Every time somebody interacts with something that I created for PlanetScale, I want them to be able to walk away having felt like I became a better engineer, I became smarter, I learned something new regardless of whether they ever buy a database from PlanetScale.

I'm joined today by Ben Dicken, a developer educator at PlanetScale. You've probably seen one of his articles because they have blown up and they're absolutely amazing. We dive into Ben's background, his process, and we talk about what you could learn as a DevTools founder thinking about content. Your content has been really blowing up on Twitter, but it's just it's just really good. It's not that, like, click based stuff.

It's like you look at it, the more you dig into it, the more effort it's obviously put in and the more we learn. And so I I want to ask a little bit about your background because I know it's it's quite unusual.

Yeah. Yeah. So I I have a bit of an interesting path with how so right now, I'm a developer educator at PlanetScale, and a lot of what I do, like you said, is creating content that is engaging to developers and also meeting with customers, doing customer trainings, so some more of, like, practical hands on stuff. But my background's a interesting combination because I started my career as a database engineer at a different database startup. So I was in that for a little while, and then I actually went into academia for a little while.

So I was a faculty member at a computer science department teaching everything from intro programming to operating systems and system software. And so from that is where I went to planet scale. So I kinda had this interesting combo of, like, I've done formal teaching and learned all about formal teaching methodologies, right, training the next generation of of programmers as well as being an engineer myself. And now this in my head, it's kinda like the perfect job because I get to blend my, like, experience of teaching. But also, one of the things I had to learn coming to PlanetScale is I have zero marketing training.

Right? And so this is, like, kind of a marketing role. Right? So you have to be able to, like, know, k. I wanna create excellent high quality blogs or videos or documentation or just general, like, resources for engineers to use our product, but how do I do that with excellence, but also in a way that's, like, engaging and draws people and attracts people to your product, your website, whatever the case may be.

And so, yeah, it's an interesting combination. It's an interesting path that I've been on, but I'm actually really thankful for it because it's given me all the skills that I need to hopefully do really well at this. So

Yeah. I I mean, I'm pretty sure you probably, like, probably, like, everyone listening here is gonna start just hitting up every computer science department and trying to hire hire the lecturers.

Yeah. It's very I mean, I it was unique because, you know, typically what you think of in a in a computer science faculty is like, I was way below the average age of the faculty there, which is fine. Right? Like, but a lot of people that in academia are you know, have gotten a PhD and they're sort of lifelong academics and all this. And I was an academic for a while for a couple years, but it's it was kind of a unique placement for me at least at that point in my career.

Okay. So you're you're saying that that's maybe the average computer science department is not filled with bands that people can just start having.

You can look for them. Go for it. Yeah.

Could you maybe just like before we dive in too much, your stuff has been doing from the outside, seems like it's just been driving an absolute ton of traffic to planet scale probably from Twitter. But also then, I think when you read it, you're like, woah. These guys. This is incredible. Like, the caching blog, that was the one that really blew me away.

I think I don't know how I discovered it, but it is like it's actually not even an article almost like it's almost like a a game or like a teaching resource. Mhmm. It's interactive. It's not just an article. It's got like these incredible animations, but then explanations are really good as well.

There's not even animations, like, actual, like, reactive Yeah. Animations. Like, you click and see how long each request was based on, like, whether it was cached or not and stuff like

that. Exactly. Yeah.

Maybe you could share a little bit, like, just these have been seems like they've been very impactful.

Yeah. Yeah. They've they have been very impactful. So the the first before I started writing these for PlanetScale, I've done some stuff with interactive articles. Have a couple old ones on my personal blog and some other work that I've done in the past.

But the first one for PlanetScale was it's about a year ago. I was just experimenting for fun because I'm a nerd and I just like to code in my free time too. But with some visualizations for b trees, which are kinda like if you know databases, that's like the most important data structure for databases. All of your indexes usually are built in b trees. So I was just experimenting with that for fun using d three JS.

And eventually, I was like, oh, I'm building all these cool interactive things. What if I could, like, integrate this into a blog? And it took a lot of work to make our blog, like, be able to support that. And when we finally published that, and it wasn't, you know, it wasn't to align with a product launch or anything, it was just like a for fun side project I was doing, It did super well. Like it like you said, kinda blew up on Twitter.

It was on the front page of Hacker News for a long time. It got like, I don't know, 80,000 page views in the first week or two, something like that. And so I'm like, okay. This is something I like and I had a lot of fun doing. And it's something unique.

I mean, are other people that do this. I draw inspiration from some other people like Sam Rose or I'm not gonna pronounce his name right, but Bartoz, Shyschanowski, something like that. There's a few other people that do these, in my opinion, even better than me, like interactive articles. But it's still very unique, and I kind of have my own unique style and twist to it and stuff. And so it it worked.

I'm like, well, I'm if if we can keep doing these, I'm happy to do them because I like them, and they and they work. And so Sam, our CEO, also really liked it and realized like, hey, people really love these. They enjoy the excellent content. And I think the other thing too is like in this day and age, it's very easy to put out just an average blog, especially with AI. Right?

You can ask AI to write you a draft and then you tweak it a little bit and you can ship a whole blog in a couple of hours if you want to. Right? But it's like how do you separate yourself from just the the massive amounts of average content or below average content that's out there and this is one of the ways to do that. So anyway, we started continuing to do that. I've published a couple since then.

I don't have one coming out every week because they take a long time to make. Each one is many many hours of work. It's usually over the course of a month or two, not a month or two just dedicated to the blog, but coming back and forth to that blog. So yeah, they're really cool. I enjoy them and I I guess people like them.

So

And by the way, when you're saying that, the one the other creator reminds me of this like, have you ever seen like Free Blue, One Brown?

Yes. I love him. Yeah.

Yeah. Yeah.

Really interesting. And you know, he's different format on video, but a similar kind of thing where he I actually listened to a podcast with him recently with Lex Friedman. He was on there like a couple years ago, so I was listening to an old episode. And he's similar in that like when he makes a video on something, he spends a long time. He dives deep, he reads papers, all this kind of stuff, and then he starts to formulate his ideas and put them into video.

I may not go quite as deep as him, but it's similar. Like it takes a lot of research and planning and building these visuals to make them look good. So yeah.

Scaling DevTools is sponsored by WorkOS. If things start going well, some of your customers are gonna start asking for enterprise features. Things like SSO, skin provisioning, role based access control. These things are hard to build, and you could get stuck spending all your time doing that instead of actually making a great dev tool. That's why WorkOS exists.

They help you with all of those enterprise features, and they're trusted by OpenAI, Vercel, and Perplexity. Here's what Kyle from Depot has to say about WorkOS.

We use WorkOS to effectively add all of the SSO and SCIM to Depot. So for us, we can effectively offer SSO and SCIM, and it's like two clicks of a button, and we don't ever have to think about it. It's like one of the best features that we can add to Depot, and it's super affordable, which effectively allows us to, like, break the SSO tax joke and essentially say, like, you can have SSO and SCIM as, an add on onto your monthly plan. So it really allows smaller startups to essentially offer, like, that enterprise feature without a huge engineering investment behind it. Like, it's literally we can just use a tool behind the scenes, and our life is exponentially easier.

Thanks again, WorkOS. Back to the episode. And then sorry. It's just like really one side point, but this is not really about scaling a DevTool. But it's cool that, like, you've, like, somehow, like I felt like this is a really this is probably, like, the best I I can imagine if someone's, like, a a college professor and they see this content.

It's the kind of thing they would, like, actually want to share to their students because it's probably, like, quite hard to find such good content because it just wouldn't make sense maybe in in house skills, like, financially for someone to spend so much time as a teacher, but it does make sense as a marketing exercise. And it's like just bringing something cool to the world that exists because of, startups and I don't know. Just

Yeah. I mean, it's it's surprising to me because, like, there's even sometimes, like, people who I know who are really good engineers. Right? Like, it's not like, oh, they don't know anything about this stuff. They're good engineers, but they say, like, hey, Ben.

I read your blog and I, like, learned something new about processes or whatever, this or that, or it clarified things for me. And I'm like, wow. Like, I can it's not just about like, try to to write these blogs where, you know, someone who's not that familiar, maybe someone who's not even in the tech space could read it and be like, okay. I learned some things. Like, maybe there was parts that were confusing or whatever, but I learned stuff.

But then also to try to provide value to someone who's like an engineer and maybe already knows like half of this stuff. Right? But can you provide new perspective or, again, the the unique visuals to like reinforce concepts? And so that that's makes me very happy when I can, like, both have someone who doesn't know much about it, read it, and be like, wow. That was super interesting.

And someone who's, like, an engineer that I respect and go like, wow. That was really great. Like, I even learned something from that. You know? And I get that pretty often, and I'm kinda surprised but thankful for it.

Yeah. That's awesome. That's awesome. Okay. So let's say someone's listening to this and they're like, okay.

What's what's your process here then? Like, how how do you go about creating a a great article?

Yeah. That's a tough one. So, yeah, like I said, one, it takes I I think, one, a lot of research, but also, like, you actually should care about, like, what you're writing about. And and that's one of the things I'm thankful for at PlanetScale is, you know, my boss Holly and Sam, they give me a lot of leeway to, like, come up with these creative ideas. Right?

Rather than just, you know, sometimes, hey. We need x y or z documentation, whatever, but then there's also freedom to be like, Ben, we trust you that, like, you're gonna create something really interesting, really engaging, something that is gonna reflect positively on our brand. And so that's what I get to do with some of these. Like, for example, the other another one that I wrote was one on IO devices and latency. So this is one where I simulate hard drive disks and solid state drives.

And so this is one that we actually wanted to time with the launch of PlanetScale Metal, which was earlier this year. And initially, we planned on launching that sort of late twenty twenty four. And I was, like, under a big time crunch to, like, get it written, and I was, like, cramming on this and it was fun, but like I was like having to push really hard for this deadline. And then we ended up actually postponing it to the next year. Not for bad reasons, it was for good reasons and and strategic and all that.

But what that meant is all of a sudden I'm like, oh, I have like a couple more months to just like work on this and really like make it good and refine it. And so I worked on other stuff, of course, in the meantime. But it was it was fun to be able to like that one I was like really able to take my time because I could work on it for a couple days and then I could leave it alone for a while and come back and then work on it a little bit. And that's, I think, actually my favorite one because I think the visuals in that are the most interesting, and and I think that's the one I've gotten the most positive feedback on. I know, like, sometimes you have a time crunch, and sometimes you have to do something under a time crunch, and sometimes you're not gonna make interactive visuals.

But I think one of the things to like, my perspective, because I've been in an engineering role and I've been in a teaching role now in this role is like, I genuinely think, like, would an engineer want to read this? Because what at PlanetScale, right, we're a dev tools company. We're a database company. Right? And if you're in that space, like, that's who you want to attract to your website or to your brand to learn more about your product even if it's indirectly through, like, an article you're writing or something.

And, like, I think not enough people sit down and be like, would would it you know, I'm trying to help PlanetScale grow as a database company. Would a database engineer find this interesting? Or if you're building some other kind of dev tool, right, would a DevOps engineer want to do this? Or would a front end engineer want to read this? Is this gonna provide educational value for them?

Because one of one of the things that I really believe is, like, every time somebody interacts with something that I created for PlanetScale, I want them to be able to walk away having felt like I became a better engineer, I became smarter, I learned something new, regardless of whether they ever buy a database from PlanetScale. Hopefully, do, but I still want there to always just be, like, real educational value. And I think that comes through, and then people start to respect your brand more and associate it with high quality content, talented engineering. And we have some of our engineers that publish blogs too, and some of them are, like incredible writers. And I think encouraging your engineering team to like write about the interesting stuff that they're doing is also a great way to like build that reputation.

Curious, Ashley, when you work with engineers on writing, is there any are there any kind of common themes that, you know, you bring from, you know, working as a lecturer, studying education probably educate how, like, educational theory and stuff at all, like, plus just doing a lot. Is there anything that you kind of commonly say to people?

Yeah. I mean, it's actually funny. There's an engineer who we recently published a blog for now. He he used to be at PlanetScale. His name is Vissent or v m g.

He works at Cursor now. But he's actually one of the I think

I saw those. Yeah. Was he Did

you see that published yet?

He's in Spain or something?

He's in Spain. Yeah. Yeah.

Yeah. I saw him. I don't know. Yeah.

But he he's actually like someone who I also really look up to as a writer, because he is just he he's very smart, but he's actually also just like naturally very good at writing. Like, I was reviewing one of his posts and I'm just like, I'm blown away at like how you're you're explaining so much, but you're also doing it in a way that is, like, accessible to someone even if I'm not, like, super familiar with the project you're working on. And that is something that, you know, it's hard. You can't just, like, teach it in a day, but I think it's actually extremely valuable, not just if you're in, like, a dev ed role, but, like, every engineer, if you learn to write really well, that can pay, like, huge dividends. Right?

Whether it's for your personal blog or that to, you know, promote, like, the products that you're working on at your company. Because I think it's actually very I would hope it's very fun to, like, you put all this effort into building a product, and then if you can learn to write really well about it to, like, show off what you did and, like, all the cool engineering that went into that, that's something that's really cool, and that's something that I and my boss, Holly, are always trying to, like, encourage our engineers, like, write about the cool stuff that you're doing assuming you can talk about it unless it's a secret project or something. Because it's like, hopefully, like, it's fun for you to reflect back on what you did. But also, again, like the outside world, when they see something like that, that builds a positive association with your brand. Like, wow.

This company is really talented engineers. Because when people come to PlanetScale, it's like, yes, they're buying our database product, but they're also kinda buying into just like, we have a lot of talent density in the company. Right? People who have worked at huge large scale databases. And so when you, like, kinda show, like, hey.

We have a very, very talented group of people that builds trust. Like, these people know how to run databases. Like, look at all of this work that they're putting out. Like, I can tell they really know it. And so that's especially in the dev tool space.

Right? Because you're selling to other engineers and they wanna know that you have an engineering team that has their back and is gonna operate like whatever whatever you're doing for them really well.

Yeah. That's a that's actually a good point. And I felt like it's almost like in a way, I felt like some engineers would be like, oh, okay. You've got these engineers that did this at x company. Like, oh, well, maybe I don't even care that much.

Like because I'm like but then they start reading the article and it's like, oh, okay. This person really knows their stuff. Yes. Wow. Like, I don't know.

Because I feel like engineers are often quite like, oh, like, that doesn't mean anything on its own. Like, kind of like Right. Do do you know what I mean? Like

Yeah. Yeah. And well, and that's the other thing too is is as an engineer, I think it's easy to undersell how cool the stuff is that you worked on, which I think is what you're getting at there. For them, it's just like, oh, I just built this cool thing and it works and it does great. But, like, sometimes they don't realize, like, how many cool innovations that, you know, went in to building this database infrastructure or networking infrastructure in our case.

Right? Stuff like that. And so, again, it's it like, it's difficult because also if you're an engineer, you have all these engineering responsibilities. But when like, it it would be I would be doing my job if I never had to write a blog post again if I could just get everybody else to write cool blog posts about what they're doing. Right?

So anyway, I I love that kind of stuff. And then again, I appreciate that kind of content from other companies as well.

Yeah. Do you have any tips actually on getting helping engineers or like prodding engineers to write from posts?

Yeah. Nothing specific. I I get, like, some of it is just depends on the person. Right? Some people love to write.

You know, I was talking with our CEO, Sam Lambert, recently. Just yesterday, I think, about writing and and good writers that are also engineers. And he was talking about early GitHub culture because he was a part of early And like everybody he said was a good writer. Like every engineer there was good. And a part of that is because it was a remote first company where, like, when you were building a new product, designing something, like, you buy force, had to be really good at writing.

Right? Because you're communicating over whatever, you know, Slack these days or whatever it was back then. So and and we have a similar thing. Like, we are a fully remote company, and so we actually do have a lot of good writers in the company, but a part of it is we're so focused on building products that sometimes it's hard to pull back and and do the writing. So I think that's that's a part of it.

It's just like how are you prioritizing you know, marketing is very important, but engineering, obviously, if you're a dev tools company, it should be the priority. And that's what we do at PlanetScale. Like, we are very 90% of our company is engineers. Right? So, like, our marketing and sales is very small and nimble, and we focus on just building really good products.

And so I think it would be cool though to, and and that every company has to decide this. Like, how much effort can you take from an engineer and put it into things like writing or creating documentation for your products? And that's another thing too. Like, we have like, we ship very fast, and so sometimes it's hard for us to keep up with, like, documentation. Like, how do we document all this great stuff that people are putting out?

And so it's very cool when engineers are like good at writing and can document the stuff that they're building to show it off to the world.

Mhmm. Mhmm. Thinking about like what the kind of big takeaway for people listening, I feel like it's like it's almost like that you had that Paul Graham quote that was like classic Paul Graham quote. It's like how to get good start up ideas. I I would go down.

I'm probably gonna say get us wrong and get, like, quoted by the guy saying I didn't say that. Sorry. Anyway, like, become how to become good at start off ideas is to be someone who has good start off ideas. And Yeah. Yeah.

Is to become sorry. To become someone who has good start off ideas. Yes. It kind of sounds like a stupid quote when you first hear it, but, like, kinda makes sense in that it's not really like a shortcut or, like, just like, you kinda just it's hard and, like, become a good writer, like, write good content. It's like, become someone that's good at writing.

It's like just probably just the process of like hundreds of hours or whatever of just actually doing it, trying to make it better, getting feedback, like, I

don't know. Yeah. It just takes a lot of practice. And again, like, my perspective is maybe a little different because I I literally like, when I was a faculty member, I spent a lot of time with like, there was a consultant who would who would come into our department and talk about different teaching strategies and different ways of running a classroom and creating content, you know, in an academic environment. But it's just like, I, for so much of my life, have already been thinking about like, how do you teach and how do you communicate?

Not that I'm amazing at it, but just like I've spent a lot of time thinking about it. So maybe I need to write more about that. Right? Writing more about like how to teach would be kind of a cool thing. There's something else I was gonna say, but now I'm blanking on it.

Well, while while you're blanking, can I just say can I just absolutely disagree with you that you definitely are amazing at it?

And if

you're not amazing Well, you. Well, hope of the rest of us got. So you are. But basically, it was like how to become how to write good content. You've been thinking about it for a very long Yeah.

It's it it's thinking about it a long time. And the other thing too, and and it just takes a lot of work. Like, I spend a lot of time on these blog posts. Like, just for example, when we did our Postgres GA a week or two ago, that's we released this processes and threads blog that I wrote soon after that. And, like, I was up super late, like, on Monday night, like, working on this, like, you know, till 11:30 or whatever.

I guess that's not super late for some. But for me, that's kinda like because I'm an early riser. And, like, it just takes a lot of work. Like, if you're actually gonna build really effective stuff and this doesn't just apply to, like, interactive blogs. This is just like being successful in your career in general at this point.

Like, it takes going above and beyond. Like, the way I like to think about it is, like, the standard or the bar should be exceeding the bar. Like, this the bar shouldn't be meeting the bar. That sounds like an infinite cycle or whatever, but, like, that's the way that I think about it. Like, every time I do something, it's like, okay.

Great. That blog was successful. I'm just like, okay. How can I do better the next time? Right?

Not like, oh, great. We did it. Right? Now let now let's kick back and relax. So that's it's that's a part of it.

It's like the mindset. Like, how hard are you willing to work for this stuff? Yeah.

Yep. And that comes easy, I guess, because you like and it's something And

and you yeah. Liking it makes it help a lot. Yeah.

Yeah. Awesome. I think those are the two. It sounds like are you are you happy with those as takeaways? It's like

Yes. Yeah. It's being passionate about it, working very hard at it, and then the working very hard at it also. The other thing I will say is like being someone who's willing to take input from other people. Like some people just kinda wanna do their own thing and don't want feedback, don't want input.

I am, like, constantly asking people, like, review this, give me feedback, tell me how I suck because I know something in here sucks. So tell me, and I wanna make that better. Right? Like, some people have a hard time taking that kind of feedback, but I want Bring people to yell at me per se. If you wanna yell at me, fine.

But I like want feedback because that's the only way you get better because we all have blind spots.

Yep. A 100%. Why do you think that most blog posts just die in the ether? You talked about stuff like ask whether an engineer would like to read this. Mhmm.

I think even that like it would be hard. Like I feel like some people just have different standards of what like, some people are like, yeah. Of course they would. Yeah. That's not Yeah.

Of course they would. I wrote it. Right? It's gotta be great. Yeah.

Yeah. It's it it's interesting because in marketing and I've learned everything I know about marketing from Holly, so shout out to Holly. But there's a there's a lot of different ways to think about, like, what you're creating for your audience. So there's, like, the stuff that you create for general awareness. Right?

Which that's the stuff that you want to, like, have large reach. Right? Go viral on Twitter or whatever. Right? But like that's less about like educating someone specifically on your product, but just again, it's that like building a reputation like of good engineering content.

So for example, in in our case, we do both MySQL and Postgres now databases. And so there's so much that can be written anywhere or videos created or docs created about those two databases just in a vacuum. Right? Because those are tech that have been both around for, like, thirty or forty years. And there's a lot of interest just out there in the community, like, regardless of even if you're not a DBA, just like engineers who are like, oh, yeah.

I know my back end is Postgres, but I don't I just use an ORM, so I don't even think about it that much or whatever. Like, there's so much to write about comparing and contrasting those or about, like, the pitfalls of logical replication in Postgres versus replication in MySQL, like, all of this kind of stuff. So I think, like you were talking about, like, putting yourself in the seat of the engineer and thinking about, like, if you're someone who runs a web app, you probably use MySQL or Postgres. And what are all the things that I can talk about that engineers are probably gonna either face this problem, then I can write about how you fix this problem or, you know, performance, at least for us, is a big thing. A lot of people come to us for performance because we have this planet scale metal with local attached disks in AWS and GCP.

So I like to do a lot of benchmarking stuff. Like, in fact, there's something I'm working on right now where Postgres eighteen just recently came out with some really cool new IO improvements to asynchronous IO, and a lot of people are talking about it. But what I am working on now is some detailed benchmarking comparing the previous version 17 to 18 so that instead of just talking about the concepts, like, one of the things engineers wanna see is, like, how is this actually gonna impact my database? And so I wanna, like, provide an incredible resource for people to reference on, like, how that's gonna affect things. So I think that's like if you can get your head in the in the mindset of the customer and, like, what are the problems that they're gonna face?

What challenges are they gonna run into? And then, like, if you can be the company that solves the problem, even if it was just like, hey. They Googled and they landed on your page and they found a blog post where you help them solve this, like, key performance problem or this key reliability problem or whatever it is, all of a sudden, they have this association in your head like, wow. PlanetScale must be really smart. Right?

Because they helped me solve this problem. So it's, like, hard. And maybe it's because I've been an engineer and and I've been in education where, like, maybe I have a little bit of an easier time getting in that mindset. But that's what I would encourage is, like, find people, your engineering teams, whatever, where, like, what are the problems that our company is trying to solve? And then how do we communicate that and represent the company as that we're the company that helps you solve these problems.

Yeah. That's that's really helpful. Is there a visual aspect to it as well? Because it seems like, you know, I I think he even launched, like, an extension of, Bulls Generate, Which we Yeah. We should clarify as, like

We should clarify that statement. Yeah.

I'll let you clarify that.

Yeah. Somehow, I became known on Twitter as the the bouncing balls guy because I I did. And some of them have just been kinda cheesy things. But, like, legitimately, like, you're asking, like, is there something to the visual aspect of it? And I I actually took a course in college, a data visualization course, and that was one of my favorite courses because it talked about the technical side of it, like how do you actually write the code to do data viz, but also like sort of the human perception and human psychology part of it, which is like, you know, down to little things like what colors you should and shouldn't use together or what kinds of things are engaging and what kinds of things aren't, what visuals represent things well.

And I'm not saying I always follow those like formal principles, but ever since then, I've been very fascinated by visualizing things and, like, making things interactive. Like, you see even for, like, when I publish benchmarks, I often publish them with, like, some interactive visuals where you can compare and contrast benchmark results. So I think that's, again, one of the things that, for one, separates you from from AI slop. Right? Like, at least yet, I can't just write, hey, AI.

Like, write something that simulates a CPU for me, and it it can't just spit that out. It probably will be able to in a couple years. But it also is just maybe it's just a little bit of my style. Right? And, again, I've been inspired by other people, but that's, something that, you know, I I get, like, feedback time and time again that people appreciate that it's unique or that they can sort of try things out themselves.

And it it attracts people. Right? It attracts attention, but then they they they can be attracted by that, but I actually want to also give value, not just, like, show visualizations and stuff. So have them stay for the actual value that you're providing.

Yeah. Come for the balls. Stay for the understanding of databases.

Man, we need a we need a better term for that. You know?

Yeah. That that's that's awesome, mate. Is Lou from TL Draw said, like, similar stuff around, like, make it visual. And I think, like, a lot of the stuff that really blew up for them. I mean, they have, some level of advantage here and that their product is just extremely visual.

And Yeah. Yeah.

Or even you think about like people like Theo, if you ever watch his streams, right, like or his videos on YouTube, he's maybe not interactive, but he's constantly pulling up like Excalidraw instead of TLDraw. Right? But in like diagramming things out and all of this kind of stuff and that's like, you know, he's way more successful than me, but, a niche that he's found of, like, wow. Engineers really enjoy, like, seeing things laid out visually, and he has a really good ability to, like, sort of on the fly or maybe he's planned it ahead of time, but, like, come up with a good way of visualizing a concept. And so that's a really good practice if you're gonna write is also to learn how to, yeah, like, visualize things in a unique way that will, you know, provide more value than just writing about it.

Hell, yeah. I'm excited. When's the course dropping?

When's the course dropping? Oh, good question. Yeah. You're giving me lots of ideas, Jack.

Sorry, Sam. Don't know. But Ben becomes a millionaire selling courses on that. We'll see. We'll see.

Okay. Amazing. Okay. So we're gonna try and do this. So it's, like, podcast friendly, but I am gonna pull up an article that you've written.

Yeah. Okay. So we have the IO devices and latency article that Ben wrote in March 2025. And maybe you could kind of talk us through probably less the very, very specific things that people will need to see, but, like, maybe, like, how you structured the article and, like, anything interesting.

Yeah. So the the context of this is this, we were trying to time this when we came out with PlanetScale Metal, which does things very differently than most cloud databases. Most cloud databases are on network attached storage, which we do as well, but we also have ones in the big cloud providers that have direct attached SSDs, right, which is kind of a unique thing. But it provides really good performance for our customers, and we tell, like, a lot of incredible stories. So this was all about IO devices to kind of celebrate, like, hey.

We're launching this cool new thing, and let's also provide you know, we mentioned metal a couple times in the blog, but for the most part, we're not talking about the product. It's just we're providing cool educational value in an interactive way about this stuff. And so, like, even that first visual up there at the top, right, that's kind of inspired by, you know, the same bouncing balls. It's bouncing squares instead. Right?

But, like, one of the things that actually if you're gonna do something interactive, it's very nice to have something that quickly grabs people's attention. Right? Because if you just go to a page and see like a wall paragraph of text, like let's be honest, 90% of people like read a paragraph and like, oh, shoot. This is like this is too much. Or or, you know, I have to go do something else or whatever.

Right? And then leave. But this is a way to, like, very quickly, that actually and, again, I know if you're listening, it's a visual that basically visualizes this the latency of an NVMe read on local disk versus a network attached read, and it kinda grabs your attention. You're like, oh, am I gonna learn about this in the article? Right?

Like and so that's kind of the goal of doing something like that. It's like instant education, but also like attracting your attention. And then we get into kind of the actual details of the blog. Right? So if you go down essentially, what I did is and this this took a while.

So, you know, if you're gonna do something like this, brace yourselves for a lot of work. But essentially, visualizing, like, it was kind of supposed to be like a a a brief history of how we have stored things in computers, like, in a permanent way. And so some of the earlier versions of, like, hard drives were these big tape systems. Right? So I wrote a little simulator for tape systems where the the spindles spin back and forth and you can see tape reads and writes happening.

I actually have a good friend who works on tape systems for one of the big tape companies. So it was kind of fun to, like, get to talk to him about real tape systems sometimes. So this is Incredible. Fun. Right?

But, like, just this one. Right? This took, like, many, many hours to, like, get this simulation down and make it look good in light mode and dark mode and to, like, make it work on all browsers. Like, there's all these, like, fiddly things, right, that take a while to do. But, yeah, like, just this.

And, again, like, most people look at this, and they're like, I have never seen this in a blog before. Right? Like Yeah. Or or maybe they have if they've been to Sam Rose or someone like that. But so yeah.

And then I you go down the list and then there's another you know, we talk about this and and there's ways to to look at the timing of latencies for reads and writes and you can kinda interact with it. And then it's essentially similar for, like, you do this on this version of it, and then there's another section below that's kinda the same thing, but simulating hard drive, like spinning disk hard drives. If you get on to kinda the next section or we could link her here and then we do the same for SSDs. So

Yeah.

Yeah. Sorry. Just and that's kind of the the interesting thing too is a lot of what this one was about is, like, learning how these things work, but the I love to talk about latencies because that's what PlanetScale is known for, very low latency databases. And so when you're talking about the timing of things, it's like we could just draw pictures of this, but let's what if we used actual human time to measure it? Right?

So a lot of these visuals, like, can click on a button and it shows you, like, how long this thing actually takes, and you have to wait there and sit and watch the device, you know, do the read or do the write. So that was kind of a fun aspect of this. It's like a play on time, almost. Yeah.

Yeah. It's incredible. Yeah. Well, okay. Sorry.

Sorry. Yeah.

Oh, you said you hadn't actually looked at this one before. Right? Or maybe not gone all the way through it.

Seen. It's like, no. No. No. This is the one I hadn't yeah.

This is just incredible. I yeah. Wow. I I don't even know how you would get this working on mobile.

The other thing too, so for those who do visual stuff, data visualizations, all of these visuals were built with d three JS, which I will say is a huge abuse of d three j s. Because it's really meant for like data viz or doing line plots and all of this kind of stuff. And I basically just used its tweening functionality to, you know, simulate hard drive disks and stuff like that.

It's absolutely incredible. Okay.

Yeah. Yeah. So again, just like I'm kind of reinforcing throughout the whole blog, like I'm giving this history, but we're also talking about the latencies of devices. So I'm trying to like reinforce these points. And again, as someone who is, you know, I am in, like, technical marketing.

Right? So as I'm writing this, I am thinking, like, one of the points I'm trying to drill home is, like, you want as low latency as possible, especially in the database world. Right? This is a core piece of infrastructure for companies from a startup that just started yesterday to the hyperscalers. Right?

Like, latency is bad. We wanna get rid of it. So I'm, like, driving this point home the whole time, the same the very same week that we launch an ultra low latency database product. Right? So it's kinda like you're almost building a narrative.

It's a true narrative. I'm not lying to you. But it's like you you're that's kind of a part of what you need to do. Right? You need to like build a narrative for your audience and be like, hey.

You wanna eliminate latency? And we're we have these like very engaging visuals that show that. And as you can see, it's very long, so you just keep scrolling forever.

Yeah. This is like honestly unbelievable, the amount of stuff in her. It's like incredible.

Yeah. And this is even something unique. So like this that little visual you were just on there, there's this concept when you're using EBS, which that's like if you're in AWS running servers there, most of your computers store things on EBS volumes. It's a network attached volume. But there's this concept of you have a limited number of IOPS, but you also get this burst balance where you can sort of accumulate IO operations in a little, bank.

And then if you need to burst and do a bunch of IO to disk, you can, like, use your reserved IOPS. But it's a concept, like, I think a lot of people, unless you're, like, a hardcore AWS engineer, just don't know about. And so not only do we talk about it, but I kind of visualize, like, if you slide that slider all the way to the bottom, you have this you have this, like, bank of IOPS that, like, slowly starts to yeah. That way on that side. Yeah.

It like slowly starts to accumulate and this is visually like, this is what's happening in AWS, you know, it's just an integer counter or whatever. But I'm like visualizing it and showing you like, hey, if your server isn't doing many IO operations, you're gonna build up a balance of IO that you can then use later. Again, it's just this, like, one of these niche little concepts in AWS that a lot of people don't know, and I got to visualize it in a fun way. And then if you suddenly slide it to the other side, you'll start to see that, okay, now now my server's doing lots of IO and I'm draining the IO balance that I have. Right?

That's very cool. It's like the So It's like in breaks or whatever, like almost like the is that the regenerative? That's what it reminds me of anyway. Yeah. Very cool.

So yeah. And yeah, all of this was again, like this this wasn't literally six months of just me writing this blog. There's lots of other stuff I did in the meantime. But this one was unique in that because we had this this delay in when we were launching, I had started it and it didn't get launched till six months later. And so I had time to really like refine it and go back and forth with it, which was fun.

It's actually fun because I feel like most people say like, don't, you know, put like a CTA in your article. But there is like a very strong at the bottom, it is a very strong, like, the solution is metal. Yeah. Plan is beyond just and and it works so well because, like, it's like if you were all the effort put into this article and there's, like, so much in there and it's like, people understand the challenges, I guess. Right.

Yeah. Like, can tell we're not BS ing you. Like, you know, I'm not saying I'm the world expert in latency, but I did lots of research. I had our engineers review this. Right?

Like, we put a lot of effort into trying to teach you, like, how latency works in the cloud and and a little bit of the history of it. And then we're like, hey. Like, we have something that solves this problem. So if you use if you're if any of this struck a chord with you, like, you should look at Metal. Metal.

And of course, like, when we launched, there was a banner at the top of our website, so you saw it when you first showed up and all that. But Yeah.

This is great. This makes total sense like as well. Like, that's in this case, like, that it because it is, I guess, the solution that I need to read the article. But yeah.

Yeah. It's it's a good long read, so only read it if you have a lot of time. But you can also just go through and click buttons and have fun.

True. Which is very fun so far. Okay. That was amazing. Is there anything else you wanna tell us about this article?

About like the creation or like any any fun stuff?

Yeah. I mean, it was you know, this is one of those things that I think Sam, our CEO, kinda had the original idea for this and was like, hey, Ben. It would be really cool. Like, he had seen what I'd done before. And he's like, we're doing this metal thing.

It'd be cool if you did a blog that talked about devices and latency and all this kind of stuff and really drove the point home. But then after that, it was like, we trust you, Ben. Go figure it out and do something cool. Which is which is awesome. Right?

Like, I got a lot of time to just, like, tinker with things and think about, like, what would be a cool way to express this. And this one in particular, I I think this is my favorite one that I've ever done. And, you know, I know it's like, okay, vanity metrics and all this stuff. But like this article, the day you know, it was a couple days after we launched, did really well on social media and all that. Right?

Got lots of likes, but this went to the front page of Hacker News, the orange website, which some people have mixed feelings about. But the fact of the matter is, lots of traffic was driven to our website. Right? Like like many tens of thousands of yeah. Even just within the first couple of days, like page views of people coming to our website.

And, again, we're educating. We're trying to provide, like, real good value here. But, also, they're seeing, like, oh, PlanetScale. They have this new product. They're a fast database.

Right? And, like, if you're it doesn't matter if you're an engineer at a small company, the founder, the CEO, whatever, like, you want that. Right? Like, one of the hardest things like, you're Google, everyone already knows what Google is. If you're Amazon AWS, every engineer already knows what AWS is.

So one of your hardest challenges from the marketing perspective is how do people know you exist and then start to know what is your product and then start to know what problems you solve. And so even though this might seem like, oh, it's just hacker news and whatever, this is helping solve that problem for planet scale. I mean, of people know what planet scale is, but there's some that don't. Right? And so every time we can build trust with the audience by giving them really good education, and then, you know, let them see our website, let them see our incredible products, like, that helps the company grow and succeed.

I felt like this was a good summary of that as well. I was like, can I just say that I love how informative this was that I completely forgot it was to promote a product?

Yes. I remember that comment. Interactivity. That was actually my favorite comment too. Yeah.

I was like, oh, that feel, that's exactly what I was going for, and you expressed it so beautifully. Yeah. Yeah.

This is marketing.

Yeah. For those just listening in audio, yeah. There's a comment right at the top of the hacker news, and it says, it it's so good I forgot that you were, like, marketing a product to me. And I'm like, yes. That that was the goal.

Yeah. Amazing. I think we're coming to the time, Ben. This was incredible. You're doing amazing work.

Amazing work in terms of marketing. Amazing work in terms of just putting amazing stuff out into the world. So it was really fun to hear a little bit behind it. So it's an episode I wanted to do for ages. So thank you so much for your time.

I really appreciate it.

Thanks, Jack. This was a lot of fun. I enjoyed talking with you about all this nerdy stuff, so appreciate it. Amazing.

And, yeah, congratulations on doing so well. PlanetScale, people should go check it out. Metal, Postgres. Yeah. Go to go

to planetscale.com. That's what I'll leave you with. Check out our really fast databases.