Latest / Automated with Brian Heater / How Democratizing Access to Robotics will Benefit All: Intrinsic’s Brian Gerkey on Open-Source Software

Transcript

How Democratizing Access to Robotics will Benefit All: Intrinsic’s Brian Gerkey on Open-Source Software · Automated with Brian Heater

• 0:00
  SPEAKER_00: You can build a system that looks good in a spreadsheet, looks good on paper.
• 0:04
  SPEAKER_00: If it's not usable by the people who are going to be on the shop floor interacting with it, if they don't understand the value that it brings to them, then they're going to reject it.
• 0:13
  SPEAKER_00: And if you can't make the case to the management of a facility that their workers are actually going to get value out of this, then you're never going to make the sale.
• 0:41
  SPEAKER_01: This week we are digging into Alphabet's physical AI play, open source hardware, robotics, and a whole lot more with Brian Gherky of Intrinsic and Open Robotics.
• 0:53
  SPEAKER_01: It's a really great conversation.
• 0:54
  SPEAKER_01: I learned a lot, and I hope you will as well.
• 0:58
  SPEAKER_01: So like, subscribe, and enjoy.
• 1:02
  SPEAKER_01: We talk a lot about what's coming up next in automation on this show, but if you really want to see the future in motion, you've got to be there in person.
• 1:11
  SPEAKER_01: Automate 2026 is where the world's leading innovators, builders, and dreamers come together to show what's possible.
• 1:19
  SPEAKER_01: Robots, AI, machine vision, motion control, you name it, all automation under one roof.
• 1:26
  SPEAKER_01: Register for free at automateshow.com to join us in Chicago June 22nd through the 25th.
• 1:32
  SPEAKER_01: We'll see you there.
• 1:34
  SPEAKER_01: I like to ask you some people who have been um around the industry for a long time, because I I get this feeling myself.
• 1:41
  SPEAKER_01: You know, I've been covering robotics for like seven, eight years now, where there's really this sort of like, oh, oh, it's happening moment that we've been experiencing over the past two years.
• 1:51
  SPEAKER_01: So does it feel like all of the things that you've been kind of talking about and working to are sort of starting to, in a certain sense, come to fruition?
• 1:59
  SPEAKER_00: I think I'm seeing the same thing that you are.
• 2:01
  SPEAKER_00: And I I've been waiting for this for 25 years since I've been working in robotics, waiting for it to happen.
• 2:09
  SPEAKER_00: And there is a certain sometimes you feel a little bit like uh Charlie Brown going after the football.
• 2:14
  SPEAKER_00: Like, is it is this really the time?
• 2:17
  SPEAKER_00: I in retrospect, I can say that I think we had some big moments along the way.
• 2:21
  SPEAKER_00: Like, you know, circa 15 years ago, we figured out autonomous mobile robots like AMRs and warehouses factories.
• 2:29
  SPEAKER_00: That that's that's working pretty well right now.
• 2:32
  SPEAKER_00: Um I think the the next big step is how do we combine mobility and manipulation and figure out how how to handle the much greater variety of things that people want robots to do.
• 2:46
  SPEAKER_00: And is that about to happen?
• 2:48
  SPEAKER_00: I I I think we're getting there's there's good science.
• 2:51
  SPEAKER_00: Yeah, I'd say that a lot of things have come together from the hardware's getting better, more reliable, lower costs.
• 2:58
  SPEAKER_00: There's uh there's bet greater willingness on the part of especially some pretty old industries to like adopt new technology and kind of change the the way that they think about solving some of the problems they've had for a long time.
• 3:11
  SPEAKER_00: And then as a robot software guy myself, I'll say the software's gotten much better.
• 3:16
  SPEAKER_00: There are just we've got more greater ability to soak up a lot of the the variability and the challenges that you have in these tasks in the software in a way that you just you couldn't do before.
• 3:28
  SPEAKER_01: Yeah, I I I guess something to important to point out here is that this sort of thing, like it doesn't just happen at once.
• 3:34
  SPEAKER_01: It isn't like, oh, you know, that this is like the moment for robotics.
• 3:38
  SPEAKER_01: Obviously, it's going to come in waves, but this this does to me feel pretty um pretty different than all those those those moments in the past.
• 3:47
  SPEAKER_01: I and I'm curious at that point, at the point when you were really sort of starting your career early on, how would you describe the state of AI and robotics?
• 3:58
  SPEAKER_00: Oh man, I so I got into robotics in the mid-90s as an undergrad.
• 4:04
  SPEAKER_00: There was a guy named Jim Jennings, was a new professor to the CS department.
• 4:08
  SPEAKER_00: Uh, I was at Tulane University in New Orleans, and he showed up and started a robot lab.
• 4:14
  SPEAKER_00: And I, you know, as an eager undergrad, burgeon, you know, uh budding computer scientist, I saw these robots and I volunteered to work in his lab.
• 4:25
  SPEAKER_00: And I just got such satisfaction out of writing code that made something move physically in the world.
• 4:31
  SPEAKER_00: That was just fundamentally different from all the other things.
• 4:33
  SPEAKER_00: Like all the other things you do with software, generally you get the right, the right letter, number, or color on a screen.
• 4:39
  SPEAKER_00: And with robots, you get to move things physically in the world.
• 4:42
  SPEAKER_00: So for me, it was just fascinating.
• 4:45
  SPEAKER_00: At the same time, the robots were capable of just about nothing, right?
• 4:49
  SPEAKER_00: They were, they were lab tools.
• 4:50
  SPEAKER_00: They were experimental tools that we had to help us try to answer scientific questions about uh multi-robot collaboration or motion planning, things like that.
• 5:00
  SPEAKER_00: And we wrote these papers saying, you know, oh, one day there will be all these robots in these factories and these warehouses doing all this useful work.
• 5:07
  SPEAKER_00: But I don't think we really expected it to come, you know, anytime soon.
• 5:11
  SPEAKER_00: It was really an academic endeavor.
• 5:13
  SPEAKER_00: Like that was famously a time of, it was one of the AI winters that we talk about, where there had been a huge level of interest in combination of expert systems, early versions of neural networks, and they had all kind of failed to meet expectations.
• 5:32
  SPEAKER_00: And so the 90s was a time of kind of reconsidering like what what kind of intelligence can we really expect from artificial systems?
• 5:41
  SPEAKER_00: It had a very different feel from how it does now.
• 5:43
  SPEAKER_00: At the time, it was, I really thought about it as an intellectual academic exercise.
• 5:48
  SPEAKER_00: Like we did it because we thought it was interesting, not because we were convinced it was going to be useful.
• 5:52
  SPEAKER_01: Aaron Powell Actually, walk me through the the the this concept of of AI winter and whether there are any parallels or I guess potential parallels to where we are right now.
• 6:03
  SPEAKER_01: Is it to a certain extent, is it kind of overpromising and under underdelivering?
• 6:08
  SPEAKER_01: Obviously, there's always gonna be an element of that in tech.
• 6:10
  SPEAKER_01: And obviously, like, listen, you know, no one's gonna say that to a certain extent, like they're hyping the hell out of AI right now.
• 6:18
  SPEAKER_01: Is it that AI didn't deliver on the things that it was promised, and therefore people kind of just moved away from it for an extended period of time?
• 6:26
  SPEAKER_00: I think there are a few factors at play.
• 6:27
  SPEAKER_00: You use the word hype, which I think is just right.
• 6:28
  SPEAKER_00: And I think here we I mean we don't have to reinvent anything.
• 6:31
  SPEAKER_00: It it's the Gartner, I think it's the Gartner hype curve is often used to describe how these things unfold.
• 6:36
  SPEAKER_00: And it and it's not just AI, it's it's any new technology.
• 6:39
  SPEAKER_00: There's a there's a huge amount of excitement and expectations that are frankly unrealistic, and then there's a trough of disillusionment, and then eventually you figure out how to make it useful.
• 6:50
  SPEAKER_00: So I think that that's applying to AI and to some aspects of robotics today, just like it does to any new development.
• 6:58
  SPEAKER_00: I think an additional challenge that we have with AI is that you know, AI is a term that's been around since the 50s.
• 7:06
  SPEAKER_00: Today we use AI to describe a very, a relatively narrow part of what, you know, over a broader time horizon you would consider to be AI.
• 7:15
  SPEAKER_00: Like I study.
• 7:16
  SPEAKER_01: It's funny you say narrow because I think of AI as being incredibly broad, just the way that like it seems to describe everything at this point.
• 7:22
  SPEAKER_00: Yeah, but today if you if you say AI today, I think there's a good chance you're referring to a particular approach to building relatively large neural networks to solve problems in a particular way.
• 7:35
  SPEAKER_00: And that's that's what I mean by narrow.
• 7:37
  SPEAKER_00: Like that's that actually excludes a whole bunch of things that I think can genuinely be considered AI, which are related to motion planning and search.
• 7:45
  SPEAKER_00: And there's all these other things that have happened along the way.
• 7:49
  SPEAKER_00: And that's that's kind of the curse of AI, is that once you solve a problem, uh, it's no longer considered AI.
• 7:56
  SPEAKER_00: Like once it, once it gets into use in a serious way, I mean, you you look at like a lot of things that you use today, whether that's whether that's, I don't know, Google search or it's a Netflix recommendation system, those are all AI-based now.
• 8:12
  SPEAKER_00: Um we just don't call it AI anymore.
• 8:14
  SPEAKER_00: And so the label, it's like the horizon keeps moving.
• 8:17
  SPEAKER_00: So the bar keeps getting higher, and we kind of forget that the things that that they start to work, they become reliable.
• 8:24
  SPEAKER_00: We no longer think of them as having that magic aura of AI.
• 8:27
  SPEAKER_00: And so that that gives you the additional appearance that like you feel like it didn't work when because you stopped paying attention to it and it took a while to get into production, but then it ends up actually being things that we rely on.
• 8:38
  SPEAKER_01: Yeah, I wanted to pull on a thread something you mentioned earlier when we were talking about hype cycles, uh unrealistic expectations.
• 8:45
  SPEAKER_01: And I know obviously everybody has their own idea of what we're dealing with right now.
• 8:49
  SPEAKER_01: Is there anything at the moment that we're looking at that just really feels a lot further off than we're being led to believe?
• 8:58
  SPEAKER_00: As a roboticist, I look for ways to make things useful uh as close as possible to today.
• 9:06
  SPEAKER_00: And so if I look at what are the problems that we can we can solve today, I look around and I see we've got, like at Intrinsic, we're working mostly with robot arms.
• 9:16
  SPEAKER_00: They're newer and better than the ones that came before, but frankly, they look kind of like the robot arms that have been around for decades.
• 9:21
  SPEAKER_00: The thing is, they can do amazing things if we add better sensing and better software to them.
• 9:27
  SPEAKER_00: And so I think that there's there's a huge amount of extra productivity, new problems we can solve with those systems.
• 9:34
  SPEAKER_00: Uh combining those with mobile systems in some way opens up new opportunities as well.
• 9:40
  SPEAKER_00: An area where I would, I personally think that it's going to take a little bit longer than some people are currently forecasting to get into use would be the much more mechanically complex systems.
• 9:52
  SPEAKER_00: And here I'm thinking about legged systems.
• 9:55
  SPEAKER_00: So legs are great for solving certain types of problems.
• 9:59
  SPEAKER_00: There are lots of cases where it's not clear that you need them.
• 10:02
  SPEAKER_00: And so the trade-off of making figuring out how to make those work really reliably, uh, you've got to, I think you need to be realistic about how long it's gonna take to do that and what's the what's the the the benefit, the payoff that you're gonna get.
• 10:15
  SPEAKER_01: That's so interesting because like like 201, just about every single person I ask this question to says uh manipulation, says dexterity, that that's the real difficult thing to crack.
• 10:26
  SPEAKER_00: I think that's also a hard thing to crack, but I think that it's where if you ask me where where we're overpromising, I think that I think as a as a as an industry, I see greater overpromising on legged locomotion.
• 10:43
  SPEAKER_00: Then I I I at least I have the sense.
• 10:45
  SPEAKER_00: I mean, maybe you could tell me as you know, a as an observer of all this, like I I don't feel like as a as a community we're overpromising on the manipulation.
• 10:53
  SPEAKER_01: So you you mean like in terms of like how actually sort of useful it might be day to day versus like how far along it is, perhaps?
• 11:01
  SPEAKER_00: Yeah, I think that we're I mean, frankly, I see I see that like legged systems are getting very reliable and but it's not clear to me where the where that high value use case is going to end up coming from.
• 11:16
  SPEAKER_01: Aaron Powell I I really did cut sort of want to pull on the the the complexity problem, you know, in terms of this is kind of been this is something I've been really focused on lately.
• 11:26
  SPEAKER_01: Um, we see a lot of really uh highly produced videos coming through, um robots that look like they're incredibly capable in the field, in warehouses and in homes.
• 11:38
  SPEAKER_01: Mobile manipulation is is obviously sort of a a a really big part of that.
• 11:45
  SPEAKER_01: I I as as somebody who works with a company that uh to um that deals so closely with like picking and and placing, granted, like largely stationary, but why why is that such a difficult problem to solve?
• 12:02
  SPEAKER_00: Picking and placing things, if you want to do the same thing again and again with really, really high precision and you know, we want to do it really efficiently, that actually is solved.
• 12:14
  SPEAKER_00: Anything that you're willing to put enough money into to build up a bespoke assembly line for you can automate.
• 12:21
  SPEAKER_00: You can automate just about anything.
• 12:23
  SPEAKER_00: Uh the challenge is are there's a huge amount of production that happens.
• 12:29
  SPEAKER_00: It has higher mix or it has different volume characteristics.
• 12:32
  SPEAKER_00: And so it's not, it's not economically viable to automate it.
• 12:35
  SPEAKER_00: And it's that it's handling that variability.
• 12:37
  SPEAKER_00: That that's really the challenge.
• 12:39
  SPEAKER_00: If if you almost any production system you can imagine, if you're gonna have the same parts coming through, we can build a system, we can build a what I would consider like a classical automation system to handle the manipulation, the picking, the placing, the part handling.
• 12:54
  SPEAKER_00: If you want to do what a lot of manufacturers want to do today, which is I want to run uh I want to run 10 of that part, and I want to run 100 of that part, and I want to run uh 50 of that part, and I want to do all that in a single shift, and I want to have it be automated, that's really hard today.
• 13:13
  SPEAKER_00: And it's being able to react to that variability that is the challenge.
• 13:17
  SPEAKER_00: And that's where you know our approach is to soak that up in the software.
• 13:22
  SPEAKER_00: That's where we can we can build and we have built perception systems that are able to, with, you know, basically just give me an image and give me a prompt in the form of a CAD model, and I can find where that object is in the image, and more to the point, I can have the robot go and pick it up, and it's a part that we never saw before, we didn't train on in advance.
• 13:42
  SPEAKER_00: That's something the the result of that is that you don't have to fixture all this.
• 13:46
  SPEAKER_00: You don't have to, and you don't have to bring in the system integrator who installed your robot and have them reprogram it for you when you're switching from what handling one type of part to handling another type of part.
• 13:56
  SPEAKER_00: And it's those sort of those sorts of barriers are what has made it historically difficult.
• 14:01
  SPEAKER_00: Just the the cost of changeover is so high that most manufacturers think about these systems as being basically single purpose.
• 14:09
  SPEAKER_00: And you can see this in the business model.
• 14:11
  SPEAKER_00: You look at like what does it take for a factory to agree to buy a robot cell?
• 14:15
  SPEAKER_00: They usually want a return on investment that's on like a one to two year time frame.
• 14:20
  SPEAKER_00: That's not because the robot is gonna break down over one to two years.
• 14:24
  SPEAKER_00: The useful lifetime of these systems is easily five, seven more years.
• 14:29
  SPEAKER_00: The calculation they're making is well, I'm not buying a multi-purpose device.
• 14:34
  SPEAKER_00: What I'm buying is a like a fixed appliance that only is programmed to build exactly the one part I'm building right now, the one product.
• 14:43
  SPEAKER_00: And when I change my product line next year, I basically have to buy and program an entire new system.
• 14:49
  SPEAKER_00: And so the value of that system I bought before is, you know, basically amortized down to zero, depreciated down to zero.
• 14:56
  SPEAKER_00: With better software, you can start to think about these systems in a totally different way.
• 15:00
  SPEAKER_01: Yeah, and even like seasonally, I'm I'm guessing that there's a big change to a lot of these systems, that they want to be able to shift them to a certain extent on the fly.
• 15:08
  SPEAKER_00: Oh, absolutely.
• 15:09
  SPEAKER_00: Look at how cloud computing has evolved, right?
• 15:12
  SPEAKER_00: There was there was a point when you would all of you would you would use one computer, one machine to do a particular job, and you would you would put your software on that computer.
• 15:23
  SPEAKER_00: Originally that was you know on on premises, and then you might put it in a data center.
• 15:28
  SPEAKER_00: Now you don't think about individual machines.
• 15:30
  SPEAKER_00: You think about you have a compute workload, you have a set of kind of tasks you need done.
• 15:35
  SPEAKER_00: You kind of throw that at a data center, and it has in-house a few different types of machines.
• 15:43
  SPEAKER_00: You know, some of them have a lot of GPUs, some have a lot of CPUs, some have a lot of storage, some it might have TPUs or other sorts of uh processing uh technologies.
• 15:52
  SPEAKER_00: And there's there's a way to allocate your workload across those machines.
• 15:56
  SPEAKER_00: Those machines are are multi-purpose.
• 15:58
  SPEAKER_00: They can take on one job in the morning, one job in the afternoon, a different job next week.
• 16:03
  SPEAKER_00: What if we could do that for manufacturing?
• 16:05
  SPEAKER_00: What if instead of having the these like fixed assembly lines that just do one thing again and again, imagine uh a factory which is a it it's it's this it's full of these repurposable work uh resources.
• 16:20
  SPEAKER_00: And so I can toss my manufacturing, my production job at the factory, and it's deciding uh there's an allocation or an orchestration system that's deciding, well, okay, that task can be done by that work cell over there, which was doing something different yesterday.
• 16:33
  SPEAKER_00: It's going to do this today, and we'll we'll it'll depend on where the inventory is available.
• 16:37
  SPEAKER_00: And we can start to treat these as like a matrix where we can adjust the the flow of product through this matrix in a way that looks very, very different from the traditional assembly line that we've had since the the time of Henry Ford.
• 16:50
  SPEAKER_01: Aaron Powell It strikes me that maybe kind of one of the key thesis statements for intrinsic is that a lot a lot of these problems are software problems where we thought maybe they were primarily just hardware problems.
• 17:04
  SPEAKER_00: There are still hardware problems that we need really smart people working on.
• 17:08
  SPEAKER_00: Um we can talk about we can talk about grippers and hands in particular.
• 17:12
  SPEAKER_00: Um, but yeah, I think that there is I think the the robot arm hardware, the sensing is really good.
• 17:21
  SPEAKER_00: Uh our ability to use that sensor data is getting better all the time.
• 17:25
  SPEAKER_00: I mean, we used to rely more on depth cameras.
• 17:29
  SPEAKER_00: Uh modern, modern AI-driven perception is relying increasingly on just you know single, single color images without necessarily explicitly trying to calculate depth data.
• 17:40
  SPEAKER_00: So the hardware, I'd say the hardware is willing, and historically the software has been weak.
• 17:46
  SPEAKER_00: And now we can we can really make some step changes there.
• 17:50
  SPEAKER_00: The area where we still do need significant advances, just and here I'm just talking about you know robot manipulators, is the so-called end effectors, the hands, the grippers.
• 17:59
  SPEAKER_00: We're still by and large using relatively simple, like parallel jaw pinch grippers or suction grippers for just about everything that is done in a day in, day out, reliable way.
• 18:11
  SPEAKER_00: Because that's the technology that's been really well proven.
• 18:13
  SPEAKER_00: It's totally hardened.
• 18:15
  SPEAKER_00: And we often adapt processes to live with the fact that we have these simple grippers.
• 18:22
  SPEAKER_00: There's interesting three-finger grippers, there's interesting, you know, more dexterous systems that are modeled after human hands.
• 18:29
  SPEAKER_00: They are not nearly as widely used yet.
• 18:32
  SPEAKER_00: We know that we need those capabilities and we need people working on the hardware there.
• 18:37
  SPEAKER_00: And then, you know, as soon as it's working reliably enough that we could put it into production, we're happy to take it on.
• 18:42
  SPEAKER_01: Yeah, it's striking me talking to you that um maybe like a big part of the reason why manufacturing and logistics ended up being such a good fit for robotics and automation is because all of these problems are math problems.
• 19:01
  SPEAKER_01: One of the reasons why I've really enjoyed writing about robotics is every time you write about a different company, like you learn about a different vertical, right?
• 19:07
  SPEAKER_01: Like suddenly, you know, you're learning about farming or you're learning about the ocean.
• 19:12
  SPEAKER_01: And obviously, working at a place like intrinsic, you know, I'm sure that you know a lot about manufacturing that you didn't before, but at the end of the day, abstractly everything boils down to numbers.
• 19:24
  SPEAKER_01: And that's, I assume, something that your entire team knows really well.
• 19:29
  SPEAKER_00: I can tell you that I and my team know a lot more about CNC metal cutting and sheet metal cutting and bending uh than I ever thought we would know.
• 19:41
  SPEAKER_00: Because you do have to go deep into these, into the verticals in order to build something that's useful.
• 19:47
  SPEAKER_00: I see what you're saying about it being about the numbers.
• 19:49
  SPEAKER_00: I will tell you, it is also, I mean, this sounds cliche, it's also about the people.
• 19:54
  SPEAKER_00: You can build a system that looks good in a spreadsheet, looks good on paper.
• 19:57
  SPEAKER_00: If it's not usable by the people who are gonna be on the shop floor interacting with it, if they don't understand the value that it brings to them, then they're gonna reject it.
• 20:07
  SPEAKER_00: And if you can't make the case to the management of a facility that their workers are actually gonna get value out of this, then you're never gonna make the sale.
• 20:15
  SPEAKER_00: And so there's, especially in these situations where we're like we're in the process of asking, in many cases, these established industries and some of the customers are relatively small.
• 20:27
  SPEAKER_00: You know, uh a lot of the a lot of the places where automation has the opportunity to be taken up the quickest is in small sort of mom and pop machine shops.
• 20:38
  SPEAKER_00: You really need to meet them where they are and offer them something that they're gonna, they see the value in it and they need to have confidence that they're gonna be able to understand how to use it.
• 20:48
  SPEAKER_00: You know, are they don't want to be on the hook for making being it having to call you to come back and deal with every potential problem.
• 20:55
  SPEAKER_00: So there's a there's a relationship and a business uh aspect to it, which I I've learned, especially over the last couple of years, you just can't ignore.
• 21:04
  SPEAKER_00: Yeah, otherwise you you can end up building what you think is the perfect technology and just nobody ends up using it.
• 21:10
  SPEAKER_01: Yeah, it it it must also be an interesting problem to work with, as you said, like Monpa versus a large corporation.
• 21:18
  SPEAKER_01: And I suspect that one of the things that you know they would love to have is one of these kind of um, you know, multi or I hate to use this, but general purpose robot, you know, something that can be like adapted to various things so that they don't have to buy all of these, all of these different systems and as seasons change and products change that it can be adapted as well.
• 21:41
  SPEAKER_01: Um one of the things, obviously, that we're all talking about right now is is physical embodied AI.
• 21:47
  SPEAKER_01: And a big point of contention, I guess, that I've been hearing from a lot of folks in the field is this conversation around creating uh you know, one large generalized model.
• 22:00
  SPEAKER_01: That kind of works across embodied or small specialized models.
• 22:05
  SPEAKER_01: And I got the sense when I talked to Torsten on your team that maybe intrinsics philosophy approach to that up to this point has been largely more specialized.
• 22:16
  SPEAKER_01: Is that fair to say?
• 22:18
  SPEAKER_00: Yeah, Torsten Kroger's our chief science officer, and you know, we look to him to really set the tone for how we how we think about pulling in these new technologies.
• 22:29
  SPEAKER_00: And the the term we're using uh now is uh AAFI or artificial functional intelligence.
• 22:36
  SPEAKER_00: And it is, I'd say it what it reflects is our approach to wanting to provide these functional components that are usable today.
• 22:48
  SPEAKER_00: So if we think about how we've historically uh tackled programming in these robot systems, usually it's it's a standard systems engineering approach, right?
• 22:55
  SPEAKER_00: You take a you take a big problem, you decompose it into pieces, and then the pieces connect to each other in some way.
• 23:01
  SPEAKER_00: So you have a piece that you know takes in the image from the camera, and you have another piece that processes the camera to get depth data, and then another one that processes the depth data to find the object, and another one that decides where you're going to grasp and motion planning and so on.
• 23:15
  SPEAKER_00: And you have this kind of distributed systems approach.
• 23:17
  SPEAKER_00: What I think of us as doing is saying, well, look, there are this, there are these new ways that we can solve the, we can basically provide those functions.
• 23:25
  SPEAKER_00: We can solve those problems in different ways.
• 23:27
  SPEAKER_00: And so often what we're doing is taking a modern machine learning AI-based approach, and we're taking out one of those existing blocks and we're popping in uh an AI-powered version of it, which is better, more reliable, more robust, uh lower cost, easier to use.
• 23:42
  SPEAKER_00: Now, over time, what we're seeing is that there are these adjacencies.
• 23:46
  SPEAKER_00: Like, you know, you talk about uh grasping, do I need to explicitly compute the where an object is in the scene and then pass it to a grasp planner?
• 23:57
  SPEAKER_00: Or can I ask a slightly bigger AI system to just go ahead and tell me where to grasp?
• 24:04
  SPEAKER_00: Right?
• 24:04
  SPEAKER_00: There that you can start to skip these steps, and and we're exploring that as well.
• 24:08
  SPEAKER_00: Now, if in the fullness of time these this coalescence keeps going on, that the blobs keep getting bigger to the point where you end up with more or less one blob that is doing solving the whole problem.
• 24:23
  SPEAKER_00: If we get there, sure, that might be the right approach.
• 24:26
  SPEAKER_00: I would, to me, that's not an end in itself.
• 24:29
  SPEAKER_00: I I think that you know, there's a there's a you you can try to draw parallels here with like, you know, the search for AGI, which is, but that's more of like a that's like a philosophical goal to me.
• 24:41
  SPEAKER_00: Uh that's not that may or may not produce useful technology.
• 24:45
  SPEAKER_00: And I'd say we're really focused on producing useful technology, if and when there's this one model that can control the robot, especially if it can control all the robots, then I would like to think that at intrinsic, we've we've laid all the groundwork so that we're ready to host that that AI model and connect it out to the physical world.
• 25:07
  SPEAKER_01: Is that approach this somewhat in contrast to, you know, um obviously intrinsic is part of is a part of Alphabet, Google is a part of Alphabet, um, you know, DeepMind is a part of Google.
• 25:20
  SPEAKER_01: Um are are they sort of starting with a blob first approach with Gemini, would you say?
• 25:26
  SPEAKER_00: It's a matter of time horizon.
• 25:28
  SPEAKER_00: So if you look at, you know, Alphabet has a has a portfolio of uh efforts that it invests in.
• 25:36
  SPEAKER_00: And I think about intrinsic as being the much more commercially oriented, product focused, focused, getting these systems out to people today, kind of effort.
• 25:49
  SPEAKER_00: And that that means that we're gonna make those, uh, we're gonna make decisions like we, if if there's a smaller functional block that we can swap in today, we're gonna do that.
• 26:01
  SPEAKER_00: When our partners who I think about as being kind of upstream in the, you know, they're I think they're further back into kind of the the in the pipeline between research and and development and technology deployment.
• 26:13
  SPEAKER_00: They're, you know, DeepMind is is uh a world-class research organization that's doing a bunch of work that over time we're gonna figure out how to take on board and then put out into the world.
• 26:25
  SPEAKER_00: I mean, our in in fact, the the uh our one-shot vision model is the result of a collaboration between intrinsic and deepmind.
• 26:33
  SPEAKER_00: We took some of the work that they did, built on that, extended it, trained it in ways that are relevant for the kinds of problems that we're trying to solve.
• 26:40
  SPEAKER_00: So I I think that's gonna that's gonna continue.
• 26:43
  SPEAKER_01: Yeah, it's interesting.
• 26:44
  SPEAKER_01: And I I know obviously this predates your your time with uh with intrinsic and Google, but um I I I get the sense that maybe this is uh sort of a big part of what's happened to a lot of these what had been uh you know moonshots or or um alphabet X companies is um this push to get from sort of like the the the abstract moonshot to hey, what can we do right now to actually make this like a company that is actually making that is making a product?
• 27:15
  SPEAKER_01: What can we do to actually like graduate this from the lab?
• 27:18
  SPEAKER_01: And and perhaps that's what a big part of what started to sort of separate intrinsic, the idea of of intrinsic from what would have been like Google Brain or Deep Mind or everyday robots at the time.
• 27:32
  SPEAKER_00: Like you said, I wasn't here at that time.
• 27:35
  SPEAKER_00: Um and but the way it is worth um recognizing that we've got we've got Alphabet betting on what we're doing here.
• 27:46
  SPEAKER_00: And so what that means is that for that for this to be meaningful for Alphabet, success needs to be big here.
• 27:53
  SPEAKER_00: I think of it as a a very big game that we're playing.
• 27:56
  SPEAKER_00: So there's, you know, success for us here looks like having a very broad, very ubiquitous use of the kind of software platform that we're building.
• 28:07
  SPEAKER_00: Now, to me, to get there, we have to go through these, all the intermediate steps.
• 28:12
  SPEAKER_00: You don't jump from the ground directly to the moon, right?
• 28:14
  SPEAKER_00: You've got to, there are these steps along the way, some of which in the moment look like they look kind of pedestrian.
• 28:20
  SPEAKER_00: You know, you're you're working with a small machine shop and you're figuring out how to get a robot in there that solves a problem for them so that they can run a third shift overnight because they don't have enough employees uh to be able to meet their production demand.
• 28:36
  SPEAKER_00: And those those sound like kind of you know one-by-one sorts of interactions, but you have to go through that in order to learn everything that you need so that you can scale up and figure out how to meet, like actually win that bigger game.
• 28:51
  SPEAKER_01: This actually kind of gets back to I think what what you were talking about before when we were discussing bipedal robots.
• 28:57
  SPEAKER_01: But you know, I've seen again a lot of companies right now that are talking about speed to development of a bipedal system.
• 29:04
  SPEAKER_01: And it's impressive, right?
• 29:05
  SPEAKER_01: I mean, it's impressive to see a company build hardware in a year that, you know, from nothing that can suddenly walk, but that's a different goal than building a company that has ROI and can actually like has a saleable product.
• 29:21
  SPEAKER_00: We should not ignore how amazing it is that it's possible to do that in such a short time frame, right?
• 29:25
  SPEAKER_00: And I think a lot of credit here goes to um folks like uh Mark Rabert and Rob Plater, who you know really pioneered a lot of this work at Boston Dynamics for literally decades.
• 29:35
  SPEAKER_00: And now the things that we we can kind of take for granted that, yeah, if you wanted to go from scratch and build a quadruped or a biped, we know how to do that as a as a community of practice, right, in the in the robotics industry, which is extraordinary.
• 29:50
  SPEAKER_00: Um, I think you've raised the right question, which is okay, that these these systems, they're clearly going to be areas where they are useful.
• 29:58
  SPEAKER_00: But what like what specifically are the use cases?
• 30:00
  SPEAKER_00: What's the value proposition to get people to uh to take these on, uh take these systems on and deploy them?
• 30:06
  SPEAKER_00: And I think that's a case where uh we we still struggle a little bit.
• 30:11
  SPEAKER_00: And uh it's probably kind of like what we were talking about earlier, you know, the kind of let's say one big artificial generally intelligent model versus a collection of smaller functional ones.
• 30:22
  SPEAKER_00: I think this is also a question of time horizon.
• 30:24
  SPEAKER_00: I have no doubt that the legged systems are, we're gonna find the right places to use them.
• 30:29
  SPEAKER_00: I think we're still working on that right now.
• 30:31
  SPEAKER_00: And in the meantime, I'm I'm perfectly happy to be providing value to people today with the, you know, kind of the arguably less sexy robot arms.
• 30:42
  SPEAKER_00: Uh, but I I would argue that they're, you know, they're they're doing really useful productive work.
• 30:48
  SPEAKER_01: Yeah, this is one of the many things that, and and people who watch the show will know that um Willow Garage is a longtime fascination of mine.
• 30:56
  SPEAKER_01: And this is one of the many things that's that's interesting to me, is is there did, again, in hindsight from somebody who is not at all involved, but there did seem to be maybe a little bit of um tension between uh the idea to be a pure research facility and the idea to to commercialize to bring to bring something to market.
• 31:16
  SPEAKER_01: Is that is that a fair way to kind of categorize Willow Garage's existence?
• 31:22
  SPEAKER_00: It was never exactly clear what uh what the mission of Willow Garage was.
• 31:29
  SPEAKER_00: And what was amazing is the the autonomy that we had to kind of discover and construct that mission as we went along.
• 31:40
  SPEAKER_00: Like in the early days of Willow, um, I mean, this has been written about, but there were, you know, the early days, there was actually an autonomous car project, there was an autonomous solar-powered ocean-going boat project, and there was the what became the personal robotics project, which, you know, that that eventually created Ross, uh, created the PR2.
• 31:58
  SPEAKER_00: And there was a point where Willow was doing all those things.
• 32:01
  SPEAKER_00: And that was kind of the birth of it.
• 32:03
  SPEAKER_00: In like 2008, we narrowed it down to look, we think personal robotics, the personal robotics program, as we called it, that's the big bet that we want to take.
• 32:11
  SPEAKER_00: And that was where we kind of constructed the mission on the fly.
• 32:14
  SPEAKER_00: We said, look, we think the most, the highest impact thing we can do here is to accelerate the the state of technology development in this uh in this industry by providing the smartest, most talented people in the world with the right hardware, but not only hardware.
• 32:32
  SPEAKER_00: They need software too.
• 32:32
  SPEAKER_00: Let's give them the best hardware and software so that they can start to solve some of these hard problems.
• 32:38
  SPEAKER_00: And that's that was what we did with the the PR2.
• 32:40
  SPEAKER_00: And we gave those, gave the first dozen or so out for free, but then we sold them.
• 32:46
  SPEAKER_00: Um, I mean, it wasn't, it was never gonna run Willow Garage out of on revenue selling those robots, but we did sell another few dozen of those robots.
• 32:54
  SPEAKER_00: The lasting contribution, arguably, has been the software.
• 32:57
  SPEAKER_00: And in that case, we decided early on that a fundamental proposition is that we're gonna make all that software available open source.
• 33:04
  SPEAKER_00: And that became Ross.
• 33:05
  SPEAKER_00: Along the way, we were, we were driven by this desire to have impact as opposed to.
• 33:13
  SPEAKER_00: So I wouldn't say it was exactly like, is it more of a research organization or is it more of a commercial organization?
• 33:18
  SPEAKER_00: It was an impact organization.
• 33:20
  SPEAKER_00: Where the debates were, were like, what what type of impact should we have?
• 33:25
  SPEAKER_00: And that, and there was, and there definitely was tension there because you know, impact is a powerful word, but it's also an empty vessel in some ways.
• 33:31
  SPEAKER_00: You can pour whatever you want into it.
• 33:34
  SPEAKER_00: And so there definitely was was tension, especially in the later years to try to figure out like, okay, we've we've shipped the PR2s, Ross is a has achieved critical mass.
• 33:44
  SPEAKER_00: Like, basically, what's next?
• 33:45
  SPEAKER_00: How do we pick the next thing to have impact from?
• 33:48
  SPEAKER_00: Was there tension at the time around open sourcing?
• 33:52
  SPEAKER_00: Not at all.
• 33:52
  SPEAKER_00: No, that was a and that's a huge credit to Scott Hassan, who uh who you know uh founded Willow Garage.
• 34:00
  SPEAKER_00: He had had his own early commercial success in the first internet boom in the 90s.
• 34:06
  SPEAKER_00: You know, he founded uh e-groups and then went was uh instrumental in the early days of Google.
• 34:12
  SPEAKER_00: And what he saw was that they were able to do the things that they did in the 90s with new internet companies because they started from a really strong open source base.
• 34:25
  SPEAKER_00: It was what you would often call the Lamp Stack Linux, Apache, MySQL, and PHP, Perl or Python, depending on your preference for programming language.
• 34:33
  SPEAKER_00: And what that meant was that if you had an idea for a company like eGroups, which became Yahoo Groups, or even Google, you don't have to start by writing or even buying an operating system, a web server, a database, or a compiler or interpreter.
• 34:48
  SPEAKER_00: That stuff is just available for you for free and at very high quality and maintained by a global community.
• 34:55
  SPEAKER_00: And he, Scott, saw Willow as the opportunity to provide that same, you know, standing on the shoulders of giant starting point to the robotics industry.
• 35:06
  SPEAKER_00: And so it was a fundamental proposition from the beginning that what we wrote was going to be open source.
• 35:12
  SPEAKER_00: And even more so, more specifically, if you get if you're an open source license nerd, it was gonna be permissively open sourced.
• 35:18
  SPEAKER_00: And at the time we used the BSD three-clause license.
• 35:22
  SPEAKER_00: Today, along the way, we switched to Apache 2.
• 35:24
  SPEAKER_00: But it basically it's not a copyleft license.
• 35:27
  SPEAKER_00: It's not a GPL-like license.
• 35:28
  SPEAKER_00: It's a it's a license that says, look, here's the software.
• 35:32
  SPEAKER_00: We hope you find it useful.
• 35:33
  SPEAKER_00: You're welcome to take it, change it, incorporate it into your product, redistribute it, and you are not obligated to give anything back or share your results.
• 35:42
  SPEAKER_00: We really hope you do, and you end up with a small fraction of people contributing back.
• 35:46
  SPEAKER_00: But it was that permissive open source, while at the same time being high quality and well-supported and well-maintained aspect of Ross that caused people to be willing to incorporate it into their products to build companies around it.
• 35:58
  SPEAKER_01: You had given away, you know, the first several PR2s.
• 36:01
  SPEAKER_01: And I remember very distinctly the first time that I met Melody Wise was uh in person at Fetch, and she was showing me the research robot that they were working on.
• 36:15
  SPEAKER_01: And it really, in a lot of ways, it was kind of um I I guess like their successor or their their take on the PR2.
• 36:23
  SPEAKER_01: And um very small overall percentage of of what the company was doing, and and certainly not on Moneymaker, but what she said to me, and this has really stuck with me, is like, you know, that's kind of like how that's how you get in, right?
• 36:37
  SPEAKER_01: Like you get that, you know, you you get people to work on your system, right?
• 36:41
  SPEAKER_01: And then, you know, you get you get really smart, like, you know, kids in universities um uh you know working on it, and and you you've kind of like hooked them to a certain extent.
• 36:51
  SPEAKER_01: So then that perspective, it made a ton of sense that uh, you know, the the PR2 would be going out in the world and that you would be releasing ROS in that way.
• 37:00
  SPEAKER_00: And there's a bunch of other aspects.
• 37:01
  SPEAKER_00: I mean, another really important aspect of Willow Success, and a lot of this credit goes to Melanie for kind of creating and running the program was our intern program.
• 37:10
  SPEAKER_00: So we, in just in terms of involvement, so it's one thing to give give the robots away, to give the software away.
• 37:16
  SPEAKER_00: We went a step further and had at any time a huge number of interns coming from the best schools, the best labs in the world.
• 37:24
  SPEAKER_00: We hosted them at Willow.
• 37:27
  SPEAKER_00: And the proposition for them was it, you know, these are these are some of the best people come in the, you know, they're in grad school labs all over the world.
• 37:35
  SPEAKER_00: They could go work at big tech companies for the summer.
• 37:38
  SPEAKER_00: If they go do that, then they're working on something that's secret, it's proprietary, they're gonna have to leave it behind when they go back.
• 37:44
  SPEAKER_00: And our proposition to them, our requirement, in fact, was you come here, everything that you work on here is gonna be open source.
• 37:51
  SPEAKER_00: You're gonna be pushing it directly into open source repositories.
• 37:55
  SPEAKER_00: And then we sincerely hope that when you go back to your lab, you keep working on it.
• 37:59
  SPEAKER_00: And we started to think about them as a kind of a contagion vector.
• 38:02
  SPEAKER_00: They would take what they learned about Ross, take it back to their labs, and get their labs using it.
• 38:07
  SPEAKER_00: And that was, I mean, I did not appreciate it at the time in retrospect.
• 38:11
  SPEAKER_00: I mean, it was a it was a master stroke how like the impact that that program had.
• 38:18
  SPEAKER_01: I really wanted to get your take on, you know, again, as somebody has has been in this a while and has been so closely connected, um, specifically through uh open robotics to to Ross on what the state of the open source uh robotics, I guess, seen at the time.
• 38:39
  SPEAKER_01: You know, there's been a lot of really interesting um uh a lot of really interesting things have transpired over the last several years.
• 38:47
  SPEAKER_01: You know, Hugging Face has gotten involved.
• 38:49
  SPEAKER_01: They've got that, you know, like basically like a GitHub like repository.
• 38:53
  SPEAKER_01: And um, you know, obviously there are a lot more cheaper robots.
• 38:58
  SPEAKER_01: So that's probably in some ways a little bit closer to kind of that original vision, I guess, of a PR2 in terms of like like accessibility.
• 39:06
  SPEAKER_01: Um we've seen some projects.
• 39:09
  SPEAKER_01: Yeah, I was I was rooting for the K-scale team over in the Bay Area.
• 39:12
  SPEAKER_01: I don't think that quite worked out.
• 39:14
  SPEAKER_01: But um yeah, I'm I'm I'm I'm I'm curious what what do you think about the the state of it right now?
• 39:19
  SPEAKER_01: Nvidia, obviously like a player in that scene too.
• 39:22
  SPEAKER_01: And uh what role um intrinsic is sort of playing and I guess helping to facilitate some of that?
• 39:31
  SPEAKER_00: Well, first of all, I think it's the the state of that community, I think is really strong.
• 39:36
  SPEAKER_00: Um I saw this firsthand at uh RustCon, our the developer conference that we run every year um that I we did in Singapore this year.
• 39:45
  SPEAKER_00: And now when I say we, I should also explain I'm CTO from for intrinsic, running the engineering team.
• 39:50
  SPEAKER_00: I'm also still the chair of the board for the Open Source Robotics Foundation, which is that that was the original entity that uh became open robotics that we created in 2012.
• 40:00
  SPEAKER_00: And that's still that's that's the organization that you know really owns Ross as a as an open source project.
• 40:08
  SPEAKER_00: Um so uh in my role there, I helped to put on uh ROSCON.
• 40:12
  SPEAKER_00: We had you know approaching a thousand people get together in Singapore for three days and talk all about the latest developments in the use and development of uh open source software for a huge variety of work uh use cases.
• 40:28
  SPEAKER_00: And there was a there was at unsurprisingly a lot of attention paid to how do we take this system and make sure that it's it's fit for purpose when we bring in these new AI models.
• 40:41
  SPEAKER_00: So when we're we're doing this more like uh basically learning uh trained systems versus kind of classically programmed systems, where is where is Ross working well and where do we need to improve it?
• 40:54
  SPEAKER_00: And those are really great discussions to have.
• 40:56
  SPEAKER_00: And also, they're also presenting on you know what are the latest applications that they're putting out in the field.
• 41:00
  SPEAKER_00: So I think the the state of the community, I felt really good.
• 41:02
  SPEAKER_00: That's it's one of my one of my happy places each year is to end up at uh ROSCON and just see all the different things people are doing with it.
• 41:09
  SPEAKER_00: From intrinsic side, we had a booth at ROSCON where we showed off some of the work that we've been doing to integrate Ross with the intrinsic stack with Flow State, which is our that that's the intrinsic product that is an end-to-end simulation development deployment system for uh for today for for manipulators.
• 41:31
  SPEAKER_00: So we were showing that off.
• 41:32
  SPEAKER_00: We were also showing presenting on the work that we just do contributing to the RO ecosystem.
• 41:39
  SPEAKER_00: So I've got, you know, within my engineering team, there's part of that team is just dedicated full time to making open source contributions to ROS, to gazebo, to open RMF, to the the um the build infrastructure, which actually produces all of the binary packages that most of the users out of the RO ecosystem end up installing and running on their systems.
• 42:02
  SPEAKER_00: And that to me is a really important commitment.
• 42:06
  SPEAKER_00: It's rational as well.
• 42:07
  SPEAKER_00: Like it's I don't think of it as a as an act of altruism, because if it were, it would be fragile, right?
• 42:12
  SPEAKER_00: If if we thought about it as like a, if we think about it as a gift, it's always something you could decide not to do.
• 42:18
  SPEAKER_00: And I think about it as a as a completely rational, self-interested investment because the the goodness of like we we depend on that ecosystem and its health increasingly for our own business.
• 42:32
  SPEAKER_00: And this is this is in the end why companies contribute to open source.
• 42:37
  SPEAKER_00: They're only going to do it in a persistent, durable way if it is important to their business.
• 42:41
  SPEAKER_01: You you and I you know spoke a number of times when the acquisition was happening.
• 42:46
  SPEAKER_01: Um, and I'm sure that you got I'm sure there was a lot of freaking out happening in in certain circles.
• 42:53
  SPEAKER_01: I I'm seeing that play out in a similar way with uh with Arduino and Qualcomm right now.
• 42:59
  SPEAKER_01: And and and like I get it, right?
• 43:00
  SPEAKER_01: And I'm sure that you get it too, as you know, as as a nerd and somebody who's been involved in in those circles as well.
• 43:06
  SPEAKER_01: And you you had to be sort of you had to be an ambassador, right?
• 43:10
  SPEAKER_01: I mean, you had to be one of the people who's out there sort of explaining, like, here's how things are gonna go uh moving forward.
• 43:16
  SPEAKER_01: Were those were those difficult conversations?
• 43:20
  SPEAKER_01: Was it was it difficult to explain to people why you were making that move?
• 43:25
  SPEAKER_00: Sometimes yes, sometimes no.
• 43:27
  SPEAKER_00: And I remember you, I think you interviewed Wendy, uh, our CEO and me at an event around the time around that time a few years ago.
• 43:35
  SPEAKER_00: And I, if I recall, what I what I said then was um, you know, look, I I in my view, this is the this is gonna be the best outcome for that open source community that we've created.
• 43:49
  SPEAKER_00: But you shouldn't just trust me, you should come back and check the out check the result later.
• 43:53
  SPEAKER_00: And and that's what I would say is, you know, now we're uh let's see, that we're three years.
• 44:00
  SPEAKER_00: Now, actually.
• 44:03
  SPEAKER_00: Three years later.
• 44:04
  SPEAKER_00: And I would say, let's go look at how's how's the Ross community doing?
• 44:10
  SPEAKER_00: Are people still coming to ROSCON and showing off their results?
• 44:15
  SPEAKER_00: I would look at the Open Source Robotics Alliance, which is a new, that's a trade association, a membership-based trade association that Open Robotics created.
• 44:23
  SPEAKER_00: And I would look at the the huge uptick in membership that we've got there.
• 44:26
  SPEAKER_00: Companies, and you know, intrinsic is a member, yes, but we're one of several dozen uh companies that have come and are paying to be part of the Open Source Robotics Alliance to have a seat at the table to help steer the future of the community.
• 44:41
  SPEAKER_00: So it was it were there difficult conversations at the time?
• 44:44
  SPEAKER_00: Did people, did people freak out?
• 44:46
  SPEAKER_00: Did people think like, oh my God, this is gonna, you know, uh Google is buying Ross and everything is gonna go to hell?
• 44:53
  SPEAKER_00: Yeah, some people thought that.
• 44:55
  SPEAKER_00: We went out, we talked about it, we were honest, we were open and transparent.
• 44:59
  SPEAKER_00: But and again, I said, look, the the the proof of the pudding is gonna be in the tasting, you know, come back later and and check the result, and you can decide whether you think we followed through.
• 45:08
  SPEAKER_00: In my view, we did.
• 45:10
  SPEAKER_00: Um, you know, if somebody has a different opinion, I'd love to, I'd love to hear about it.
• 45:14
  SPEAKER_01: I'm not gonna get all like flowers and sunshine about this, but again, one of the things that I've really appreciated covering this, obviously, you know, it's a bit these it's a business, these are all business.
• 45:25
  SPEAKER_01: There, there is competition, not not everything that uh happens um in and around companies is is is great and mutually beneficial for everybody.
• 45:36
  SPEAKER_01: But I do compared to a lot of other things that I've covered, compared to consumer electronics, for example, which I covered for a number of years, um people seem genuinely impressed with each other's work.
• 45:52
  SPEAKER_01: People seem to, you know, like working together, like collaborating.
• 45:56
  SPEAKER_01: And a lot of that stems from the fact that at the end of the day, it's still a like a pretty small community that everybody went to school together, everybody worked at the same like four or five companies together, everybody knows each other, and it like at everybody's probably gonna end up working with each other again at some point.
• 46:15
  SPEAKER_00: That's one of the things that keeps me in this industry, in this community, is all those, all those parts of it.
• 46:21
  SPEAKER_00: I think it is a it, especially if you've been doing if you've been working in robotics for a reasonably long time, like more than 10 years, then you almost certainly got into it not because you thought it was gonna change the world or it was gonna make you independently wealthy.
• 46:39
  SPEAKER_00: You got into it because you love the technology and you're just fascinated by robots.
• 46:44
  SPEAKER_00: Uh, and so because all those other things, they didn't seem possible or realistic back then.
• 46:51
  SPEAKER_00: And I think that that, the especially the folks who've been in it for longer, we kind of we set a norm, right?
• 46:58
  SPEAKER_00: We were, we, we worked together in a variety of different ways, including through a lot of open source development.
• 47:04
  SPEAKER_00: And we established a kind of a standard of behavior and we got together at conferences and we got to know each other.
• 47:10
  SPEAKER_00: I I think this is frankly not not unusual among young industries.
• 47:15
  SPEAKER_00: Perhaps we've just done it and held on to that spirit for longer than it usually happens.
• 47:20
  SPEAKER_00: But I think that aspect of it is one of the reasons that we all do stay engaged, is why we don't go off and uh try try something else, right?
• 47:28
  SPEAKER_00: I'm not you don't you don't have uh you don't have people jumping off to to other industries.
• 47:33
  SPEAKER_00: They they really stay committed to what we're doing here in part because of those those relationships.
• 47:38
  SPEAKER_01: Well, Brian, um always a pleasure.
• 47:40
  SPEAKER_01: Thank you so much for taking the time.
• 47:42
  SPEAKER_00: It was it was my pleasure.
• 47:44
  SPEAKER_00: Happy to be here.
• 47:45
  SPEAKER_00: Thanks, Brian.
• 47:47
  SPEAKER_01: Really great conversation.
• 47:48
  SPEAKER_01: Thank you so much to Brian Gherkey and uh thanks to Leonard for helping set that up.
• 47:53
  SPEAKER_01: Thanks to you as always for tuning in.
• 47:55
  SPEAKER_01: Please like and subscribe.
• 47:57
  SPEAKER_01: And don't forget to subscribe to the automated newsletter over at automated.fm that drops every Thursday morning with exclusive features that you can't get anywhere else.
• 48:09
  SPEAKER_01: Thanks again, and we will catch you just about this time next week with another episode of Automated.