Have an observation? Hit record, or drop a file here — it might end up on the next episode. The passphrase is said at the start of every episode.
0:00Hi, and welcome to observations. Today, I want to talk about the oldest trick in the tech industry, and especially in how to make money in the tech industry called commoditizing your complement. And I'm gonna feature two articles, um, one by guern, which is, which I'm just gonna read the intro Vibe because? I mean, it's a one article, so it's like, um, 10K words, but the intro is pretty good, and then I'm gonna jump to the strategy letter number five by Joel and software by Joel spolsky, which he wrote in 2002.
0:35After being at Microsoft and discovering this pattern? Spelling out this pattern in a blog post. Observations is a format I want to create, which is basically me talking to a microphone. I'm just pulling out my recorders. You can probably tell from the audio quality, like my phone recorder right now.
0:57I don't even have a microphone, so if you want, if you have observations and want to share. Just send me a like, open your recorder as well, and send me a voice message, and I might feature you in the next episode. Once I figure out how audio recording works and how to splice and stuff, or it will just cite you.
1:18But yeah, um, maybe you can also reply on Mastodon or wherever? Um, I might have some website in the future right now. It's just a list of episodes. And the audio. Okay, let's jump in. So, ongoing.net complement very nice URL laws of tech. Commoditize your component. Tags AI, economics, economics, Insight, p***, Google.
1:46A classic pattern in technology economics identified virtual sportsky as layers of the stack attempting to become monopolies while turning other layers into perfectly competitive markets, which are then, which are commoditized in order to harvest more of the consumer surplus discussion and examples. Read this passage in the original (opens in a new tab)
2:04So, especially the example section is going to be interesting, but I think it's gonna be a bit too much for this episode.
2:11This was first published in 2018 and then update in 2022, so I bought a three year period. In which Gwen added more examples and is considered finished by. Um, certainty. Highly likely importance. Five. I think it's like a out of 10 system that they use. Um.
2:35Introduction Joel spolskin 2020 in 2002. For 24 years ago
2:42identified a major pattern in technology, business, and economics. The pattern of commoditizing your complement? Read this passage in the original (opens in a new tab)
2:49So as an interjection, there's like a this word commodity right commoditizes a verb based on the on the known. Commodity and? Observation I made is that the German language doesn't really have?
3:05Um, an equal word. Commonly known, and I don't think there's there's probably commodity or something used in economic circles, but we have different terms that mean. Just a piece of like foundational material, but doesn't mean commodity in. Um, so, so I would say a commodity is something that. There's multiple Alternatives that are about the same price.
3:37And take. They are replaceable. So if you go to the gas station and you, you tank something. And it's gonna be roughly the same fuel every everywhere you go, but the producers of the fuel might be a different company or even the gas stations, a different gas station. But the fuel serves a commodity, right?
3:56Same goes for. You could say the browser is a commodity, right? Firefox, chromium, Safari, they all Implement more or less the same API. You can display all the web pages more or less the same, so the browser itself. It's not really a commodity because there's only three of them, but you get the idea like, right?
4:19Um. Like Windows and buildings like Commodities. There's multiple producers. Um, who fight for the cheapest price or the best value, so the price will also? Go down towards low margins in in a healthy environment.
4:42So, there's no really equivalent word for this in German. There's good, there's vava, which is just generic for trading trading item basically. Um, there's resource resource. But there's nothing like, oh, this thing is like? Commodified doesn't exist, right? Um, yeah, that was an interesting observation I made, and I feel like that also influences how, how Germans think about?
5:10Um, especially, like? Day-To-Day. People who who don't have an economics background to think about? Um, the world. Or rather, they they don't really think about how to commoditize stuff. Not very much. But maybe that's just me being in a bubble. Okay, let's continue. So, the pattern of commoditizing a component.
5:36It's an alternative to Vertical integration, and
5:39it companies seek to secure a choke point or quasi Monopoly in products composed of many necessary and sufficient layers by dominating one layer before string. So much competition in another layer above or below its layer that no competing monopolus can emerge. Prices are driven down to marginal costs elsewhere in the stack.
5:58Total price drops and increases demand, and the majority of the consumer surplus of the final product. Can be diverted to the quasi-monopolist. No matter how valuable the original may be or how one, how much one could charge for it, it can be more valuable to make it free if it increases profits elsewhere.
6:17A classic sample is the commodification of PC Hardware by the Microsoft OS Monopoly to Read this passage in the original (opens in a new tab)
6:22judgment of IBM and benefits of Microsoft. Um.
6:31So, I noticed that the intro is a bit. Is a bit unclear. As in, it says, this is an alternative to vertical integration. Semicolon init company scene. So, I think the this whole paragraph was about commoditizing your complement. Um. Whereas vertical integration is. You have control over every part of the stack and thus.
6:58You can decrease margins in every single layer, and then, like what Apple does with with their projects, is they create products that span from they design their own Hardware, they go via the operating system. They create their own devices, and then they sell the whole ecosystem, including the apps and everything.
7:18But I think that's going to be an example about. Apple and the App Store, and like app developers as an example of commoditizing their complement later. Going on. This pattern explains many,
7:33many, otherwise odd, or apparently self-sabotaging Ventures by locked by large tech companies into apparently irrelevant Fields such, such as the high rate of releasing open source contributions by many internet companies or the intrusion of advertising companies into smartphone manufacturing and web browser development, statistical software, fiber optic networks, Municipal Wi-Fi, radio Spectrum auctions, DNS.
8:00Google. They are preemptive attempts to commodify another company elsewhere in the stack or defenses against it being done to them.
8:11Um.
8:14And then he goes on. X Microsoft product manager George Bosque, 2002 strategy letter. 5 the economics of Open Source discusses a patternly saw in technology companies, software, and Microsoft in particular. While he does not cite, it's likely he was influenced by Carl Shapiro and Google economists held variants bestselling 1999 technology economics book information rules, a strategic guide to the network economy, which discusses the idea extensively.
8:45A definition. Make yourself a monopoly by growing the markets around you. I Read this passage in the original (opens in a new tab)
8:51accept the definition and examples below. Emphasis original and mostly most links added. Okay, so I think now is an excerpt for Charles polskis. Articles, so I would say, let's switch over to. Um. To the strategy data and then start.
9:13Thinking about this a little bit. So, June 12, 2002 by Joel sposky, strategy letter number five. Tax CEO Tech news.
9:23When I was in college, I took two intro-economics courses, macroeconomics, and microeconomics. Macro rule of theories, like low end unemployment, causes inflation that never quite stood up to reality, but the microstaff was both cool and useful.
9:40It was full of interesting Concepts about the relationships between supply and demand that really did work. For example, if you have a competitor who lowers their prices, the demand for your product will go down un. To match them. In today's episodes, I'll show one of those Concepts explains a lot about some familiar computer companies.
10:02I show how along the way I noticed something interesting about open source software, which is this. Most of the companies spending big money to develop open source software are doing it because it's a good business strategy for them, not because they suddenly suddenly stop believing in capitalism and fell in love with freedom as in speech.
10:22Every product in the marketplace has substitutes and complements. A substitute is another product you might buy if the first product is too expensive. Chicken is a substitute for beef if you're a chicken farmer and the price of beef goes up. The people will want more chicken, and you will sell more.
10:40A compliment is a product that usually buy together with another product. Gas and cars are complements. Computer hardware is a classic complement of computer operating operating systems. And babysitters are complement of the night fine restaurants. In a small town when the local five-star restaurant has a two. For one, Valentine stay special.
11:00The local babysitter double their rates. Actually, the nine-year-old get robbed into early service. All else being equal. Demand for a product increases when the price of its complements decrease. Read this passage in the original (opens in a new tab)
— Strategy Letter V: The Economics of Open Source, Joel Spolsky
11:17So? I guess very? Definition of this feature.
11:26A compliment being something that is required for your Pro you to sell your product, right?
11:36So, if you think about? If you're a programmer and?
11:45A new program you want to program something? Maybe your compliments are the libraries you use or the operating systems you are running on? Or, if you are? Building a website like Google does our search engine, then the complement is probably the thing that is going to display your website and.
12:05Um, the internet providers that are gonna transfer your website to the customer.
12:13And if you're in the ad business and you? You notice that there's going to be a lot more videos uploaded to the internet, and you really, really feel like there's going to be a big slice of the internet. It's just being videos. People consuming video material. Instead of reading blog posts like, we are right now.
12:35Then, you probably want to buy a company that's gonna. Provide the infrastructure for serving video on the internet, because that's your compliment, right. You want to serve advertising on videos, so you need to be inside of the video delivery process.
12:57Or maybe, if you, if you're advertising business, you go as far as saying, oh. But Broadband internet is important for people to use. And if other providers are starting to, like, then you complement is gonna be either. Um. Laws that make delivering content like? Being specific and who you allowed conductibility to be levered by, like, net materiality.
13:28Very important for you because? Um. That's your your lifeline. Like, if you are Google or Facebook and net neutrology doesn't exist, and suddenly, maybe some other competitor can come into. Um, can, like, buy up. The domain provider. Um. And say, huh? If you serve my, if you prefer my packets over Google's packets, then I'm gonna pay you directly.
13:58And suddenly the market for advertising is. Is under attack. So, like, basically. And there's like a.
14:12Like, somebody can drive a wedge between what you're selling and who you're selling to and how you're selling it, right? That's like the the downstream side of things. And then there's the Upstream side of things. Like, what are you? What do you require to even build your product in the first place?
14:35So, there's gonna be a lot of examples later. But I think it's very important, especially for software development developers to think about economic reasons for things like browsers being open sourced or. Um, steam valve going into Linux as. Their operating system, as opposed to Microsoft Windows. Um. Or also, like why social media sites?
15:06Two things they do, why they do things the way they do them. Um. And why Microsoft will suddenly publish lots of Open Source, but like, even by companies that they just open source afterwards, right. Like, GitHub is probably something related to that which I haven't really thought. About. Um.
15:29Maybe it's a different thing actually?
15:33Okay, let's go on. All else being equal. Demand for product increases when the price
15:39of its complements decrease. Let me repeat that because you might have dosed off and it's important demand for a product increases when the price of its complements decrease. For example, if flights to Miami become cheaper demand for hotel rooms in Miami goes up because more people are flying to Miami and need a room.
16:01When computers become cheaper, more people buy them, and they all need operating systems. So, demand for operating system goes up, which means the price of operating Read this passage in the original (opens in a new tab)
16:10system can go up. So, but what if somebody comes along and says, ha? Demand for mobile phones is gonna increase, right? Because everybody's gonna want to have a mobile phone like an iPhone?
16:23Um, but Apple has got all the best mobile phones. So, everybody's buying into Apple's close ecosystem? And also, if we if suddenly somebody starts building mobile phones and somebody else comes up with an operating system. Then, we won't be able to sell our advertising on the apps in the operating system.
16:42So, what if we instead? Build the middle layer if we start building Android and. Open source, so there's like no market for anybody else to go into the into the business of selling operating systems for mobile phones, and also we can bind all of the hardware producers to use Android because it's the only.
17:03Readily available platform that can be used, and that also has all of the search engine stuff already built into it so. So you can see how the thinking at Google goes if when they, when they notice that, oh s***, the mobile phones are coming, and how do you, how do we react to that, right?
17:27But also. There's this law that everybody is talking about right now, which? Of course, I forgot already. What? What the name was? Uh, like it starts with G. Um.
17:44I will, I will. I will remember not in a bit. Which is basically that there's these kind of. Unlocks of? Um. Unlocks of of a price point where something becomes cheap enough that new cat Loop new kinds of categories of usage. Um, suddenly become. Available rights. So, for example, the the internet isn't like the the public internet wasn't possible before.
18:12There was a widely distributed. Um. Phone Network, obviously, because you needed modems, and then you needed a widely disputed personal computer in people's home. And then suddenly the internet becomes possible outside of University servers. And that also makes online advertising possible and. And you know, there's these. These unlock unlocked points where something becomes cheap enough that people.
18:40Move into it almost and then. They can discover new. Possibilities. Or, for example, Netflix is only possible. Um. Because of high bandwidth. In most developed countries or even like 4. And Broadband's mobile internet.
19:06Which is also only possible because? You get these. Flat per month payments for unlimited, mostly unlimited bandwidth and not metered by the. Microsecond as before.
19:31So basically. If something becomes cheaper, the demand goes up. But also, what I'm saying is that at certain points, the demand goes up more than linearly, like. But there's unlock points that make things. Basically. Explode. Because. For example.
19:54Um. I don't know home fitness trainer. Nobody would buy a home fitness trainer for 2 and 1/2k, but once a home fitness trainer costs under 300, suddenly people will start buying them. Maybe I don't know, probably not because it's kind of stupid. Um.
20:16Uh,
20:16at this point, it's pretty common for people to try to confuse things by saying, aha. But Linux is free. Okay, first of all. When an economist considers price, they consider the total price, including some intangible things, like the time it takes to set up, re-educate in everyone and converting system processes.
20:34All the things that we like to call total cost of ownership. Read this passage in the original (opens in a new tab)
20:39Also, often it's a mode around existing products like Microsoft object office has. These modes, like they do something that is not like that, is very specific to their. Um, their own products and. Before anybody can migrate to something else.
20:58The other solution has to implement basically all of their feature set so that all the possible files can be. Um. Executed by this other software, right. Which is why it's so hard for people to migrate away from Microsoft Excel or to LibreOffice or whatever. Um, but then, of course, you get these.
21:20Good enough solutions, which is like for 89%, 98 percent of all the use cases. A Google sheet is enough for simple. Spreadsheets and the extra functionality of being able to share, like, live. Edit multiple people in the same. Document means there's. Enough of an advantage to using that that people will switch away from Microsoft Excel.
21:46And. Towards the Google sheet stuff.
21:53Which also has this interesting corollary that just re-implementing something s*****? As a free, open source software is often. A losing battle because? Um, the existing companies have the money the existing companies have the. The advantage of being probably earlier in the game, of course. The second mover is always a bit easier because they can just re-implement everything.
22:19But. Um, which is not true when I'm talking about LibreOffice, because that was OpenOffice, which was before that. Star office, I think, which is a stone all piece of software from the early 90s. Um, but I mean, you get the idea, right? Like, this person can Implement something pretty quickly.
22:41And just copy the features, but then it will be very hard to get people to move over from the paid for product once they're already. Are sold on edge. Versus if you have some other product, which is a bit less than the original, but has some super cool extra feature.
23:01Maybe that will be enough to make people move over, especially the people that are not experts. Like, you won't be able to convince somebody to move from whim, like their favorite editor or IDE to some other IDE, unless you give them a really good reason why they would move.
23:18Because they're so settled in their own ways, like the women people will always have their own motion key bindings and and their whim config, which they probably already had spend a thousand hours configuring. And the EMX people are the same. So? If you want to sell a new product and you really like vs code, you really have to make sure that it has.
23:38Unique advantages, like ease of setup or total provider integration or plug-in ecosystem. Trump's these existing? Um, solutions.
23:54Okay, I'm totally.
23:59Off track here. The article still goes on on a little bit. So, we were about. Oh, Linux is free, but of course, it's not free because people work on it. There's intangible stuff, like, how long it takes for people to migrate from the win32 IPI to some unique space API?
24:16Like, like all of this friction points, so, so the total price of using Linux, instead of some other operating system, is a lot more than? Oh, well, it's free. Secondly, by using the free S and B
24:30argument, these Advocates try to believe that they are not subject to the rules of Economics because they've got a nice, nice zero.
24:36They can multiply everything by. Here's an example when slash dot asks Linux developer mosha bar, a few children's comments would be compatible with existing device drivers, he said. They didn't need to proprietary software goes at the Tariff of U.S dollars 50 to 100 per line of debug code. Okay, that's the wrong way to read that 5200 US dollars line of Riverbank code.
25:00No, such price applies to open source software. Moshi goes on to claim then. It's okay for every Linux kernel revision to make all existing drivers obsolete because the cost of rewriting all those existing drivers is zero. This is completely wrong. He's basically claiming that spending a small amount of programming time, making the kernel backwards compatible, is equivalent to spending a huge amount of programming time, rewriting every driver because both numbers are multiplied by their cost, which he believes to be zero.
25:32This is a primer for C fallacy. The thousands of millions of developer hours it takes to revise every existing device driver we're going to have to come at the expense of something, and until that's done, Linux will be once again handicapped in the marketplace because it doesn't support existing Hardware.
25:51Wouldn't it be better to use all those zero cost effort, making gnome better Read this passage in the original (opens in a new tab)
25:55for supporting new hardware? This is funny. It's very funny because wouldn't it be better to use this to make gnome better? Like, say that in 2002 is, like, oh oh, you're such a like. Sweet, my sweet summer child.
26:17Ignore me is exactly the thing that you don't want to put as an example here because they tend to rewrite all of their f****** libraries every. Okay, sorry, I'm swearing it like there. Just rewrite everything every 5 years, and everything else breaks like this. I think I currently have, like gtk 2, 3, and 4 on this system.
26:36And they all behave differently, and like dark mode works on gtk4, but not on 5, which means that some applications switch to dark mode when I switch, but then some don't because somebody decided they it's better to break backwards. Compatibility to get some new fun features. Tend to support old software, I guess.
26:57So, yeah, gnome is the wrong example here because they exactly have this. They, they don't behave like a homo economicals they behave like. Some random guy has some ideas and more time to implement it than other people, and then just doesn't give a s*** about everybody else.
27:20Yeah, of course. This is personal. Personal gripes.
27:27So, yeah, this is basically don't break compatibility. If you don't have to add any cost? Oh well, I guess. The actual answer would be don't break compatibility until you've done the math and figured out how much value you're destroying by destroying compatibility, which I guess nobody does in the open source ecosystem because they all think economic economics is for stupid people and.
27:56Anybody with a business degree doesn't actually know anything? Which is very funny. It's very funny, like in a sad way. But I guess that's what this is for as well. Debug
28:08code is not free, whether proprietary or open source, even if you don't pay cash dollars for it. Read this passage in the original (opens in a new tab)
28:13This is also important for the AI discussion, like, just take an economic view on this, and like.
28:23The backcode is not free. Weather propriety or open source or operators. I might add
28:28even if you don't pay cash dollars for it. It has opportunity cost, and it has time cost. There's Read this passage in the original (opens in a new tab)
28:34a finite amount of volunteer programming handling Talent available for open source. Okay, so this call paragraph.
28:41Right, this? Not outdated, just the whole equation changes, right? So right now, you get this. You get this argument by many people, which. Oh, we should don't use AI. Your software will be some. You will think through everything way more. Way more strongly, you will. Not go down bad paths as much because?
29:10Um, you will have thought about what you want to implement a lot. And so I, I can speak from experience. That's just not true. Like, I spend so much time on stupid little projects that. That I had to implement a lot of code for. And then, after five years later, I'm.
29:28Why did I spend so much time on this? I could have done so many other things in that time, right? I can say, oh, girls. I learned something by doing that. Like, I learned this program language or this framework or whatever. But if you think about it, you could have, like?
29:42I could have created a company or maybe also have. Had other things to learn in the in the meantime, right? So, so the opportunity cost thing is really overlooked. And that's also because? It's. You know, it's personal for most people like they spend a lot of time on this stuff.
30:02And when somebody tells them, oh, you could have done like five other things in the same time, and you would have learned like 10x more. If you had tried out more things? Because you had this machine that will just give you the ability to experiment a lot faster. They would probably be, like, huh, why did I waste all this time, right?
30:23Which is also it's a spiritual thing, but? Yeah, just enjoy things as they come and don't.
30:35Don't cry over lost. Potential potentiality, but also don't be stupid, like. Like, unless you really have fun writing code by hand, maybe to the? Like, maybe? Like, do the math and think. Is this really worth it? And then maybe do something and go. Make something from wood or something instead.
31:03Touchgrass. Okay, so back, there's a final. There's
31:08a finite amount of volunteer programming Talent available for open source work, and each open source project competes with every other open source project for the same limited programming resource, and only the sexiest projects really have more volunteer developers that they can use.
31:22Summarize I'm not very impressed by people who try to prove wild economic things about free as and beer software because they're just getting divided by zeros errors. As far as I'm concerned. Open source is not exempt from the laws of gravity or economics. Read this passage in the original (opens in a new tab)
31:39That's something I'm realizing more and more that, like.
31:43Things that happen. Like, okay, I shouldn't say things so much, like many.
31:50Observable. Symptoms that you can. Like Intuit, maybe? Actually speak to some. Maybe hard to hard to actually. Um. Put numbers on, but also like, deeper. Fundamental things that? A laws of nature in a way that is more. A lot more vague or. Um, cloudy, then nebulous. Oh, gravity! Towards the Gravity center, the center of gravity, right or, like, um, these are the formulas for space-time.
32:36But they're. I guess they're economic, right? So, so you have these chaotic systems, and you want to figure out what the? Um, what the trends are? What happens if you change these one or two or maybe five parameters and maybe?
32:55Um, get an intuition about how things interact with each other. And like this. Oh God, that's I think it was called courthouse now. No.
33:10Huh? Maybe not. God wins law.
33:21No, that's the Nazi one. Go to Winslaws the Nazi one.
33:30God has law. No, that's the target one.
33:39That's also very important is. Um. Whenever something becomes a Target, it will be games and can't be used as a measure anymore. Are we seeing that a lot? A Peter Principle. Mcnamara fallacy. I, I can't. I can't find the law right now.
34:12Words law de Morgan's law? No, those are all, not the laws I'm looking for.
34:25The Germans Paradox. Now, we now we are at the right one.
34:34So, instead of leading to a fall in consumption, if you improve something, it will lead to a rising consumption. But I. In like? Not make that? Maybe like a linear, or like, maybe? Um, logarithmic curve. I would make that into, like a step function, right? So there exists.
35:01On the curve of something getting cheaper where suddenly adoption explodes because a whole new use case is being found that wasn't possible before and suddenly it's possible.
35:14Like, for example, roads being paved over right. Something I learned today is that? Bicycles, bicycles, cyclists, were the people who first argued for Paving overalls because. Um, yeah, it's. It's a slog to drive a bicycle on unpaved road, so basically, this were the ones that set. Now, we should.
35:39We should probably pave all of the city Interiors. And then, suddenly, bicycle user exploded. Which? Cars possible, right? And then cars replace bicycles mostly, but that was a different thing. Um. Open source is not exempt from the laws of gravity or economics economics economics. We saw this with easel, our stitcher,
36:05the company formerly known as VA Linux and a lot of other attempts. Read this passage in the original (opens in a new tab)
36:08This is, this is probably a history lesson because. I mean us digital. Tells like I I, I've heard the name, but I have no idea what easel is. I have no idea what the company formulated known as VA Linux. This, but I guess the pattern is the same something is going on, when,
36:27which very few people in the open source world really understand a lot of very large public companies, which with responsibilities to maximize shareholder value, are investing a lot of money and supporting open source software, people paying large teams of programmers to work on it.
36:44And that's what the principle of complements explains once again. Demand for a product increases when the price of its complements decreases. In general, the company's strategic interest is going to be together the price of their compliments as low as possible, the lowest theoretically sustainable price would be the commodity price.
37:03The price that arises when you have a bunch of competitors offering indistinguishable Goods. So, Read this passage in the original (opens in a new tab)
37:08that's what I was talking about. At the very beginning, it's like a commodity is a thing. That is marginal price very low because lots of people are competing for selling the most the most of it, right.
37:23And also, it's very exchangeable, so if you sell it, somebody else says they're gonna try to make a better price for it. Create like a better supply chain so they can sell at a lower price than you. And then you have to. Try to. I'll compute them again. Which drives the prices down until things are very commoditized.
37:48So, for example, in Germany, we have these Supermarket chains like Lidl and Aldi. And they're in a bitter fight right now. For butter prices of all things? Which means they're really, really, like the bother prices are really in the. Um. How do you say that they're like? Yeah, you can buy a lot of butter for a few Euros.
38:13Um. Which is probably also because they cross Finance that from the other products, but? Basically, I want. I don't actually know why they do this. Because, for example, in France, you have the opposite that you have a few giant giant. Grocery chains that basically all collude to just set the prices at very high.
38:37Best prices so? Your butter is gonna cost three $4 per 500 G instead of 150 as it currently costs. In Germany. Or, for example, another example would be the the mobile phone providers in France. Also colluded to have. Very, very s*****. Monthly, like gigabytes per like euro per gigabyte.
39:06Um.
39:10How do you say like their contracts were super s***** and then one billionaire came along and built this call company called free, which? B**** outbid, Everybody by, like, 100%, so it was like 50%. Like it was half of what other people, what other providers gave you for the same money, and then everybody else had to start.
39:33Pulling. Like lowering the own prices and improving the offerings which started. Cycle where everybody outbid the other company and? And the process went lower and lower, and nowadays is basically commoditized, like, doesn't really matter where you go or the they are. They have very low margins for this stuff.
39:52And you basically pay a nice price for mobile phones. Mobile phone contracts. Which is not the case in Germany at all. I think it might have started to happen because you. Basically forced everybody to allow roaming in all of the use, so you can just have a contract in another country.
40:15And use it in Germany as before, as if it was a local one, which means German providers are suddenly forced to actually compete with with each other, and like with all of the other providers in Europe to a certain point. Which is why prices are suddenly. Turning into commodity prices, which is where you want to go.
40:34If you are like? Near liberal ecosystem, but for some reason it doesn't always happen.
40:45So? Smart companies tried
40:48to commoditize their products complements. If you can do this, demand for your product will increase, and you will be able to charge more and make more. When IBM designed a PC architecture, they used off-the-shelf Parts instead of custom parts, and they carefully documented the interface between the parts evolutionary.
41:06IBM PC technical reference manual. Why, so that other manufacturers could join the party as long as you match the interface? You can be used in PCS. IBM's goal was to commoditize the add-in Market, which is a complement of the PC market. And they did this quite successfully. Within a short time, squillions of companies sprang up, offering more memory, offering memory cards, hard drives, graphics cards, printers, Etc.
41:32Cheap add-ins meant more demand for PCs. When IBM licensed the operating system PC dos from Microsoft, Microsoft was very careful not to sell an exclusive license. They made it possible for Microsoft. This made it possible for Microsoft to license the same thing to compact and the other hundreds of oems were legally cloned the IBM PC using IBM's own documentation.
41:56Microsoft's goal was to commoditize the PC market very soon. The PC itself was basically a commodity with ever decreasing prices consistent. Increase in power and fierce margins that make it extremely hard to make a profit. The low prices, of course, increase the amount increased. Demand for PCS meant increased demand for their complement and the stores.
42:19All else being equal. The greater the amount for a product, the more money it makes for you. And that's why Bill Gates can buy Sweden, and you can't. Read this passage in the original (opens in a new tab)
42:33So, this is also why valve is trying to create. Headsets and why they create like? Based on Linux operating systems. And why they release. Um, the steam box, which is like, basically, just a PC with their Steam OS on it. And, like, very nice small form factor from form factor.
42:58Because they make money selling games on Steam and like making it so that their steam platform is taking 30 percent of everything that people sell. Um. So everything that increases the? Pc gaming Market basically is a win for their for their re. And also their profits. Um, which is why they sell those steam boxes then.
43:24They don't actually sell them for under value. They sell them for exactly what they, what it cost them to, make them, and also plus like a little bit of modern. Which is why they are so expensive, but also other people are going to come around. Other companies are gonna start creating their own steam boxes, which is already happening, and then everybody can just install Steamworks on their devices and make their own steam box.
43:46So, this is like really increasing the margin of the steam ecosystem that the the value of the steam ecosystem. And there's like, no real reason for people to use any other platform at this point. But also they don't care if people like if it's possible that people install like the.
44:06The EA platform or their the the Gog platform on the steam box because they know that, like, they have the? Vast Network Advantage right now! So their share of the pie is gonna grow faster than other platforms share of the pie and the whole pile is gonna grow, right?
44:24So, as long as the whole pie can grow? Um. Consistently, and also maybe the slice of the pie grows. Um, that's a win for them.
44:41This year, we're in 2002. Remember, Microsoft's trying to do it again? The new game console, the new game console, the Xbox like history Rhymes, right valve, is realizing the new game console, which is the steam box in 2026. In 2002, Microsoft is doing it.
44:59The Xbox uses commodity PC Hardware instead of custom parts.
45:04The theory. Was that the commodity Hardware gets cheaper every year, so the Xbox could write down the prices? Unfortunately, it seems to have backfired. Apparently, commodity PC Hardware has already been squeezed down to commodity prices, and so the price of Mac in Xbox isn't declining as fast as Microsoft would like.
45:24The other part of Microsoft's Xbox strategy was to use DirectX the graphics Library that can be used to write code that runs on all kinds of video chips. The goal here is to make the video chip a commodity to lower its price so that more games are sold where the real profits occur.
45:42And why don't the video chip vendors of the world? Try to commoditize the games somehow. That's a lot harder. If the game Halo is selling like crazy, it doesn't really have any substitutes. You're not going to go to the movie Read this passage in the original (opens in a new tab)
45:55theater to see Star Wars attack on the of the Clones, and this, oh man, Star Wars attack of the Clones was was current, like was current in 2000, that's so crazy.
46:06You're not going to the movie theater, you
46:08see, Star Wars attack of the Clones and the side. Instead, he would be satisfied with a Woody Allen movie. They may both be great movies, but they're not perfect substitutes. Now, who would you rather be a game publisher Read this passage in the original (opens in a new tab)
46:20video chat vendor?
46:23So, this is something that big entertainment. Like, in the meantime, big entertainment has really tried to make this happen. Um, like you can, probably at this point, replace most Netflix. Drama series or whatever series with a series from the same kind of. Jeong, and you will have a perfectly mediocre.
46:47B******* experience, and I guess most people are happy with that. Or you can. I mean, you can go to any kind of superhero movie and you will receive a perfectly Bland. Default experience, or you can watch any kind of Star Wars series game. Lego Star Wars whatever and you will also get this kind of.
47:10Star was feeling from it, so they are really trying to make these things Commodities. And I think, because it's Arch, they're gonna it's gonna backfire, like they're gonna. Yeah, they're still making a lot of money from this, but People are getting really sick of this kind of trying to.
47:28Commoditize art. And especially with AR with AIR. Now people are going to be. Yeah, super linearly. Be annoyed by this, and I think the newer generation is already going back to trying to find the interesting. In this stuff and maybe not spend all of that money on. Yeah, AAA quadruple A whatever.
47:56One can hope. But yeah, so like some things are not easy to commoditize. Like big entertainment. Complexes are always trying to do that, like with rock stars or with MTV stars or whatever, but Um, the pendulum always swings back pretty hard, I think. Whereas with the technology, it's a lot easier to.
48:20Create Commodities like? Whether company a or Company B supplies with your aluminum is probably doesn't really matter to you. Um. And I mean the ultimate Commodities energy, right? This electricity, which is why? Solar Revolution and the EV Revolution and the Battery Revolution are so important. And the the obvious Next Step, but it's like, it's not even that.
48:50It's happening because it's the right thing to do in scare quotes, but it's and well, because the work liberers want you to do it. But it's just happening because? There's like economic incentives to do it, and it's happening despite the best efforts of. Big oil and stuff. Try to avoid this.
49:12Which is also why Trump? Mr Donald Trump first mentioned EPE is the. Accidentally. Um, help climate change by invading Iran.
49:26Which is kind of funny if you think about it, but she's probably like his best. Um. Yeah, this Invasion was probably the best thing to happen to the globe, and it's probably more important than many other presidents. In a positive way, even although he's obviously a b******* president, but?
49:47This probably helps. More than people realize.
49:56Yep.
49:59In it by by complete accident, of course. Okay,
50:04headline IBM spends Millions to develop open source software myth. They're doing this because Lou justner, read the new Manifesto and decided he doesn't actually like capitalism reality. They're doing this because IBM is becoming I.T Consulting. Company ID Consulting is a complement of Enterprise software. Thus, IBM needs to commoditize Enterprise software and the best way to do this is by supporting open source lo and behold.
50:31Their Consulting division is winning back with a strategy. Headline Netscape open sources their web browser. Oh, this is ancient history myth. They're doing this to get free source code contributions from people in cyber cafes in New Zealand reality. They're doing this to commoditize the web browser. This has been Netscape strategy from day one.
50:52Have a look at the very first Netscape press release. The browser's freeware Netscape gave gave away the browser so they could make money on servers. Browsers and servers are classic, complements the cheaper Read this passage in the original (opens in a new tab)
51:05to actually browsers and Advers are complements, but I mean guess back then, people thought that browsing server was a compliment, which didn't really work out.
51:13Which is, I guess why Google won out? Because they realized around the same time that. A search engine can make a lot more money.
51:24The cheaper the browsers, the more servers you sell. This was never as true as it was October 1994. Netscape was actually surprised when MCI came to the door and dumped so much money in their Labs that they realized they could make money off the browser too off of the browser too.
51:40This wasn't required by the business plan. When Netscape released Mozilla as open source, it was because they saw an opportunity to lower the cost of developing the browser so they could get the commodity benefit benefits as a lower cost.
51:58Later, AOL Time Warner acquired Netscape, the server software, which was supposed to be the beneficiary of commodity. Browsers wasn't doing all that well and was jettisoned now. Why would AOL Time Warner continue to invest anything in open source? It's an entertaining entertainment company entertainment companies, so I was talking about them are the complement of entertainment delivery platforms of all types, including web browsers.
52:24So, this is the Netscape, the, the not the Netscape, the Netflix story, right? Including web browsers. This joint conglomerated strategies interest strategic interest is to make entertainment delivery web browsers, a commodity for which nobody can charge money. Read this passage in the original (opens in a new tab)
52:44And then Internet Explorer happened, and then everything was blocked for about 10 years.
52:49Yeah. So you can still? Hindo that plan by having a product that is widely used and stagnant. Like Microsoft did, which is how they made Windows happen until Windows Vista happened. Yeah, so it's like it's a struggle all the way down, right?
53:13So, it's interesting to read these kind of articles 20 years later. And then have the actual how these things turned out. The the actual history behind this? And then compare it against the arguments. Um. And he goes on.
53:28My argument is a little bit tortured by the fact that Internet Explorer is free as in beer.
53:32Oh, he was talking about this. Microsoft wanted to make web browsers a commodity, too, so they can sell desktop and server operating systems, so you can see here that two conglomerates are trying to compete for the same ish commodity, but for different reasons. They went a step further and delivered a collection of components which anyone could use to throw together web browser.
53:53New planet, AOL, and Juno use these components to build their own web browsers. Give, like currently everybody's using chromium. As a foundation to build their own web browsers. Giving, given that IE is free. What is the incentive for Netscape to make the browser even cheaper? It's a preemptive move they need to prevent Microsoft getting a complete Monopoly in web browsers, even free web browsers because it would, theoretically, because that would theoretically give Microsoft an opportunity to increase the cost of web browsing in other ways, Say by increasing the price of Windows. Read this passage in the original (opens in a new tab)
54:29And notice Windows is completely free, I think. Because it's now the commodity. Two other things, which is probably.
54:42It's not the server. I mean, it's also like Azure, but. What is Microsoft Windows, the commodity tool that they're trying to commoditize? I guess it's like a mode for people to stay on. Because Linux on a desktop which is? Yeah, but switch, like, oh, it's okay, so? Microsoft is also trying to sell advertise.
55:07I mean, they're kind of. They don't really know where they're going, right? So, they're like trying to sell advertising in the start menu. They have Bing as their web search alternative to Google. Um, they have Linux on the desktop trying to take their cake and it's kind of doing that right now.
55:27With valve for the gamers and with, like? Other people by that are building distros for power users, and I mean, I and Ubuntu for my grandmother 10 years ago, and it was perfectly fine, and she had enough. Capabilities to write her books on it, and the printer worked so like?
55:52There wasn't really. There wasn't really a need for Windows even like a decade ago. Unless you're doing like 3D power CAD software stuff for like Final Cut Pro, but then you were using Mac OS anyway.
56:13Yeah, so I mean Windows is is a commodity at this point and? Some and something that's waning, I would say.
56:24But that's from my own point of view, who has been using Linux on the desktop for 15 years plus. Um and I only use Windows for gaming and it's fine-ish. I don't know. I don't even have it logged onto my Microsoft account even though they try to make me every three days, but Yeah, as long as they can click away.
56:49Please, please finish setting up your PC thing. It's okay.
56:56Okay,
56:57my argument is even more shaky because it's pretty clear that Netscape in the days of Parks stale didn't exactly know what it was doing a more likely explanation of what Netscape did. Is that up of management was technologically technologically inept, and they had no choice but to go along with whatever scheme the developers came up with.
57:14The developers were hackers, not economists, and only coincidentally came up with a scheme which served their strategy, but let's give them the benefit of the doll, I mean. Netscape doesn't exist anymore, so they clearly didn't really have a plan. Headline transmit a highest linels pays them to hacker Linux myth.
57:35They just did it to get publicity. Would you have heard of transmitter otherwise, reality transmitter is a CPU company, the natural complement of a CPU is an operating system. Read this passage in the original (opens in a new tab)
57:45Transmitter wants os's to be a commodity. Headline sudden HP Peg Simeon to hakong gnome myth. Sonich peer supporting free software because they're light bazaars, not cathedrals.
57:57Also, like they open source star office into LibreOffice. I don't know how that happened, but I think there was also Sun, so guess the same move for Microsoft Office. Reality. Sun and HPR Hardware companies. They make boxing
58:10in order to make money on the desktop they need for windowing systems, which are complement of desktop computers to be a commodity.
58:19Why don't they take the money and they're paying Simeon and use it to develop a propriety proprietary windowing system they tried? This sun had news and HP had New Wave, but these are really Hardware companies at heart with pretty crude software skills, and they need windowing systems to be a cheap commodity, not a proprietary Advantage which they have to pay for.
58:40So, they hide the nice guys as Simeon to do this for the same reason that sunboard star office oh, here we go about star office. Open source to commoditize software and make more money on Hardware. Read this passage in the original (opens in a new tab)
58:53Also, this is. Kde with Nokia, I guess. Who, which? Yeah, I think that was their strategy for.
59:06Mobileish phones KDKD? But yeah, KD is probably the same for some other for some other company. I mean, in Java, right, headlines,
59:18Sun develops. Java new bytecode system means write once, run anywhere. The bytecode idea is not new. Programmers have always tried to make the code run on as many machines as possible.
59:28That's how you commoditize your complement Read this passage in the original (opens in a new tab)
59:30for make for years micros, but that's also C. C is one of the most. Commoditizing their complement language, which is, it is an abstract. I mean, it signifies a machine that just can run. Somewhat efficiently everywhere, right? And because of undefined Behavior, you can make it.
59:54Pretty efficient on most systems, even for weird bite size and stuff.
1:00:01For years, Microsoft had its own P code compiler and portable windowing layer, which let Excel run on Mac, Windows, and OS 2, and on Motorola, Intel, Alpha mips, and power PC chips. Quark has a layer which runs Macintosh codes on Windows. The C programming language is best described as a hardware independent assembler language.
1:00:20Oh, here we go. It's not a new idea Read this passage in the original (opens in a new tab)
1:00:22in too software developers. Also wine on Linux and. No WSL Windows. What's called the thing that makes Linux apps run on Windows and also the whole Unix tool chain run on Windows?
1:00:38If you can run your software anyway, anywhere that makes Hardware more of a commodity.
1:00:43As Hardware prices go down, the market expands, driving more demand for software and leaving customers with extra money to spend on software, which can now be more expensive. Sons enthusiasm for wara? Whatever that was, is, I'm strange because sun is a hardware company making Hardware commodity is the last thing they want to do, oops.
1:01:08Sun is the loose Cannon of the computer industry, unable to see past their raging fear and loathing of Microsoft. They adopt strategies based on anger rather than self-interest. Son's two strategies are a make software commodity by promoting and developing free software. Star offers Linux Apache ignom or Apache was something.
1:01:32Np make Hardware commodity by promoting Java with with its bytecode architecture and vora. Okay, so on Pop Quiz. When the Music Stops where you're going to sit down without proprietary advantages in Hardware or software, you're going to have to take commodity price. The commodity price, which barely covers the cost of cheap factories in quarter whatever some Indian Town.
1:01:58Offices in Silicon Valley. But Joel Jared's son is trying to commoditize the operating system, like transmitter, not the hardware, maybe. But the fact that Java bytecode also commoditizes. The hardware is some pretty significant collateral damage to sustain, Read this passage in the original (opens in a new tab)
1:02:15which I guess explains why Sun went down and like got sold to Oracle for scraps and scrap metal and.
1:02:22Whatever they did to Java after that. Um, Sunset, it, I guess. Um.
1:02:33I had a thought about this a few seconds ago. My Hardware commodity, oh yeah, AI. So, lnms lms don't have a mode. Like, I don't understand, like? It's pretty obvious to anybody who knows how these things are not work under the hoodest. You just need a lot of compute power, and that basically, and a few people that you can buy for like a million a year.
1:02:59Just by paying more than your than the other companies. So, you just need a lot of investment money, and you can spin up a company, and also there's a lot of very smart people in China that will do it for the stage. So you get deepsea, you get Kimmy, you get whatever Alibaba Cloud?
1:03:17Um, you get rock.
1:03:22So there's like zero modes. Basically, anybody that can type text to to some, like, it's a pipe text into curl remote server returns Json Affair. And it's like, very simple to implement technically, and that's why everybody is doing it right now. Also, you can. Quantize these models to fit on small commodity Hardware already.
1:03:51Like, you can run some models on. Um, Mac minis or on your MacBook or your newfangled arm or AMD chips system on on a chip. So, yeah, so these things are like they have zero modes, so they're they're already Commodities, and they already have been Commodities for, like, at least two years.
1:04:15So, this whole hype is basically. Either. They're gonna figure out how to make. The Trump Administration create laws for them to have artificial, like how software patterns work or how? The law system works is basically create artificial modes that are not. Inherently there by abusing the society societal system.
1:04:44Um, to make things. Economically viable, which is probably what entropic is trying to do or GPT, or like open AIS trying to do. But then they won't have success because the Chinese are already releasing these things.
1:05:06Three basically to download, so this this is not going anywhere, and I feel like. I have no idea what this is, like the only people who are winning are the hardware producers basically and even that. Like, because of llms. Like the the advantage that Nvidia had with their proprietary whatever they it was called?
1:05:29Their libraries stack basically. Isn't there anymore? They already retreated on the Linux front by saying, oh yeah, we're gonna open source our drivers and maybe put some most of it on the graphics cards, but? They're already retreating. Um, on that front, so? I feel like graphics card are gonna be com commoditized.
1:05:52There's people building Asic. Um, chips that you can just plug into your PCS that have like a burnt in model on them, so. Plug a piece of Hardware into your PC, like with the old graphics cards like, essentially, like. Like a Bitcoin mining rig, right? You plug this in, and then you have a model that you can talk to at, like 10, 100 times the speed of emulating on graphic cards.
1:06:19So there's like? Now, these things are already Commodities. You can't tell me any other on otherwise. So, I'm not sure how the economics are going to play out there.
1:06:35An important thing you notice from all these examples is that it's easy for software to commoditize Hardware. Just for a little hardware abstraction layer like Windows NT, hell, and T Health like Windows and T's hell, which is a tiny piece of code, but it's incredibly hard for Hardware to commoditize software.
1:06:52So, this is the same thing as commoditizing art, right? Like, a piece of art is a piece of art is a piece of piece of art, and you can't replace a Monet with some cheap knockoff AI generated money. It's not, it's not the same. It's like a single instance of this thing, and maybe you can make a print of the.
1:07:11Thing, but Um. Yeah. Software is not interchangeable, as staff is the star office marketing team is learning. Even when the price is zero, the cost of switching from Microsoft Office is non-zero until the switching cost becomes Read this passage in the original (opens in a new tab)
1:07:26you. And that's the thing with llm stack, the switching cost is zero.
1:07:30Like, that's why there's open code has integration of like a hundred different providers because they are already like the switching cost is or has always been zero for. These things is token and token out. Doesn't matter whether it's like a PDF or a PNG like, this is all completely replaceable.
1:07:49And maybe you get like a little degraded results, but nothing fundamentally unfixable.
1:08:00Until the switching cost becomes zero. Desktop office software is not truly a commodity. And even the smallest differences can make two softer packages, a pain to switch between, not true anymore, not true anymore. Despite the fact that Mozilla has all the features I want. I'd love to use it if it only avoid if only to avoid the whack-a-mole pop-up ad game.
1:08:18I'm too used to heading L to D to go to the address bar, so sue me one tiny difference, and you lose your commodity status, but I've pulled hard drives out of IBM computers and slammed them into Dell computers and boom. The system comes up perfectly and Read this passage in the original (opens in a new tab)
1:08:32run and run the face as if it's still the old computer.
1:08:36Hard drives are f****** magic, like spinning discs. The package you get? Was like 12 spinning discs and read heads in them. Is complete like that. These things even manage to exist in the physical world is a complete. It blows my brain, right? I have like a 12 terabyte drive right in front of me, which I haven't managed to put into a.
1:09:03A case yet. And like that, you can store this much. Data on like this little physical space is. Incredible and the amount of? Capitalist, you know, like the the amount of?
1:09:22Orders of magnitudes of improvements over this short of amount of time, like in the in the like. When this article was written, the biggest drives were like. A gigabyte, probably. And now it's 12 terabytes. It is absolutely baffling. Um, and even like the smallest. The smallest USB drives are like?
1:09:46100 gigabytes no? Which is crazy?
1:09:55Yeah, I've pulled hard drives out of IBM computers and slam them into the their computers and boom. The system comes up perfectly and runs. If it is as if it was still in your computer, but the thing is, even though these increases have been happening, the interfaces have always been the same, which means these things are commodity Hardware.
1:10:12Like, even though only like three providers in the world can build these things. There's the commodity Hardware because there could be another. Another provider that builds them, maybe a bit little bit worse, maybe only like a gigabyte instead of like, maybe only like a terabyte instead of 10 terabytes, but it's still a commodity.
1:10:33Um, a real black box, right? A replaceable thing?
1:10:39Um, Amos Mickelson, the CEO of Creo. What, what kind of company is? Korea
1:10:44told me that every employee in his firm is required to take a course in what he calls economic thinking. Great idea. Even simple Concepts in basic microeconomics go a long way to understanding some of the fundamental shifts going on today. Read this passage in the original (opens in a new tab)
1:10:58So, yeah, I mean. A lot has been happening, and everything's kind of still the same like the the principles are still there. Uh, once you know how to look for them, you're going to see them everywhere. I feel like, which is why I want to talk about this. Yes.
1:11:14Um. Thanks for listening.