WEBVTT 00:00:00.000 --> 00:00:11.000 I'm not going to convince you, you're not going to convince me. We will just become even more convinced in our own righteousness and also convinced that the other party is a jerk. On the specific thing I disagree. 00:00:11.000 --> 00:00:29.839 And I'm going to be a friend you will hear from my life now. Hello, welcome to episode 45 of the Metacast Podcast. This is Arnab. And this is Ilya. And in the Metacast Podcast we talk about the journey of our 00:00:29.839 --> 00:00:49.840 startup, the Metacast app. Today in this episode we're going to talk about different product development methodologies. We're specifically going to talk about Lego Block style of building products versus model airplane style of building products. And of course towards the end we're also going to talk about what Ilya and I are listening to. 00:00:49.840 --> 00:01:18.840 So let's get started Ilya. What's this Lego blocks and model airplanes? Lego blocks is an approach when your software can do a lot of different things. It's how you're configurable. So think about it like an Android phone. You can configure every small thing in there. And then you can create a custom experience. That's what Android users tell me that they don't like about iPhone is that iPhone by design is not a Lego block. It's a model airplane kit where it gives you an opinion. 00:01:18.840 --> 00:01:38.840 So in opinion, it's set of settings and defaults. And this is the only way it fits together. Yeah, Apple's way is the only right way to do things on an iPhone, which works for 99% of users who use iPhones. Yeah, everybody else just goes in uses Android because they want that extra configurability. 00:01:38.840 --> 00:01:55.840 So they are functional because on one hand if you do a Lego block approach, then you basically tell your user that we trust you the user to do whatever is right for you from the pieces that we give you right from the platform components if you like. 00:01:55.840 --> 00:02:05.840 But not only trust you, we also require you usually to figure out how to compose these pieces together and come up with something that works for you. 00:02:05.840 --> 00:02:19.840 Yeah, whereas a model airplanes is like take it to leave it. Let's just how do you think things should be done. And you have no choice what your choice is to go elsewhere. What type of person are you? Are you the Lego blocks person or the model airplanes person? 00:02:19.840 --> 00:02:35.840 I think when it comes to different products, I prefer different things. So for a phone, I definitely prefer the iPhone experience. I have switched to Android in the past and spent maybe four or five years on it and then come back again. 00:02:35.840 --> 00:02:53.840 But that was purely because Android for a while was where I had a iPhone in terms of a lot of features. And I was always like, whoa, this is there. This is there and that's there. But eventually the whole configurability did catch up to me where it's like too much to think through. 00:02:53.840 --> 00:03:08.840 Whereas the iPhone maybe legs behind in terms of like cutting as features, but the things that are there are polished and it works because it's been built for a specific kind of use case in my. 00:03:08.840 --> 00:03:14.840 Yeah, I'm kind of imagining the night key logo with this logo and it just works. 00:03:14.840 --> 00:03:29.840 So why are we talking about this today? Because when we started building medacas, I think very early on that was one of the things we were wondering about without going too much into the details or into the future. 00:03:29.840 --> 00:03:43.840 We do have like in our possible amount of things we could do in the podcasting space that interests us as a company. There is the app that we're starting to work on, which is a consumer product. 00:03:43.840 --> 00:03:53.840 There are also products or tools for like creators, podcasters. There's also like various other things that we've found or do. And eventually we decided to start with this. 00:03:53.840 --> 00:04:02.840 Yeah, I feel like both of us kind of like the model airplanes when it comes to consumer products. So things that just work. 00:04:02.840 --> 00:04:08.840 Opinionated simple. I think simplicity is a big thing for both you and me. 00:04:08.840 --> 00:04:13.840 Yeah, there's actually a very good book by Ken Sigal, I think his name is. 00:04:13.840 --> 00:04:16.839 Oh yeah, insanely simple. 00:04:16.839 --> 00:04:22.839 Insanely simple? Yes, that's the book about Apple. And I really like the point that he makes there that simplicity is actually very hard. 00:04:22.839 --> 00:04:37.839 It's easy to make things complicated, but it is very hard to make things simple. And even as we're building the podcast app, we are kind of rethinking what podcasts experience could be from the ground up. 00:04:37.839 --> 00:04:48.839 But also obviously, we need to have the features that other apps have so that users can seamlessly transition from what they use, even though we try to innovate and improve those experiences as well. 00:04:48.839 --> 00:04:53.839 And I think if you have discussions that we had, you and I had, or should this be configurable? 00:04:53.839 --> 00:05:00.839 Or should the same default be just good enough? I think you have a tendency to be more like, let's make it a configuration. 00:05:00.839 --> 00:05:04.839 And I'm like, if it doesn't work for 10% of the users, well, we can just go elsewhere. 00:05:04.839 --> 00:05:15.839 Or we can eventually make it configurable because I feel like we are converging towards that approach where we just have a simpler experience that's sort of opinionated for most part. 00:05:15.839 --> 00:05:20.839 And it depends on now our own kind of taste and the things that we've heard from users. 00:05:20.839 --> 00:05:29.839 And one of the revelations for both of us that we had when we were doing some user research, we were talking to about 10 users or so in a few months ago. 00:05:29.839 --> 00:05:36.839 And none of them used their podcasting apps the same way. They were all doing little different things differently. 00:05:36.839 --> 00:05:47.839 That was like, wow, how do we satisfy all of those use cases? And I think that's when it may be tempting to become a LEGO block company or a LEGO block product. 00:05:47.839 --> 00:05:48.839 Everybody can use it. 00:05:48.839 --> 00:05:54.839 Yeah, these 10 different use cases. So we need to add these 10 different things and 10 different settings. 00:05:54.839 --> 00:06:00.839 And then, well, guess what? It doesn't work for anyone because people just can't figure it out because it's too complicated. 00:06:00.839 --> 00:06:07.839 Because there's a difference between complex and complicated. I really like that distinction. Our app is complex. Like the backend and all that. 00:06:07.839 --> 00:06:11.839 Complex, but it's the internals. The user shouldn't have to see your care about it. 00:06:11.839 --> 00:06:18.839 Exactly. Yes. But it's possible to make anything complicated. It means it's not very well thought through. 00:06:18.839 --> 00:06:21.839 There are many ways to do the same thing. 00:06:21.839 --> 00:06:25.839 Yeah, a good example is AWS, which is actually both complex and complicated. 00:06:25.839 --> 00:06:32.839 But the complicated part of AWS is like, what I was living three years ago, it was 170 services. Now it's probably over 200. 00:06:32.839 --> 00:06:36.839 And they keep invading the new names. I was looking at LinkedIn this morning. 00:06:36.839 --> 00:06:42.839 And there was Adam C. Lipsky talking about party rock built on the top of Amazon bad rock. 00:06:42.839 --> 00:06:47.839 I don't know what bad rock is. But it sounds like I'm supposed to know. 00:06:47.839 --> 00:06:57.839 When I joined AWS in 2015, you were there already for a couple of years by the time. The team we were on, AWS console team, was about 30 people, maybe 20 people. 00:06:57.839 --> 00:07:05.839 It was a very small team. The whole team, including engineers, PMs, TPMs and managers, right? And our JVL manager. 00:07:05.839 --> 00:07:11.839 And I think there were two product managers, the three engineering teams, one TPM. 00:07:11.839 --> 00:07:22.839 When I started the whole AWS developer experience, which is console and SDK and tools and CLI and the websites, that was all like about 10 people. 00:07:22.839 --> 00:07:34.839 So I don't know around 2015 when you joined. I don't remember what the state was, but I think you're right about 10 people for the console platform side with like two PMs. 00:07:34.839 --> 00:07:42.839 Now there were about, I think 20 engineers by the time I joined just for the console, the other tools was already a separate team separate org. 00:07:42.839 --> 00:07:47.839 But it was really cool talking to our boss Craig who actually started the console. 00:07:47.839 --> 00:07:52.839 And it was like one team was building the UI for the entire AWS. 00:07:52.839 --> 00:07:57.839 Had this tab experience with like five or six services in there like S3 EC2 and others. 00:07:57.839 --> 00:08:04.839 Then AWS was growing too rapidly, too many new services were coming up. So they had to like kind of platformize all that. 00:08:04.839 --> 00:08:12.839 And then they created a new UI with like a list of services. And then at least two large, then they became a grid of services. 00:08:12.839 --> 00:08:18.839 And then that grid of services became a scrawlable grid of services. And that like wouldn't fit even the large screen. 00:08:18.839 --> 00:08:26.839 And then they added those groups of services. Then they had to add search. And it's almost impossible to find anything right now if you don't know what you're looking for. 00:08:26.839 --> 00:08:33.839 Especially that there are multiple services that do the same thing just slightly differently like EC2 and light sale, fargate and lambda. 00:08:33.839 --> 00:08:39.839 You look at this and you're like how do I choose between the two. And this is what makes it complicated. So it's not just complex. 00:08:39.839 --> 00:08:45.839 So actually I wonder, think about this right. It will start as a very simple Lego block experience. 00:08:45.839 --> 00:08:52.839 Yes, like EC2 and S3. But then it became almost like every Lego block is a model layer play. 00:08:52.840 --> 00:09:02.840 So I think there is the bedrock. If you the bedrock of AWS Lego blocks the EC2s S3s lambda and all that. 00:09:02.840 --> 00:09:17.840 And then now as the years have progressed and the customers of AWS have requested the one thing great in AWS is they don't build anything without actually talking to customers and validating that there is a need for something like this. 00:09:17.840 --> 00:09:25.840 Right. Like you don't just build and throw away things. Is this how they build Firefox. That's not AWS. See that's Amazon. 00:09:25.840 --> 00:09:32.840 True. AWS I feel like that culture is very much there where you do try to validate with customer. 00:09:32.840 --> 00:09:43.840 Earlier as customer demand appetite grew for higher and higher function services. So the bedrock EC2s and S3s and all that is not. 00:09:43.840 --> 00:09:50.840 I also want a Sage maker. You'll have to tell what Sage maker is to already snorkese and to me as well. 00:09:50.840 --> 00:10:01.840 It's a way to like build and find you and run your machine learning models on AWS. Everything in AWS you could of course do it yourself in EC2. 00:10:01.840 --> 00:10:12.840 You could run your own database, but you could also use the managed database products that AWS. So over time I think we see this growth of managed services on top of the bedrock services. 00:10:12.840 --> 00:10:22.840 Which is fine, but like you said, it has become complicated because there is a whole like insane amount of services available. 00:10:22.840 --> 00:10:30.840 Which is not necessarily a bad thing for AWS. It is bad for the starting up user who doesn't know what AWS. 00:10:30.840 --> 00:10:36.840 It's a steep learning curve to like figure out what are all the things possible here. What should I use? 00:10:36.840 --> 00:10:46.840 I think you brought up a very good point as to why this happens. Customers ask for things and write different teams responsible for those different products. 00:10:46.840 --> 00:10:53.840 Let's say AWS created LightSale, which is a simplified version of EC2. Very, very simplified version. 00:10:53.840 --> 00:11:01.840 It's almost like a shim on the top of EC2 that just has a few knobs, whereas EC2 has probably hundreds of knobs. 00:11:01.840 --> 00:11:09.840 But then a customer is like, I really like LightSale, but I want this thing that EC2 has. EC2 is too complex for me, but this thing is just right. 00:11:09.840 --> 00:11:22.840 But I need that one thing. And then the other customer needs another thing. And then if product managers say yes to everything, then you end up building a clone of EC2 under a different name. 00:11:22.840 --> 00:11:31.840 I think LAMD is probably a good example here, because I think you can run your own containers there. And then how is it similar or different from Fargate or ECS? 00:11:31.840 --> 00:11:40.840 I am a big fan of LAMD but I do agree recently it has gotten very close to what Fargate is or what ECS is. 00:11:40.840 --> 00:11:51.840 And all these products, at this point you almost need a guide that gives you a framework of like, I want this performance and I want this kind of availability, so I should go look at this one. 00:11:51.840 --> 00:11:55.840 Which is a complex framework for every decision to come up with. 00:11:55.840 --> 00:12:03.840 Yeah, I wonder is this inevitable for platforms like AWS to become this kind of bloated, complicated. 00:12:03.840 --> 00:12:11.840 What I am imagining is like, so we've got lots and lots of LEGO sets, which my son used to do, but he doesn't do them anymore. 00:12:11.840 --> 00:12:26.840 So we have like, I don't know how many God knows how many boxes of those semi broken LEGO sets we have. And my younger son every now and then we just take them and just in huge just like spread those LEGOs all over the floor. 00:12:26.840 --> 00:12:34.840 And you just have this massive LEGOs all over the legroom that if you step in them and it hurts, they all different colors. 00:12:34.840 --> 00:12:40.840 You look at this and like, what do I do with this? It's so chaotic. So that's how I imagine it. 00:12:40.840 --> 00:12:47.840 Yes, in some way, like to somebody who was a coming there completely new, you'd be like, how do you make a decision as to what to use. 00:12:47.840 --> 00:12:53.840 And actually, I think it might be easy if you are a small company. If you build a startup, you just pick two or three services, you just build on them. 00:12:53.840 --> 00:13:06.840 But if you are coming from a legacy environment where you have a bunch of stuff and you probably need to use 10 to 20 different services and you have scale where cost makes a difference performance makes a difference. 00:13:06.840 --> 00:13:10.840 But also cost and performance will depend on the interaction between the things. 00:13:10.840 --> 00:13:18.840 And I think this is where a good solution architect kind of role comes in. That's where AWS has invested quite a lot, right? 00:13:18.840 --> 00:13:25.840 Because it has gotten quite complicated in choosing how to use it and not just what to use, but how to use it. 00:13:25.840 --> 00:13:30.840 There's a lot of configuration and choices available. So ultimately you throw people into the problem. 00:13:30.840 --> 00:13:41.840 I think because these decisions are such case to case specific, the framework that I talked about, let's say like I want less than five millisecond latency. 00:13:41.840 --> 00:13:51.840 And I wanted for 99.9999% of my requests coming in. Then maybe lambda is not a good choice because you can't guarantee that if you want that guarantee. 00:13:51.840 --> 00:13:58.840 But you're sacrificing by not using lambda, you're sacrificing something else and it's going to cost you more to build that solution. 00:13:58.840 --> 00:14:10.840 But making these decisions are so product and each use case specific, then I do think right now you definitely need a human in the loop to come up with that. 00:14:10.840 --> 00:14:19.840 Do you know of any good examples of companies that started simple and remained simple while actually becoming big? 00:14:19.840 --> 00:14:21.840 Not Google. 00:14:21.840 --> 00:14:28.840 Based on our recent experiences with like Google workspaces and all that, it's beautiful. It is pretty simple. 00:14:28.840 --> 00:14:33.840 If things go well, it's when things break. It's like, oh my god, it's insane. Yeah. 00:14:33.840 --> 00:14:47.840 I don't know good examples. However, I do think AWS and Google have avoided a typical pitfall that you tend to see in platforms, which is you build a platform. 00:14:47.840 --> 00:14:59.840 You start to get good customers and big paying customers with a strong voice and you start to customize your offering for those customers. 00:14:59.840 --> 00:15:06.840 And before you know it, it's five years down the line and you are actually offering like 10 flavors of the same product. 00:15:06.840 --> 00:15:14.840 And I feel like AWS has done a great job of not going into that pitfall because this is a bit before my time. 00:15:14.840 --> 00:15:20.840 But if you remember, Amazon used to actually host e-commerce websites for other retailers. 00:15:20.840 --> 00:15:21.840 Oh, no idea. 00:15:21.840 --> 00:15:31.840 Yeah, like target.com was actually built by Amazon. It was called m.com and it was a big deal. But each of those was a specialized thing. 00:15:31.840 --> 00:15:36.840 And eventually everybody figured out that this is not going to work out. It's not going to scale up. 00:15:36.840 --> 00:15:43.840 It's kind of a tangential topic. I used to have a service business when I was very young, like 20 years ago. 00:15:43.840 --> 00:15:48.840 So yeah, we were building websites and custom software for people. I hated doing that. 00:15:48.840 --> 00:15:54.840 There is something about that kind of consulting work where you are doing custom things. 00:15:54.840 --> 00:15:59.840 And then the next project you have is similar to the previous one, but slightly different. 00:15:59.840 --> 00:16:05.840 And you just keep doing the same thing, but a bit differently. It drove me insane. It's a point. 00:16:05.840 --> 00:16:09.840 And I'm like, I just can't do it anymore. I want to do a product that is standardized. 00:16:09.840 --> 00:16:13.840 And actually, maybe that's where those like blocks come from, right? 00:16:13.840 --> 00:16:22.840 You can't build a custom thing for every customer that wants it, complex website, for example. So you give them building blocks that they can use to build with themselves. 00:16:22.840 --> 00:16:26.840 Or other companies can recombine those building blocks into a website. 00:16:26.840 --> 00:16:41.840 But inevitably you have to offer configuration. And I think that probably makes sense for these platform offerings, especially tools that a developer will use to build services with, like AWS or Google Cloud. 00:16:41.840 --> 00:16:50.840 Configurability. So if you want to meet the needs of different types of customers, you have to give them configurability. 00:16:50.840 --> 00:16:57.840 Yeah, where do you draw the line of what's too much configurability versus what's acceptable configurability? 00:16:57.840 --> 00:17:07.839 And I'm thinking about services like maybe Squarespace, Shopify, WordPress, where it all started as like, just click a few buttons and it just works. 00:17:07.838 --> 00:17:14.838 And you're done. And now it's like, I think Shopify has like its own language that you use to define templates. 00:17:14.838 --> 00:17:21.838 It's probably based on some other language, but it's like language that I've not seen before. It's like a common industry use thing. 00:17:21.838 --> 00:17:25.838 WordPress has also this plugins that you have to code up yourself. 00:17:25.838 --> 00:17:30.839 And then you look at those services and you're like, I want to do just a little bit more than what they allow. 00:17:30.839 --> 00:17:38.839 But then you have to learn all of that extra stuff. At some point, it might feel like, oh, it's just easier to treat myself on something like AWS. 00:17:38.839 --> 00:17:42.839 Then she's like, it'll be a little bit of a proprietary stuff that they offer. 00:17:42.839 --> 00:17:52.839 So I think what you're getting to and what you started with is, can we think of a thing that started as super simple, has grown, but has stayed simple. 00:17:52.839 --> 00:17:58.839 And in some ways, I feel like Apple ecosystem meets that definition. 00:17:58.839 --> 00:18:07.839 Even products built by Apple. Yes, they all work with each other. It has grown in terms of complexity and what complicated, right? 00:18:07.839 --> 00:18:13.839 In terms of complexity, there's a significant more complexity than the initial iPhone today. 00:18:13.839 --> 00:18:19.839 Your iPhone seamlessly works with your watch, with your iPad and your MacBook. 00:18:19.839 --> 00:18:24.839 They're all Apple products, of course. And maybe that's why they're continuing to keep it simple. 00:18:24.839 --> 00:18:34.839 But this, like, I can copy things on my phone and paste it on my MacBook and I can do the other way or I can use the camera without any like physical link between the devices. 00:18:34.839 --> 00:18:45.839 Or I can swipe like one app from one device to the other one. These things are complex to build, but they're not complicated user experience rights. 00:18:45.839 --> 00:19:00.839 Apple's products though are very underwhelming. Most of the specific apps, like Apple notes or Apple mail, Apple podcasts, they are very minimal to the point that they're very usable in my opinion. 00:19:00.839 --> 00:19:13.839 Like Apple mail, it doesn't integrate with Gmail enough. I don't like Apple mail. I like how it looks. It's like nice and minimal until you want to do like labels and other stuff that you can do in Gmail and it just doesn't support that Apple calendar is the same way. 00:19:13.839 --> 00:19:31.839 Actually, Apple notes, I use just for one specific thing. We have a shared shopping list with my wife, which works like magic. Actually, the reason why I like it over what I use over bear is when you check an item in Apple notes, it moves to the bottom. 00:19:31.839 --> 00:19:40.839 And if you uncheck it, it moves back up. So that's the only reason why I use Apple notes. Otherwise, I feel like it's in a very underwhelming piece of software. Apple podcasts are the interesting one. 00:19:40.839 --> 00:19:57.839 They've improved it over time. But then they start adding a lot of discovery features and stuff. Yes, our app was down because Firebase was down. And I went to Apple Podcast and like maybe I should find something to listen to Apple Podcasts. And I just got so overwhelmed by the chaos that I saw there. 00:19:57.839 --> 00:20:01.839 It was very much so an Apple in my opinion. 00:20:01.839 --> 00:20:11.839 That's in terms of content, which is a big part of the user experience, for sure. I mean, Apple Podcasts is a software for presenting you with somebody else's content. 00:20:11.839 --> 00:20:19.839 Yes. But in terms of the flows and all, there's still it's very minimal. You can't create your playlist or like you can do much there. 00:20:19.839 --> 00:20:31.839 There's not much here. Yeah, yeah. So you mentioned bear and Apple notes. There's also we want to talk about I think in this specific Lego blocks versus model airplanes. The Lego blocks here is notion. 00:20:31.839 --> 00:20:42.839 Oh my gosh. Yes. It's a database and a note picking up and like a spreadsheet and it's everything. People love it. I can say many people love it. I'm not one of them. 00:20:42.839 --> 00:20:56.839 I used it quite a lot for a while. And I think eventually what you said that the specific thing was the fact that ultimately we are using this as a document archive. 00:20:56.839 --> 00:21:10.839 That's what we're using notion for. But it makes text editing very cumbersome because everything is like a block. And that kind of I eventually felt that I feel like as a very flexible database kind of thing. 00:21:10.839 --> 00:21:19.839 Like if you want to build your own to do list or you want to build your own, let's say library of books that you've read in your life. Those kind of things. It's really good. Right. 00:21:19.839 --> 00:21:28.839 Custom forms. You enter something into the form. It's very easy to use. And it goes into a database. Now you can search through all that filter. All that's great. 00:21:28.839 --> 00:21:42.839 A note taking or document writing app. It's not very good. Like as a wiki for your company. I don't think it's a very good product. Yeah. So it's more like an organizing tool for a person's life. So to speak. 00:21:42.839 --> 00:21:53.839 It could be for a company also. It's a structured data. Like it could be your CRM tool. I feel like for a startup. You want to keep track of your customers or leads or things like that. 00:21:53.839 --> 00:22:10.839 And you don't want to set up like a sales force or pay for all that. It's a very simple tool and yet very customizable. And it has all the bells and whistles you could need for like filtering slicing, dicing data like you would in a database while not having to do all of that work. 00:22:10.839 --> 00:22:18.839 And you would need to create those experiences for yourself using the building blocks that notion provides or using templates that it provides to you. 00:22:18.839 --> 00:22:34.839 And the alternative is you can pay for specialized tool like sales force, for example, or like a note taking up or Google docs. I mean, the writing in notion is just horrendous in my opinion because of the blocks problem because you can select text with keyboard or across multiple blocks. 00:22:34.839 --> 00:22:36.839 That's what killed it for me. 00:22:36.839 --> 00:22:51.839 What finally dawned on me and I think this is what you probably struggled with is because it adds mechanical disruptions to the act of writing because you have to think about blocks and structuring and all that. 00:22:51.839 --> 00:22:58.839 It takes away from this. This is a sheet of paper. And I just want to write and I'll figure about the formatting and all that later on. 00:22:58.839 --> 00:23:12.839 I found that very difficult to do in notion because just the product itself would force me to think about the structure of the document and bolding something or not or creating sections and all that first before I actually start writing. 00:23:12.839 --> 00:23:21.839 Whereas when I'm writing on a piece of paper, rough notes, that's not how I do it. I just write down whatever. And then I figured out, okay, how do I want to structure this? 00:23:21.839 --> 00:23:33.839 It's interesting. It's when the document, I mean, the software enforces a process on you and it doesn't do it explicitly. It does it implicitly just because of the way the interaction is designed. 00:23:33.839 --> 00:23:43.839 It's interesting. I noticed that in Google docs as well about a year ago, I think Google docs introduced page less mode where you can switch between paginated document versus page less. 00:23:43.839 --> 00:23:56.839 Actually, I think we saw that even when we were at Amazon, we were using quip. They remember quip. Actually, it's a good example of a opinion tool. A Salesforce body, I think Salesforce can abandon it from understanding. 00:23:56.839 --> 00:24:05.839 It's just very simple document experience. You just type text in there. There is no way to change funds. There is no way to change funds. You can change the theme for the entire document. 00:24:05.839 --> 00:24:19.839 But all you have is normal text, heading one, heading two, heading three, I think that's all you have. And like, bold in italics, but you can't use different funds. You can't align text in different ways. You can't insert tabs. None of that exists. That exists in word. 00:24:19.839 --> 00:24:23.839 In a way, word or Google docs, they are like a Lego blocks, right? 00:24:23.839 --> 00:24:30.839 Microsoft word for sure. Yeah. There are some macros and stuff there. Word is one of the most amazing pieces of software, I think. 00:24:30.839 --> 00:24:37.839 Because they've managed to introduce all of that stuff while still keeping that blank sheet experience, like you said, the paper experience. 00:24:37.839 --> 00:24:50.839 Actually, that might be a good example of the thing that you ask. That has become super complex, has many offerings. But yet, if you want to get started with it, it's a simple, I can type things in it and be done. 00:24:50.839 --> 00:25:06.839 Yeah, actually, the interesting thing about word is that for the basics, it's all just right in your face, all of those buttons for like bolding text and all that. Actually, Microsoft might have set these standards for this, even with the first Microsoft Office, whenever it came out. 00:25:06.839 --> 00:25:21.839 Perfect, I think it was a software which Microsoft, I think, acquired at some point. But then if you want to do something complex, it's all available. But then you have to work to get it out of that complexity pit that it has. Actually, that's why they created Clippy. 00:25:21.839 --> 00:25:33.839 I listened to that interview with Steven Snuffsky, I think. But I may be wrong. So that guy was the person who was the VP of Microsoft Office when they introduced Microsoft Office 95 that had Clippy. 00:25:33.839 --> 00:25:43.839 There was a two-part episode where he was talking about some people who were asking for a lot of new features for word, which were already available in the product. And it just couldn't find them. 00:25:43.839 --> 00:25:49.839 And yeah, they didn't think of anything better, but introduced an artificial intelligence agent in 1995. 00:25:49.839 --> 00:26:00.839 Well, before it's time, I think in today's world, like with chat GPT or Bard, that's an excellent thing, right? This is a typical use case. Like, how do I do this in word? And it will tell you exactly. 00:26:00.839 --> 00:26:07.839 Well, I guess what Microsoft is doing this again with OpenAI, right? So yeah, the Clippy was very artificial and that was very intelligent. 00:26:07.839 --> 00:26:16.839 It was also interruptive. It wasn't a thing where you went to it and asked for it. It would interrupt and say, hey, it looks like you're writing a letter to your mom. 00:26:16.839 --> 00:26:18.839 So let's me help you with that. 00:26:18.839 --> 00:26:21.839 And you're like, look at it like a cash legend and stretch sheets. 00:26:21.839 --> 00:26:29.839 Yeah, but I think that the three don't do me. Then when I was listening to this podcast is how complex products are just difficult to use. 00:26:29.839 --> 00:26:35.839 And people just can't find features that you think, all we add this, we add that. And then nobody can find this. 00:26:35.839 --> 00:26:41.839 Even though they need it, like, how do you make the discovery more easy? And that's probably a problem with Lego blocks. 00:26:41.839 --> 00:26:52.839 Actually, the interesting thing about Lego blocks, right? When we say Lego blocks, we actually mean like a bunch of Lego blocks spread over the floor that you can use to build anything. 00:26:52.839 --> 00:26:56.839 Whereas actually, if you buy a Lego kit, it's a guided experience. It is a kit. 00:26:56.839 --> 00:27:02.839 It's a step by step. And you come out with an airplane or something at the end, right? 00:27:02.839 --> 00:27:06.839 But then you can break it and build a car from the same parts if you want to. 00:27:06.839 --> 00:27:10.839 So Lego actually does both things. Lego is a company. 00:27:10.839 --> 00:27:23.839 One thing I'll go back to is a specific example of Lego blocks versus model airplanes that we were talking about a few days ago for Metacas was the playlist experience we want. 00:27:23.839 --> 00:27:32.839 I think there are tools available that allow you to create your custom playlist with lots of different settings anyway you like. 00:27:32.839 --> 00:27:41.839 And there are tools like Apple podcasts that don't have playlists. You just pick an episode and listen and be done with it, right? Thank you. 00:27:41.839 --> 00:27:49.839 So we were trying to do something in the middle. And this is the discussion we had. Like, should we have custom playlists? If so, what would be the user experience? 00:27:49.839 --> 00:27:57.839 And I think eventually we fell back on keeping it simple. There's going to be just one listen later default playlist. 00:27:57.839 --> 00:28:06.839 Yeah, it's like your queue of episodes you want to listen to after this one. And you curated yourself. It's pre-created for you. You don't have to create it. 00:28:06.839 --> 00:28:14.839 You don't have to figure out how where to go to create a playlist and all that. And later on if we need to, we can still introduce custom playlists. 00:28:14.839 --> 00:28:23.839 And we believe that not many people are actually going to use the playlist itself. Forget about like creating custom playlists. 00:28:23.839 --> 00:28:30.839 Yeah, it's always a question like, who do you optimize for? It's an interesting question because for a small company, let's say like overcast, right? 00:28:30.839 --> 00:28:41.839 Build by one person. I think they charge like 10 bucks a year. It's a very cheap app. It has all of these sophisticated features like those custom playlists and custom RSS feeds and all that stuff. 00:28:41.839 --> 00:28:49.839 That's probably a very small percentage of podcasts listeners value. But because this is a one person company, it doesn't have to be billion dollar company. 00:28:49.839 --> 00:28:55.839 Whereas for us, we are three person companies, so we probably need to be bigger just to be able to pay ourselves, right? 00:28:55.839 --> 00:29:02.839 But but also like we'll talk more about that when we launch, but our value prop is such that we need to have a lot of users. 00:29:02.839 --> 00:29:11.839 We don't need to have 50% market share, but something on order of like, I would say 2 to 5% market share. This initially, it's kind of this feed spot where it's like a lot of money. 00:29:11.839 --> 00:29:16.839 But at the same time, it's a lot of usage because we have a beautiful network effect in our app. 00:29:16.839 --> 00:29:20.839 Not another network effect. It's more like economy of scale. A heavy bit of an economy of scale. 00:29:20.839 --> 00:29:29.839 I think we could easily survive on like half or 1% even given that at least 5% of our users are like paying customers. 00:29:29.839 --> 00:29:42.839 However, at 2 to 5% yes, that vision that we have becoming the Google for podcasts or Wikipedia for podcasts that becomes real. 00:29:42.839 --> 00:29:48.839 Yes, because then we will have enough revenue to support the infrastructure that investment for this. 00:29:48.839 --> 00:29:55.839 So yeah, whereas if you take Google or Apple, they need freaking 30% share to justify anything because of their scale. 00:29:55.839 --> 00:30:01.839 For example, Google is killing the Google podcast app, which has how many million users? 00:30:01.839 --> 00:30:04.839 I forgot the number but millions. Yeah, a lot. 00:30:04.839 --> 00:30:08.839 Yeah, and they're trying to make people move to YouTube music podcasts. 00:30:08.839 --> 00:30:10.839 It's like blow to wear. 00:30:10.839 --> 00:30:16.839 Which is following Spotify basically. It's like Spotify has music and it has podcasts. So let's do it. 00:30:16.839 --> 00:30:21.839 Yeah, like music and audio books and I don't know, they probably add like audio messages there because it's also audio. 00:30:21.839 --> 00:30:25.839 Why would you go for audio anywhere else? Which is the reason everything is Spotify. 00:30:25.839 --> 00:30:30.839 There is a trend to go for these super apps for sure. 00:30:30.839 --> 00:30:37.839 It's also actually cultural, right? So perhaps emerged in Asia but never really caught in the US or Western Europe. 00:30:37.839 --> 00:30:41.839 No, actually I don't know any super app that is successful here. 00:30:41.839 --> 00:30:45.839 We talked a lot about all of these products but I don't think we've come up really with principles. 00:30:45.839 --> 00:30:49.839 Maybe that's what we should try to do. Like how do you make the decision? Let's say you're starting a company. 00:30:49.839 --> 00:30:56.839 How do you make a decision? Are you a legable company or are you a airplane model kit company? 00:30:56.839 --> 00:31:02.839 I'll give you my instinct, my answer for it. I think every person is probably going to have a slightly different answer. 00:31:02.839 --> 00:31:06.839 I thought you were going to say every person's going to agree with you. 00:31:06.839 --> 00:31:10.839 Every person is going to have their own Lego blocks to answer this question. 00:31:10.839 --> 00:31:20.839 But as a new company, especially if it's a small company, I don't think it's a good idea to go build a platform. 00:31:20.839 --> 00:31:28.839 Because you don't really know what you're building for and you're probably going to end up with trying to build everything all at the same time. 00:31:28.839 --> 00:31:34.839 For a small company, what's most important is to figure out, is there a need for this product? 00:31:34.839 --> 00:31:42.839 And is there a specific, maybe niche even set up customers who are excited by this, delighted by this product? 00:31:42.839 --> 00:31:48.839 If so, then you can start to build on that. But you need to arrive at that first. 00:31:48.839 --> 00:31:51.839 Yeah, I wonder if venture capital plays are all here as well. 00:31:51.839 --> 00:31:56.839 Because you could build a product like bear for example. And just be totally COVID. 00:31:56.839 --> 00:32:02.839 Whatever market share it has, probably very small. But it's very much loved by markdown people. 00:32:02.839 --> 00:32:07.839 Whereas if you have hundreds of millions of funding, you just have to expand. 00:32:07.839 --> 00:32:12.839 And then you start building a platform just because you have to justify growth at all costs. 00:32:12.839 --> 00:32:22.839 Yeah, but typically venture capital for this kind of a thing would not even come through until you start to see some success or at least in today's world. 00:32:22.839 --> 00:32:31.839 Like three years ago, yeah, maybe just based on your name or whatever you've done in the past, you could like a humongous ton of money and go build a team and build something. 00:32:31.839 --> 00:32:40.839 But the point about trying to validate what you're building is true regardless of the size of what you're kind of build for that. 00:32:40.839 --> 00:32:51.839 I think the first approach almost always is going to be the model airplane kind of thing because it is cheaper to build those than a Lego block thing. 00:32:51.839 --> 00:32:54.839 Then you break it down into Lego blocks. If you need to. 00:32:54.839 --> 00:33:08.839 And actually would it help building your model airplane to actually use service or interact or right where you can basically build it out of internal Lego blocks that then later with some extra work you can externalize. 00:33:08.839 --> 00:33:12.839 That's pretty much how Amazon built it will be us initially. 00:33:12.839 --> 00:33:19.839 So this is a good segue. I mean, the venture capital stuff into what we've been reading and listening and watching recently. 00:33:19.839 --> 00:33:30.839 And I've just finished watching V crashed. Oh my god. This is such a fascinating story. I know how much of it is fiction versus reality. But it's a story about the we work. 00:33:30.839 --> 00:33:47.839 So it's a story about Adam Newman, who was the co founder of we work and how all of that grew from small co-working space into an empire that spans all of the countries and continents and all that while burning through cash like crazy. 00:33:47.839 --> 00:33:53.839 It shows how the desire for insane growth really changes. 00:33:53.839 --> 00:34:02.839 Culturacy changes people how like people just do insane things just so that they can grow at the rate that they're expected to grow. 00:34:02.839 --> 00:34:09.840 And it's not a pretty story. And now I think as of this recording we crashed did it go bankrupt or the world acquired with somebody? 00:34:09.840 --> 00:34:12.840 They filed for bankruptcy couple of weeks back. 00:34:12.840 --> 00:34:23.840 So the company that I actually don't know the facts, but in the series when JP Morgan was talking to them about the IPO that pricing was at $63 billion dollars. 00:34:23.840 --> 00:34:26.840 The valuation. And then now it's worth a few hundred million probably. 00:34:26.840 --> 00:34:33.840 It's an incredible story of just how it crashed and burned. Now it's become popular to I guess bash those high growth startups. 00:34:33.840 --> 00:34:38.840 There was a great series about Uber. There was a great movie about a Blackberry. 00:34:38.840 --> 00:34:44.840 What was that super pumped? Super pumped yes. And then the Blackberry one is really cool. 00:34:44.840 --> 00:34:54.840 It's basically like how you have all that success. Super growth. But then it all like either the whole thing crashes or specific people crash and burn while doing that. 00:34:54.840 --> 00:35:04.840 But it also like again I guess showed me that culture of growth at all costs. How it could be really not really providing any value to anyone at the end. 00:35:04.840 --> 00:35:13.840 I haven't watched we crashed yet. I want to I do feel that that is a very valuable and valid product. 00:35:13.840 --> 00:35:17.840 Not we work specifically but co working spaces. 00:35:17.840 --> 00:35:23.840 Especially in today's world like post COVID where most companies are pretty much hybrid. 00:35:23.840 --> 00:35:28.840 You don't know when people like working from home once in a while coming to work once in a while. 00:35:28.840 --> 00:35:40.840 I think this sort of thing can really work well. But I think this also highlights the grow at any cost kind of mindset that you're forced into. 00:35:40.840 --> 00:35:44.840 I think similar to Uber we work it elevated the whole industry. 00:35:44.840 --> 00:35:50.840 Actually airlines industry is kind of in the same same realm where it's really valuable to the society. 00:35:50.840 --> 00:35:55.840 But you can't build a profitable business out of it. Nobody has figured it out yet. 00:35:55.840 --> 00:36:00.840 I think I read somewhere that cumulatively the airlines industry is not profitable. 00:36:00.840 --> 00:36:06.840 I think Uber they probably still don't have any profit today. So we work was never profitable. 00:36:06.840 --> 00:36:12.840 Interesting right this is helpful and useful. But if you can't make money out of it who's going to pay for it. 00:36:12.840 --> 00:36:18.840 Yeah. What's the other I finished the Elon Musk's book Walter Isaacson's Elon Musk biography. 00:36:18.840 --> 00:36:30.840 It's a fascinating read. At least into it. It's 20 hour audio book. It's very long. It gets very dark in some places I had to take like a one month break or so from the book because it just was going to dark. 00:36:30.840 --> 00:36:40.840 But what I started to appreciate after the book is just how complex a personality Elon Musk is. It's not easy to be Elon Musk. 00:36:40.840 --> 00:36:43.840 Mark your real I think is the word. 00:36:43.840 --> 00:36:49.840 Yeah, the aspergers and potentially bipolar that he was talking about there. That's not fun. 00:36:49.840 --> 00:36:56.840 It shows you what you need to have to achieve this kind of greatness by society standards and the cost of it. 00:36:56.840 --> 00:37:03.840 I wouldn't want to be Elon Musk. I wouldn't want to fulfill that kind of level of ambition at the cost that this comes at. 00:37:03.840 --> 00:37:18.840 I started reading I think I was about five hours or so into it and started seeing some of what you're talking about and then you told me that it gets really dark and I stopped it and I moved to the hitchhiker's guy to the galaxy because that was mentioned in the Elon Musk. 00:37:18.840 --> 00:37:20.840 Pretty big time. 00:37:20.840 --> 00:37:21.840 Multiple times. 00:37:21.840 --> 00:37:22.840 Yeah. 00:37:22.840 --> 00:37:35.840 Anyway, I'll pick it up again. So recently I found two podcasts that I really like the first one I'm going to say it's fall of civilizations. So I listen to a lot of like history. 00:37:35.840 --> 00:37:47.840 I love history. Hardcore history by Dan Carlin is one of my I think all time favorites. So his episodes are like five six are each episode very very detailed. 00:37:47.840 --> 00:37:56.840 I'm so year a new episode comes out because it takes so much research and storytelling fall of civilizations is a little bit like that. 00:37:56.840 --> 00:38:12.840 It's definitely shorter a little less detailed or storytelling. But there's so much fascinating like history pieces to dig into right now in the collapse of the Mayan civilization thing and the Mayan civilization at a outset. 00:38:12.840 --> 00:38:19.840 I don't know about but I've never dug into the details of it. Why did it collapse? How big was it? It's fascinating so far. 00:38:19.840 --> 00:38:22.840 So it's all about the civilizations that have fallen. 00:38:22.840 --> 00:38:23.840 Yeah. 00:38:23.840 --> 00:38:36.840 And there's a lot of like 15-ish chapters right now or episodes and they're doing a book about it too next year. So that'll be interesting like with maps and everything that they talk about in the podcast. 00:38:36.840 --> 00:38:45.840 The other one is Trevor Noah famous comedian right and I love his stuff usually he has a book. 00:38:45.840 --> 00:38:57.840 I forgot the name of the book but if you haven't read it you must that is like a hilarious and yet very insightful into the human psychology kind of mix of books. 00:38:57.840 --> 00:39:12.840 And he was the host of the daily show and so famous coming anyway so he has started a podcast called what now it's presented by Spotify but somehow it's not exclusive to Spotify so yeah I started listening to it. 00:39:12.840 --> 00:39:24.840 I heard it on Justin Jackson's podcast I think the podcast industry news where I think he was speculating that Spotify can make some of its exclusives non-exclusive. 00:39:24.840 --> 00:39:32.840 Actually like Amazon Music has a bunch of podcasts that are produced by Amazon Music but they are available everywhere like the Metallica podcast for example. 00:39:32.840 --> 00:39:34.840 It just has Amazon all over it. 00:39:34.840 --> 00:39:38.840 I mean that's how podcasts usually are they're available everybody. 00:39:38.840 --> 00:39:45.840 Yeah I just don't know what economics with if Spotify pays them to be exclusive but then it like makes them non-exclusive I don't know how it works. 00:39:45.840 --> 00:40:03.840 Yeah I don't know. Anyway so his podcast the whole idea of that is in 2023 in today's world we are very polarized not just US pretty much every place you take every culture you take every discussion is very polarized. 00:40:03.840 --> 00:40:15.840 I disagree. Yeah. Where if I tell you that you know what I don't believe in that then you immediately like okay I befriend you I'm never going to talk to you. 00:40:15.840 --> 00:40:28.840 This is becoming very common you can't go to family dinners anymore because you don't agree with something that your brother-in-law talks about or your parents have a belief and you don't agree to it and you no longer it's become that fractured. 00:40:28.840 --> 00:40:35.840 I think what he's striving for is going back to that discussion kind of mentality that we had. 00:40:35.840 --> 00:40:51.840 Yes like I don't agree with what you're saying but I'm going to listen to what you're trying to say from your perspective and then we'll have a discussion about it and maybe at the end of it we won't still be like we'll come back with it but I'm not going to like befriend you over this. 00:40:51.840 --> 00:41:00.840 Befriend or different. Yes, yes. Befriend is the opposite or like befriend is I want to be friends with you. 00:41:00.840 --> 00:41:10.840 It's an interesting phenomenon let you're describing right because people became a lot less tolerant but maybe it's also because the sides became much more extreme. 00:41:10.840 --> 00:41:19.840 I think the problem arises where you're of certain belief I'm you know how a different belief and therefore we are enemies right which is based on that. 00:41:19.840 --> 00:41:34.840 However what might be difficult is if you and I have different opinions my approach is that we should just ignore that topic altogether just not discuss it because like the moment we start discussing it that creates the ground for disagreement and an argument. 00:41:34.840 --> 00:41:48.840 In most of those I think issues are so complex they cannot be resolved like I'm not going to convince you not going to convince me we will just become even more convinced in our own righteousness and also convinced that the other parties are jerks. 00:41:48.840 --> 00:41:58.840 On the specific thing I disagree and I'm going to befriend you Ilia from my life now you're a polar opposite on this. 00:41:58.840 --> 00:42:14.840 Yeah, I feel like there is space to have discussion as long as both sides are tolerant of each I think it comes from trust and respect for your friends and your family regardless of what their beliefs are. 00:42:14.840 --> 00:42:32.840 Sure you have a belief we can talk about it we don't have to talk about it we can talk about it if we want to but that doesn't mean that I'm going to like completely eliminate you from my life because of your specific belief on a piece of thing that I don't know if it concerns both of us. 00:42:32.840 --> 00:42:47.840 Yeah, I feel like when you actually start arguing about the specific thing then that starts to eat at the trust that you have in other things because like if somebody believes in something stupid from my perspective and they argue about it. 00:42:47.840 --> 00:42:56.840 You need to find that like okay this is getting into a territory where we are losing trust for each other it's not worth doing that. 00:42:56.840 --> 00:43:05.840 Yeah, okay so on that very serious note shall we close this out yes, so are now where can people find us. 00:43:05.840 --> 00:43:17.840 So the podcast is at Metacast podcast.com we also have our app landing page right now at Metacast.app when the app comes out if you sign up for it on that landing page will let you know. 00:43:17.840 --> 00:43:37.840 And we're in close beta so if you're interested if you have not signed up already let us know we can get you signed up you can send us an email with any feedback that you have for this podcast any kind words or bad words or your polar opposite beliefs at hello at Metacast podcast.com. 00:43:37.840 --> 00:43:47.840 You know what I started doing recently like when I get an email like unsolicited spam email I reply marked as spam thanks and marked as spam. 00:43:47.840 --> 00:43:57.840 And what does that do? I think it tells these people that their email was not welcome and also it gives Gmail a signal that these people as spammers. 00:43:57.840 --> 00:44:04.840 I mark as spam very frequently but I don't actually reply to them saying marked as spam. 00:44:04.840 --> 00:44:13.840 I am just being a bit more aggressive here but yeah anyway so if you send us any polar opposite feedback do wear your marked as spam. 00:44:13.840 --> 00:44:23.840 Because as John don't go or say on one of his podcasts we don't care about your constructive feedback you only want to see good comments. 00:44:23.840 --> 00:44:27.840 No I'm just kidding constructive stuff is always welcome. 00:44:27.840 --> 00:44:34.840 And we can't prevent destructive stuff even if we wanted to so everything is welcome. 00:44:34.840 --> 00:45:00.840 Thanks for listening bye. Bye.