the flute. Let me.
Speaker B:Yeah, it could be.
Speaker A:It makes me wonder like on A little sort of side sort of note is, you know, it'd be interesting. I'd like to know if there's studies out there between, let's call it for a moment, musicality, for example, versus measured against dancing ability.
Because one would expect that people who are good at music. I've seen you play the piano. I think you're fantastic. I mean, you'd expect there to be a sort of a positive link, but, I mean, it appears not. Right.
But yeah, I'd like to know about that.
Speaker B:Fraser.
Speaker C:What.
Speaker B:What skill?
Speaker A:I'm not sure yet.
Speaker B:Okay, we'll show it.
Speaker A:Peter, Peter, Peter.
Speaker B:Yeah.
Speaker C:There's a very, very specific skill I would like to.
Speaker B:Is it wheelies on a bmx?
Speaker C:No, it's not wheelies, it's bunny hops.
Speaker B:On a mountain bike.
Speaker A:It's not far off, right? I mean, basically, yes is the answer.
Speaker C:It's basically a bunny hop, more or less.
Speaker A:Start.
Speaker C:No, it's not a wheelie per se. It's what's actually called a manual. A wheelie is when your bum is on the seat. A manual is when you're off the seat. I can. I can manual.
I cannot bunny hop. So bunny hopping. So feeling his very specific skill. So.
But it's kind of like it's one of those fundamental things in mountain biking that lets unlocks a load of other things like. Like jumping over things safely, getting down big drops and all sorts of, you know, things that you want to do.
So, I mean, I'm pretty fit at the moment, reasonably athletic on the bike. That's no longer my blocker. My blocker is now being some of the technical skills, like bunny hops.
Speaker B:Well, Peter, I've actually asked a friend of mine who's a very talented choreographer to give me some dancing lessons.
Speaker C:Okay. Yeah. So can they give me mountain bike lessons?
Speaker B:No, but what I'm saying is if I'm going to learn to dance. If you learn to bunny hop.
Speaker C:Yes. Yeah. But luckily this skill acquisition is within my reach. There are lots of places I can go to get specific coaching on mountain biking.
Speaker A:Yeah. I mean, this feels like achievable. It's not a dream.
Speaker C:Yeah.
Speaker B:You know, Fraser, what ludicrous, unachievable skill dream do you have?
Speaker A:Yeah.
So in the spirit of the way you've answered the question, I'm going to answer in the same way, which is you actually didn't answer the question that I wanted.
Speaker B:Right. You're going to say you never do that. I know.
Speaker A:Which, to be fair, I mean, you're due yeah, because what I wanted to know is what's been lost, what sort of skills been lost, which you would like. So I'm gonna ignore that and ask a different question which I was thinking of, which is what skill has been imparted to you or you impart to others.
That's kind of a favourite one. I'm gonna answer that question and my skill is this and it's one that I impart to others.
Speaker B:Do you. Okay. You like to think and it's to.
Speaker A:Do with photography in the sage guru.
Speaker C:Kind of role that you have.
Speaker A:Exactly.
Speaker B:Go on, hit us with your big lesson then.
Speaker C:Take the lens.
Speaker B:You've got a massive audience now so it's time to transfer that as widely as possible.
Speaker A:Yeah. So if I'm able to up photography and what we should be looking to achieve, what we're looking for to make a good photograph or great photograph.
What we mean by that is really simple, I think. Any thoughts, any ideas?
Speaker B:Yeah. Basically a black and white photo of an old man close up. That's. That's how. That's award winning photos. He's got to be really wrinkly.
Speaker A:Yeah.
Speaker B:Right.
Speaker A:It's really, really simple, I think. Light, it's just light.
Speaker B:Right.
Speaker C:Okay, so fairly fundamental.
Speaker A:Well, it's in the name, right. I mean it's light. You've got to be looking at what the light's doing.
Speaker B:Well, have you seen that film the Fablemans which is about Steven Spielberg? It's semi automatic.
Speaker A:I know of it, but I've not watched.
Speaker B:Well, there's this scene where he finally joins a studio and he goes next door and it's John Ford played by David lynch. And he says basically if the horizons. If the horizon's at the top, that's good. If the horizon's at the bottom, that's good.
If the horizon's in the middle, it's fucking boring. And it's like. That's actually this actually pretty good advice.
Speaker C:One third, two thirds.
Speaker A:Yeah, that's a whole frame type thing going on. But yeah, light. Always be looking for the light, I think.
Speaker C:I mean I get it, but what I. But I don't get it at the same time it's like what does that mean though, beyond turn the lights on or make sure there is light?
That doesn't transfer to me any sense.
Speaker B:He wants to sound mysterious and good and he doesn't always.
Speaker A:Yeah, I mean, yeah, I can answer that but I'm not gonna answer it now. But yeah, I'll send you a link, some of my lessons there you go. All right, we'll stop there. Yep, I enjoyed that. I thought that was quite fun.
So you've been listening to the Cognitive Engineering Podcast brought to you by Aleph Insights and produced by me, Fraser McGruer. If you haven't already, please like and subscribe. We try to release an episode every week or two.
Until next time, goodbye.
