Richard Moot joins us to discuss Changelog helping Square launch a developer pod and the excitement around MCP (Model Context Protocol) servers. What might it foretell about the future of human/robot relations?
Featuring
Sponsors
Heroku – The Next Generation of Heroku “Fir” is coming soon — Fir is built on a foundation of cloud native technologies and open source standards, ensuring portability, interoperability, and a vibrant ecosystem for your applications. Cloud Native for everyone, for the next decade and beyond.
Retool – The low-code platform for developers to build internal tools — Some of the best teams out there trust Retool…Brex, Coinbase, Plaid, Doordash, LegalGenius, Amazon, Allbirds, Peloton, and so many more – the developers at these teams trust Retool as the platform to build their internal tools. Try it free at retool.com/changelog
Augment Code – Developer AI that uses deep understanding of your large codebase and how you build software to deliver personalized code suggestions and insights. Augment provides relevant, contextualized code right in your IDE or Slack. It transforms scattered knowledge into code or answers, eliminating time spent searching docs or interrupting teammates.
Notes & Links
Chapters
Chapter Number | Chapter Start Time | Chapter Title | Chapter Duration |
1 | 00:00 | Let's talk! | 00:41 |
2 | 00:41 | Sponsor: Heroku | 02:43 |
3 | 03:25 | OSCON & Friends | 01:35 |
4 | 05:00 | Years in the making | 00:46 |
5 | 05:46 | What held them back | 04:01 |
6 | 09:47 | The Square/Block relationship | 01:48 |
7 | 11:35 | Naming the pod | 07:29 |
8 | 19:04 | Inspirational content | 06:44 |
9 | 25:49 | Content strategies | 02:12 |
10 | 28:00 | Sponsor: Retool | 02:41 |
11 | 30:42 | Learning to host well | 04:01 |
12 | 34:43 | Happy pod parents | 00:44 |
13 | 35:27 | Adam can't say the URL | 01:59 |
14 | 37:26 | The determining factor | 02:34 |
15 | 40:01 | Podcasting advice | 03:51 |
16 | 43:52 | Getting real | 04:35 |
17 | 48:26 | Intro to MCP | 05:00 |
18 | 53:27 | What MCP servers look like | 03:15 |
19 | 56:42 | Sponsor: Augment Code | 03:26 |
20 | 1:00:07 | The open API era | 04:01 |
21 | 1:04:09 | High stakes | 02:54 |
22 | 1:07:02 | An AGI hypothesis | 01:26 |
23 | 1:08:28 | What's the next race? | 01:16 |
24 | 1:09:45 | OpenAI vs the world | 07:46 |
25 | 1:17:31 | You have to be there | 03:06 |
26 | 1:20:37 | Where's the tollbooth | 02:55 |
27 | 1:23:32 | If you take ChatGPT away... | 01:58 |
28 | 1:25:30 | The highway analogy | 01:15 |
29 | 1:26:44 | The horse and buggy analogy | 02:01 |
30 | 1:28:46 | Ejection seats | 04:24 |
31 | 1:33:10 | Small army of personal agents | 00:43 |
32 | 1:33:53 | Galaxy brain meme 👀 | 01:15 |
33 | 1:35:08 | Time to pull the plug | 00:21 |
34 | 1:35:29 | Bye, friends | 01:40 |
35 | 1:37:09 | Coming up next | 01:10 |
Transcript
Play the audio to listen along while you enjoy the transcript. 🎧
Well, friends, we’re here with one of our potentially oldest friends; not quite the oldest, but pretty old, Richard…
I don’t know, I don’t think I’m that old…
Not in age, but in –
How old is he?
[laughs]
Not in age, but in age of relationship. We met – Jerod, do you remember this? We met him back with Shannon at the booth, at the Square booth at OSCON.
Yes.
Yes, OSCON.
Forever ago, basically…?
Forever ago.
I remember Shannon more than Richard, if I’m being honest, but… I’m sure Richard remembers you more than he remembers me.
Maybe. I don’t know.
I do remember that. That was forever ago. I mean, OSCON - it’s not even real anymore.
I know.
People don’t even know what it is. It’s the open source –
We should have a moment…
…convention, or conference.
Yeah, we should have a moment.
All those good O’Reilly conferences used to be so much fun.
Yeah, they closed the whole section down. As soon as COVID hit, OSCON got canceled. And not just OSCON, but O’Reilly events, the entire wing of their business just gone forever.
Yeah.
Do we miss it? I miss it.
I do. I do miss it.
I do. I haven’t been to Portland since, I don’t think.
Yeah…
But All Things Open has stepped up, in my opinion.
And Raleigh, though… But Raleigh – is Raleigh a comparable city? No offense, Raleigh…
Raleigh’s cool.
Raleigh is cool. It’s got the best steakhouse, right?
I’ve never been.
It’s not as weird…
Sullivan’s… All day at Sullivan’s… Anyways, let’s lament later for other things. Okay, so Richard, we produced a podcast together.
We sure did.
And I’ve got to say, this is – again, with the years… This is years in the making. How long, how far back was it when I was “Y’all need to do a developer podcast for Square ?” How long ago was that?
Oh, geez. That’s definitely got to be years at this point.
I’d say four. That’s what I’m thinking, four.
[laughs] He asked the question, but he already knew the answer.
Well, I’m guessing it’s four. I don’t even know.
Yeah, I think you’re probably right. I think I remember we were talking about that… I remember talking about it with Shannon a little bit, and I think we just weren’t ready to pull the trigger on that. But I’m so glad that we finally came around. You were able to convince us.
What do you think held you back? Be specific with what a large team, a large platform Square … How do you think about new content that, that brings in people? And it was not a sales mechanism. I know marketing was involved in it, and that’s always a good thing, because you want their blessing and involvement, obviously. But how difficult was it to think about creating a new content piece that you had to nurture and sell inside? What were the holdbacks from you?
I think it was difficult to come around on it because – so when I joined Square , I kind of dove straight into taking over content for the blog, and revamping how we were going to be writing content, switching from really short blog posts to longer form, in depth… Really trying to build some meat for developers to chew on when they’re coming to learn about the platform. And then when we launched the YouTube channel, doing that zero to one, it was a lot of stuff that we had to learn in the process. And I think it just made us a little bit hesitant. We had a small team to figure out “How do we actually scale up a podcast?” And I guess at the time I probably should have just understood that I could have leaned on you all a little bit more, because that was actually essential to getting this up and running, was trying to just go with experts who can actually tell us what we should do, what we shouldn’t do… Because when we were doing that for the YouTube channel, it was we were just flying blind. We did a lot of things wrong, but eventually got things right.
[00:07:29.07] So I think that, and then the combination of that was trying to understand what is the overall story. I would admit I was a little bit hesitant to think that I could source all of the guests frequently enough to have a constant pipeline of episodes going out… So I think that’s kind of what held me back. And looking back, it’s like, I should have just started way earlier. I mean, I should have been thinking about blog content. The best time to start a blog is five years ago. The second best time is right now. And I think the same can be true of a podcast. Just start making the content. You’re going to learn, you’re going to find the right stories, you’re going to find what works… And we’re already seeing quite a few people inside the company, a few people outside the company who really what we already produced with some of these episodes.
Well, that’s a vote of confidence, for sure. I think you’re spot on with when should you start, because I feel – you know, a lot of brands reach out to us because we have a great megaphone to developers across the globe. And for good reason, right? We come every week, every day bringing great content. We’re in the trenches and we can help brands reach these developers in meaningful ways… But we can only go so far with that. And I think there’s a story that brands have to tell. I think that’s the thing that I recall most, is Square has a developer story you’re not telling. And if I have to repeat that more than once, I’m “Richard, Square has a developer story you are not telling.” And one of the many ways you could do it, obviously, is via a long form, authentic conversation, a.k.a. a podcast. So I was a big proponent of that.
And then if you think about Block, the Block level - y’all have this larger… I don’t even know how to describe Block, really. And I don’t even know how to describe Block compared to Square . I just know you have a big, big mission. Jack Dorsey is a big personality to even have at the helm… So that’s one thing. But then big vision, big dreams, big ideas. And ultimately our idea, or at least your idea/ my idea was “How can we not just do that for Square , but for Block as well?” So…
Yeah. That’s the dream, isn’t it?
What’s the Square /Block relationship? I don’t even know it.
It’s kind of funky, because I think most people think of it as separate companies. And I think they were kind of operated functionally that way… But as far as I’ve understood this, it’s all Block. And each one of these other things are brands that are part of Block. But I joined before we became block, so it’s I still have this thing burned in my brain that I work for Square … Recently that’s changed a little bit, so I work on all Block products, doing developer relations now.
So yeah the relationship is each one of these divisions kind of focused on a different audience in the economy. Square is focused on businesses, and trying to run small businesses, midsize businesses… We even have some enterprises. And then cash is all about consumers; trying to give financial tools to people who are “Hey, I want to go invest in Bitcoin”, or “I want to go buy stocks”, or “I want to send money to my barber.” It really helps essentially act as a consumer banking product. And so that’s kind of where there’s a little bit of – getting both sides of the counter is kind of what we talk about.
And then we have our other brands, Tidal with streaming… And that’s kind of focused on artists, and we’re trying to figure out what ways can we enhance artists… Because they kind of operate their own business. And then we’ve got some of the more out there things, like Spiral doing Bitcoin grants, and open source Bitcoin, trying to really transform Bitcoin into the digital currency of the internet. And so yeah.
So it’s the Square Developer Podcast. Was there a conversation with you all about “Should this be the Block podcast?”
Oh, my gosh…
“Should this be the Square podcast?”
Richard’s, he’s opening up the name… Oh, my gosh…
[laughs]
Oh, man… He’s trying to find out what –
How much can you share about that? I know some of this is BTS…
[00:11:49.18] “How can we get Richard potentially in trouble?”
Oh, no. I don’t want to do that.
But no, no, no. It’s not gonna happen. That won’t happen. No, definitely, that was what I was – when I was originally talking to Adam, that’s really what I want to do. And that’s still in the back of my mind. I still want to be able to do that type of a podcast… So it’s not out of the realm of possibility, because I feel there’s a lot to tell in terms of developer stories. And I don’t want it to just be focused on stuff that we’re doing inside of Block. I want to talk to people about how are people leveraging AI, with AI clients, or building MCP servers, or connecting different tools to one another, or implementing Bitcoin dashboards… I think we just recently launched our own open source framework to show your holdings as a company.
And so yeah, that was what I originally was wanting when talking to Adam about this podcast… But in terms of how we were actually going to be able to get a podcast started, it looked like it was going to make the most sense doing it for Square … Because we ended up finding – we had a lot of stories of sellers, what we like to call seller developers; people who are tech-savvy, they start their own business, but then they start finding our APIs and building their own stuff for themselves, and really extending things.
So I think that’s kind of where we wanted to sort of get the foot in the door, of like “Let’s get a podcast started to show that this has legs.” And then my hope is that we can sort of capitalize on that and do something a little bit bigger, a little bit more expansive.
I thought your question was more around the name, Jerod, because we’ve put some effort into this name, too…
Oh, that was a long debate…
That’s part of it, too.
Much longer than it should have been.
Oh, it’s a part of it? How much can you share about that part of it? I know that was a – I would say a point of contention. I would say that lightly, from my side… Just because I’m so passionate. It’s not because y’all were combative. It’s just because I’m so – I’m like “We could not have had a five-year, four-year sales cycle”, let’s just say, of me suggesting something to you all, you finally decide to do it, we start doing it, and it’s like “Whoa, hang on a second. We can’t call it that.”
I don’t want to sell anybody down the river on this, but that was frustrating. I would definitely say that. And I was definitely in your camp on this, of like I wanted to have a more fun, better name. I remember you were really pushing on Square dev.fm.
Still am.
And I thought – I mean, I still think that’s a great name. I feel like I say that and somebody’s going to go snag that domain and we’re gonna be like “Oh… Well, there goes that.”
We’ll bleep it out.
Yeah, bleep it out.
We’ll bleep it out. They can’t know for sure.
Let them figure it out. Read my lips. But that was definitely a front runner. I can’t really go into all of the details of why that one wouldn’t work… It’s mostly having to do with domains, and trying to acquire domains, and going through some sort of internal process…
I’m actually quite happy with the Square Developer podcast. I know it’s very on the nose, it’s very straightforward… It is what it is, and I think that that works for developers. My original – this is what I don’t know if I would get in trouble for, but whatever… I originally wanted to call it Local Host, because we have a big push on trying to build local tools for sellers to be able to connect with cash users, to be able to give local offers… So that you can sort of see your coffee shop down the street - they have an offer in cash app that if you go over here and use cash, you can get 10% off your latte, or something. And I just love being able to tell these stories of like “This is the local host hangout to talk about sellers, and developers, and what people are building on Square.” That did not fly well with – I mean, once you get all the other folks involved who want to sort of put their little touch on it… It went through a very lengthy review process, trying to see “Is this going to conflict with some other copyright trademark?”, whatever… And so eventually, we came all the way back down to the Square Developer podcast.
[00:16:11.19] Gotcha.
Well, now I feel bad, because I feel like that was me. I pushed against that. Wasn’t that me?
[laughs]
Against the Square Developer Podcast originally?
Against Local Host.
Local Host, yeah.
I mean, I see valid arguments against it. I thought it was fun. But I could also see how it’s not really self-evident like “What is this podcast about?”
I think it’s a cool name for a podcast, but not for a podcast from Square. Like, if Adam and I were starting something about networking, maybe Local Host would be cool. But even for Changelog. One of the things that we had was all these shows and different brands. JS Party, and Go Time, and Practical AI, and The Changelog. And every time that you come up with a new show, you want to be like “It’s still us from the Changelog, but now we have a new name.” And there was people that don’t even know, “JS Party was you guys?”, you know… And one of the reasons for our simplification back down is like “We’re just the Changelog now. Everything’s Changelog.” Yeah, we have Changelog & Friends, we have Changelog News, but it’s all Changelog. And so that’s just more straightforward.
And I think for Square, if it’s like Square has a developer podcast now - yeah, it’s not as creative, but so much easier to find… It just connects the dots for people, and so I think it’s the right choice… Even though probably, Richard, deep down inside of these meetings I would have been on the local host side most likely, because it is just cooler and more interesting.
Yes.
And when you zoom out, too – so we haven’t gotten to this part of the story yet, but there is the Bottom Line, which I believe is an online publication by Square, which is essentially the… I would probably describe this as the media hub. It’s blog content, it’s written content, it’s video content…
That’s a good name, Bottom Line, for a transactions-focused, you know, business…
Yeah. Running the business… Focused on the bottom line…
You care about it, right? “What’s the bottom line?” I like that name. And so that’s kind of a umbrella over the top of this, is that right? Over the top of the Square Developer Podcast, is the bottom line is –
Yeah, yeah. So they actually have –
…a bigger thing, that it’s one of?
Yeah, so the bottom line is kind of focused on everything for Square sellers, or, you know… It’s basically our very large publication that has all different kinds of stuff in it. They have other podcasts on there, that are say like more business-focused, or SMB-focused. I think there’s one that was called Tipped… I cannot remember any of the other ones for the life of me right now. I live over in the developer world.
Paying It Forward?
Yeah. Paying It Forward. They’ve had these other podcasts, and so this kind of gets folded in with that umbrella. I think part of the idea is that rather than having this just on our technical blog, we wanted to have this in a place where more sellers can be familiar with like “Hey, we have a lot of solutions outside of what you might see in an app marketplace.” Or you can hire people to build stuff. Or you can build stuff yourself. So I think that’s kind of the idea, that it would get a lot more exposure over there.
I’m a big fan of content that you discover as an individual, and you get inspired by. So I just imagine someone who is building the beginnings of their business, and they’re thinking “Gosh, taking payments is one thing.” But then it’s like, “Well, how can I dream about the future of this business? I have a background in software, I have a friend who’s in software, and they want to be my co-founder…” However it plays out, I feel like someone in this world could discover the Square Developer Podcast and pick it up as a business owner and figure out how to dream with Square. That’s what I always envisioned. Not just like “Let’s just feed developers more developer content”, but more like “How can we inspire new ways to think about the future of commerce leveraging Square?” And people discover that content and those who are doing it, and that’s what the show kind of capitalizes on. I was dreaming for that kind of thing, not just “Let’s nerd out.” Which is always cool, but at the same time, “Let’s nerd out and also bring somebody into the fold that may not have normally found this content, or saw how this franchise business is pretty scrappy with their tech, but really polished with Square.”
[00:20:16.29] Yeah, I mean that’s spot on. I mean, I couldn’t have imagined what we would end up talking about on the podcast from the beginning to when I actually got to sit down and interview these folks. There’s one that – Rhea Lana is a great one. They were just like “Hey, this started out in a garage. I just wanted to be able to sell stuff,” to expanding into 120 franchises. And it’s just building on the scrappiest way possible, and they were able to build stuff completely on their own.
The other one that really impressed me – I don’t think the episode’s come out yet, but it’s this company called HeadPinz. They’re a family entertainment center in Southwest Florida, and they only started using Square during COVID. They started using our Square online product to basically do curbside pickup, to be able to like “Hey, we need to be able to keep the lights on here. We’re going to do curbside pickup. Nobody can come in here and do bowling, but we have food. And so we’re going to be able to sell food through Square, using their online offering.” From there, they just started using – they found out that we had APIs, SDKs… They were like “Hey, we want to build this kiosk solution, so somebody can come in, buy arcade value, load it up on a card, and pay for it using a Square terminal.” So they started just expanding and trying out all these different new things… And at the end of it now, they’re just diehard Square users. And they say “Square is the source of truth. If we can’t push the data back into Square in some way, we’re just going to use a different product.”
And I never would have found out about these stories until I sit down and I talked to them… Because I think the way that we ended up getting in touch with them is somebody within our company recorded a promo thing that was completely devoid of the developer story here. They just happened to find out “Oh, they’re working with this one API. Let’s go put them in touch with Richard and see if he wants to talk with them.” And I was like “Yeah, of course.”
So getting to service these stories has been really great, because I’m talking to people in our community… They just even are eager to understand how are other people building on this platform. And I feel like this is a much better way to be able to get people to understand how this is working… I can definitely say, doing DevRel, I could sit here and talk about all the different ways that you as a developer can build on this platform… It’s only going to go so far when I actually have somebody who’s like “No, I’ve actually built this business on here. You can go see it, you can go try it.” That is just 10 times more powerful. And the podcast is a way to basically be that vehicle to let people tell that story.
That HeadPinz story is a great example, because we edited all these shows, but I’m also mastering the shows, and I’m listening to them and I’m chaptering them, and I’m in these details… And I recall this part. I didn’t listen to every episode end to end, but this HeadPinz one I was actually really impressed with. And I think it was closer to the end, if I recall correctly, they were saying something like “Once we realized how data flows through Square and how it works and all this stuff, essentially”, I’m paraphrasing, of course, “then we’re just thinking “How can we get all the data in there? If it can’t go through Square , we don’t want it.” It was as if it was this layer that helped them ensure all the data was correct, could be accessible via an API etc. And it became where they wanted all the data to go through Square, and it was its source of truth. I was like “You can’t get that kind of story just randomly talking to people.” How do you expose those kinds of ideas and that truthfulness and that honesty with how they leverage a platform? You can’t pay for that. So that’s not something that you go and do a case study on, or something like that.
[00:24:19.22] You get sort of these great customer stories that are inspirational and spark curiosity for others, and they’re trapped in these weird, I don’t know, content silos where it’s like it’s in this case study, but the case study didn’t have any life. It’s only this printed document, it’s only via a PDF kind of thing, or this one single webpage, which almost nobody ever goes and reads… And you’ve got these great stories out there, and I’m like “Wow, this HeadPinz one was an example of what we dream for this conversation, for this podcast.”
Yeah. And I would even add to that, there’s – it wouldn’t work if I went out there and I told people “You should be using Square as your source of truth”, because I feel anybody hearing that would be like “Yeah, of course you would say that. That kind of puts me into this position of now I’m just entirely reliant on Square.” And there’s probably instances where I wouldn’t recommend that some people use Square as their single source of truth. But I would like for people to know that some people have done this, and it’s been wildly successful for them. They have simplified all of their analytics, all of their understanding of how the business is being run… Their marketing department can now say “Hey, how do we get people to book racing while waiting for their bowling appointment, and doing more arcade? What kind of promos can we mix together?” They now have access to this full customer experience every time somebody sets foot inside of one of those HeadPinz locations.
How do y’all come to these interviews, in terms of what’s your – obviously you’ve found that one because somebody inside of Square had already connected with that customer… But is there an editorial aim that you’re shooting for, and then how do you execute on it, etc? Asking for a friend. [laughs]
Yeah… I wish I could tell you that we have a really good process. It’s very much us getting together with a bunch of folks and saying “Alright, who do we know that we can –” and we just kind of blasted out feelers to a bunch of people who we know were building something, working on something… Being part of the developer community for Square I’ve built relationships with a few people over the years, and so first round is going through those people and being like “Hey, would you to be on the podcast to talk about how you built your kiosk for Shake Shack, or how you built out the whole ordering system for Chase Center?” And then kind of just going from there. Once we had a few people into the book, then we can kind of at least use that as the “Hey, we already have these folks bought in…” But yeah, the sourcing was kind of all over the place, just trying to send out a bunch of messages to people that we knew in the community, or partners… And so yeah, it was kind of all over the place. I was glad that we were able to get as many folks as we did. I feel like once it’s out there, maybe we have a little bit better chance of getting people to want to be on it.
Yeah, totally. Once you have a reputation, it’s much easier. We have a much easier time nowadays getting guests than we used to, when they’d be like “Excuse me, who are you?” Of course, Square already has the name, so that helps… But they’re like “Square has a podcast? Okay…”
I mean, you could imagine – I remember vividly the puzzled look of the first one or two guests that I had on, and had to be clear with them “Yeah, you’re the first one. This is –” And they’re like “Oh, okay… I have no idea how this is going to go.” And I’m like “Me neither. But we’re going to figure it out.”
“Me neither.” [laughs] That’s funny.
Break: [00:27:59.13]
Even the interviewing process is a journey to learn. Can you talk about becoming a host of a podcast, what it takes to before, during, and after?
I think that you kind of set me up pretty well in understanding, I think, from a few times when we had sort of informally recorded some things, had some conversations… It kind of gave me a little bit of understanding. What I’ve found worked really well for me is – so we would set up a pre-production call with anybody that was going to be a guest… I’m going to sort of let it out that that was also the point where I was trying to figure out “Is there a story here?” or do we need to kindly say “Hey, this isn’t probably a good fit. Let’s find somebody else.”
I don’t think there’s actually anybody that we ended up having to say no to, which was great… But there’s that pre-call to just sort of have an exploration of “What are you using? How are you using it? Tell me about the business.” Those were all pretty essential, because I would sort of build out my themes of conversation to fall back on as we were talking… And so it was definitely very much a learning process, because I think my first few recordings I was too self-conscious of trying to keep the conversation flowing… And I think as time went on, I got more comfortable of letting the conversation meander. There’s probably one or two episodes in there, I’m not going to call out which ones, but I’m sure you know from having listened to some of them… But they were way too Square-heavy, and in retrospect I really wish that we talked about more than just that.
I think some people even coming onto the podcast were just assuming “Oh, I should just talk about Square stuff, because I’m a partner of Square, I’m using Square… I should just talk about Square stuff.” And so I eventually started trying to push the conversation out into other directions, just to keep it a little more variety, help them sort of relax more.
But yeah, I’ve definitely found that – what you said originally, it took me a while to actually come around to it, of being like less planned, more organic. I think the one thing that I did take to heart - before I would sit in these pre-production calls and people would ask… They were like a larger company. They’re like “Can you please send us a list of the questions that you’re going to ask on the podcast?” And I was like “No.”
Hard no.
“There’s no list of questions. This is just going to be a conversation. It’s going to sound really weird and really canned if we send you questions ahead of time, because I don’t even know if I’m going to ask all those questions.”
Well, we get that question once a week. Every other week.
I’m sure.
I mean, I’ve got text expander answers for that one… But it’s always a nice way of saying no. [laughs] Some people just want an idea of where we’re going to take it, and so I can provide a couple of like “Well, here’s the things that I think are interesting… But I can’t give you questions, because I haven’t written any yet, nor will I have by the time the show starts. I’m going to go ahead and just think some thoughts”, and then do what I think the best podcasters do, and I try to do, and Adam does as well, is listen. That’s the key as a host, is to listen and react, versus “My next question is” and then read it off of the next card.
[00:34:12.00] Yeah, that sounds [unintelligible 00:34:12.04]
It’s like, no one wants to listen to that.
I did find that one thing that was also super-helpful is clarifying to people ahead of time, like, “We’re not here to make you look bad. We want this to be good. We want this to be enjoyable. We want to look good, we want you to look good.” If you say something that you didn’t like, I would always tell them “Just stop, tell me that you want to re-say that differently, and pick it back up.” You don’t have to be worried, “Oh my God, I’m going to say something that’s going to get me in a lot of trouble.” It’s not that big of a deal.
I’m like a happy parent over here right now, Jerod.
I don’t know, you’re like preaching right back to the choir here. I mean, Adam, it sounds like something that we say.
[unintelligible 00:34:45.06] to the world and graduated and succeeded.
[laughs]
Yeah.
I mean, everything he’s saying is literally things we say…
For sure.
…as a script, but it’s not a script, because it’s become what we share with people, which is just what you said. It’s so cool. I love that. It makes me feel like the literal years of investing into it has paid off… Because I’ve always cared for you, Richard, and our relationship with Square, and working with you and Shannon and others there, [unintelligible 00:35:18.15] and others… Just an awesome team behind Square. Over many years. And to now be at a point where we can go to the bottom line, we can go there and listen to all the shows and all the episodes, we can go to… I really wish there was a domain though for the podcast, because I can’t say it right now…
Even if it just redirected, it would be cool.
Yeah, I know. I’ve told Richard this.
I’ll go start this secret mission internally.
Well, you’ve just gotta be able to say it on the air and have people [unintelligible 00:35:44.28]
Put it into this chapter right here, “Adam tried to describe our podcast and couldn’t send people to it directly, because there is no URL.”
Okay, go to squareup.com/us/en, but only if you’re, of course, in the United States. Then the-bottom-line/podcast/the-square-developer-podcast. Easy.
Too easy, Jerod.
Rolls off the tongue. [laughter] That’s what I said – I said to the entire team, I was like “This will roll off the tongue when we read this out.”
Yeah.
It’s so good. Anyways, the point is is the child has grown up, and I love that. I mean, I think when you have these conversations, it’s so impossible to say “Here’s the questions.” It’s in my opinion so much easier to produce and to listen to when you know it’s a real conversation. It’s formed in sort of topics, of course… Just like this - our topics are “Let’s talk about the show we produced, and let’s talk about AI, and MCP, and APIs, because you thrive in that world…”
Yeah.
That’s our two bullet points for what to talk about during this Changelog & Friends. And so you can’t just tell people “Here’s the questions. Let’s go down them one by one.” That’s not a podcast. A podcast is an authentic conversation, that is just long form. You can’t hide behind the answers, you can’t, I guess, be hidden, I suppose… Not that we’re trying to not expose anybody, but… Even with them - you gave them room to talk about their own tech stack, even… Like, with HeadPinz in particular.
Yeah.
And many of them, where they were just gushing about their own stuff, and it was cool. I love it.
Yeah. In fact, you just now reminded me… When asked earlier what was sort of the determining factor when we figured out that we wanted to do a podcast, it was just – okay, so I’m going to run you through this fluke of events that happened.
Okay.
So you had participated in this one year… You remember our Unboxed event.
Yeah.
[00:37:48.09] We had you come in to interview Jack, and have that whole conversation… It’s one of our better performing things that’s on our YouTube channel, just because everybody loves hearing from Jack.
And me.
[laughs]
And you, of course.
And me.
I mean, you were able to poke and prod him about Rust, and get him to sort of geek out… So yeah, what ended up happening was one of the years we were running Unboxed - I won’t go into too many details, but it’s kind of chaotic getting set up for a large event that, and so a lot of things change last minute. And so originally, we were supposed to be filming something else the day or two before unboxed, with some of our partners, some of the DevRel team… And then tons of things started falling through, and so we had to completely change what we were filming. And so it turned into me just doing interviews with several of our community developer friends. And I think we called these - I’m going to butcher this - like Dev Chat Sessions or something like that. If somebody goes to Square Dev on YouTube, you can find the Dev Chat Sessions. That to me was the first step towards doing a podcast, because when I sat down and did the interview, I actually just had very loose questions, and… Basically, it was in a podcast format, but we just had a full film crew there filming it like it’s some sort of partner interview.
And I think when we saw how well that was received, even within Square, and from the partners who participated, that to me was the “Okay, I’m a hundred percent convinced we can actually just do a podcast.” Because the podcast is actually – I don’t want to say it’s easier, but it’s easier in terms of “Hey, I’m in my home office/studio type thing. They can be anywhere. We’re not having to fly everybody into a single location.” And so yeah, that is actually what totally convinced me that we should be doing the podcast, and like this is what the content should be. So that was kind of my revelation.
Gotcha. So now you’re on the other side, you have a season in the can… I think episodes are coming out weekly…
Yeah.
It’s four years later or whatever and you’ve successfully created and branded and shipped a podcast. So being where you are now, can you speak to anybody who might be back where you were a few years ago, thinking “Maybe we should have a podcast”? What were the hard parts? What was expected/unexpectedly easy, is it all worth it? …for another platform company to go through the building process and the creation process. What are the hard parts? What are the easy parts? etc.
I would say the hard part – I mean, Adam, you became familiar with some of the hard parts… My number one recommendation is just start small. And when I say small, I mean with a small team. Don’t try to get –
Too many cooks?
…too many cooks in the kitchen. Yeah. Just too many cooks in the kitchen, everybody wants to go in a different direction… And I think that really, just start doing it, even if – so even when we first started doing the podcast here, I think one thing that helped a lot… And I’m nearly positive this was Adam’s recommendation - we actually just interviewed, brought in somebody internally to run a kind of test episode or two… And I think it’s just, if you go and you do that, you can just build at least the tiny bit of confidence that you need that this can work. And you can hammer out all those little things that you might be worried about ahead of time.
The other thing that I would just say - it’s actually a lot easier than anybody might be making it out to be. You’ve just got to start, just sit down, have a conversation and record it. It might not be great the first time, but when you just sit down, you have the conversation, you record it, you can have a sense of what went well and what didn’t go well… And it’s weird that I’m saying this now, because we ran into this same exact thing when we launched our YouTube channel…
[00:42:05.16] When we first recorded episodes of the YouTube [channel], we overproduced way too much, and we realized “Okay, we can throw out half of the work here.” We were way overthinking it. And I think even before doing this podcast, we were way overthinking it. So the thing that I just can’t help but continually think is “We should have just listened to Adam and trusted him more about it that - just start it. The format will start to make sense.”
And I think the only part that I think I still got hung up on was what the content is going to look like. And I think in retrospect, you just need to start inviting a few guests, and start recording… Because the content will start to make more sense as you start making it. Because you might think “Oh, this is what I want the content to be about.” You start having the conversations and the interviews with people, and then you realize it keeps going in this other direction. And this other direction is actually not what I thought originally, but it’s working.
I always think of this analogy a lot of the time; I’ve said it to my team… I don’t know if it works well, but I always say “The map is not the territory.” So the map is an abstraction of where it is – so there’s a very big difference between reading on a map of where you’re going, and actually walking the path. And so in actually walking the path, you can see “Hey, there’s a tree falling here.” Or “Hey, there’s not a lot of cover through here.” Maps don’t tell you these things. They can kind of just tell you an idea of “Oh, I can get from here to here.” But it’s very different actually going and walking that path. And I think when you actually go and make the podcast, that it’s very different than when somebody tells you how to make a podcast.
There’s a lot of analogs to software too there. It’s like, you can’t really know until you’re in there what actually needs doing, what’s not going to work. You have to get real - just to go back and borrow some 37signals… Getting Real, which was a good book title. The territory is the reality, right? The map is the abstraction. And so it’s a helpful tool, but until you’re actually in the real world out there where that tree is… You know, the map didn’t have the tree on it. Well, I missed something. You don’t actually know. And so you have to just get going.
It’s not full fidelity. It’s just a map.
That’s right.
Exactly.
And then you actually create the thing, and then you’re like “Oh, well, that wasn’t very good, but I’ve got some ideas for my next try”, and you just iterate, and shine… And don’t go back and listen to our early episodes, please. Although people do that. [laughter] And I’m like “What are you, a glutton for punishment?”, you know…
Yeah. I think too with you, Richard, that we had a – so our audience is aware of this. Our ad styles are uniquely different than I think other ad styles are, where we really go inside of a company and have deep, authentic conversations, and pull out clips from that, and produce them around the story. So that’s how we do it. And you and I have had many conversations across – we don’t just sit down for 20 minutes and hammer out the content… We’ll obviously have various conversations throughout, and sometimes we’ll have an hour-long session and 20 minutes recording. But I had learned enough about you and learned enough about how you operate to have confidence, and what I think your potential was. And so my idea wasn’t just “Oh, Square should have a podcast”, it was like “No, Square needs you to produce the podcast, because you have all these unique insights into DevRel, and what Square is, and this evolution from Square to Block, and… I don’t know what your title is on the inside around APIs; I know that your title is mainly around DevRel… But I think you’ve also been in charge of the API or APIs generally for many, many years now… So you kind of [unintelligible 00:45:51.11] yourself.
So you had this depth and background and ability to talk to people and share stories that I just was like “Man, you are an underutilized asset. They can do some cool stuff for Square. Let’s make it happen.” And so I had complete confidence in you from back, back in the day.
[00:46:13.08] Well, it’s so very warm to hear that, because –
It’s so warm. I’m just like a blanket.
[laughs] I feel like I spend so much time just steeped into these ridiculous conversations… I mean, they’re not ridiculous; they’re important, but they feel ridiculous. The conversations around API design, API… So yeah, that dual title is like I lead our – it’s called API Working Group. I kind of externally just say it’s like an API design team. They kind of set the standards, guide teams on how to build APIs in a cohesive, consistent manner… And yeah, I think that actually spurred from – our dev rel team just kind of got frustrated at times - I guess the nicest way of putting that - where we would have an API, make it all the way out to being put into the developer’s hands, and then the developer’s like “This doesn’t work.” Or “This isn’t well-designed.” And we were just like “This doesn’t make sense. Why are we putting this in their hands?” And so then the design group at that time would consult with us pretty regularly… And it eventually got to a point where we were like “Okay, one of us needs to actually be a part of this group, and helping you make these decisions, rather than just being consulted every once in a while.” And so we ended up – I mean, eventually I ended up leading it as things changed over…
And so we’ve kind of formalized this into a company-wide group in terms of API design… Because [unintelligible 00:47:41.11] actually has their own APIs now, and they decided to adopt the standard. And so yeah, it’s something that I don’t talk about up until recently, more publicly, and I’m hoping to actually talk a lot more about API design, especially after I participated in – we had this sort of internally, where we started building… Everybody just went crazy on building MCP servers. And the light bulb kind of clicked for me when we built an MCP server for some of Square’s stuff, and I was like “Oh, this is can be so powerful, but you kind of have to have that foundational layer in order for it to be capitalized on.”
Well, let’s unpack that. So Model Context Protocol, it’s all the rage. Everyone’s shipping servers… I saw GitHub’s official MCP server is hitting the top of GitHub trending… I saw Shopify MCP, Amazon MCP servers… You’re on LinkedIn talking about MCP… So what’s all the hype and excitement about? It seems like a new style API? I haven’t actually looked into it. I know it’s for LLMs to get context, but why do we need something new? Why can’t we just use the APIs that are already out there?
You know, that is the great question that I feel like – there’s many people asking the same question, like “Why can’t you just hand a bunch –” Like say “Here’s an Open API spec. Hand it to your LLM and say “Go. Build this thing for me”, and look at it.
Exactly.
And part of it is because the LLM clients need some way to know how to call tools. And so this is kind of trying to standardize what the tool calling is. So the way that I sort of describe MCP is it’s like the SDK for robots. And so you have these well-formed ways of saying “Hey, this is how you call the payments endpoint”, or “This is how you call the catalog endpoint.” Because I’ve tried this before, you can sort of just hand an Open API spec to the LLM. Usually, it’s too large or unwieldy for it to really kind of parse out the right things for the particular context… And so MCP kind of formalizes the way that you want the LLM to interact with your platform.
[00:50:08.07] Okay.
And it also is trying to formalize things beyond, say, just REST APIs. So I’ve seen some folks who actually just use it for tapping into snapping visual elements out of a computer, and sort of using that to interact – I saw a demo of somebody actually having it work with an Android app, using some sort of indirect visual reasoning where it’s basically taking a snapshot of the UI, and then figuring out where to “Oh, press this button”, and then gives an instruction to press a different button. So it can go well beyond just traditional REST APIs.
Yeah. It seems like on the face of it, except for that beyond stuff, it’s kind of delivering on what the promise of REST was, at least with the HATEOAS stuff, where it’s like, you should be able to give a single endpoint to a programmer, and the programmer can through tooling discover the entirety of their abilities. And it never really worked out, it always ended up being a spec, or go read the docs. Or make the docs really good, which was a step up from bad docs. But still, it was never – like, you’re still just basically reading URLs and finding nouns and like “Okay, now I’ve got to add this, and nest that, and then I get what I want, and I can post” etc. And that was never what we wanted it to be, but it was better than, I guess, an RPC endpoint with a list of methods you could call.
Yeah.
And this seems like it’s maybe taking that concept and saying “Well, when the LLMs need their context, we’re going to give them everything they need to actually do that without a human reading the docs, or without an LLM even having to read the docs or scrape the site.”
Yeah. I think what you touched on there is like – I’ve been thinking about HATEOAS a lot recently… For people who built their APIs in that way, I feel like they’re definitely advantaged when trying to integrate or adopt AI, because it’s self-documenting.
Right.
It can figure out how things are linked together… I feel like they just have an easier time with integrating this type of stuff. One thing I would actually – we had this argument… Well, not really an argument; slight disagreement in how to view APIs going forward… Somebody within our company said “Oh, we should design our APIs to work better for robots.” And I was just like “Well, here’s where I would disagree on that.” And mainly that all of the AI LLMs have been trained on APIs, SDKs, documentation that’s all designed for people. So it is trained to try to understand it in the human-like way. And so I feel like trying to design your API to work in a more robotic way makes it less human, less usable, less useful overall. And I think that it just means that you should build better, more well-defined, and clearer APIs going forward. Because if we eventually hit this dream of AGI, you don’t need to have a robot-specific API; you can have a human API. So either a human or a robot can use it.
Interesting. So what does an MCP server look like? Describe it to us.
Sure. So at the base level, an MCP server will first sort of define all the available tools that you want to give to it. So in the case of say Square - I’ll give an example - you might say “Oh, there’s a catalog, a customer, a payment tool”, and then it sort of describes how these can be used in different contexts. And so when you ask your LLM, “Hey, can you go build me a menu for a taqueria in my Square account?”, it’s going to go to the MCP server and say “List all the tools”, and then figure out “Okay, of these tools, which one of these kind of matches with something that they’re asking me?” And then it’ll go call the tool, and then the tool will sort of give an additional description of what else can you do with this. And then it kind of figures out from there how to actually accomplish the task that you want.
[00:54:24.14] So the thing that I always think about MCP servers is it’s actually ridiculously simple in what they actually enable, because you’re just sort of creating this tiny little abstraction layer of saying like “Oh, here’s all the things that we’re giving you access to.”
And it’s all just JSON at the end of the day, or…? In terms of actual interchange formats, it’s just typical stuff, there’s no special sauce?
Essentially. So I think there’s two – there probably is three. So this is the part where I’m like – I’ve only sort of worked with server-sent events is one format. So that’s when you can have a remote MCP server. And then you also have command line ones. So you might have a command line tool that – I’ve seen a lot of them are just calling NPX, and then whatever this… Basically, it’s like a CLI app that’s sort of packaged together for it to call in and get all those different tools. And so those are kind of the two main ways… But I think the most common interchange format is probably JSON, that people are sending stuff back to the LLM with.
And how do you plug these into your clients then? Does it depend on the client, maybe?
Yeah, it does depend on the client. So when we were building an MCP server for Square during this Hack Week… Goose is our main open source AI client that we’ve created here at Block, and it has basically sort of the standard config - and I think it was originally YAML, it might be in JSON now - that basically just sort of says “This is the name of your extension, this is the interchange format”, is it going to be server sent events or is it going to be command line? And then you just kind of say “Here’s the command.”
I’ve seen versions of this where people actually are just running a Docker container locally on your system, and that’s actually what it’s communicating with… And so I think Claude has its own standard config file… But for the Goose desktop client we have a way for you to just do a deep link URL. So you just click that button, and it immediately starts pre-configuring that into your desktop client and installs the extension for you.
Break: [00:56:44.02]
I’m thinking back to the early days of Web 2.0 and REST APIs - or let’s just call them web APIs - and the spirit of innovation, discovery, and creativity with open… Not Open API the spec, but open web APIs, specifically around Twitter and Flickr, and FourSquare, and that group of social, cool tech companies. It was really the heyday for mashups, and trying new things…
Yeah.
And it was really fun. And I feel like some of that excitement right now – I’m getting flashbacks to that with MCP servers, because there’s a lot of cool demos right now. You can plug it into darn near anything. You’ve got an MCP server… You know, I saw a guy who just plugged it into Postgres. And of course, his LLM can speak SQL, but it’s just better. It seems like it can just do better with not having to write SQL just to hit the MCP server. And you can just talk to your Postgres database.
And then I’m curious about like – well, what happened the last time was the good old days got old, and then the companies grew up and they needed to make money, and they shut things off, and they locked out third party developers… And it just got very hostile eventually. Sometimes out of necessity, sometimes you don’t know, but it happened… And I’m curious your thoughts on that parallel, because it seems like MCP and letting your LLM talk to X, where X is not the social network, but X is all of the things that have MCP servers, is really cool right now, and could produce some really interesting use cases, and let some awesome hacks out. And then I’m wondering, is there a plan for when that spirit of openness changes? What are your thoughts?
Yeah, what you touched on – I think Angie Jones, my new manager, who leads Block’s DevRel, just gave a talk on these kinds of mashups. Like, it’s very reminiscent of the mashups of the heyday of 2.0.
Yes.
And I think – what was the example they had? It was like Google Maps and Craigslist. And that actually was what spurred Google to create open API platforms for the Maps product. So there’s a lot of very interesting innovations that ended up coming from this… But yeah, the way that I view it now is we’re kind of in the peak of the hype cycle. You’re seeing this massive explosion, and people just producing all these various things… And the recoil I’m sure that we’re going to be seeing is all of the anti-patterns. There were probably anti-patterns before, but we just decided to do them again in this new way.
So I think one thing that you talk about is it’s probably useful for somebody to be able to talk directly with a Postgres database. I’m sure that there’s some people that find a lot of value in that. But it’s also very quickly going to – you’re going to realize “Well, should I enable other people to talk to my Postgres database?” And that’s when I would say “No. There’s a reason that we had these abstraction layers created, because we don’t think that everybody should just be talking directly to each other’s Postgres databases.” That’s why we created those interfaces.
Right.
And so I think that that’s what we’re going to see start happening. I think we saw some early signs of this… People identifying security vulnerabilities with MCP servers… It’s probably something that’s not well-defined within the protocol. But I think that’s been true of a lot of standards as they’ve been developed. I think even in the first version of OAuth there were security vulnerabilities within the format.
For sure.
And it unlocked a lot, but there’s always going to be ways that people are going to find why it’s not going to quite do, or should do everything that we want it to.
[01:04:09.04] Right. And the stakes are higher. Of course, with Auth the stakes are high, but now when it’s like plug an LLM into your life, every facet of your life, whether it’s your bank account, or your FourSquare account, or whatever your Square account - when there is a vulnerability, there’s just so much opportunity for bad actors that definitely we’ll see some trouble.
And it’s a brand new thing. I mean, I think Anthropic came out with it back in November, and of course, they had the task of getting other people besides Anthropic to actually adopt the protocol… Which - it was open and I think designed to be not an Anthropic thing from the start, which is a great way to get it going… But here we are in April and it seems like - will it be the protocol to rule them all in this next age of agentic things? Not sure, but right now it seems like it’s got a lot of momentum behind it, and it quite possibly might be. And so it’s going to be vulnerable until it gets robust. I mean, that’s what happens. You have to face the real world for a while, and find all the obvious things, and work out the kinks.
Yes. And that’s part of the reason why when I was writing about it on LinkedIn – I think that the platforms that are going to be able to leverage this the most, but also last and survive some of the downsides a little bit better are the ones that are already invested in really good API platforms. If you already have that built in, like with OAuth, authentication, permissioning, if you have that basis and then you’re wrapping your MCP server around that, you have a lot of the safety guards already put into place. My concern is for the people who just haven’t done that base level of work, and they’re just like “Hey, we’re just gonna stand up an MCP server. It talks to Postgres. It talks to this other thing inside our internal things.” That’s, I think, going to be a recipe for disaster in the long term. You might see some initial short wins, but I think over time it becomes just a huge exploit.
I think that we’ve debated this inside, with a couple other engineers, in my company… Like, “Hey, it’s really cool that we built an MCP server that can talk to Square’s APIs… But why stop there? Why not have it talk to some of our internal services?” And I had to point out to folks, I’m like “Well, if we create an LLM agent that can talk to Square’s internal services, that effectively makes them public.” Because now I can go – and I’m sure somebody could end up doing some sort of prompt engineering to basically dissect what are all the different tools it’s calling internally, and be like “Great. I’m now going to go use those to write my own program on top of those. I don’t need the LLM to do it for me.”
How about this as an idea - what if AGI is essentially a self-building MPC server, that the LLM talks to discover what it needs; it doesn’t have what it needs, so it makes new stuff, and it calls the same MPC server to discover more of itself. So it’s like, it needs to do something, can’t do it, builds it, can’t do it, builds it, needs to do something, can’t do it… It’s like this constant talk between the LLM and the MPC server.
Well, I’m going to be perfectly honest here, so I’m kind of just shooting a shot in the dark… With Google’s new agent to agent protocol that they’re trying to propose, one of the theories could be that you can have your agent go to this hub of agents and be like “Hey, they’re asking me to do this thing. I don’t really know how to do it, but I’m going to ask all of these other agents in here. Can any of you solve this problem?” And then one of them goes, “Oh yeah, I can.” Maybe that’s the way that you have this agent to agent hub…
[01:08:04.27] It’s like Upwork for agents, you know? It’s like “Hey, I’m looking for a designer.” [laughs]
Yeah.
“I’m looking for this function. Can’t [unintelligible 01:08:10.22] this function.”
It’s like Fiverr for agents…
Exactly.
…so you can throw the bids out there.
“I’ll do it for five tokens.” [laughter]
Right. Race to the bottom…
“I’m trying to hack this Square MCP server… Can somebody help me?” “I’ve got an agent that does that…”
What’s the race? …you know, given MCP and what’s happening, what is the next arms’ race when it comes to that layer of the implementation? Who’s next? What’s the next big thing?
Past MCP, or with it?
Well, I mean, if it’s burgeoning, what is it going to enable? What’s the next big thing? I mean, you kind of touched on it with Web 2.0 and access to APIs that had freedom and innovation…
Well, that’s why I was concerned… The providers may change their heart about the thing. Obviously, Richard, you are one, so that’s why I was throwing it to you… Because the spirit of innovation and openness – and Square obviously would be more useful if I could plug my agent into it and get information ad hoc, and have it do things for me without having to code it up myself etc. However, this leads to the client gets super-powers. And so whoever controls the client - and right now Open AI very much controls the client, because ChatGPT has hundreds of millions of users, and it’s starting to become the Google of this era in terms of just the default choice… And then Square becomes this thing that plugs into ChatGPT and now Open AI holds your interface, and you’re just commoditized… That’s the concern, isn’t it?
I mean, it is a concern, but I think we’re seeing – I mean, I want to believe always that open source will find a way to win. And I think the open-sourcing of models, the trying to incentivize folks to releasing the weights and their measurements… Sort of insisting on the transparency, so that we can have confidence… Because I think that’s a precarious relationship to be managing, if you don’t open up a little bit and build that transparency… Because I think that you can end up triggering a huge defensive move from everyone, where they go “Oh at first signal we realize they’re exploiting our relationship or our data.” Everyone’s going to be “Alright, we’re cutting off, we’re pivoting over and we’re going to either do it in house, or we’re going to go pick a different provider. We’re going to do it open source.”
And so I think it’s a tough thing to manage over time… I mean, one thing that is benefiting us all right now is at least there’s a market of LLMs for you to be able to select from. Some work better in certain contexts… You know, ChatGPT is really great for sort of general purpose, Anthropic is really building stuff that’s like super-awesome for developers… And we’re just hoping that these open models can actually do more and more.
I’ve definitely been dabbling and trying to run an LLM locally on an old gaming laptop… It’s fun, but it’s not practical yet. I’d have to have a way more powerful machine, and I don’t really want to spend that much money on it.
Right. Yeah, I mean, Adam and I have been doing some local stuff as well, and the – what did I call it, Adam? The gravitational pull? The tractor beam?
The tractor beam, yeah.
…that I think ChatGPT has is their product design. And it’s the old open source versus proprietary, capitalized company, who’s laser-focused on just… What they’re doing so well is they’re rolling all the latest advancements into a product. And what all these other things are is their models that you can use with a chat app UI built by one dev to scratch their own itch. Obviously, you can also have capitalized companies that use the open stuff to create a competing product, and that’s happening as well. And I just think that – I agree with you a hundred percent that it’s great that we have a diversity of options, and I think when it comes to models, they’re being commoditized.
[01:12:22.24] I do not think Open AI will hold a lead over the open source world that’s meaningful for very long. Although LLaMA 4 just dropped and it’s disappointing. But DeepSeek is really good. You know, there’s things happening. Right now Meta seems to be not impressing with their latest efforts… But anyways, that I think open source wins, but that’s implementation. Interface is what matters at the end of the day, how people are going ahead and using these things, which is why Apple’s so seriously dropping the ball right now. Because they have the interface to everything, and it’s terrible. It’s called Siri, and they can’t ship intelligence in it.
[laughs]
It is. Think about the opportunity they had with Siri. They were so far ahead. They have the device, they literally have the interface… They have a five-year lead – I mean, Alexa was the next one that got some penetration, but Siri was there, and they got millions of people talking to Siri. And then Siri was just sucking over and over again, for years. And it could have been so much better. And I’m glad kind of that that happened, because they’d have a lock on everything. Everyone would be talking to Siri. At least now it’s a separate company, with ChatGPT being the new Google. I mean, I have people in my life who just – and these are not techies, who are like “Did you ChatGPT it?” It’s the new Google.
I think OpenAI right now, even though in their technology, in their models and stuff it’s not really going to hold a moat, but in their product and their momentum. And so all these people creating MCP servers might just be making ChatGPT, the muscle, the superpower client. And so that’s why I feel like – as well as likening it back to that previous decade, where everyone’s like “Yeah, mash up our API with everybody else’s.” And then people started doing it and they’re like “Wait a second, you’re not coming to our website anymore, because you’re just using our data to create better products or different things, and so now we’re going to have to charge you for that.” Obviously, you don’t represent Square and the entire company, or Block, you represent a small portion of it. But I think when it comes to MCP servers, there has to be some trepidation in allowing the full – not the full, but a lot of functionality that you provide to be used by people that maybe don’t even know what Square is. They’re just using it through their ChatGPT.
Yeah. I mean, I know that for a lot of what we’ve been doing we have safeguards in place. There’s a lot of folks out there who use AWS Bedrock to do hosted models, be more confident that their data isn’t being shared elsewhere, and trying to have some of these protections… But I think, yeah, over time – the one thing I think about what gives OpenAI… I’m trying to, and I’m not super-confident in this, that they actually – a lot of their moat right now is definitely their LLM model. But the interesting thing to me has been recently with these smaller – I mean, essentially smaller updates, that are just enough to cut out features that you find in other essentially ChatGPT wrappers.
Exactly.
[01:15:47.05] When they say in their desktop client “Hey, we’re going to actually apply code directly into your VS Code.” You’re like “Yeah, that’s clearly a cutout for Cursor.” And it’s like oh, the more you just bridge the little gap, and they can just choose to solve little parts of the problem, it’s like “Well, we’re preventing them from eating too much of our lunch. But we’re going to continue to–” I mean, it’s in a place where they don’t even necessarily need to be front-running everything. They can actually let open source and all these other things kind of frontrun things, and then pick and choose the things that go “Oh, people really this piece of that puzzle? Let’s actually just go bake that into ChatGPT.”
Right. Yeah, they can sherlock the industry, as Apple would always do with their third party developers.
Yeah.
Yeah, I think they’re very well-positioned to do that, and I think that they’ve shown competency in that. I say that from my personal experience, because I’ve been trying to leave ChatGPT, and the product keeps pulling me back in. Most recently, the image generation stuff, which is a different thing altogether. It’s multi-model now, and so there’s DALL-E involved… And I don’t even have to care about that. It’s like, it pulls me back in, and while I’m here, I’m just starting to talk to it again. And I’m like “Oh, it’s better than it was last time I was here.” And so it makes it harder and harder for me to use what I had been using, which is LLaMA, and I’ve been trying DeepSeek, and I’m using Google’s Gemini for coding, and trying out all these different things… But at the end of the day, it’s like, ChatGPT just keeps getting better and better and better.
I feel like they’re running that Netflix or HBO kind of playbook…
Yes…!
…where it’s like as soon as the season’s about to be over and you’re like “Alright, I don’t need to be subbed to this anymore.” “Oh, wait, we rolled out the hot new thing.”
They drag you back in. Yeah, they’re dragging us back in.
You can’t go away.
Totally. So I don’t know. Obviously, we don’t know what’s going to happen with all these MCP servers, and I think it’s similar to Google search back in the day, where it’s like you have to be there. Like, you can’t not be there, because then you’re irrelevant. And obviously, OpenAI is not the only people that will be using these MCP servers, so it’ll serve all the needs of everybody. And so I think it’s smart, at least for now, to be like – like, what’s Square’s approach right now? You guys have the hackathon, you tried some stuff… Is there going to be an official thing shipped, or…? Do you know?
Oh, I mean… Well, now you’re asking the things that I can’t quite –
Okay…
Uh-oh… TMI… TMI…
So the thing I can say, because - I mean, I did work on this… I know that from the DevRel side we’re definitely going to be pushing for getting this out there.
Yeah.
Somebody from our Goose team actually did a first version of a Square MCP server. In fact, when we were going to do this hack week we looked at that and we were like “Yeah, this is great, but it’s only covering a very small portion.” And then somebody on our team had a really innovative idea on how to do this, and basically cover the entire service area in a single shot. And so we definitely want to get that out there and get people using it.
The thing that I still keep in mind - and this isn’t reflective of what everyone else thinks within Block - I think it still requires a certain level of care. And I don’t want to be the doomsayer of what ends up happening, but I can’t help but think like when you enable MCP servers for all these different API platforms… So I’m going to ballpark a metric from a long time ago, but say the global number of developers was 13-15 million people. And this is a number I’m pulling from years ago, so it’s probably wrong. When you enable this MCP server to connect into Claude Desktop, ChatGPT Desktop, plugging into anything else, the number of how you viewed developers before has just exploded.
Totally. Yeah, good point.
[01:19:40.12] Like, 100x. And that’s just thinking about how a regular person can now go enable this stuff. But that’s also not accounting for the very savvy developer who goes “Well, why spin up one agent when I could spin up 20 agents to go do work for me?” And so it’s just going to be this exploding problem. And so the thing I keep thinking about is “Can all of us handle the volume of a bunch of bots going around calling all these things?” Are we just going to see a huge influx of DDoS, and…? I mean, it’s hard to understand what the second order effects of this are when you have just tons and tons of agents being spun up everywhere.
Yeah. Well said. Good thoughts. Exciting times, interesting times… Not sure exactly how it’s all going to shake out, but that’s what makes it exciting and interesting, right?
The question I think about is where’s the toll booth? That’s where you make the money at, right? You’ve got a need, and you want to create your own version of a toll booth, and to get access you have to pay X. I suppose if you’re thinking about that from a commercial standpoint or enterprising standpoint, it’s like, what is the next toll booth when it comes to this scenario? You’ve got agents that are going to go and spawn, let’s just say, like you had just described… And that work was normally an individual opening up a book, or going to the library, or talking to so-and-so from the county about how regulations work. However these MPC servers, MCP, MPC… Gosh… Anyways. Anyways… These MCP servers act. Because you’ve got agents that are going to go out there and scour the whatever to figure out information. And it may just be a dead thought, or it may be a new thought that spawns a new thing, that’s now this next layer of whatever… And it’s so hard to even describe it. So if you’re lost, I’m also lost.
The point is you’ve got all this traffic happening on the internet, all this API traffic, all this database traffic that is kind of probably not from a paid user, but it’s like data that’s important to the world. Really hard to see where the toll booth is going to be at for this one, though. I can’t personally see where it’s going to be… Yet. I’m thinking about it though. I’ve got some agents trying to figure it out.
I mean, it’s interesting – I think you make a good point, because the one thing I was thinking about… I used to be very allergic to this; still kind of am. When you – I mean, I’m not opposed to monetizing APIs, but I feel like when you monetize APIs, you have to kind of go about it in the right way, because otherwise developers will never use it. But now with MCP servers, I kind of go “Well, it costs almost nothing for the developer to now go use your APIs”, because they can just have an agent go do it for them and call them for them, and the barrier to adopt adoption is much, much lower. And so it does sort of beg the question of “Okay, if we’re going to have, say, 10X more traffic, do we need to actually put monetization in here in some way?” Or go to the ChatGPT approach, I guess - we pitch our remote MCP server, and it’s five bucks a month, and you get near unlimited calls or something to connect to any of your other AI tools.
Right. Yeah, I think that seems somewhat feasible and likely… But I can’t see the future. I’ve got some agents working on it though.
Yeah. Nice line.
Thank you.
I heard somebody say recently - and tell me if this is how you all feel. They said “If you take ChatGPT away from us right now, we are dead in the water.” I’m paraphrasing, but it’s a version of just doom. “We don’t know how to work anymore. We have learned how to work because of the speed of trial and error and the speed of access to information and the speed of the user experience of asking and getting and this sort of volley back and forth…” I heard someone say recently that they said “If you took this away from us, we wouldn’t know how to work anymore.” Is this how you all feel, in any way, shape or form?
[01:24:10.29] No.
No. I think it would suck. I mean, it would definitely suck, because I think that there’s many ways that it has accelerated us… But I would argue that we’re not quite at a point where we would be like “Oh my gosh, we can’t function.” There might be some companies that might not be able to function because they let too many people go thinking that AI was going to automate everything… But I think there’s still quite a few of us who’d be like “Yeah, I still know how to write code by hand.”
So you’re only doing the code context though, and that’s the challenge here…
I’m not. I’m doing the whole life context. I’m answering whole life. My answer is still no. Because I’d just fall back to where I was a couple of years ago, which was just googling stuff, and looking at YouTube videos. For me so far - and I’ve used it extensively - aside from the joy aspect of the image generation, which for me is pure joy and I love it, and I don’t want to go back to a life before I could turn someone into a walrus if I want to…
[laughs]
…aside from that, I can just go back to googling stuff. It’s mostly just speeding me up. Like, it’s answering my questions, but I’ll find the same answers. Maybe it’ll take me a half an hour versus 30 seconds. So I don’t want to go back, but I don’t feel like I’d be dead in the water, or I’d be a fool, or this “I couldn’t do my job.”
[unintelligible 01:25:24.24]
Yeah. It’s not that groundbreaking.
Well, let me ask you a different question.
Okay.
Same question, but different, because this is actually a good analogy, I think, potentially. Okay, so you’re going from your house, let’s just say, into town; wherever into town is for you.
Sure. Sure, sure, sure.
You’re going from your house, into town. The old way was these back roads, right? Stop sign, and so-and-so’s house, and…
The scenic route.
Right. And you arrived at in-town.
Sure. Then you learned about this new thing called the highway. Right?
[laughs] Okay. Fast. Dangerous.
And you went there so much faster. This is how I liken it. And I don’t disagree with what you’re saying, I just wonder how truthful it is, or how truthful you’re letting yourself be. Because once you’ve gone from your house to in-town the highway way… Sans traffic, okay? Just bear with me here. Don’t put all the worst stuff in there. The speed version of it, the direct access, the get off the wherever and now you’re in town.
Yeah.
When you go from here to there the old way, the scenic route, the 30-minute route versus the 10-minute route via the highway, and then you have to go back to that old way, it does hurt. I don’t care if you can still do it; it still hurts.
Sure. Yeah, but I said that in my answer. I said I would suck, I don’t want to go back. I’d like the fast route more.
But here’s a different kind of analogy where I think it would – I want to sort of tweak what you had there. In the transition from a horse and buggy to a car, there was a transition period where people were just not getting horses anymore, and everyone’s getting cars. But imagine some sort of key failure, like all of a sudden we have no more fuel for cars. So we have all these hunks of junk everywhere, and now we’re in this state of there’s not enough horses for people to be able to get from point A to point B, and now everyone is extremely disconnected. And so you can’t get from where you are – going from San Francisco to LA, now you’ve got to hike, because you have no way of really getting there, and you can only take what you can carry. I think there will be a point in time where we might get to that point where we rely on AI for so many different components that if you suddenly shut it down, that getting from point A to point B might actually be essentially impossible, because it would take too long.
It becomes impossible not because it’s true impossibility, but because it’s like “Gosh, no. I’m never going that way again. I went the 30-minute route to in-town and that was no fun. I’m just never going back there again.”
It’s like, five years from now I want to go check my Square dashboard and they’re like “We don’t have dashboards anymore. We’re just the MCP server”, you know?
[01:28:13.03] Exactly. Yeah.
[laughs]
You are literally describing what I think can happen, because that’s what people are proposing. AI has created – everything is just voice chat, so you’re just going to be using Super-Whisper to say “Hey, go ahead and update my business hours for the upcoming holiday.” And then suddenly, you turn off AI and it’s like “Wait, how do I actually go update my business hours? There’s nothing in the dashboard here for me to go do this.”
Right.
We could end up in that format, where as soon as somebody doesn’t have access to AI, it’s like “I don’t actually have an interface anymore.”
We have to keep the ejection – we have to be able to eject.
Yeah.
Think about automatic cars… There’s a way of doing things manually…? I don’t know, maybe there’s not anymore. Who was I just talking to, where there’s a – oh, if a car’s battery is out, for instance, and it has an electric start, there has to be a way to start the car without the battery. Or at least to shift the gears, sorry. To shift the gears; to get it into neutral. We had somebody stuck in reverse, and their car died…
For real?
Yeah. And it’s like, you’ve gotta be able to – and then a car guy came along and was like “You see this little patch right here? You take this panel off and you stick a screwdriver in, and you turn it, and you can shift the gears manually.”
There’s gotta be a way to do it without the battery, even though a thousand times out of a thousand you’re just going to use the battery to start the car. And so I think we have to do that with these things. We can’t go all agentic, even though they’re going to be the way in 10 years or whatever it is, because the agents will fail. Something’s going to go wrong, and there has to be an injection process. Square has to still have a dashboard, Richard. You can’t just get rid of it.
Yeah.
Let me ask you a question, Jerod. Or Richard, too. You can both answer this question. Okay, let me think of a person… Do either of you have aunts or uncles? Both of you, right?
Yes.
Just say yes.
“Just say yes.” [laughter]
Of course you’ve got somebody in your –
Why don’t you just answer it for me then? Yeah, we do.
…in your genealogy, right? Okay. Tell me your uncle’s number. Your uncle’s phone number.
505 0505.
You do not know. Do you really know your uncle’s number, Jerod?
That’s actually his phone number.
Okay. So a different uncle –
I just doxxed that.
…you don’t have it memorized, because it’s some sort of special number.
Isn’t that the best phone number in history? Can you imagine that his phone number was – and he’s passed away now, so… It was 505 0505. I mean, he didn’t even pick that. It just came to him. Isn’t that amazing?
That’s a cool number.
Anyways, aside from that - yeah, I wouldn’t know it.
The point is you forget to remember phone numbers because your phone has them in the context. You have it in the context, so there’s no point of – you’ve now just obliterated the need to ever remember. Now, my wife’s phone number I know. My brother’s phone number I don’t know by heart. Who else’s phone number do I know by heart? Not that many. I mean, a small few. Sub five.
My parents, my wife… That’s about it.
Right. Sub five. And the reason why is because the phone or the thing or the cloud holds the data, and it’s so accessible. And like you said, a thousand out of a thousand times you’re reaching for the phone to get the number, because it has the information.
But how would you get your brother’s number if you didn’t have the phone?
Yeah, exactly.
Yeah. How are you calling anybody without your phone?
[laughs]
I could not reach my brother. I’d be like “Okay, where’d that phone go?”
“We’re driving to Pennsylvania”, right?
“Where’s the internet, so I can get access to this contact card?” Because I have not personally remembered the phone number in my own brain. I’ve relied on a different brain, my second brain, or whatever brain, or whatever access to data. I mean, we’re going to be that at some point, right? Don’t you think that’s going to be the case, is that there’s just certain things you just don’t have to remember anymore, because for the most part it’s going to be there? Like, there’s my phone.
[01:32:10.03] Cool.
Call brother. No. I stopped that from happening. [laughter] It could totally do it, just right there.
“I’m sorry, I can’t do that. Here’s what I’ve found on the web for Call Brother.”
“Would you me to call ChatGPT for you?”
Yeah. [laughs] Sure, when it comes to memory, I think you’re right. When it comes to functionality, there are certain things that you’re going to have to be able to do. You know, like, you can file your taxes manually with paper still, even though no one’s doing that.
Right. Sure.
So I think there’s permission-critical things in life. Of course, if you lost your phone and you need to get ahold of your brother, you’d probably find somebody between him and you who knows his phone number, and you’d say “Hey, can you give me his phone number?” and then you’d have it. So there are ways, but yes, we are going to probably forget a bunch of stuff, and just ask LLMs. And then if they don’t have the answer for us, we’re going to be like the people in Idiocracy, you know?
It doesn’t exist. Yeah, it doesn’t exist anymore. It’s no longer true. Like, that’s – yeah. if the agent can’t find it… I’m thinking of this thing now, like, gosh, do you think we’re heading to a point where you have an agent, I have an agent, and everyone has a small army, if that’s even the proper word to use?
Why just one? Yeah.
Of agents that we just now command…
A host of agents.
Have your agent do it.
Yeah. Your agent talk to my agent. We’ll do lunch.
Right. And it’s not a human agent, it’s just some – it’s what you refer to as your agent. It might actually have some humans behind it, but generally it’s not a human. It’s some sort of software.
I mean, I look forward to having a personal assistant… What I don’t look forward to is my personal assistant learning exactly how I do my job, and then replacing me.
Right…
Yeah. Like, “Oh, I’m just your agent for now. I’m shadowing you, Richard. I’m learning all your – oh, Richard’s not needed anymore.”
That reminds me of my latest Galaxy Brain meme, that I don’t think I actually shared with anybody. You guys know the galaxy brain one, where it’s like a progression of smarter, smarter, smarter, smarter? So the first one, which is like the small brain, was I programmed the computer, and then the bigger brain is the computer helps me program it. The next bigger brain is I help the computer program itself, and then the galaxy brain is the computer programs me. [laughs]
That’s where we’re heading.
That’s where we’re heading, man.
And that’s the truth, too. Let me tell you how that happens. Algorithms. When you go to the For You page, that’s called programming.
Oh, that’s true. That is programming.
That’s the way the computer’s programming you. It’s larger agents, that are human potentially… Potentially… [laughter] Putting out things, deciding what content makes sense to you in the bubble you’ve put yourself in. And it’s usually the world – man, this is getting doom and gloom, man… The world seems to be opt in, but it’s mostly by force.
You only – you know what I mean? I mean, it’s, for the most part…
And only the privileged get to opt out.
It’s true.
You’ve got to be in a position to be able to opt out. Alright. Well, let’s close on this then… If the algorithms are programming us, I think it might be time to pull the plug. This might be it.
Oh, by the way, we’re producing podcasts that serve this… I’m just kidding. [laughter] Serving this algorithm.
We’re part of the problem around here…
Oh, man… Do we just get de-platformed off YouTube from this?
Maybe. Maybe.
At least demonetized.
So I’m going to throw out a couple of URLs. You can at least go to youtube.com/@quaredev. I don’t know if that’s case-sensitive. Let me check it, because we’ve had some instances where case-sensitive was needed before.
Yeah, I don’t think it should be.
Yeah… Just search for Square Dev on YouTube. You’ll get there.
Yeah, just youtube.com/squaredev works as well. All the pods are there, videos that Richard mentioned are there on YouTube… Even that infamous conversation I had with Jack Dorsey is a scroll or two back, sitting at 18,000 views. Let’s get that out. Come on now.
Yeah.
That’s not enough.
It’s an interesting conversation. It’s still relevant today.
It still is relevant today. There was more to that story, but we’ll let it go for now. Samuel Jackson, call Siri and get us some directions to our favorite restaurant, please. It ain’t going to happen. It’s been fun… It’s been fun, Richard. Thanks for giving us the chance to help produce the Square Developer Podcast. I think it’s a super-cool show. I’m glad that we finally got to a point where we can do it. I’m glad to have the season rolling out as we speak, a lot of good stuff on there. Very proud of you and the team behind the scenes making it happen. Could not be happier. Could not be happier. Thank you so much.
Man, thank you. I mean, I really appreciate all the help that you’ve given us through this… And I’ve got to say, it’s also – it’s fun being a guest. I mean, I want to try this out a little bit more. It’s almost more fun than being a host.
Almost. Almost.
Almost.
Almost.
Well, that’s it. Bye, friends.
Alright. Bye, friends.
Bye!
Our transcripts are open source on GitHub. Improvements are welcome. 💚