Open versus Closed Standards

Speaker A: 00:00:00

Hello and welcome to the Cognitive Engineering Podcast produced by me, Fraser McGruer, for Aleph Insights. In this series of podcasts we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm here with Nick Hare and Peter Coghill of Aleph Insights and this week we're discussing what three words. Peter, I've got some kind of awareness of what three words, I think I've seen the adverts, but tell us more about it and maybe then tell us why we're talking about it.

Speaker B: 00:00:34

Okay, what three words is an online mapping and reference service. What it does is it provides a unique reference for unique places on the earth. Every three square meter block on the earth has got a unique three words and they're all fairly common words and they're common words within different languages, so different languages have different sets of words, but every place on the earth has a unique reference. The idea is that it provides a nice, convenient, human-readable way of referring to a place. You could be in the middle of Salisbury Plain and say, I'm at fish yellow basket or something like that and then you'd be able to pinpoint exactly where you are within three meters, which is enough to find somebody who's unconscious or something like that. The use case of their sales literature site is for exactly that kind of reference, where you can provide over the phone, in natural language you can give three words which indicate where you are and then emergency services could come and find you, but also other use cases like package deliveries and things like that. What it does is it provides an alternative to other geographical referencing systems such as lat long or postcodes or addresses like postal addresses that is intended

Speaker A: 00:02:00

to be convenient, so that's what three words is. Okay, so that's what it is and it sounds really good to me. I mean, when I saw the adverts for it, I thought that looks cool.

Speaker C: 00:02:13

It's a nice idea. Incidentally, there's no such place as fish yellow basket, but I can get you a fish fellow basket which is in Lower Silesia. Okay, there we go. I just made

Speaker B: 00:02:23

that one up. So, it is a nice idea. I can totally see the benefit of being able to over the phone quickly give a reference and you sort of pinpoint exactly where you are. That's the use case that happens kind of quite regularly. I sense a but coming. But, yes, but. Now, there are certain things that I don't like about it. There are certain concerns I have but I will come on to them. So, let's explore why it's a nice idea first. So, in comparison to lat long. So, lat long can give you any degree of precision that you want. You can just keep adding more decimal places to your lat long and get ever more accurate. So, that's sort of technically better, but because it's numbers, big strings of numbers are quite difficult to remember, quite difficult to say. So, imagining over the phone, you've got a scratchy phone line to the emergency services. You're dropping the old figure here and there could put you hundreds, if not thousands of miles in the wrong place. Whereas words, because they're words, they're common words, there's a sort of language. Our brains are really good at processing language, especially when we can't really, even when we can't really hear very well, there's a sort of natural error correction with spoken language, with spoken words, which we don't have so much with numbers. So, that makes it a nice idea. What I don't like about it is that it's a proprietary closed standard. How these words are allotted to places on the earth is not published anywhere. And indeed, What3Words is a profit-making organization whose job is to use this algorithm that they've invented, that they don't share with anyone, to make a profit from it. So, they sell this as a service. So, it's free for basic use. You can go on the website and use it for free. But if you want to use it for a business, you would have to pay them a license fee to use it. So, there's inherent problems with that, especially considering the sort of safety use case, where you're trying to locate people in the middle of Dartmoor. If your emergency services hasn't paid the license fee, or isn't current on the license fee, they won't be able to look up these locations that they're being provided. Well, that's no good. So, there are other sort of interesting informational kind of constraints or problems with it that I see. So, for example, I'll just pick two random squares that are somewhere in London together. So, two adjacent squares, you have PuttyPlugWished is one. And then very adjacent squares, HostVividEngine. So, we've only moved three meters, but the words

Speaker C: 00:05:12

are completely unrelated. It's not even like they're in alphabetical order. It's mental. I think, see, unlike Peter, I think it's a completely terrible idea. And I do, obviously, we want to talk about standards and open standards and closed standards. That's the interesting bit. But I think just before, just on the issue, instead of Peter picking random squares, I can give you my own address, right? OpposerMusePose. OpposerMusePose? Yeah, exactly. This is supposed to be easier to read down a phone line. Yes, it's OpposerMusePose. That's my front door, right? But don't worry, because my hallway is FloatTownJump. And my front garden is WorthMouthTicket. It's bonkers. What this is, is someone who's only, you know, lived in an office, in a cubicle in Silicon Valley, thinking, hey, I can become a millionaire this way.

Speaker A: 00:06:06

And has not really thought about- Also known as the space wanker git go.

Speaker C: 00:06:09

I mean, there's all- I'll fuck right off. Yeah, I don't, I mean, first of all, I mean, it doesn't do Z coordinates, as far as I can tell. So if you live in a tower block, you are SOL, because all it will do is give a sort of rod going all the way through your building. It's not robust, any more robust to errors, as far as I can see. It's worse, in a way, because, you know, if you, yeah, if you're sending something to that long, and you get, you know, the second digit wrong, well, you're going to be in the right, you know, sort of 100 miles. But here, you know, if you get fish mellow basket instead of fish fellow basket, you're going to be thousands of miles away in the wrong country. So there's no indication of proximity. I mean, and the other thing, you obviously can't use it for anything machine readable, it's much worse than even an address in some ways.

Speaker B: 00:06:58

And that's the sort of nub of their business model, is you have to use their service to look up between-

Speaker C: 00:07:06

Yeah, so I mean, given that it's the kind of thing, Peter and I could sit down and probably write an alternative to this, you know, using free dictionaries in about 25 minutes. It's baffling that anything got this far and got as much press as it did.

Speaker B: 00:07:22

Yeah, there are some fun alternatives. So what three birds uses British bird names. That's an open standard. And then what three words is a slightly more serious example. But yeah, people have produced alternatives to it.

Speaker A: 00:07:35

So but also, it sounds like, you know, it's the kind of thing that has been developed in a vacuum, more or less, you know, it's been walled off. And because one of the advantages of doing things in a more open manner and developing in an open manner is, you get greater sort of, you get a greater breadth of input, right. And, and so there's the sorts of issues that we're talking about, if that being developed differently, or was able to adapt in an open way, then you could get around these things, all these things. I mean, this is what we're talking about.

Speaker C: 00:08:06

I mean, that's the thing I was going to ask Peter, right? It seems to me, this is it's quite ironic, but there's no standard definition of an open standard. So I just wondering what, whether Peter's got a favourite definition, because it seems there's hundreds, there's loads.

Speaker B: 00:08:22

Well, it's, it's not really a sort of a definition per se. But yeah, there are there are lots of different, different standards defining organisations will define an open standard differently. But in essence, what I mean by an open standard is one that how it works, and how you use it is free public information that's fully open. So there'll be GitHub pages, which actually provide you with code, which instantiates the algorithm, there'll be documents that describe how that works. And there will be, there will probably be a community around it, or maybe not, but that's the minimum for us to open standard. Basically, it's the absence of any proprietary licence that you must pay for or agree to in order to use it. And certainly the absence of any cost of using it.

Speaker C: 00:09:14

But but also, I think the there's also so I think a lot of open standards have a kind of open participation, don't they? So that you know, that if the standards need to change or be added to, that's a fairly sort of public process that that outsiders can join in. Whereas, you know, you're at the whim really of what three words if they decide to change what they're doing, or in a way that doesn't suit your need. Or, you know, like you mentioned different languages, I mean, what a nightmare that your address in French is different to your address in English. You know, you've got no control over that. You just can't.

Speaker A: 00:09:47

So let's get down to this. I mean, what we really want to talk about is open standard versus closed standard, right? Yeah, which is better. Yeah. And, you know, we're about a third to a half of the way through this already. So where do you want to go with this? I mean, all I mean, the only thing I can think of immediately that sort of the I mean, closed standards that and if this is the right thing is Apple is famous for doing this stuff, particularly, let's say iTunes and so on. And whenever I've tried to music, move music away from iTunes, and I haven't used it for more than a decade round for specifically this reason. It's just such a faff. It's so difficult to do that because of all the blocks that are in place effectively. And I mean, kind of what we're talking about

Speaker C: 00:10:29

I think things weirdly, it seems that things tend to open up as they go along. So there's quite a lot of open standards. So there's the World Wide Web, USB, HTML, the things that, you know, MP3, I think in most countries, and obviously things like text files and CSV files. But even things like I think, Mike, the Windows API for a long time was closed. But I think it's now open. And obviously, like printer cartridges, similar example where, you know, used to be I think it was illegal at one point to sell, you know, things that were compatible with your Epson printer. Now all that happens is my Epson printer nags me mercilessly if I put in a non Epson cartridge, but also the same with case cake up, you know, the Nespresso cartridge thing that that is now effectively an open standard. I think anyone can manufacture those. And so I think over time, things, you know, tends to become open. But the you mentioned Apple, I mean, they're famous for it, right? So is it good business? I think that's the first thing we I'd be interested in is, you know, is, is he actually a good everyone wants what's called vendor lock in, right? This is a really good way of getting vendor lock in if there's a standard that only you can use, you know, even down to Apple's flipping screwdrivers, which you know, you can only open their things with proprietary Apple screwdrivers. I mean, it's all massively Apple only, you know, the whole thing, but you know, their share price is up, you know, has gone up by a factor of five at the last five years, whereas Google, which is a lot more open has only gone up by a factor of three. So, you know, is it working for them? I don't know why

Speaker B: 00:12:11

it would be. Well, investors love patent based economic rent. So if you can sort of get some, if you can, if you can patent something, or put a copyright on something and make it popular, that's a very investable proposition, because you have exclusivity there. You have to go to that vendor to get that thing. And investors love that. That's a really kind of classic investor model. And I think your point again, I think lots of things do tend to open up. So a standard will develop often around a proprietary software solution or tool or a way of process of doing things. And as more people adopt it, it becomes constrained by that single sort of point of purchase. And many companies see the benefit of actually then opening that standard up to invite other suppliers to provide services around their products, or indeed, other products around their service that they want to provide around that standard. So it works in certain situations to be profit making to actually open things up somewhat counterintuitively. But it really, I think it really boils down to that sort of protectionist. If I'm making money out of this thing, and I can charge people to use it, I'll keep doing that, because that's nice, easy, safe.

Speaker C: 00:13:38

dn't. And we, you know, in in: 1988

Speaker B: 00:16:32

Yeah, I think, yeah, I think that's what you want out of life, you know, it's definitely certainly in the when you're in the early days of a standard, there's a case to be made for making it proprietary, because you then you can, you can centralize a small community of developers to make it better and make make it, you know, promote it. But as you say, Nick, it's probably not a very scalable model, because you can't then have a marketplace of vendors providing services and products around that around that thing as freely as you would with an open standard. But apart from the sort of technical and business constraints, there's a sort of, there's a sort of sense of there's a moral objection to it as well. Is it because it closed standards, big, big companies, Facebook's and the Googles and the apples, they can they can they can create a standard before breakfast, and they can sort of they can, over the months and years, kind of impose it on their user base. So it the standards do benefit the incumbents if it's if it's fully if it's closed or very highly proprietary. And it doesn't, it doesn't allow, it doesn't allow competitors to come into the market doesn't allow a blocks the innovation by third parties. So there's a sort of moral objection to it as well.

Speaker C: 00:17:55

Yeah, I feel like it's a fragile, a fragile possession, though, isn't it? A monopoly based on a closed standard is it feels like a vulnerable, vulnerable proposition. It's not you're not there, because what you're selling is better. You're there because what you're selling is the thing that everyone has to use because they've got no choice. And all it all it takes is, you know, an incumbent to come up with something a bit better, a bit cheaper. You know, and innovators will will try and adopt it. I mean, I mean, yeah, if you take Frazier example of Apple, I mean, that's one of the reasons I did test Apple as a country company. I really like and I detest the fact they have a fanatical following that baffles me that so many people can can appear to feel that they have a kind of personal relationship with this soulless company, you know, which survives entirely on marketing as far as I can see. And and I yeah, I've I use a thing called Media Monkey, which is like a, you know, paid for media management thing. But you do everything you say. It's all you've got total control over it. The idea of having, you know, of having Apple tell me whether or not I can listen to my own file, which I've paid for on, you know, on another machine is infuriating.

Speaker A: 00:19:06

Listen, by the way, there's only one thing that's more annoying than Apple fanboys, right? Which is Apple haters. Okay. Because I just get so fed up. That's what they want you to do. Nearly everything I possess in terms of gadgets and so on is Apple, right? No, wait, wait, no, no, no, wait, wait, wait. Yeah. No, it's the eye Kool-Aid. A lot of it is just because actually, for my work, I this everything from my work that is given to me is is an Apple anyway, right? But when you're a creative like me, that's, you know, it's just why I've created. Yeah, that's why I don't know how to use computers. That's why I love saying that. Because as soon as I say that everyone goes, Oh, you know, but wait, I'm just I want to I do

Speaker C: 00:19:53

want to probe on this. I is do Apple make things that are better for creators? Or is it simply that creators can't be bothered to learn how to use a real computer? What's the answer? Because everyone says it? Oh, yeah, I've got an iMac because I'm a because I'm a designer. Well, is it better? Is it? I don't bloody know. Because it's budget hardware at premium prices. No, no, no, no, no,

Speaker A: 00:20:15

listen. What I do know is if we take the example of a tablet, let's say is I do know that picking up a what Apple tablets called? iPads? iPad. Thank you. Yeah. Picking up an iPad and just starting to use it I find way much easier than than any other tablet. Yeah, you're like a little kid. Oh, I'm just poking it and it's doing something. And this is just and this is and but when it comes to like tech geeks like you lot. As soon as we get to this sort of conversation, you go but this one's got this many gigabytes and rams and and it's got the it's got the flange gidget here and oh what you're so stupid you don't have it almost gets down to a conversation about grads from tech degrees. Just get all kind of Oh, you know, just because you studied English, literally you're you know, you don't know what you're on about, right? That's just because those like engineering types just can't get laid at university and they're just jealous of all the arts guys just like shagging all over the place. Note the Apple fanboys resulted to ad hominem attacks. So anyway, I just wanted to get all that out of the way, right? Look, where are we on this? We need to finish shortly. I don't quite understand. I mean, all we've said so far, right, is closed is sort of okay. If you're starting up, although you're gonna be competing against, as you said, Peter Peeple, like Facebook, etc, you can develop stuff closed in house really quickly one that but if you can sell it off, closed in, that's better for investor locking, I think you said, and two, there's something slightly morally objectable to that. But let me just finish over three overall open sources better. But what have we said? I mean, that is that it? I sorry, there isn't a good kind of

Speaker C: 00:22:01

solid economic argument against kind of closed anything really, which is that you know, from an economic perspective, effectively, you have a kind of prisoners dilemma. Let's say you've got VHS and Betamax, you know, fighting it out, and they're just spending a lot, they are fighting over the same market share and wasting money is, you know, what's known as rent seeking, where there's not they're not adding value through this activity, because all they want to do is increase their market share. You know, the in the same situation where you had both of them producing the same effectively the same standard, but but able to compete on in the same market through things like quality, and price, then you're going to get better outcomes, you're gonna get more efficient outcomes. So there is an economic argument, there's nothing, you can't make it illegal for people to invent a closed standard. But you know, I in a way, I think, as consumers, we should we have a kind of obligation to ourselves to to try and choose the thing with the open standard if we can, because it makes us worse off. If we if we if we keep buying, you know, products from companies, I don't want to get into that whole debate again. But you know, products from companies where you have to buy their screwdriver to undo the bloody

Speaker A: 00:23:17

I've never bought a bloody Apple screwdriver in my life. I don't even know what it is. Anyway,

Speaker B: 00:23:21

it was a related topic, not not particularly picking on Apple again. But there is a there is a movement, particularly in America, and in the end in the EU as well, which is called right for repair, right or right to repair. So it's it's react a reaction to commercial consumer products, the iPhones, the iPads, the whatever PCs you buy, being designed to fail or being designed, you know, having planned obsolescence that are built in, and then being designed in such a way that repairing them is more expensive or impossible than just replacing it, right. So they are they are, I think there's a recent ruling in the EU that the computer hardware can be sort of computer units. So if you buy a whole laptop or a whole desktop is going to be designed in such a way that it will last 10 years. And that that in order to make that happen, it's got to be sort of repairable. So you've got to be able to swap out the hard disk or change the battery and that sort of thing. And now that's sort of that that's being fiercely resisted by the big tech companies for obvious reasons, because Apple and others make lots of money by controlling the entire supply chain for the entire manufacturer, but also the repair chain. So you can't get an iPhone fixed just anywhere, you have to go to a proper Apple supplier, Apple approved developer repair shop. Now they will they'll say that that's because they can control quality, that's their that would be their their line. But actually, it's really just to control everything. So they they get license fees and they there's not a proliferation of genuine parts being put into fake phones and vice versa and things. So yeah, so that's that's that's an interesting related topic. It's sort of topical.

Speaker A: 00:25:10

Something just occurs to me, we need to wrap up shortly, but something we can go here or not, we don't need to talk about this. But aren't you at Aleph Insights, you develop proprietary stuff,

Speaker C: 00:25:19

right? Not really. I don't think we've got any. We're not in possession of any standards where people have to come to us in order to understand what our report means. I don't. It's not you know, we were things we publish you don't. No, I wouldn't say we are. What about your AGA project?

Speaker B: 00:25:35

Was it AGA? Is that right? So well, that's Yeah, I think there's a distinction between intellectual property and open standard. Yeah, you're right. I mean, I mean, you know, I guess we've probably

Speaker C: 00:25:46

published how more or less how it works. And it's only based on stuff which is, you know,

Speaker B: 00:25:51

if you like public domain anyway, our instance of how it works is, is proprietary is proprietary. You're what? You're what? instantiation of the concept is, is, is private to us is our property. But the idea is something that we could we wouldn't or couldn't patent. Okay,

Speaker C: 00:26:12

think of it like a book Fraser that in order to read Great Expectations by Charles Dickens, you do need to buy a copy of Great Expectations by Charles Dickens. But you'll find that the words he uses are very much words that you're allowed to use. What is it with us this morning?

Speaker A: 00:26:27

Why are we so bloody argumentative? Why are you trying to piss me off? No, I just think it's a with that? Is that right? Yeah, yeah. All right. We're gonna stop there. If you've got any thoughts or suggestions for topics, you can email us at podcast at aleph insights.com. We'd love to hear from you. Also, if you've enjoyed the podcast, what should people do? What should listeners do Nick smash that like button? Yeah, smash that like button. So don't forget to like and follow us on your chosen streaming service. So thank you as always for listening to the cognitive engineering podcast. I'm Fraser McGruer being here with Nick Hare and Peter Coghill of Aleph Insights. Until next time, goodbye.