Episode 415

full
Published on:

11th Nov 2026

It Just Doesn't Work

Why can you buy a toaster that can't toast a whole slice of bread? A listener’s dismay sets Fraser, Nick and Peter off on an investigation into everyday products that fail at their basic job. From dripping teapots to unpeelable bacon packaging, why do familiar design problems persist?

We distinguish between things that wear out, things that are badly made and things that don’t work even when they work. Along the way, we discuss planned obsolescence and the Phoebus light bulb cartel, survivorship bias and why five-star reviews can tell us so little about long-term reliability.

Peter introduces the “doorbell paradox”: we want more sophisticated products, but adding features also creates more opportunities for failure. And Nick offers a surprisingly optimistic conclusion: perhaps good design has already solved more of our everyday frustrations than we notice?

Links:

Tas's offending toaster and the partially-toasted bread it produces.

Appliance brand reliability survey from Which?.

The 'Bathtub Curve' of failure rates.

A Redditor achieves the impossible and opens a packet of bacon using the pull-tab.

Lifetime of electrical appliances becoming shorter and shorter (2016)

The Phoebus Cartel

The Teapot Effect



This podcast uses the following third-party services for analysis:

Podtrac - https://analytics.podtrac.com/privacy-policy-gdrp
Transcript
Speaker A:

Hello and welcome to the Cognitive Engineering Podcast brought to you by Aleph Insights and produced by me, Fraser McGruer.

I'm here with Nick Hare and Peter Coghill of Aleph Insight.

On this podcast, we look at a wide range of topics from an analytical viewpoint.

And today we're discussing stuff that doesn't work.

Nick, go for it.

Speaker B:

Well, this was a suggestion sent in by loyal listener Task, who's got a sad tale about a toaster?

Speaker A:

Oh, dear.

Speaker B:

Yeah, Tas bought a toaster in:

Speaker A:

Nice.

Speaker B:

And it was the best toaster he'd ever had in his life and it just recently died.

Speaker C:

Oh, yeah.

Speaker B:

So he had to get a new one and he was very disappointed.

Speaker A:

Oh, dear.

What happened?

Speaker B:

Well, it's.

You know how toasters are supposed to toast all your bread?

Speaker A:

Yeah, yeah.

Speaker B:

Apparently this one only does the bottom two thirds of the slice of bread.

Speaker A:

Okay.

Speaker B:

I mean, I think that's not 2/3 of a toaster.

I think that's 0 of a toaster, personally.

So I'm with him on the disappointment, but his question is, why can you buy things that don't work?

And he's got some other things listed.

So vacuum cleaners with what he describes as the suction power of an asthmatic pensioner, kettles that drip from the spout when you pour them, washing machines that leave slimy residue inside, tumble dryers that stop tumbling before the clothes are dry, and TVs with 23 million functions on the remote, but it has to be 2ft away from the telly to work.

And oddly, this came in the same month that our colleague Chris complained about his new doorbell.

Apparently, he goes through doorbells at an incredible rate because they never work.

And I'm not aware of this because we've got a doorbell which just was there when we moved in and has never required dealing with.

Speaker A:

Yeah, I mean, Chris has got quite powerful fingers, but keep going.

Speaker B:

I mean, that's the thing.

It could be.

Speaker C:

Why is he ringing his own doorbell, then?

Speaker A:

Yeah, yeah.

What's going on with Chris's life, though, anyway?

Speaker B:

It does tell a story of its own, doesn't it?

I have an issue with my DVD player because I had to buy a new one.

Again, bought one in:

Conked out last year, had to buy a new one.

It's terrible for half of my DVDs, right.

The older ones, it plays the video in a tiny square in the middle of the screen because it can't apparently understand old DVDs.

Not just that, but when you turn it on, it it says, choose an app.

And one of the apps is DVD player.

And I'm like, that's literally your job.

I shouldn't have to tell you to be a DVD player when I turn you on.

It's so annoying.

So, I mean, basically, yeah.

Why are there things available that don't actually work?

Speaker A:

Okay.

Absolutely.

Right.

And just to think, just to set out this stool a little bit further, I guess, things we're gonna be looking at if we do this like we usually do.

Speaker B:

Right.

Speaker A:

We're going to be looking at.

Well, what does it actually mean, I'm guessing, for stuff to not work.

We're going to sort of nail down some definitions.

Speaker B:

Yeah.

Speaker A:

And then we're going to sort of work out, well, is this really the case?

Speaker B:

Right.

Yeah.

Is it a perception or is it reality?

Are things really that bad?

Speaker A:

And depending on that bit of analysis from there, does that feel about right?

Okay.

All right, cool.

All right, so how are we going about.

Speaker B:

Well, I just want to propose, first of all, just a hierarchy of types of not working.

Speaker A:

Okay.

Speaker B:

Right.

So we.

We have type 1 failures, which are where the design and the construction of the thing are good.

It's just bad luck.

Things go wrong eventually.

Like Tas's old toaster, my DVD player, they did a good job if they lasted, you know, 15, 20, 25 years, that's good.

So, you know, that's.

There's nothing wrong.

It's just failure that is going to happen eventually anyway.

Right.

So.

So that's.

That's kind of like good failure.

Yeah, failure where we can't blame the construction or design.

Then we've got type 2 failures where the design is fine, but it's the construction or the components are bad in some way, so it's just been badly made.

So it fails because of the quality of the construction.

Then we've got type three.

And this is where I think we're starting to get into what we want to talk about, which is where the design is.

The thing actually does perform as designed, but the design is actually.

The design is such that the failure will actually happen.

It's like the failure is designed in because of the nature of the way it's actually designed.

So, I mean, things like push taps that don't stay on, Right.

Where you actually.

That's intrinsic to the design of that thing is that eventually that's going to get worn down and stop sticking.

Right.

So that's.

It's kind of like that's a failure.

Which is intrinsic to the design.

And then I think really the key thing, this is the main category of thing that we probably want to talk about, which, which is the type 4 not working, which is where the design is fundamentally broken.

So even when it works, it doesn't work.

It just does not perform as intended.

And that's where I think a lot of these things fall.

Like a toaster, which is only two thirds of the height of an average slice of bread, is not designed to be a toaster.

It just, it can never be a toaster.

Like what are you toasting?

Tiny bread?

Speaker A:

Well, yeah, I was gonna say it.

Speaker B:

Could be, you know, what's the.

Speaker A:

Yeah, and actually I misunderstood Taz's issue with that toaster.

Yeah, it could be that Taz needs to rethink his bread strategy.

Or it could be.

And actually I misunderstood.

I just thought that it kind of worked.

But only the filaments at the bottom.

I know this is not how filaments work.

Speaker B:

Well, yeah, I could be part of it.

I mean, we'll get on to more detail about toasters.

Speaker C:

The toaster needs a post mortem, I think.

Speaker B:

I mean toasters, toasters require a whole, I think probably a whole podcast of their own.

But, but, but I mean there's all kinds of things people complain about with toasters in general, you know, but like in this specific case, if the toaster cannot accept if the top of the bread is sticking out the top and it's a normal size of bread, it just doesn't deserve to be called a toaster.

Yeah, yeah, absolutely not worthy of the name.

Speaker A:

No, it's not.

It's a disgrace.

Speaker B:

So, yeah, basically type three kind of the failure is like designed in and type 4 doesn't work even when it works.

And so just a few examples of these sorts of failures before you do.

Speaker A:

I can't help but wondering which kind of failure I am, whether I'm a three or a four.

Speaker C:

But you were set up to fail.

Speaker A:

Yeah, yeah, no, yeah, I'll talk it all through later with my therapist.

Speaker B:

Exactly.

Speaker A:

Anyway, yeah, go on.

Speaker B:

So these are the common, some very common things that people made about.

We've got hand soap dispensers where the tube isn't long enough to get to the bottom of the, of the thing.

So there's always soap left over.

Probably my biggest daily bugbear is those pull off plastic lids, you know of like bacon and ham where they would never the tear ever the little thing either you can't, it's too strong to pull up or the Tab just rips off.

Speaker C:

Yeah.

The resealable ones that just tear halfway down.

Speaker B:

Oh, yeah.

I mean, I hadn't thought about resealable, but the resealable things, which are actually just a sticker and the sticker doesn't stick.

Jugs whose spout doesn't pour properly.

I mean, why do they exist at all?

Hand dries in public toilets that don't do anything.

Saucepans whose handles get too hot to touch.

Cling film cutters.

Speaker A:

Oh, my gosh.

Speaker B:

Yeah.

They're completely useless.

Speaker A:

It fills me with rage.

So.

Speaker B:

Exactly, I think.

And.

Speaker C:

And so you need a Wrap Master.

They are great.

Speaker B:

What is that?

Like a thing you put your cling film into?

Speaker C:

Yeah, you put your cling film onto.

It has a lid with a blade that's hidden in a sheath and you pull the.

You open the lid and you pull the cling film out, put the cling film everywhere and then you just click.

You just clamp the lid shut and it slices it nicely.

Speaker B:

Perfectly incredible.

Speaker A:

It means not for the first time,.

Speaker C:

There's a free advert for Rap Master.

Speaker A:

I can't help but observe that Peter's got this whole life thing.

Speaker B:

Why aren't we getting sponsorship from Rapmaster?

Speaker C:

Maybe.

Maybe they'll reach out now.

Speaker A:

Yeah.

Speaker B:

So I think we can all say that this is something we're quite familiar with.

And actually, if anything, I want to rouse people out of their slumber.

Why are we putting up with this?

Speaker A:

Yeah.

Speaker B:

I mean, like, Tas, I.

You know, it is enraging, actually, that you can buy things which purport to do something and they don't.

They couldn't do because of the way they're designed.

Yeah.

And I think, you know, the fact that when I open bacon, I don't even try.

I just reach straight for the knife and open it.

That way I don't even bother with that pull tab.

The whole thing is a charade and it sickens me.

Speaker A:

And it's not your first rodeo.

Speaker B:

You've been around a bit.

Speaker C:

You've been disappointed enough by those tabs.

Speaker A:

It is.

Speaker B:

I'm like one of those.

You know, there's people who have a couple of bad relationships and then they just.

That's it.

They just determined they're not going to let them do that again.

Yeah.

And that's how I feel with those things been burnt.

Yeah.

So that's it, really.

I mean, I guess if only we had a professional engineer here to help us understand what's going on.

Speaker A:

Yeah.

And indeed.

So I don't know if you're ready to jump right in, Peter.

Speaker C:

Yeah, but.

Speaker A:

Yeah, with some data.

But before I do, I guess one of the things that when you were talking through type three, type four failure is to the extent that we're talking about conspiracy or not or just incompetence.

But are we ready for that yet or.

I don't know, I suspect we might come to that.

But anyway, go for it, Peter.

Speaker C:

Well, I've got a few stats on the they don't make them like they used to kind of question.

So the German environment agency, the UBA, did a quite a comprehensive study in the last 10 years or so.

I'm going exact date for it, but they found that appliances are dying due to defects, more so than they used to.

So within the last, within five years it rose from 3.5% of appliances.

I think these are sort of electrical appliances.

This is washing machines and toasters, etc.

% in:

Speaker B:

Is this like faulty out of the box?

Speaker C:

Yeah, no, this is breaking within warranty, I think.

Right.

% in:

So quite a significant jump.

Speaker B:

So three times as many of these things are breaking.

Speaker C:

Yes, that's quite.

Speaker A:

have had this conversation in:

But anyway, carry on.

Sorry.

And by the way, I like how we are aligned with the German consumer in our disgruntlement.

Speaker B:

You sort of imagine that Germans actually have very high standards.

Like in the UK we'd consider that as fine.

It's when it gets up to a 50% failure rate we might start to worry.

But Germans are more exacting, aren't they?

Speaker A:

Yeah, yeah, quite.

Speaker C:

And about one third of those surveyed were unhappy with their appliances service life.

So there's a general sense that with all their appliances.

Yeah, with all their.

Well, they should, you know, there's a general sense that appliances should last longer than they do.

Counterpoint though, the EU repair center data, I'm not quite sure what that is, but they put the average washing machine service life at 12.6 years.

Okay.

Speaker B:

Which is pretty good.

Speaker C:

I mean, thinking how much I use a washing machine, it's on at least every other day.

Speaker B:

Right.

Speaker C:

So it's getting hammered, you know, it's busy.

So that's.

That feels like quite good value for money.

Speaker A:

And there's some quite powerful forces at play with a, with a washing machine.

Speaker C:

Yeah, they're not trivial things.

They're big motors and lots of water, you know, electrics and water famous.

They aren't.

Speaker B:

I mean, if you.

If you ever see a video of a.

Of a naked washing machine, a washing machine with the outer casing, it's pretty terrifying.

It's like having a motorbike in your kitchen.

Speaker C:

You know, always remember to remove the shipping bolts because it gives you this.

It gives you an indication of how violent it really is.

For the shipping bolts.

Speaker B:

Yeah, but actually, don't do that.

Speaker A:

Always.

Speaker B:

What?

Root them.

Speaker A:

Did you say remove them?

Remove them?

Yeah.

Yeah.

No.

Speaker C:

But another interesting finding from these studies is that actually it's not all to do with crappy devices.

Quite a lot of the churn is to do with people wanting upgrades.

So upgrades, not sabotage.

So, yeah, so three quarters of those surveyed desired a better device.

The desire for a better device is pivotal, and it covers that for even things like mundane things like fridges which just sit there and stay cold.

Speaker B:

Who upgrades their fridge?

Peter?

Speaker C:

Maybe they need more.

Well, they need more.

They need more egg cult holders for their morning eggs.

Speaker B:

I guess I just.

It's funny you saying that.

It's just made me realize that under no circumstances would I ever want to upgrade white goods.

I just sort of think I'm.

I'm never gonna want to do that.

I'll wait till they break, and then I'll buy the new one.

It's like the idea of saying, I'm gonna get rid of this perfectly good fridge because I want one which has.

Has got.

I don't know, AI in it or something.

It's just.

Well, yeah, I think.

Speaker C:

But I think there are people.

There are people like that, though.

Speaker B:

Yeah.

Speaker C:

You know, new smart.

A new smart fridge with new.

New features comes out and that.

So they get a fixated anyway.

Speaker A:

So the issue may be consumers moving the goalposts.

Speaker C:

Right.

Speaker A:

And increased expectations.

Speaker C:

Yeah, okay.

But.

But finally they.

I wouldn't write off the possibility of underhand dealing, though there's interesting historical precedent for it.

A very notable case was the Phoebus Cartel.

Phoebus Cartel, which won't go into too much detail.

It's very well covered by lots of people on YouTube, et cetera.

But it was a cartel formed in:

So at the time, there was Osram Phillips, Tungsten, aei, Tokyo Electric, General Electric.

They decided that because light bulbs at that point were getting pretty good and the manufacturing technology was getting really good, so a cheap bulb could be manufactured that would last a very, very long time.

And this was not making them enough money.

They don't want a Light bulb that's going to last 10 years.

They want a light bulb that's going to last a year, so you have to replace it all the time.

be manufactured to fail after:

The aim eventually got the doors blown off and they got in a lot of trouble.

Speaker B:

Yeah, I mean this is one of those sort of conspiracies where the ramifications have been.

So, you know, in terms of public trust in corporations, it's just one of those things where you think it's so incredibly short sighted, ironically for a light bulb consortium, you know that.

And it's because everyone always said, I remember when I was a kid, you know, I remember everyone, everyone says, oh yeah, light bulbs, you know, that they're designed to fit, fail, they want to sell more light bulbs in it.

And of course that means now everyone thinks that that's why things fail.

It's a default, a default explanation for these things.

Speaker A:

Well, sorry, well, that and of course,.

Speaker B:

No, but, but I mean the other things like Apple have been rumbled, haven't they, about deliberately slowing down their older phones and those kinds of things.

So this definitely does happen.

But, but I think so I had a quick look at which to Apple's.

Speaker C:

Credit there, I mean, I think there's not.

Their motive was partly good because they figured that their consumers care more about battery life, their phone lasting a whole day, than they do about the performance, sort of the absolute performance of it.

And that was the trade off.

They deliberately slowed the phones down to elongate the length of the battery as the battery capacity, as the battery aged, it loses capacity.

And so if they kept it running full tilt, it might last only half a day when the phone's five years old.

So they stretch it out to make sure it lasts a whole day.

Speaker A:

Okay, before you come in, Nick, so just summarising what you've said there, Pete.

So we've sort of delved into is this happening and why?

Speaker B:

Well, yeah, I mean the trouble is the data we've got on failures which are quite measurable.

So things like people returning things, warranty claims, those kinds of things.

I mean this is giving us some insight into the trends about potential changes in the quality of products and those kinds of things, but it's not quite giving us direct insight into this question of the things that work as intended, but what they're intended to do is.

Doesn't work.

Speaker C:

Yeah, I mean there's also a bit of survivor bias going on as well.

You know, Taz remembers Their toaster that lasted a long, long time because it felt exceptional and you'd expect, you know, things fail.

Things fail according to a distribution like,.

Speaker B:

And Tasha's lucked out with like this.

Speaker C:

Incredible lockdown with the lucky toaster.

Speaker B:

Five pound lucky toaster.

So nothing else.

By regression to the mean, every subsequent toaster is going to disappoint him.

Speaker C:

Other, all other toasters of that particular skew probably fails.

Speaker B:

Will never see toasters like these again.

Yeah, yeah, yeah.

So again I looked at similar stats to Peter, but so, I mean there's a lot of variation within brands I looked at which, you know, obviously investigate these things and for washer dryers, some brands, one third of them develop a fault within seven years and in some it's like 5%.

So like depending on which brand you get and guess what, the more expensive ones are less likely to fail.

So I think that might give us an insight into why cheap buy twice.

Well, I mean, absolutely.

Or potentially buy three times.

So there is that and I think we'll definitely get onto the issue of cost.

But yeah, so that's, I mean true on a lot of the categories of things they looked at is basically some brands, cheap brands fail more to a significant degree.

So yeah.

Something else which is probably just worth mentioning for interest really is what's called the bathtub curve of failure.

So when Peter mentions that his, the, you know, the average lifespan of one of these things is 12 years, that what you might imagine is that there's a sort of normal distribution where, you know, things tend to fail at the 12 year point.

But actually with, with these kinds of devices what you tend to find is that there's a lot of things that fail early on because they're defective and then there's things that then fail later on because they start to just get wear and tear.

So actually, you know, it's one of those things if you've got a toaster that's already lasted for five years, it's probably going to last for another five years.

The big uncertainty about a toaster is the first piece of toast that's going to give you lots of information about it.

I think so anyway.

Yeah, I just, I'd mention that it's quite an interesting fact about, about failure rates.

Speaker A:

Yeah, I mean this is a bit like why a first birthday of a child is celebrated so much.

Speaker B:

Because kids go wrong in the first year.

Yeah.

Speaker A:

And once you get past first year.

Speaker B:

Yeah, no, I mean that's absolutely right.

Speaker A:

Yeah, yeah.

Speaker B:

No, I mean humans definitely have the same, same bathtub curve.

Yeah.

Speaker A:

Okay.

Speaker B:

So I suppose the question is what you would think is that in this day and age where we all have access to tons of information online about the quality of the thing that we're about to buy, that this problem would be solved, that you wouldn't be able to buy a toaster that doesn't toast the toast because it's like, well, surely the reviews will do that job.

And Tas actually mentions that he bought the one which was like the biggest seller on Amazon with also good reviews.

Well, exactly.

How is it that we.

That that you're still able to somehow online sell a toaster when everyone can see and respond to and rate that toaster?

Speaker C:

Yeah.

Speaker B:

Peter, what's the.

Speaker C:

It would be very helpful, very informative if that.

If the review system also included how many returns were.

Because it might be the biggest.

It might be the biggest seller.

But is it the biggest returner?

Speaker A:

I mean, they do have that.

They don't.

I think it's relatively opaque.

But they do include that on Amazon.

Speaker B:

Right.

Speaker A:

This item is often returned, you know, that kind of thing.

But it doesn't give you.

Speaker C:

It's not grand, does it?

It tells.

It says 1 million units sold, 999,000 returned.

Would be quite informative, wouldn't it?

Speaker A:

Well, I mean, but it would save something like that.

This thing is often returned.

Speaker C:

But anyway, yeah, you'd think it'd be in Amazon's interest to be a bit more transparent about that.

Speaker B:

Well, what annoys me is that there is no way of what there should be is a sort of Bayesian adjusted rating.

You know, probably.

I mean you'd think it would be in their interest to do this.

But where you have at the moment, if you sort by rating, what comes top because of small sample effects is things where there's only five reviews.

Speaker A:

Quite.

Speaker B:

And they all happen to be five stars.

It's like I have to do the trade off of do I want a four star where there's 10,000 reviews or a four and a half star where there's five?

Speaker A:

It's always.

I think I personally always go for the larger number.

Speaker C:

There used to be websites that did that, but I think Amazon changed their rules about what information third parties can publish from Amazon anyway.

But yeah, I wanted to address the problem with reviews generally.

Speaker A:

Great.

Speaker C:

So I see a couple of problems.

So the simple summary 5 star review thing that you get on Amazon and other places, it's not very helpful as one reason we've described is that you don't if you've only got A few reviews, then you.

It doesn't give you much data.

But also those reviews usually filled out by people.

Well meaning they want to give a bit of a review.

But unless they've detected a failure or fault with it right away they're giving a positive reaction because they've just got this new flashy thing with new flashy features and they're really excited about it.

They're having the unboxing effect is making them really positive about it.

What you really want as a consumer is the review of, of a toaster that's five years old.

You want someone who's used a toaster, bashed it about a bit for five years to find out is it still good.

Speaker B:

That's right.

And perhaps.

So what you're saying is we should be randomly sampling people who bought a toaster and requiring them to write a review about forcing them because otherwise you've got this selection effect.

So only people who are really unhappy or super happy.

And of course the other problem of having you often you'll read these reviews and it'll be like one stop because it arrived a week late and it's like that is not what I. Yeah, you review.

Speaker C:

Yeah.

You're reviewing the delivery product and also.

Speaker A:

You often get that like I say, the new shiny thing you get.

People give you a five star review after.

After a week.

Speaker B:

Yeah.

Speaker A:

And as you say that's no good.

But the only problem is you need to incentivize promptly.

Speaker B:

Haven't used it yet.

Speaker A:

You need to incentivize.

These people have had it for five years because otherwise what kind of nutter.

I don't trust paying for the information.

Speaker C:

This is the business model of which.

So they are for people who aren't aware there's a consumer review organization, they've been around for a very long time.

They will do long term big comparative tests between consumer products.

So they'll have 20 washing machines kind of on the go in a lab somewhere being thrashed to death and they will track how well they perform and how long they last and that kind of thing.

But which is a pay for.

Is a pay for magazine.

You have to buy a subscription.

Speaker B:

I think you can, you can like get one off.

Speaker C:

You can get one off.

So that is good.

It's worth do if you are buying a.

Get a subscription when I've got to replace something big and expensive because I want to know what which says good.

The world of product review is vast.

There's like anything you can imagine a.

Speaker A:

Lot of products out there.

Speaker C:

Anything you can imagine There'll be somebody on YouTube doing videos about it.

They are not interested or incentivized in doing something that's five years old, by and large, because they need to keep up with the flashy and the new and the exciting.

So the same problem of the unboxing effect occurs with our kind of professional inverted commerce review.

Speaker B:

And actually then views aren't necessarily correlated with accuracy at all.

Speaker A:

I wonder.

I feel a touch of patriotic pride surging in my heart because it feels to me.

Which feels very British.

Speaker B:

It does, doesn't it?

You imagine sort of men with thick glasses and clipboards standing around a washing machine.

Speaker A:

I'd like to.

Yeah, I'd like to think it's not.

Speaker C:

Such a big, serious, very dull people.

Speaker B:

It does sounds very.

It feels very:

Well, I think, and I think I suppose the surprising.

The superficially surprising thing, thing and actually heartening thing is that I would have expected which to be one of the things that was completely abolished by the Internet.

And as Peter says, you know, the Internet has, if any, I mean, the review system, which should have been one of the great benefits of the Internet is the ability to share information very obviously has failed for.

For all kinds of reasons.

So.

Yeah.

So I think let's have a look at the theory of those.

Try and dive into a bit of the type 4 failures.

Right.

So I think we talked about things that break.

We talked a bit about then, which is kind of Type three things that break.

And if they break a lot, it's probably because they're badly designed.

But the thing is that they do.

They work as a washing machine, but because they've got a wonky drum placement or something, that it will eventually fail too early.

The most interesting thing, I think we want to get into the type 4 failures.

The things that don't work.

Speaker A:

They're rubbish even when they work.

Speaker B:

So let's start.

And this is where I think Peter's engineering expertise will come to the fore.

Because why have we not worked out how to make a jug or a teapot that pours properly?

Why does it always spill out?

Speaker C:

We have what the science.

Speaker B:

Someone should tell people who make jugs.

Speaker C:

The science exists, unfortunately, the people who.

A lot of the jugs that are sold aren't optimized for performance, are optimized for aesthetics.

Speaker B:

Oh, God.

People buy jugs to put on their shelves.

Speaker C:

You want to look for ugly jugs.

Start with ugly jugs.

And then.

Speaker B:

Wait a minute.

I can feel an idea for a company coming on.

Speaker C:

Certainly got a good Brand, isn't it?

Ugly jugs.

Yeah.

Speaker B:

What do we need to get right?

To get right?

Speaker C:

Well, so a bit of the science.

So it turns out that the thin lips is what you need on a jug.

You need a thin material at the very tip of the lip is key.

Speaker A:

Is that the very pointy bit?

Speaker C:

The very pointy bit.

The very end.

I'm not sure why, because that stops.

Speaker B:

The liquid curving, I guess it stops.

Speaker C:

The meniscus of the pour.

Speaker B:

You almost.

It needs to kind of cut the liquid as the liquid's coming out.

Yeah.

Speaker C:

Yes.

So there's a bit of science on that.

The fluid dynamics in this is quite interesting.

Another important thing is the coating of the spout, the inside of the spout.

Speaker B:

Hydrophobic.

Speaker C:

Yeah.

Basically you want it to be hydrophobic as possible because that will cause it.

That will also stop the meniscus from sort of rising, riding around also.

Speaker A:

I'm going to jump in here.

Speaker B:

Go on.

Speaker A:

I'm not.

I'm only a hobbyist engineer, as you know.

But we've all had.

Speaker B:

You're a jug user.

Speaker A:

I am a jug user and we've all had little kids.

I mean, our own.

And tippy cups work really well.

Okay.

Are they sippy cups?

Tippy cups.

They're the angled thing.

Speaker B:

Right.

Speaker A:

And.

And the idea I think is there's.

Speaker B:

A brand called Tommy Tippy.

Speaker A:

Maybe it's that I'm talking about.

Speaker C:

Yeah.

Speaker A:

And maybe that this is our way to just disrupt big kettle to have like this sort of trapeze.

What do you call it?

This sort of leaning over kettle jug.

Speaker C:

You know, I don't think it's the lean.

Speaker B:

The shape of the vessel isn't the key thing?

Speaker A:

Well, I think it is, really.

Speaker B:

The spout, I think it is.

Speaker C:

Have you used a tippy cup to try pouring water?

Speaker A:

No.

Speaker C:

Well, I mean, how can you say that they're good then?

Speaker A:

Well, no, they're good.

Speaker B:

Do that test first before you.

Speaker A:

Yeah, yeah, yeah.

But I think I could be on something.

I'm not saying this is a done deal yet, but, you know, I'm just, you know.

Speaker C:

Yeah, yeah.

But as you might expect, like thin materials and special specialized coatings are potentially expensive, but not always.

Not always looking the best.

Speaker B:

Are we touching on something pathological about human brains?

Market selection, then?

Speaker C:

Yeah, a bit.

Yeah.

But it's.

Yeah, it's so a lot, I think,.

Speaker B:

Is it, that humans are cheap and superficial?

Speaker C:

Yes, that's exactly where I'm going.

Yeah.

Yes.

Speaker A:

Like fruit.

Speaker B:

Right.

Speaker A:

Supermarket fruit.

Tastes crap.

Looks good.

Speaker B:

Yeah.

Speaker A:

Right.

Same thing, basically.

Sorry Peter, I interrupted.

Were you going to go on?

Speaker C:

I was just going to say for further reading, look up.

It's called the teapot effect.

er or Reiner documented it in:

Speaker B:

Amazon:

Goes back a bit, then goes back a bit.

Speaker C:

And there's loads of active fluid dynamic research going on about how to make spouts better that manufacturers quietly and happily ignore.

Speaker B:

Well, I think.

And I think the.

The.

The thing is there is a correlation, isn't there between aesthetics and performance often in that if you're investing at all in the design of a product, it's probably going to be reflected in both those.

Both those things.

Right.

Speaker C:

To a point.

I think that.

I think though purely functional things are.

Runs completely counter to aesthetics.

Speaker A:

Right.

Do you reckon?

Speaker C:

Yeah, they are misuse.

They are in.

I think you're right.

There's a correlation between the amount of design effort you put into something because that design efforts principal job is to balance aesthetics with functionality.

And if they're doing a good job they'll find an aesthetic which fits the function.

So there is a correlation there with the amount of design effort that goes in it.

But that you are balancing two diametrically opposed things.

Speaker A:

I mean we probably don't have time to go into it here but that's something I'd like to sort of delve into.

Speaker C:

I think we've covered it in other things but you know, think like brutalist architecture for example or even industrial architecture.

Industrial architecture, better example that is.

That is.

Doesn't care what it looks like.

Doesn't care what it looks like.

Think of a power station or a.

Or a.

Some sort of metal works or something like that.

It's pipes and tubes and it's ugly.

Speaker B:

And it's deliberately not to you though, Peter.

I bet you think they're beautiful.

Speaker C:

Yeah, well, yes.

I mean there is a sort of aesthetic appreciation which is.

Speaker A:

So it's accidentally aesthetically pleasing.

Speaker B:

Right.

Speaker A:

As it were to someone.

Speaker B:

Yeah, well, I guess you have to have the right eyes to see the function.

Right?

Speaker A:

Yeah.

Speaker B:

And I.

So I mean I think the problem is that you couldn't therefore conclude well what I need is an ugly jug because that is likely to be good.

What you're likely to end up with is a jug where they haven't paid any attention to the design at all.

Speaker C:

Yeah.

Speaker B:

And it's ugly by accident and it doesn't work because they haven't.

Speaker C:

You go and buy your jugs in CK Max or something.

Yeah, probably what you're gonna get.

Speaker B:

I mean I Think the answer is something like which isn't it?

But do they do jugs?

Speaker C:

I think it might be too low level.

Speaker B:

Right?

Speaker A:

Yeah, I think so.

Speaker B:

So that's teapots and jugs.

Right.

So I think, and I think toasters, possibly the same underlying issues here about, you know, cost and aesthetics and so on, but toasters, obviously a lot of people complain about setting.

Like, why doesn't the setting, why does this setting burn toast but, you know, barely touch a crumpet?

And that's because the settings, obviously it's just a timer.

It's just a timer.

So it doesn't know whether your bread is frozen.

It doesn't know if it's a bagel.

I think apparently people have tried to design toasters that actually somehow optically detect how brown the toast is, but they've never really taken off, I suspect because of cost.

You've also got the issue of like the sheer physics of it, that the outer heating elements are always going to, you know, lose heat to the outside of the toaster.

So if you have a conductive outside of toaster, then the outside of the toast is always going to be less well done than the inside.

And that's why, I mean, I flip, I'm a toast flipper.

I turn them upside down and I turn them back to front same.

And yeah, about halfway through you have to.

So that's, that's toasters.

I mean, I'm just saying that again, it's hard to get a toaster.

There's intrinsic issues with toasters.

Easy peel, packaging.

There is an intrinsic trade off between being keeping your bacon free of germs and making the thing easy to pull off.

Yeah, I wish they just dropped the pretense and just get rid of it altogether and say, must be opened with a knife.

But it, but it's.

This is obviously I found from just, from four days ago, someone posted on Reddit, I opened a packet of bacon without the need of a knife.

And the comments included, out of everything I've seen on here, this is the fakest post I've come across.

He is the chosen one.

Stop lying.

It's impossible.

It's like pulling out Excalibur.

A bucket list achievement.

You may now die happy.

And all hail the bacon whisperer.

So obviously it's understood that this is not something we ever expected.

Speaker A:

It's a thing.

Speaker B:

So, yeah.

Anyway, Peter, what's going on?

Is there a general theory of this?

Speaker C:

Yeah.

So I would like to introduce the doorbell paradox.

Oh, so this is the principle that the more partly Driving.

Why things fail sooner than they used to is that we are piling on more and more and more features onto things.

So you can't buy a washing machine or a TV or anything that's now not got WI FI in it and it connects to.

Speaker B:

It's entirely mechanical.

Speaker C:

Yeah.

And there's an app.

Speaker A:

Yeah.

So greater complexity suggests there's more to go wrong.

Speaker C:

Right.

Ten years ago, maybe more, I had a quite an old Zanussi washing machine and I was umming and ahhing about whether or not I should just replace it when it went wrong.

The engineer came out and did a quick fix and he said don't, whatever you do, get rid of this washing machine.

Because this is one of the last mechanical, purely electric mechanical washing machines.

It was completely serviceable.

Really?

Speaker B:

No.

No like electronics at all.

No computer inside your wall.

Speaker C:

No computer at all.

Wow.

Had a kind of.

The only electronics it had.

Speaker B:

Do you get people coming round to look at it?

Speaker C:

Well, it should be.

It probably, probably should.

Belongs in the science museum now in the Smithsonian.

Yeah, it was electromechanical, so I had a little.

I think the only electronics it had was a little motor that drove the clock which broke and made contact with various things on a dial to set the program.

But it did die completely eventually.

So yeah, we pile on loads more stuff and there's quite simple rules about how your rate of failure goes up as you create.

The arithmetic of how failure rates increase with the number of parts is very simple.

So if you have a thing made of 10 pieces and each one's got a sort of reliability of 99% say that, you know, there's a chance it will die.

One in every 10 years.

One in every 100 years.

The actual failure rate is 0.99.

To the power of 10.

Speaker A:

Yeah.

Speaker C:

To the power of the number of parts.

So as soon as you add more and more parts, that number starts to plummet.

Speaker B:

Or you need increasingly reliable components.

Speaker C:

All you need which is not increasingly reliable.

So.

So the doorbell, which.

Oh yeah.

Speaker A:

How's this do?

Speaker C:

What's the doorbell?

Doorbell.

All you really need for a doorbell is a button, a wire and a bell.

Right.

Speaker B:

I think we might be losing some of our listeners.

Speaker C:

I know tentacle, isn't it?

But you know, try and find a doorbell that's just that.

Well, yeah.

Speaker A:

Why do they do like the whole video camera thing?

Speaker C:

Well, yeah, yeah.

You have video camera, you have some app, you have something that captures the video.

Speaker B:

I doubt got chat GPT in there.

Yeah, yeah.

Speaker A:

You can.

Speaker C:

Yeah.

You have someone in AI.

To answer the door and chat with every.

Speaker A:

My doorbell's even set.

Even simpler than that.

Right.

Because, as you know, I live in quite a whole.

Speaker C:

An old house on a string.

Speaker A:

Well, you pull a lever.

Speaker C:

Yeah.

Speaker A:

And there's a rope and it.

Ding, ding, ding.

Yeah, yeah, that's it.

Speaker B:

And the butler lets you.

Speaker A:

The butler says, yes, sir.

Yeah, yeah, yeah.

Speaker C:

Even the simple.

Sort of.

Even the simple ones with a wireless transmitter, there's quite a lot of little.

There are loads of little electronic components in the transmitter, where the button is and then in the chime.

There's loads of stuff going on there, so hundreds of parts, even on the simplest possible.

Speaker B:

I think most people will claim that they don't want all these extra functions,.

Speaker C:

But it feels like most people, like us, will claim that.

I don't agree that most people.

Speaker B:

Well, the market.

I mean, I have to conclude the market knows what it's doing and, you know, things acquire functions that they don't need.

And so, you know, I guess in order to make the doorbell paradox, a kind of, you know, empirical paradox, what we're saying is people both want their devices to become more sophisticated and they want them to be more reliable.

And actually, those two are in conflict.

Speaker C:

I think there's a schism in the market.

I think there are people who chase the newest, flashiest thing and they want the latest camera on their doorbell and the newest WI FI standard on their fridge.

But then there are people more like us who are reacting to it, reacting to these slightly unnecessary design choices.

Speaker A:

Not for the first time.

You know, I often wonder, why can't everyone just be sensible like me?

All right, Nick, we're sort of pretty much at the end here.

Speaker B:

I'm gonna.

I think I can end on those on a happy note.

Speaker A:

Oh, brilliant.

Good.

Speaker B:

Because what I want to tell TAS is that it feels, I think, psychologically like we are every day besieged by crap design and things that don't work.

And I. I'm here to say that actually the list of those things is pretty small.

And the only reason I know that is because I went looking for as many examples as I can find, and actually, there aren't that many.

It is like the things we've talked about.

It's the toasters, the things that you have to open bacon with, and so on.

I mean, there's probably 10 or so things that crop up every time someone asks this question online.

What are the most annoying things that don't work?

And I think each of those.

I mean, we looked at a little bit of Toasters and kettles, they kind of have their own thing going on.

Typically it's because there's some intrinsic problem that is a trade off that's actually just quite hard solve or quite hard to solve at a reasonable cost.

But these things do get solved.

Let's remind ourselves that the electric toaster is itself a solution to having to toast bread on a griddle, you know, with the toasting tongs and no doubt on a fire.

Speaker A:

Right.

Speaker B:

Which would occasionally burn your house down, you know.

So let's remind ourselves that the electric toaster is a much is itself a great leap forward.

And when we look at all the things that people used to complain about, there's a huge great list of things which fell into this category and which have been fixed.

Top of the list, of course.

Tetra pack milk cartons.

Now anyone but under the age of about 40 won't know what I'm talking about.

But milk cartons used to be cardboard and to get them to open, you'd sort of pinch the spout so that it would kind of could turn into a little.

And it would often that would not work and it would then be a really poor quality.

Poorer.

It would have all the problems we've discussed with jugs.

Ketchup that doesn't come out of the bottle.

Well, that's the thing of the past, isn't it?

With the squeezy squeeze.

The squeezies ring pulls, remember them coming off the can fixed.

Because now the ring pulls are sort of stuck to the can in a way that seems to be better.

Cassette players that ate tapes.

Doesn't happen anymore because we don't have cassettes.

USB connectors which would never go in the first time you put them in.

But it's been solved with the usb.

C. Yeah.

Leaky pens.

I don't have leaky pens anymore.

Headphone cables that become tangled all the time.

Big complaint.

Wireless fixed it.

So I think somewhere down the line we can look forward to toasters and jugs being fixed too, when toaster and.

Speaker C:

Jug technology finally catches up.

Speaker B:

And then we will look back on this podcast and go, wow.

Remember the time when jugs didn't pour properly?

Okay.

So, yeah, I want to offer that note of reassurance to Tas, but also I think the answer is he's got to stop caring so much about what his toaster looks like and be willing to shell out a few more quid.

Speaker A:

Yeah, okay.

Yeah, no, I think you're quite right.

And we've never had it so good.

Right, Very good.

Yeah.

No, the cockles of my heart are warmed so toasted.

Toasted.

Yeah, they're all nice and toasty.

All right, let's wrap it up there.

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 aim to release an episode every week or two.

If there are any topics you'd like us to cover, please email us podcast@alephinsights.com thanks, as always, for listening.

Until next time, goodbye.

Show artwork for Cognitive Engineering

About the Podcast

Cognitive Engineering
Welcome to the Cognitive Engineering podcast.
Welcome to the Cognitive Engineering podcast. Occasionally coherent musings of Aleph Insights. We hope you like listening to them as much as we like recording them.

About your host

Profile picture for Fraser McGruer

Fraser McGruer