Credera is excited to announce the release of our first full Technology Tangents podcast episode: "What's Going on at Meta?"
In Technology Tangents: Conversations for Leaders in Tech, Credera's Chief Technology Officer, Jason Goth, and Chief Data Scientist, Vincent Yates, bring leaders together to discuss the modern technologies of today and how organizations should navigate the implications for tomorrow.
The discussions are fun, lighthearted, and frankly opinionated, but hopefully it gives technology leaders a sense of what matters, what to pay attention to, and what to ignore.
This podcast is available on iTunes, Spotify, and Anchor FM.
On This Episode
Meta is trying to be the biggest player in the new, immersive world of the Metaverse, but will it pay off? With Meta stock dropping 25% in last month, how should business leaders be thinking about the Metaverse? Credera leaders discuss.
The following transcript has been edited for clarity.
Vincent Yates (00:00:01):
Welcome to Technology Tangents. We get technology leaders together to discuss the important tech of today and the implications for tomorrow. Our discussions are fun, lighthearted, and frankly opinionated, but hopefully it gives you a sense of what matters, what to pay attention to and what to ignore.
Jason Goth, welcome. Happy Friday.
Jason Goth (00:00:21):
I am all caffeinated up and ready to go today.
Vincent Yates (00:00:24):
Good news, because we have a juicy one hot off the press. Facebook. Meta.
Jason Goth (00:00:28):
Vincent Yates (00:00:31):
Meta, as they're now known. That's right. It's been a year, by the way. I don't know if you know this, it's been almost exactly a year since Facebook renamed itself as Meta and started touting the Metaverse. And I think many of us are still scratching our heads about that. They have not been doing great. I don't know if you just saw the Q3 earnings here. Not a great situation.
Jason Goth (00:00:51):
That's where they spent the GDP of a small nation on development.
Vincent Yates (00:00:56):
That's the one. So I want to get into that. I want to talk a little bit about what's going on with them specifically, because I think it's interesting on several dimensions. I want to get into the Metaverse and whether strategically it is the right way to approach the Metaverse from Meta's perspective and a little bit about the consequences for developing net new technology, innovative technology in general. Sound good?
Jason Goth (00:01:18):
Vincent Yates (00:01:19):
You have enough caffeine for all that?
Jason Goth (00:01:20):
I got plenty. We had breakfast this morning and I had five cups of coffee.
Vincent Yates (00:01:24):
Awesome. Good news. So just to kick things off, I don't know if you saw this, Gerstner, a VC, venture capitalist heavy investor in Meta, recently published an open letter and I'm just going to read you a couple quotes here because I thought it was quite fascinating. He says, "I'm sharing this open letter to strongly encourage Meta to streamline and focus its path forward. Like many other companies in the zero rate world in there," he's talking about the zero rate interest rate that we've just been living with for the past decade. "Meta has drifted into the land of access. Too many people, too many ideas and too little urgency. This lack of focus and fitness is obscured when the growth is easy, but deadly when the growth slows and technology changes." And it seems that the market in general agreed with them.
Jason Goth (00:02:13):
Yeah, I think that's generally true for a lot of companies, not just Meta, there's probably others companies in the news that would fall into that bucket, but I think definitely so for Meta.
Vincent Yates (00:02:24):
Yeah, absolutely. And you look at all of big tech, if you look at the Meta's cap tech stocks recently, they're all taken a hit. They're all talking about layoffs. They're all talking about massive reductions and cost cutting measures here. And just a point, when money's free, might as well go all in., When growth seems unlimited, might as well go all in. But that world is a distant memory already, it seems in a remarkably short amount of time.
And I think that the conventional wisdom and pressman investors agree that again, their core business, Meta's core business effectively hit a wall last fall. And as a result, the team... again, I'm not as smart as Zuckerberg, I'm pretend to be, but it seems like he just said, "Eh, don't look over here everybody. We're going to pivot hardcore and we're going to build this Metaverse thing. Don't worry about our core business. We're going to build this Metaverse." And renamed the whole company as a result and then dumped an incredible amount of money into it.
I mean, this is insane. And then by the way, in those quarterly earnings, effectively all the fears for investors at that moment in time were confirmed. Because again, they had a sizable mishap back throughout 2022. Really. So in short, the facts are startling. I mean, that that's the headline and that's why I want to bring it up today is a sort of realtime case study to talk about technology in general.
But in the last 18 months, Meta stock is down 77%. I know that the stock market might seem bad to everybody right now, but just for comparison, compared to its peers, the other big tech peers are only down about 19% on average. There's 77%. The price to earnings ratio, that is how much it costs to buy the stock relative to how much they're making per unit share effectively, it's fallen from 23 to nine and now trades less than half, less than half of what its peers get in a PTE ratio, we're talking about 800 billion.
That's with a B. 800 billion dollars of value has been erased, just gone. Evaporated. Again, that's a huge number. I don't know how to think about it. So I tried my best here. That's the equivalent of $23,000 every second for the past year from its peak. So since its peak, it's lost $23,000 of value every single second for the past year. So that's all startling and I think it has some really important implications for its future in some sense, certainly for investors who have been with it. But also look, I mean, I want to be fair, I want to be fair here and intellectually honest, it does also obscure the fact that their core business is still one of the largest and most profitable businesses in the world. They generate over 45 billion in operating profits last year alone. So it is still highly profitable. But the thing that's crazy is they are investing in an enormous amount of money into the Metaverse and they're really getting beat up for it. So I'm going to talk about-
Jason Goth (00:05:13):
About roughly a third of that profit quarter every quarter is going into Metaverse development.
Vincent Yates (00:05:21):
That's right. And by the way, the reason for that is really headcount. I mean people capax both from chips of course, but really from people, just again, for a point of reference, the number of employees is three times what it was just in the last four years. So four years ago to today, it's three times the number of employees. Their core business has not grown at three x's, just to be clear, the core business has grown but not nearly three x's.
Jason Goth (00:05:46):
Well, and that's the point of a software business is I can get additional sales without additional headcount. Now the additional headcount is supposed to be developing new product and service, right?
Vincent Yates (00:05:58):
That's right. Yeah. And that's exactly right. And so they've taken this business, it's already fairly mature, it was already incredibly profitable as research said. And they've tripled it again from 25,000, we're not throw small numbers here, just 25,000 people and they have 85,000 employees now. That's a lot of people, to be at 60,000 people hired in four years just for perspective. 60,000 people hired in four years. So again, back to Gerstner's letter real quick.
Last thing I'll say about him, in his open letter, one of the things he's suggested is to trim headcount by 20%. He says just right up off the bat, cut up by 20%. Again, his point is, why 20%? I say, well look, that merely takes you back to where you were mid 2021. I mean barely at all. So they've just been doing an incredible clip. Okay.
Jason Goth (00:06:45):
Vincent Yates (00:06:47):
Growing headcount, sorry. Yeah, not revenue. That's exactly right. Yeah. Revenue's actually been, again, back to the point it hit a wall last fall and it really has been sort of down into the right rather than up into the right. Head count and costs have gone incredibly up into the right, which is kind of backward to what you normally expect. So to get into why they're doing all this, again, not really a podcast about the stock market per se, but what would motivate an organization to do something like this?
I want to talk a little bit about the Metaverse then. So we've had a pod before on the other channel about what the Metaverse is, but what is the Metaverse, Jason? Yeah, look, it's really funny actually. In Q3 earnings call, Zuckerberg got this question effectively from analysts and he, I think did a very poor job personally. That's just my personal opinion. He did a poor job of describing it. I think he kind of rambled all over the place and I think it's because it wasn't super crisp in his mind, frankly. One journalist said when quizzed by analyst about the Metaverse in the Metaverse route for new opportunity, the Meta chief gave a rambling answer that investors did not like. Right? So what is it?
Jason Goth (00:07:51):
Well, I think it's a little like the old fable of the nine blind men touching the elephant. Is it nine or five? I can't remember. But everyone touches the elephant and...
Vincent Yates (00:08:01):
They're blind and can't tell.
Jason Goth (00:08:01):
And if you think what is the Meta, far out Metaverse version of that, I always think of the Star Wars where they see the image of the other person online in 3d, we have that in 2d, you do video calling all day, right? In 2d. And so let's have holograms. Those have corporate use cases and they have potentially consumer use cases around individual to individual or maybe group calling communications. I do think that a lot of that exists up to the hologram point is a very mature market.
No one is investing to create a new Zoom. There are some early stages of the hologram. You got a question, like the value of it, but it is a use case and it is an area that could be expanded into. So that's one. Another use case would be the augmented reality use cases. These are things like the HoloLens or remember Google Glass. A lot of those tend to be around training, education. You're working on something and you can get specs or instruction or put a big X, don't touch that. That, it's got electricity running through it, whatever, or telemetry, you know, remotely see what telemetry looks like and potentially even control them remotely, robotically. Again, I think a lot of those use cases are fairly well understood and many of them exist today. And the kind of collaboration or calling or corporate collaboration use cases, it's somewhat a limited audience.
Not everyone is going to use an AR solution all day. And if you are like... Well, maybe I'm wrong on that, maybe cars, my wife's new Audi has the little heads up display kind of thing. But it's a fairly mature technology and the use cases are fairly well known, I guess is what I'm saying. And they could be expanded more into it. But these are evolutionary uses and incremental uses, not revolutionary, I guess, is more what I'm trying to say.
And then there's the gaming, obviously the Halo's and Fortnite and all of these. And that could certainly be evolved and matured to do things like sell digital products, could buy clothing to wear in the game or ads. Advertisings huge, especially when you talk about Meta advertising is a huge use case and certainly there's a lot of revenue potential there. But again, those exist, Epic Games, others, they all provide that kind of advertising use case today. And so in a lot of ways you might say, well those are the use cases, there's nothing new here. It's just an evolution of those use cases. Well, I don't-
Vincent Yates (00:11:16):
I want to real clear for one second, sorry to interrupt you.
Jason Goth (00:11:18):
Yeah, go ahead.
Vincent Yates (00:11:19):
So you've said two things here. I think you're crossing them and not conflicting them, but you are crossing them, which is like there's both the technological implementation of these use cases and that's the VR stuff, the hologram, and then they're the actual use cases by industry if you will, but gaming, entertainment, training, corporate video conferencing. Is that right?
Jason Goth (00:11:41):
Vincent Yates (00:11:42):
Jason Goth (00:11:42):
And those things exist and they're fairly well understood and they have fairly mature markets around them and players in them. And that's not to say that there isn't incremental opportunity to expand and grow those things, but it's all fairly well understood. I think. Now I think what people tend to mean when they talk about Meta...
Vincent Yates (00:12:03):
Sure. Wait, just to be clear, you're saying that that's where we are today. We have these modalities, these types of technology, we have these use cases identified per industry, per type, whether be entertainment or corporate. And again, they're all reasonably mature, but that's not what here, from your perspective, that's not what actually is saying when they talk about Metaverse.
Jason Goth (00:12:21):
That's right. Well I think some people think of those things when they talk about the Metaverse, but I think what most people think about is...
Vincent Yates (00:12:29):
There's no reason to spend that much money if you already have them. That's not what Zuckerberg's imagining.
Jason Goth (00:12:33):
And that's my exactly. And that's my point. No one's going to go... What, are they going to spend 15 billion this year on Metaverse development? They're not going to improve those things. Those are well known. What I think people mean is the kind of fully immersive world, which is somewhat a combination I think, of gaming and social media, relationship and economy. It's the Ready Player One kind of idea. And even where it is user developed, it's not... The game company develops it. And there are some games even have the idea of mods that you can, but I guess the Metaverse is the fully immersive digital world that you could enter into. And that world itself would be created by many, many people. Much like the web is, right? So there's how many websites are there? Billions.
Vincent Yates (00:13:31):
I don't know, a lot?
Jason Goth (00:13:35):
They're created by individuals or different companies or individuals. It's not all produced by one entity.
Vincent Yates (00:13:42):
User generated, which by the way, we've seen a trend in the marketing domain too. Social media has created this whole new economy effectively of user generating content and ways to monetize that to make it compelling. So you're saying this will be the same from that perspective.
Jason Goth (00:13:54):
So this is the idea of the Metaverse is something like that.
Vincent Yates (00:13:58):
Well, I have good news for you by the way. You're not alone in that perspective. The CEO of ROBLOX, Dave Baszucki describes the Metaverse as the natural evolution of communication, holistically. And that's a multi-device world. Back to your point, it might be AR might be VR, it might be... He actually includes telephone specifically. He calls out telephone specifically. And that is best thought of as a better version of text, video, and voice in a way that will make us all feel more connected. Similar, is my point.
Jason Goth (00:14:29):
What he's saying is more along the lines of those other use cases I mentioned. Telephony, AR training and those kind of things as a communication medium, not as this virtual world where you have user provided games and user provided gambling establishments or user provided movies within any virtual world, virtual movie theaters, go play virtual golf, whatever. I think that's more of the Zuckerberg Meta, Ready Player One is the canonical example there.
But I guess my thought on that is that is revolutionary and it is also incredibly daunting to create. And in fact, I don't know that it could be created within decades because, and let me explain that. If you think about the internet today, it's a big common thing for people to say, "Oh, it's going to be the next evolution of the internet." The internet was big and nobody understood it. But look what it turned into. So this is going to be big, everybody's got to get on it because well, there's nothing to get on today.
Vincent Yates (00:15:45):
Sure. So to be really clear, all these modalities, all these technologies, devices, interfaces, those are part of it. All those use cases are part of it. And what you're getting at here is that those things are actually connected together. That's a critical element of how the internet works, for example, the other thing you're highlighting is that the users will generate a large portion of this content.
Jason Goth (00:16:11):
Oh, they would need to for it to become a reality.
Vincent Yates (00:16:13):
Because it's a necessary condition for this Metaverse to actually work. And you're saying the equivalent statement that everybody always goes to is of course is well, it's like internet 3.0 or web 3.0, it's the whole thing in the next iteration of this. So exactly where you're going is my question then, which is, okay, so if this is the next iteration, what can we learn from the previous iteration of the internet that would inform what would have to happen in order to create the next revolution of this internet?
Jason Goth (00:16:46):
And that's where I was trying to get to, jumped ahead, sorry. But if you think about it, the internet's been around since the late sixties and it has had a lot of evolution. The main thing I think has driven the success of it is interoperable standards. So in the seventies we had lots of different networks and they used different network protocols. Well, we had to get them all connected. That means they all had to share a protocol. We came up with things like IP, right? TCPIP is a networking protocol that can interconnect networks.
Vincent Yates (00:17:23):
So why is that necessary?
Jason Goth (00:17:28):
It's somewhat like you make a phone call with a person that speaks another language. If you speak Spanish and I speak English and we get on the phone and we don't speak user's language, what are we going to say to each other? And networks have to be able to speak the same language. That's everything from addressing, it's like, well I have a network here. What's some machines on it? I have some networks here on the other side of the world with machines on it. How do I find them? How do I send them data? How do they understand what data I sent? Make sure that they get it in the same order and all of those things. And you know, can't send it all in one at one shot. So the messages were broken up. You have to reassemble them, you have to make sure you got everything.
Vincent Yates (00:18:08):
And just to be clear, does current internet protocol, is it not adequate for that?
Jason Goth (00:18:12):
No, no, no. I think that the network internet protocols... Well, maybe adequate, I don't know. But what I'm getting at is more of the history is like you start there, you develop that and then you have to have higher level protocols. So we want web servers and web browsers. Well, they have to be able to communicate. So we need standards like DNS so that I can say aa.com and not 16.2858. And that took those naming standards, took a while and implementations of those naming standards and we had to go through several iterations of them. So then we had email, we went through several iterations of email X400, X500 before we landed on SMTP, which works reasonably well. Some would argue, I'd say no. And then we had browsers and web servers, people forget, there were many incarnations of that before HTTP and HTML that we used.
Say my favorite was always Gopher. It was developed at the University of Minnesota. If you really want to go to the Wayback Machine, go look up Gopher as a protocol. But we had to come up with a way for a browser written by any number of people running on any number of platforms to communicate with a server written on by any number of people, on any number of platforms. That became HTTP and HTML.
Finally got codified into a more standard version by Ecma, which a standards organization. And so these things happened over 30 or 40 years to get to where you have the environment that you have today. And for lack of a better term, technologists we like to say, "Oh, we do an iterative and incremental development approach." Which is a very nice way to say, "We trial and error. We try something, it doesn't work, it doesn't work at scale. And we eventually evolve." And I think that those standards have to exist for this Metaverse in order for billions of people and millions of companies around the world to interact those. But the number of those standards has an incremental explosion when you go from this 2d medium to this 3d medium. Tim Sweeney, the founder of Epic Games, has a great talk where he spends an.. I think it's an hour and 18 minutes listing out all the standards that you would need for this type of Metaverse.
Vincent Yates (00:21:23):
Just give me a flavor of some examples.
Jason Goth (00:21:25):
So for example, rendering, lighting, so where's the light and what are the shadows? All of that happens locally. On games, it's described by the server, but the rendering happens locally. It's too slow to send sure the pixels over the internet. Physics to see objects falling.
Vincent Yates (00:21:46):
The physics engine.
Jason Goth (00:21:46):
The physics engines, those are local. Even things like your shirt, if you want to render a shirt materials, what does aluminum look like? You know, wouldn't be able to describe it. You'd have to have standards that say this is how to render blonde hair, facial expressions is another example. So those are all kinds of those rendering standards. But then there's integration or interaction standards. "I am touching this object," right? I'm grabbing your hand. If you think of haptics, things you might wear, that has to be described. How hard am I grabbing your hand? What does that feel like?
Vincent Yates (00:22:27):
And back the physics engine, the momentum conservation. Distortion, disruption, all that would have to be described in these standards too it sounds like.
Jason Goth (00:22:36):
Right? And object motion and birds, whatever you wanted to render. And there are an unlimited, almost number of them, I guess they're not technically unlimited, but, well...
Vincent Yates (00:22:49):
There's an hour and a half of them.
Jason Goth (00:22:50):
I guess it's an hour and a half of someone listing them out. Again, Tim is the founder of the company that built Fortnite, so he understands this probably better than anyone else. And none of those standards exist because today those things are very proprietary because that's somewhat people's competitive advantage. Like EA Sports or Epic Games or Microsoft or whomever.
Those things give people a better experience in the game. So there they're somewhat tightly held. Now there are some open physics things, there are a few of those like Unreal. It's something that published that has a lot of that and those can be then consumed by others. But if you were going to make it worldwide scale, then you would have to greatly, exponentially expand those standards and the implementations of those standards. And that's just not something that... So that's one aspect of it.
Vincent Yates (00:23:52):
Well hold I into that aspect. Because I think this is really interesting. Two things of this are really interesting to me. One, back to your idea that this is what differentiates coming to. I mean, I remember in high school, you were probably already working, but whatever. When I was in high school, Grand Turismo was a big deal and it was a big deal because it was this-
Jason Goth (00:24:10):
My kids love that game.
Vincent Yates (00:24:12):
Okay, there you go. See, it was a big deal because this engine, this physics and they had created, at that moment in time, this world class sort of simulator and the car behaved the way that it would behave on a race track. They send all their developers and engineers to race school and try to teach them all this stuff. It's going to integrate that back in. That was a big reason that I was super interested in that game.
And so I think that's really interesting to point out because to your point, if there's actual interoperability here, if there's a world where I could go do anything with anyone on the internet in some sense or in the Metaverse as it were, we have to have the same experience. I have to be able to do something that they see and experience the same way that I would see an experience on my side of that device. And that creates a different business model conceptually than we have for games today.
Jason Goth (00:25:04):
Right, exactly. Think of it this way, what if in websites, if I went to any website built with Microsoft technology, I had to use Microsoft's browser and if I went to any website built with...
Vincent Yates (00:25:18):
Chrome or Safari.
Jason Goth (00:25:19):
And so if you think about it, this 2d medium, it's really not that complicated. You have a page with text and you have a layout of text on a page. That's not a hard problem. You have images, you have video, you have documents to download, you have some standards around requesting those things. You have some standards around proving who you are in terms of authentication. Although most of those, there are many of those as well. But there are some, and that's really it.
There's not really anything else that the internet does. But if you think about this truly immersive world of avatars and holograms and all of the different things in it, that would require an enormous number of these standards really for anything that you wanted to do. And again, they don't exist. And I think the big challenge that we have is that it's not an entity like the National Science Foundation or some industry consortium trying to develop those standards. It's individual companies and they're totally disincented to open those up.
Vincent Yates (00:27:34):
Yeah, let's go to that in one second. The second thing I wanted to point out about standards is take email as an example. How is email transmitted today through the original standard more or less, right?
Jason Goth (00:27:47):
Vincent Yates (00:27:47):
And I don't know, I guess it's a bit neighborhoods back in the day they used to be safe and you leave your doors unlocked or so I'm told. The world's scary now and everything must be locked. And in the world of the internet, everything must be encrypted and yeah, email that contains this incredibly sensitive information, real conversations is by default not encrypted. Why is that, Jason?
Jason Goth (00:28:14):
That's probably a subject for four or five podcasts.
Vincent Yates (00:28:19):
No, I think it's because the standard is too hard to change at the end of the day.
Jason Goth (00:28:23):
It's not that the standard is hard to change. There's changes to the standard, the new standard exists. It's hard to get a million different companies to go change and upgrade to that standard.
Vincent Yates (00:28:36):
As you like to say. That's a distinction without a difference. Yes. I can change a piece of paper word doc that describes the standard. Yeah, no doubt. But part of the standards values only if... Or not part of the whole value of a standards, only if people actually adopt it and actually leverage it. And the problem is that getting people to adopt a net new standard is difficult.
Jason Goth (00:28:57):
But I mean people adopt... We've had changes from the web, we had HHTPS the SSL and we've had many changes to those new ciphers, many changes to those. And people adopt them because they see value in adopting the new standards. I don't want my credit card... The reason email has it done is because no one sees the value in-
Vincent Yates (00:29:18):
Well I guess I'm wrong.
Jason Goth (00:29:20):
No, you're right. I see the value in it. You see the value in it. My wife, she's not communicating with the customer, she's sending her mom a recipe. I'm joking.
Vincent Yates (00:29:33):
Jason Goth (00:29:34):
I'm not the cook. But she sends her friends what restaurant they're going to go to, they don't care if it's encrypted. It's the same with text messages.
Vincent Yates (00:29:44):
But [inaudible 00:29:44] by default. Okay, well anyway, we should do a different point it because maybe I misunderstand the reasons, or at least my perception of why it might be different than you. Is it worth getting into? I don't know.
Jason Goth (00:29:56):
The big company email is often encrypted.
Vincent Yates (00:29:59):
I guess either way, the point here is that defining the standards upfront is important because you have to do it. It's requirement, it's necessary, but it's not sufficient. You have to get the adoption of it.
Jason Goth (00:30:10):
That's the adoption part. Well and more importantly than either of the... Well adoption obviously is super important, but also important is the evolution of them. You have to be able to evolve them and change them because the world changes and the idea that you get it right the first time, if we ever get one of these standards the first time, it will be the first time.
Vincent Yates (00:30:34):
You don't see the exception that proof.
Jason Goth (00:30:35):
Yeah, none of them are ever right the first time. What I'm getting at is you have the combination of needing an exponential number of new standards times the fact that those need time to mature and evolve and change and that they need to do so based on usage by at scale usage. It doesn't matter if two people do it. And right now, yeah, one of the funniest quotes ever was in the Wall Street Journal the other day, they were talking about the number of people that go into Meta's Horizon World and I think the author says more people went to a Mets game this year, which is if you're a baseball fan, it's really funny, but it's...
A critical mass to use it to understand what needs to evolve and change, right? Well and so it's the combination of those three things that I think puts any realistic implementation out decade or decades from my perspective.
Vincent Yates (00:31:48):
Okay, well I want to get,-
Jason Goth (00:31:49):
But that's again, as I said earlier, that's only one of three reasons that I think it goes out decades.
Vincent Yates (00:31:54):
Okay, well I want to get to that one real quick. So pause that thought. The thing that I find so fascinating about this is, okay, so let's say that we believe there's a revolutionary technology, they very well might be. I mean it could be Ready Player One could be a thing. I have three daughters as you know, and yeah, I'm not super excited about them being on social media and the internet to begin with and I'm really not excited about them doing something that's totally immersive and even more addictive.
But let's assume there's something interesting there. Let's just assume that's true. The fascinating question then is how do you go about building it? And we've covered the first ones, you have these standards. And so my question is, if you're Meta and you believe that the current internet is already breathing its last breath conceptually and you want to be the first biggest, best player in the new world and you believe that new world is the Metaverse, how do you go about actually building said Metaverse?
And if you think about Facebook, I grew up in professionally grew up in Silicon Valley at the time Facebook, this was pre IPO of Facebook, even right when I first started my career here or there rather, and one of their phrases that was very catchy, you said for a while was you got to move fast and break things. This concept of intelligent failure and the idea there, the spirit of that was exactly what you just said, which is like we have to iterate. We can have the best ideas in the world or so we think we deploy them in production and get feedback from our users directly. And that's how we know it actually works. And when you do that through AB testing for example, we discover empirically that at least two thirds of our ideas, two thirds of the things that make it to product are best ideas and stuff we invest money on.
Two thirds of them are wrong. They actually do not help and often hurt end user experience. So at best, a third of our best ideas actually are right. And that was really the ethos of Silicon Valley back then was like you got a test, you got to learn, you do it as fast as possible and if you want a really big problem, you have to break it down into its smaller constituent pieces. And what's fascinating to me about what Meta is doing right now, it seems at least from the outside that they have ignored every core element of what they believed in that created the company, the behemoth, the success that they are today. Again, from a perspective, if you just think about how much they're investing in this, Zuckerberg said they were going to have a significant increase in their investment. It's already spending 15 billion, significant increase.
So call it, I don't know who knows what that number means exactly, but call it 25 billion. And he said, "Look, we're not going to see any revenue for at least a decade." Going to invest at least 25 billion for at least 10 years. We're talking about a total investment of a quarter of a trillion dollars, like 250 billion over the next decade to get to a place where the Metaverse might be useful and interesting and compelling so that it has more people using it than I guess the New York Mets. Is that what you said?
Jason Goth (00:35:09):
Vincent Yates (00:35:12):
That number again, I have a hard time with these really big numbers. I didn't know what that meant. Chamath actually, VC guy, really great, has a great chart he made to put it in perspective.
Jason Goth (00:35:25):
Former Facebook executive, right?
Vincent Yates (00:35:29):
Incidentally, early Facebook guy. That's exactly right. The iPhone, the Apple iPhone from 2004, 2007 when they were developing an R&D and these are inflation adjusted dollars. So the real dollars today, Facebook's spent 250 billion dollars. How much did it take to transform the entire world of the internet and how we communicate? 3.6?
Jason Goth (00:35:53):
You're saying with mobile phone development?
Vincent Yates (00:35:54):
Exactly. iPhone. R&D spent total 3.6 billion. So a hundred times that is what we think it's going to take to build the boat. The Manhattan Project, this is the thing that created the atomic bomb. This is the world's most impressive R&D venture on the tightest timeline doing something that was quite possibly impossible or perceived to be impossible at that moment in time. Best brightest, no cost mattered at all. Money was no object to the government in that moment in time.
Jason Goth (00:36:22):
My favorite quote about that is one of Einstein's [inaudible 00:36:28 , Leo Skylar said that the idea of splitting an atom was shooting a bird from horseback in the dark in a country with no birds.
Vincent Yates (00:36:39):
So that level of difficulty was solved with a mere 23 billion, all of the R&D for Tesla, there is nothing that is even comparable. Save one thing, what's the one thing, Jason?
Jason Goth (00:36:55):
Vincent Yates (00:36:56):
You nailed that. The Apollo Program, the entirety of the Apollo Program, getting a man to the moon was roughly the same 253 billion spent. This is an incredible investment before we can have anything that's materially valuable and and testable. And this feels antithetical to everything of MVP.
Jason Goth (00:37:21):
Yeah, and I think that is the exponential nature of the additional things needed. So if you think about the web, what do you need in the email? Text is good enough. What about the web? For a while, text was good enough. Okay, well then we can add images. We went decades before we had a video.
Vincent Yates (00:37:41):
Jason Goth (00:37:42):
There's just not that much that is required. The floor for what is valuable to people and you can launch and then start iterating and getting feedback on is fairly low. In this kind of 3D immersive world, and you see it today, they've tried but they've launched Horizon World, people didn't have legs and they've been getting beaten up in the press and everything else. It's just because you have a factor of a hundred or a thousand x more difficult or more things to be done. And so the MVP bar, the what's the MVP? Meaning Minimum Viable Product, the bar of what is necessary to even release something that people can use is a hundred to a thousand x.
Vincent Yates (00:38:34):
Do you think that's really the MVP? Again, MVP in my mind is exactly as small as possible. Not smaller, not bigger, as small just enough to be informative. Like test your hypothesis. Do people want this? Right? Do you have to have all those standards developed to get there?
Jason Goth (00:38:51):
I think the answer is it depends on if... We don't know where the line is of it has to be enough that people want it. Right? And I think so I think they are trying to do it. Again, they launched without legs and I think what they've learned is...
Vincent Yates (00:39:07):
You need to have legs.
Jason Goth (00:39:09):
You have to have legs. People don't want it if you don't have legs, they don't have great facial expressions, they don't have realistic, they're kind of cartoonish looking. They don't have realistic facial expressions and...
Vincent Yates (00:39:18):
Yeah, Zuckerberg did get destroyed for his avatar in fairness.
Jason Goth (00:39:21):
And people again, test and learn. Yeah, you got to have that. And I think some of the other things people were complaining about and it is the kind of interaction, it's what are policies around who can approach you and what they can say to you and that kind of thing. So I think what they're learning is that that bar is going to be way higher and it's going to again drive this need tremendous investment in these hundreds or thousands of standards. And even if they're not standard, even if Meta is owning them, but implementations or interactions for it to be useful enough.
Vincent Yates (00:40:08):
I have to push back a little bit. Like I hear what you're saying and...
Jason Goth (00:40:12):
I'm not saying it's good, I'm actually saying that is a big problem with it, right? Is that there? Yes. I think to build something, to build something successful, you have to have these smaller incremental iterations in order to learn what people want and course correct and drive it. I guess what I'm saying is the problem with the Metaverse is that you're trying to reproduce reality in a digital form and the bar there is so high, the number of things you need to iterate on. It's sort of a self-fulfilling prophecy. We have to have these thousand things where we can't do all a thousand things. So let's do this one. Well, nobody liked it. Let's do the second one. Well, no one liked it, right? If there's not something that people will come to get the feedback, you don't have the opportunity to iterate on it is what I'm saying.
Vincent Yates (00:41:10):
Well, yeah. So here's my pushback. I have a couple things. One, I think if you offered people the internet of the 1980s, they wouldn't be interested, like said differently, I think where a lot of these transformation technologies started was solving only one of the use cases. So for example, wouldn't it make more sense to just keep developing independently these platforms, these technologies?
So for example, if we think that gaming's important, well, I mean not even 20 years ago you couldn't really play games with other people online. You could do text based a little bit. And that was chess, I guess. And that was basically it. Now we have these really rich dynamic environments of Fortnite that you can build stuff, building lots of creativity, Minecraft, et cetera. Why wouldn't it be make more sense to keep developing those iteratively to fund the next iteration, to fund the next, to get better and better and better and do that while having an eye towards, "Hey someday it'd be cool if I could figure out how to integrate this with other things so I could go from Minecraft to Fortnite to whatever." And I think they're, again, right to the business problem. I don't know why you'd ever want to do that, but separate from that a second.
Jason Goth (00:42:31):
No, I hear what you're saying and I agree... That would be a good way for many companies or many organizations to work on their own pieces and grow towards it and have that, I hundred percent agree. I guess what I'm saying is Facebook Meta doesn't have any of those things. They don't have Fortnite to improve, they don't have-
Vincent Yates (00:42:58):
Jason Goth (00:42:59):
And so if they-
Vincent Yates (00:43:01):
If they give you 15 billion dollars-
Jason Goth (00:43:02):
I know, but they go buy one if they want to. Well again...
Vincent Yates (00:43:05):
Take the Adobe approach here. Look, Adobe spent the equivalent in terms of their profit to buy Figma because they're like, "Hey, we need to do this-"
Jason Goth (00:43:17):
I think we're arguing different arguments towards the same end. I'm saying-
Vincent Yates (00:43:21):
Jason Goth (00:43:21):
I'm saying it's a bad investment for Meta because the bar is so high and you're saying it's a bad investment for Meta because they should go buy stuff that gets them closer to that bar. Which I don't disagree with that either.
Vincent Yates (00:43:37):
Jason Goth (00:43:38):
My point is either way it is a bridge that is very, very far away and neither the current approach is not going to get them to that bridge anytime soon. Maybe your approach of buying game companies and buying this and incrementally evolving them would get them there faster. But I still think that, well, I think that may be right. I don't know, but I think that means it's 10 years away, not 20 years away.
Vincent Yates (00:44:11):
Sure. Well, I just think, look again back to what made Silicon Valley and during the Kool-Aid, I guess just the headline here, but my belief is that you have to take iterative purchase where to your point, you have to create something adequately valuable that people want to try it, but you can't... We should probably get Kevin Erickson back here to join us again from a strategic standpoint, but it doesn't feel like a good corporate strategy bet the bank, bet the farm on something that's feels so...
Jason Goth (00:44:36):
So I 100 percent agree with that. Because again, I do agree that the best way to build things is iterative and evolve them based on feedback.
Vincent Yates (00:44:46):
There's a great quote, by the way, a professor, distinguished Professor of Duke, Dr. Sitkin, he says, "You definitely want to have a parachute if you're going to jump off a high cliff and you want to know if there's a staircase going down so you could do it step one step at a time." And my point here is I feel like there must be steps somewhere. Does it get you all the way instantly? No. Maybe it takes longer, I don't know. But it feels like a far better strategy to me.
Jason Goth (00:45:11):
Yeah. And again, I'm violently agreeing with you, right? I'm just telling you that the point I'm making is that because there are so many of these, things that would needed, it's going to take a long time before this is well is actually commercially viable and the fact that they are not able to get enough users to evolve them is not helping, I guess is my point.
Vincent Yates (00:45:39):
Yeah, that's fair.
Jason Goth (00:45:40):
And then there's one other thing, and I was talking about the standards, but there's one other thing in this whole discussion that we're missing and that is scale. There is an entire technical problem with scale. And think of it this way, let's look at Fortnite. You can go into a Fortnite where I play a Fortnite game and they can get about a hundred people playing simultaneously on a server and they have lots of servers, so lots of people can play, but they're in groups of a hundred.
Same way with Discord and these other, you can create your own server, but you got limits, right? You're talking about a virtual world, you're talking about potentially billions of people all interact potentially in a world that can interact. And how do you provide that scale? You may have servers all over the world, you're on this server, someone else is on that server. Those things have to be able to communicate with each other, much like the phone system. And so if you think about the actual amount of infrastructure that drives the phone system all over the world, that has to be from the cable to your house to all the servers in the central office to everything else. I don't know the exact number, it's probably hundreds of billions if not trillions of dollars in investment.
But also, and this is why Bell Labs existed, hundreds of standards for signaling and location and data interchange and all of those thing and those standards for that type of realtime communication at scale for I'm shooting at you, you're over there got to somehow it has to tell you that the bullet is coming at you across servers, across the globe with latency, speed of light is a constant you can't beat. And that's a tremendous engineering problem that also has to be solved. And so I guess what I'm getting at, it's like you've got this tremendous problem of the number of standards and this tremendous problem of getting a number of users to get you the feedback with the tremendous problem of you have to be able to solve some of these scale problems before you can get that number of users on it. And I also think there's an economics issue there we can cover, but if there was a four way analogy of chicken and egg, which comes first, you have four problems that all have to get solved.
Vincent Yates (00:48:25):
They're all necessary, but none of them are sufficient by the solved. Got it. Well this is where, look, I'm not saying that spending a quarter of a trillion dollars is not necessary. I think the complexity is insane. I think the value is unclear, but the complexity clearly is be expensive to solve. Who invests that money? Is it right for Meta to try and go develop this? Somebody has to, I guess.
Jason Goth (00:48:51):
Well the majority of voting control says yes.
Vincent Yates (00:48:55):
Right? Yeah, exactly.
Jason Goth (00:48:57):
Meta, which is one person.
Vincent Yates (00:48:58):
That's right. Yeah. Zuckerberg does have 54% I think of voting authority because of super shares. But I mean seriously, if you think about Meta is going to spend a quarter of a trillion dollars, let's say successful, let's say that it's actually useful and they've solved all these problems, but in a decade, great. They have to believe that they can make it back economically. This isn't altruism, this is, "Hey, we think that the business model exists that we can make more money than we spent."
Jason Goth (00:49:30):
Well, that's what I meant in when I said the fourth issue. It's like if you think of the internet today, it was all the different components. Pieces were all developed somewhat out outside of a corporate. It wasn't a corporation. Some corporations tried, Lotus tried to build Lotus, but most things were developed through National Science Foundation, DARPA, other type of grants that were viewed at least initially as public utilities.
And whether there's a big argument, whether they should be or not, it's a different argument. But the fact is that's how they were developed. And it wasn't one company that developed the internet. And then here you have a situation where it is a company that is spending 250 billion over the next 10 years to go develop this thing. And so people say there are these big things that control the internet, but not really. I don't believe that. I mean, Google is a big site that lots of people go to. YouTube is a big site that lots of people go to and they have lots of influence from advertising and their amount of traffic. But how many websites are there? And you're able, you can go put one up today, Amazon, it's a big site. But I also buy, I'm a fisherman. I buy stuff off Tackle Warehouse. There's a little plug for Tackle Warehouse. Maybe you guys can send me free gear, but a lot of podcasters are not, they want to be objective and not take swag.
I will totally take swag from Tackle Warehouse, and be unobjective. But my point is there are millions of web properties and new ones that come every day and new ones that die every day and it's not controlled by an entity. Whereas back to the Meta example, if you're going to go spend 250 billion of shareholder money, they're going to expect a return on that. And this is one of the economic problems I think is also, this is one of my four way chicken and egg problems is people say, "Well, Zuckerberg doesn't care if the stock price loses 70 cent percent of his value. He's got control of the company and he sees this vision and he's going to do it." Right? But remember, they hire all those 80,000 people, they're able to pay them by giving them stock options. The developers, there's a lot of developers that could go somewhere else. And when their stock options are worth nothing and they get no dividends and that's a significant portion of their pay, why are they going to stay there?
Vincent Yates (00:52:10):
Yeah. I think that's the thing that I think happen ultimately. Look at some point when you erode 88 or 77% of your share value, your developers whose comp is at least a quarter, potentially up to 50%, 70% for senior people at Facebook of their compensation you've just eroded at least a third of their pay. How long are they going to stick around? I don't know. I don't think forever though is the headline.
Jason Goth (00:52:38):
And the other point is that scale and infrastructure scale, well someone has to underwrite that, usually. You don't go just drop a whole load of cash that's usually financed. And so as you run in rough times, you need to expand, you need to grow, you need to go lay fiber like Google did, lay fiber across the ocean. Who's going to go underwrite that project for you if your stock price is tanking?
Vincent Yates (00:53:05):
That's right. And your people might be quitting and et cetera, et cetera. Look, I think this is the second point I wanted to raise in Meta is investing more in Capex than Apple, Tesla, Twitter, Snap and Uber combined. There's spend an enormous amount of money on this.
Jason Goth (00:53:27):
So maybe I'm wrong, they are dropping cash and not going abroad, but I think eventually they're going to have to get some help.
Vincent Yates (00:53:33):
Yeah, they're spending an enormous amount of cash. Again, this is because it's currently supported by their core business, but their core business has been stagnant and decreasing and they're not investing in it. I think is a larger concern, is that I think this back to what we need direction back in the call for. I think the issue here is that Zuckerberg is all in on Meta obviously, and nobody's taking care of the core business, which is supposed to fund this 250 billion capital expenditure towards this R&D effort called the Metaverse.
Jason Goth (00:54:01):
And I think that's a large part of the original open letter to the shareholders.
Vincent Yates (00:54:05):
That's right, Congressman. That's exactly right. The thing I was going to say though is look, what I've learned historically right now, so the economy overall is tight. Startups are feeling pressure all over the place. You see a lot of VCs actually talking about the idea that actually this is a great time for startups and it's a great time for startups because a engineering talent is now becomes available because of people leaving Facebook, for example, on Google and they can't pay them everything anymore.
But also because when times are lean, you get laser focused on what your customers ultimately want. And that's actually really good for the business overall. The question to you is, because they're spending so much money trying to do everything all at once, I actually think they're going to be less effective than if they gave themselves an artificial cap. Like in the Gerstner [inaudible 00:54:53] letter, I don't know if that's the right number, I haven't thought about it. He suggests 5 billion a year. He says, look, you want to go make a big bet on the Metaverse, keep it to 5 billion a year because what you're going to do is you're going to force yourself to prioritize what is truly MVP. Not as not what's like what would be nice to have an MVP, but what is truly MVP to grow iteratively. And I think 5 billion, he chose the number in part because AWS was built on less than 5 billion.
Jason Goth (00:55:21):
Listen, I agree. My rant about constraints are good. I think I have a blog out on cordera.com. It's a shameless pitch.and I think it's titled, constraints are good.
Vincent Yates (00:55:36):
It's like the Mythical Man month. Same problems here. All these things are, I think from multiple dimensions we see that just dumping more money into solving hard problems doesn't actually solve them better or faster.
Jason Goth (00:55:47):
My inspiration for that blog article is we went on a family vacation, we flew actually up to San Francisco and we drove down the coast to LA and it's through Big Sur and the coast. My kids called that highway... What is that? Highway one? It runs along the coast. It's beautiful. My kids called it The Road of Death because they were afraid we were going to fall off at any point. But we stopped at the Hearst Mansion, which is fascinating, but it's a fascinating, but the floors are imported from this church in Italy and it's an incredible thing. But the most impactful thing for me when I saw it was the fact that it's not finished. They've been building it for a hundred years. It's still not finished, right? Because there were no constraints.
Vincent Yates (00:56:44):
I think it's a great anecdote and I think that's my fear with Meta. And it's the Metaverse, it's certainly not real today. Even Zuckerberg says it's not real today. He doesn't see it being real or revenue generating as far as I can tell from his sort of long meandering answer for at least a decade.
Jason Goth (00:57:03):
I think the issue, and we always try to get this back to something that's useful for our customers on this pod. So I'm sure a lot of people are listening to, "Well, how is this useful to me?" And I think the big challenge for a lot of our customers, it's looking for use cases. How can I use, I've got a product or service and I have all these different, when you think about digital transformation, what does that mean? It's like, well, I want to either offer my service or product or service and enhance it somehow digitally or change the channel to which people purchase it or change it to a digital product. And so, well, how does the Metaverse fit into that is a question that people are asking. How do I use it to advertise? Or how do I use it to deliver a product?
Or how do I get my products delivered through it? And what are those use cases that are going to help my business? And the challenge that I think we have, frankly, answering that question is like, well, I don't know. I don't know yet because I don't know what these standards and what it is. And we can look at things in these very early forms and say, "Well, yeah, we maybe could use it for that," but that may get wiped out in the next version. And so I guess my answer is it's something to consider. And I definitely think some of these other non-immersive world things that sometimes get tagged as a Metaverse, like sure there may be many use cases there on those existing things like AR for training and that kind of stuff. And yes, look at those fully immersive Ready Player One ideas. I think it may just be too early.
Vincent Yates (00:58:52):
No, I agree. I think, look, the takeaway for me is there's a couple, which is I think if you want to talk about Metaverse specifically to your point, how would you get involved? I would think about it less as the Metaverse and I'd get more specific. So back to the point, what is the modality that you want to talk about here? Is it ARr? Is it VR? Is it something else? Hologram? Is it telephone touch.
Jason Goth (00:59:10):
Your shoe company, you want to sell your shoes, a virtual version of your shoes in a game, great. There's a way to go do that. Go figure that out with Epic.
Vincent Yates (00:59:17):
That's right. And then I think the second part is think about the industry that you want to tackle here. Is it a gaming use case? Like you just said, is it a corporate training use case? Is it, think about it from these dimensions. Like, okay, I want to tackle training, I want to tackle meetings. Okay, great. And then what's the modality? Well, I want it to be voice. I want it to be video. I want it to be 3d. I want it to be... Fine. Yeah, good. I wouldn't think about it as the Metaverse holistically, I think at this point it's just too early. It reminds me a lot of my early days in my career as a data scientist and thinking about AI, every CIO is like retail therapy for CIOs. I need AI, I need data scientists. I need to go get more data, big data, big data is going to be the thing.
AI is going to be the thing. I don't even think we've really hit the heyday of AI and this actually is an interesting point that Gerstner points out in his open letter to Meta is like, look, you guys have an enormous amount of opportunities still laying ahead of you in terms of AI. What does it mean when you can generate text, when you can generate video, when you can generate images based just off of what somebody describes what somebody wants? So I think there's a huge amount that we still have to get out of the value of AI, which again was the technology that was super hot and interesting 20 years ago. If you think about the hype cycle, it's finally getting to the point where it's really valuable. I think that this is where that was 20 years ago. Effectively. The last thing I'd say it's from a product planning slash corporate strategy standpoint, I think it's from interesting lessons to take away from this as well, which is I think MVP is really important.
I think that how you go about that, you need to be very thoughtful, especially with really big bets here. How do you break them up into their constituent parts? To your point, we know there's a huge number of standards. Well, which of these do we think it's going to be required? How are we going to test these standard time going to develop these standards and how do we build an economic model around that?
And so not that of the legislators listen to us, but for legislators, I think this is an interesting question too. We've seen what happens with email and in cryptography we've seen what happens with the internet and net neutrality. We've seen the sort of debates we've had is the Loon project is satellites, is all of these other ways of trying to get internet across the world that individual companies have invested in how that plays out long term. And so it's an opportunity to start thinking through that today. If we believe that this Metaverse is real, we think interconnecting is going to be a thing, we know it's going to be a huge investment, how do we fund that so that ultimately it is part of the public domain?
Jason Goth (01:01:40):
You just brought up another topic which I didn't really prepare for, but the whole topic of regulation around some of these things. So there are going to be regulations around, how can you touch some of these? And so those also shape, so I'm going to revise my four way chicken and egg to now five way chicken and egg problem. But I agree, we have to be able to develop these things iteratively and incrementally and get enough users to give the feedback and to learn about what works and what doesn't. And that that's not only on these interactive kind of centers, but it's also on the scale stuff. And I just don't see how they're going to do it without some level, the MVP standard is so high in people's expectation. And by the way, I think that is actually driven by movies like Ready Player One and Star Wars and people have this expectation of it.
Vincent Yates (01:02:45):
Hollywood, yeah. That's why I think you can constrain to the budget. You wouldn't even try to build Ready Player One. I think that's the problem is they say, "Well look, we're going to build Ready Player One." And you've set people, you anchored them to the wrong thing because you're like, "Well, I have an infinite money, I can go build that." Yeah, maybe in 20 years if you don't go bankrupt first perhaps. But instead you said, I only have 5 billion.
Jason Goth (01:03:08):
Nobody built the hologram phone, right?
Vincent Yates (01:03:10):
Yeah. Nobody in their right mind would say, well, with that kind of much I'm going to build... No, they would say, how do I get closer to that reality without actually planning on getting there anytime soon? So anyway, that's that. I think, look, the other last takeaway I was going to say is it's a well-known secret in Silicon Valley that we have more engineers, they have more engineers than they need for a lot of those products. As we've seen with this tightening, the economy with Meta in particular, I think a lot of those engineers are going to be leaving in the case of Twitter, that they'll be forced to leave here. And so I think for executives who historically have never even tried to recruit from these areas because it was just impossible, too expensive, just no interest. I think that world is going to change imminently. So I think that's the other thing to contemplate here. And if anybody from these organization is working, we are hiring. Just so you know, Credera is hiring.
Jason Goth (01:04:04):
That's another shameless plug.
Vincent Yates (01:04:06):
Well, anyway, I appreciate it Jason. Thanks for diving in. Hopefully it wasn't too heated in a topic here, but...
Jason Goth (01:04:14):
I actually think we need to have another discussion because the one thing we didn't cover is the whole economy. And I don't mean the economics of building it and trying to recoup that and that kind of stuff we talked about. I'm talking about the in world economics and what... I think that that's a really interesting topic that might...
Vincent Yates (01:04:38):
Jason Goth (01:04:39):
Be one of those things that might be a crawl, walk, run, but maybe we could cover that in the future pod.
Vincent Yates (01:04:47):
Okay, sounds good. Well, for those of you that would like to learn more, feel free to visit the insights page with Credera, credera.com. Thanks for listening and I hope you'll join us again. Thanks, Jason.
- Technology Tangents
- Technology Consulting
- Data Science, Artificial Intelligence & Machine Learning
- Machine Learning
- Machine Learning Applications
- Technology Trends