 Hey everybody, welcome to another episode of Open Infra-Live. I'm excited to be bringing you to, coming to you live with Jonathan Grice and Amar Papadavan. Now we had a conversation earlier with Bruce Davy and Martin Casado and we all had an awesome conversation a couple of days ago. We have Bruce from Australia and Martin coming from California. So all these are some awesome networking experts. We've got Amar here with us live on the stream. So as we bring to you this conversation we had a couple of days ago recorded, the three of us will be live in the chat. So we'll be there to answer your questions in the text chat as we go. And then after we roll this conversation, we will actually come back on live and answer any other questions that we haven't been able to get to in the chat. So I hope you enjoy this conversation as much as we did. These are three great experts in networking, global connectivity and why this topic is so important to the world and open source. So let's go ahead and roll the video. Hey everybody, welcome to Open Infra-Live and we're really excited to have Martin, Amar and Bruce with us. So to get started, let's talk about global connectivity. And I think the first thing we want to talk about is just why does it really matter? What all is at stake when we talk about global connectivity? And I thought maybe Amar, you might have some thoughts to share on that. Yeah, definitely. Thanks a lot, Mark. So yeah, I think the main motivation behind how even I got involved with Magma was some of the numbers that we discovered while doing some preliminary research on internet access. So the internet has become a fairly crucial tool for improving livelihood of people and 68% of people use that to learn new jobs. In fact, there was a Deloitte report in 2011 or something that if you connect 10 people on to communication, not necessarily internet, one person moves about the poverty line. So this is actually a lot more than just a luxury. It's sort of become a necessity. And then a lot of people use that as Martin is deeply familiar with to pursue an education and also bridge the digital divide. So the biggest data point and that came across was that of the 3.3 billion people in developing and emerging markets, most of their network performance is going to degrade by the year 2023. So it's not even that we're going to continue to offer the same quality of service, it's actually going to degrade by the year 2023. So a lot of the work that we're doing in open source as well as in Magma in particular is targeted towards emerging market and Google access and areas where there's not been traditional investment. And we're trying to bridge this gap of like sliding connectivity, which is a pretty key challenge in a barrier. So some people would actually get lower performance in the future than they're already getting. So it seems like we actually have a lot of work to do to even offset that, I guess. Did Mark freeze? On this global connectivity initiative? We lost you for just a second there, Mark. What was the last thing you said? I just said, I was asking Martin what his thoughts were on global connectivity. Yeah, so, you know what I find interesting is anytime you talk about kind of like connecting the next billion or the global connectivity problem, I think people immediately view some island or Sub-Saharan Africa and like some hut and like, you know, what does it mean to take connectivity out there? And that's really a problem. That's definitely a problem. But it turns out that, you know, listen, I'm sitting in the Bay Area. I probably have to travel 30 miles to get to areas that have no connectivity. And so the problem is so much more pervasive than most people realize if there's a huge issue in the United States, something like 23 million people in the United States just under 10 percent are under connected. So from my standpoint, the Internet came very quickly. It changed everybody's lives. It's almost a necessity. But because probably how rapid the growth was, it has not been homogenous in that growth. And anywhere you look, I think that there is, you know, something to do. And, you know, having, you know, like Bruce and like Amar been in networking for the last 10, 20 years. I think there's a lot of good that, you know, we can do by focusing on this problem. And you don't have to go to, you know, Amazonia or the Philippines, you know, Sub-Saharan Africa in order to have an impact. Yeah. Jonathan, do you have any thoughts on this? Yeah, I was going to say that, you know, what Martin is talking about there, I think is something that I've definitely noticed as well. My my parents live in kind of East Texas. And it's, you know, it's a few hours from here. And I go there and visit and the network connections are always less reliable there. And I think the point that you made, which is that we've made a lot of progress, but it hasn't it hasn't been kind of even progress that has that that everyone has advanced at the same rate. One of the things that I'm talking about, my parents, that whenever I'm thinking about connectivity and the value of it, we travel a lot for work when in non-pandemic times. And in, you know, my dad, when I when I travel a lot of times, he's like, OK, well, where are you staying? You know, what hotel are you going to be in? How do we reach you? And and for me, you know, I'm just like, you have my mobile number. You know, that's how you reach me where I am. But, you know, that's not that's not the way it is for for everybody. It's a huge it's been a huge advantage, I think, for for those of us who do have that kind of always on connection to the network. Always have all of all of the people that we want to talk to at our fingertips and the data that we want. But that development of of that advancement, I think, has has definitely been uneven everywhere in the world. And yeah, in many, in many places that that you maybe don't don't think of as as the first ones that pop into your head. Yeah, thanks for the story storytelling. If anyone else has any stories about about their traveling and connect with their parents, let me know. I think, you know, it definitely affects people everywhere. And yet East Texas is not the most wired part of the of the country. That's for sure. So I think, you know, one of the things we wanted to talk about next is just kind of what are some of the big forces that are driving change across the industry more, you know, globally, in terms of the trends and what's happening in the technology side. And so I think, you know, Jonathan, do you want to do that? So I, you know, I was talking with Amar last week, actually, and and he he made a comment that the network evolutions have have always been in service of new computing platforms. And and and that was an interesting thing that that he mentioned. And so I asked him to tell me more about that. And he said, actually, you should talk to baby about. So, Bruce, I mean, is that something that you agree with? Do you think that that's that is often what drives those those major changes at the network level is actually computing? Or what are the things that kind of drive those changes? Yeah, I mean, I think it's a good point that, you know, basically, we've we've we've seen a network that evolved from connecting all the compute facilities in in the world back in like the 70s. Like that was the internet, right? It was built to connect essentially a few dozen computers that a very privileged few people had access to. And, you know, one of the biggest shifts in networking was the when we suddenly started seeing smartphones, all of a sudden there were, you know, there's a computer in everybody's pocket. And so we've gone from an architecture that was meant to connect a few dozen computers to an architecture that now needs to connect billions of computers. And in some respects, the internet's done an amazingly good job of managing that transition. You know, there was kind of one big glitch, which was that the original Internet addressing space wasn't quite big enough for several billion devices. But that one's been pretty well addressed. But the other thing is this idea that, you know, the place where you connect to the network now could be anywhere on the planet, as opposed to it being inside a lab. That's kind of the really big change. And so we've we've got an architecture that works pretty well for connecting fixed devices in, you know, big buildings. But we the architecture for connecting all these mobile devices that are spread all over the face of the earth, that's still a little bit in in, you know, evolution. And I think one really interesting thing that's happened is so we had the Internet architecture worked really well. And then we've had this parallel architecture of connecting mobile devices, and they've kind of grown up in two different ways. And now what we're seeing is they're starting to converge. And to some extent, this is what I think magma is doing, is it's taking all the things that were proven to work really well in the Internet and applying those principles to mobile networks in a way that hasn't been done before. So, you know, if you look at the architecture of today's mobile networks, it's actually not that hard to see. They started off being voice networks, and then data services were kind of grafted onto them over time. And so I think what we're seeing with magma is trying to take sort of an Internet centric view of the world and apply that out to the edge. And that's really what we need to do if we're going to connect the lines of devices all over the world. Yeah, I think we have a stat that came from a recent analyst report that's the 20, 30. And I remember a few years back when we first started talking about this Internet of Things kind of and some of this people were talking about 10 billion and 20 billion devices and that kind of thing. In this time frame, and now we're up to 50. And first of all, I'm always these numbers, I think always just big guesses, but it does seem like every time that they revise these guesses, they're getting higher. Do you think these numbers are realistic or do you think they're too high, too low? What do you think we're really going to see if we head through the next decade? Oh, this is, I mean, 50 billion seems totally believable. I mean, how many devices are in your house right now? And so, I mean, granted, we're we're at the kind of the rich end of the spectrum, I think, but the idea that there's more than one device per person, that to me seems completely obvious. And it's just a matter of how many devices per person. So to me, 50 billion actually doesn't seem that ambitious. I mean, I'd be surprised if it isn't that now, honestly, if you think about I'm just saying, I mean, it depends on I think maybe what you mean on the Internet, but if you're just talking about like end points connected to a network that have public connectivity, I'm certain it's more than 50 billion. Yeah. And then you start thinking about, you know, when we start using devices to think about autonomous vehicles, like that feels like something that's maybe within 10 years. And I think a lot of what's been talked about with 5G is just being faster, which I think is kind of actually not that interesting. And, you know, a good 4G service works pretty well, as long as you don't have to share it with too many other people. But the thing which I think is much more exciting is when you start trying to do connectivity to a vast number of devices and to do it in a way that is cost-effective and to do it in a way that deals with, you know, having low latency, there's a bunch of things you have to deal with as the number of devices gets bigger. And that's why I think this is sort of most people, when they think about the evolution of mobile networks, we have 3G, 4G, 5G. We're always thinking about going a bit faster, but I think there's much more interesting transformations happening that have a lot more to do with this increasing amount of activity at the edge. And I mean, the autonomous vehicles is one that I often use as an example, because it's pretty obvious how quickly you need to be able to get information to devices if you're going to try to communicate with them in the timeframe necessarily necessary to handle things like autonomous driving. And you like a lot more compute, obviously at the edge too, right? The network's speed increases driving the need from our compute and vice versa. And obviously, like intelligence and sort of the ML AI side of it seems like it just changes everything when you have that super low latency more so, like you said, than the raw speeds, which get the headline, but it's actually not what's disrupting the architectures. And that's why I think like when you look at the architecture of Magma or even the architecture of sort of a more standard 5G implementation, you're going to see more like a cloud computing system out of the edge. So we're used to the idea of that there's big data centers that sit in a handful of locations and you traverse a long distance to get to your nearest Amazon or Azure data center, but they really need these little mini data centers out of the edge so that you can do that local edge computing. So what are some of the specific things that are different architecturally about Magma that make it a better fit for that? What are some of those changes that tend to support that kind of more parallel distributed device type traffic versus something that's more central back and forth to fewer devices? I mean, for me, the single biggest innovation in Magma is that the details of the radio don't leak into your core. So if you look at a traditional implementation of a cellular network, you can tell by looking at its core what kind of radio there is. And that to me, it's analogous to like, imagine if you could go into the core of the internet, look at a great big router sitting inside an ISP and tell which users were on ethernet and which users were on Wi-Fi. And like the idea that that sort of access technology and the state of individual edge devices leaks into the core, I think is one of the big shortcomings of the standard sort of telco cellular architecture. And when you look at Magma, it basically terminates all the wireless protocols right at the edge. So you can then build a core that's independent of what's going on. And that has all kinds of implications. But it basically means you can build a more scalable, more reliable kind of core. And also you can say, oh, if I can get access to any kind of radio, I can I can stick it on to my Magma core. So I don't have one core for 3G and a different one for 4G and a different one for Wi-Fi and one common edge system that terminates all the radios right at the edge and then provides common core services to those to those radio devices. I also think it's worth putting out. So I just jump in here, which is Magma itself is remarkable. And we're going to talk a lot about Magma, which is great. But I think actually, Magma is also an indicator of something bigger, right? It's like almost like technologies come and fill in needs. And you can look at this technology, or you can look at the need, right? And it says something about the ecosystem. It says something about the need. And so I think that it's just kind of a good point to point that out, which is it's turnout connectivity has come to the point where it allows for something like Magma. What does that mean? It means it's an open source project that can run on other people's hardware, where you can have hobbyists, companies work on it, which means that there's enough maturity in the supply line, there's enough maturity, the developers and there's enough need from the customer base for this type of thing. And that didn't exist 10 years ago. So it almost feels like, as an industry, as a community, we're now ready for an open platform to solve connectivity problems writ large. And so I love that we're talking about Magma because itself, it really fits that need very well. But again, the reason it exists as viable is just because I think the world is not ready for such a thing. Well, that you bring up a really good point there. And it's one of the things that I wanted to wanted to get to you, which was to talk about some of those new, because I think besides the the computing platform, I think there are commercial forces that are also driving this change on the network. And, and, you know, one thing that there was a light reading article that came out just a few days back that had had some stats about the wireless Internet service provider or the wireless ISP market. This is a market that I didn't really even know existed as a as kind of like a whole thing until maybe about a year ago. But this, this article covered some stats from a recent report. It said, for instance, that the the number of customers that that the wireless ISPs have, it's currently doubling every five years. And the that's basically, you know, 73% I think is the annual growth rate there. And this is the thing that really surprised me is that last year, I think it was already over a $4 billion market in the US in in terms of the revenue for for this category, that is actually one of the I think one of the main things that that that becomes a use case for magma, you know, like you said, the part of the reason why magma as an open source project can exist is because there is now this this commercial opportunity, and there's a need. And these these wireless ISPs are are are interesting to me because they are filling some of those those spaces in in in the United States, for instance, that haven't had great coverage. And they're also going after markets that haven't always been appealing to the traditional network operators. So I think you're you know, you raise a really good point there. Where do you think are some of the other the other opportunities that that this new this new style of network is either going to enable or that that it's going to have to meet the demands of new kind of new use cases. So I want to say something quickly, and I think XMR has got a great perspective on this. But so I think there's this grand convergence, right? It's like, you know, whatever, like the samurai movie, or like, you know, like the, like, like from the different regions they come together, and they form like this kind of great group, and like in the metaphorical sense. So listen, the hardware supply chain is now ready, right? It's to the point that we can use it. I think regulatory policy has evolved so much that we actually now have a lot of access to spectrum. The software has gotten there. And then the understanding of the use cases, all of these things have gotten together now, so that we can we can deploy more cheaply than ever cellular connectivity in more varied areas ever than before. And so I think that, you know, one way to think about this is we can address, you know, known problems, which is, you know, there's the there's the well known problem of connectivity, you know, we, we know that people are interconnected, and and we can solve that problem. And that's a huge that's a massive, massive market. I mean, we're talking 10s of billions, probably hundreds of billions. Actually, if you look at just the geosynchronous satellite market alone, which you could definitely do with terrestrial bandwidth better, it's a $200 billion market. So we're talking about like markets that consume other markets, you know, for breakfast, I think trillions, I like trillions. I mean, well easily, and we're talking about connectivity, right? You know, and like, even just basically, you just like, like, like an old crufty part of this, like geosynchronous satellites, and it's huge, right? And so, and so I think there's just going to be a lot of like kind of business innovations pushing that into new areas. But then you can ask the question what you're asking now, which is, well, also, what does that unlock, right? You know, you you're almost in this economic equilibrium where you can't push this into like every device or every new corner. But now you can what does that open up? I'm certainly not in the position to predict that. But I'm so excited to see the Cambrian explosion of these new apps kind of open up because now we have the enabling technology to do that. Yeah, Amar, what do you think? What are some of the use cases that you have seen coming from magma users that are that are are are the leading edge of the Cambrian explosion? Yeah, I think, yeah, so definitely like the the CBRS market has opened up a bunch of players and you know, Can I stop you there for just a second? And can you for people who aren't familiar, you know, this is another new acronym that I had to learn recently, could you just give a quick definition of CBRS? So CBRS is a community broadband, I think radio spectrum, I believe, but it's basically owned by the US Navy. And and in the non coastal areas that is currently available for gender views by other people using this a spectrum database that gives you a like a lease. So think of this as DHCP, but for spectrum, right? So this allows a lot of people to, you know, get access to license spectrum. So what this is the thing that happens with these multi billion dollar auctions every so often where the giant companies go gobble it all up and nobody else can access it. CBRS gives a dynamic way where I can, I can actually get a lease on on a little bit of spectrum around my location here and then make use of it. And and and that's something that is that is completely new, right? In the history of sort of radio spectrum, at least in the US, I know that that there are policies around this in other countries too, but at least in the US, it's totally new, right? That's right. That's right. Exactly. Thanks for the help there. Yeah, so I think what what it actually did is that by unlocking the spectrum barrier, it's sort of it opens up one access for innovation and just echoing Martin's point, right? Like the the thing with the even the previous generation of like networking like SDM, right? A lot of the work that Martin and Bruce did was that it never actually optimized for a particular application. What it did was it built a platform that was suitable for heterogeneity, right? So that is basically what the networking platform that's going to service connectivity is going to be is is that it's not going to be a platform that's going to be in a position to predict the next killer use case. It's going to be a platform that can abstract away that heterogeneity so the killer use case gets enabled, right? And that is where I think the key investment that is going to happen is going to be. And there's obviously like, you know, at the end of the day, people need to bring home the bacon. So there are obviously like, you know, you know, connecting people and you know, paying for like private entity and those kind of use cases are like sort of the tip of the iceberg. But there's a completely new sort of like, you know, market that's getting created with wearables, right? Like wearables are all like, you know, tether devices, and they have a fundamental problem where, you know, they're thermally bomb. So you can't keep sticking CPU in there, right? So you can't do all the computation that you're like your like your smart glasses need or your watch needs at the edge, right? So at that point, you need to figure out how do I create a consistent runtime between the edge and the network and all of these things are like, you know, patterns that in some way rely on like a wireless network being in there that's sort of facilitating this. And what we think here is going to happen is that the open platform that Martin spoke about that is best suited for heterogeneity is going to be in the best position to sort of unlock the next like billion dollar use case, right? And maybe like to Bruce's point, like the Uber app, right? All it required was GPS and and like LTE and then you suddenly got Uber on your phone and then it like completely changed the taxi industry. Are there other change? I changed automotive. Not just the taxi industry. Are there other policy examples like the like the spectrum example that that besides the technology, besides the software, the hardware and kind of like the the business desire, are there other policy issues that are that are kind of, you know, keeping the old network models entrenched or, you know, harder to change? Yeah, so at least in a lot of emerging markets, a huge percentage of GDP is actually based on tax that the M&Os pay, right? Like it's basically like there's a large unbanked sector and then there's these people buying data plans and then that drives a lot of tax revenue. So historically there have been like, you know, that has sort of been something that, you know, regulation is like, you know, the telecom industry has been like heavily regulated because of that, right? So so licensing of like, you know, for an ISP in many of these emerging markets is actually pretty difficult because, you know, they don't want to like cannibalize a pretty large revenue stream. A lot of these, especially in LATAM, what we're seeing is that a lot of these governments are actually seeing connectivity more as a fundamental need. And so are sort of adapting the regulations to sort of make this a second order need as opposed to a primary need. And then by just bringing connectivity, they're seeing enough like monetary advantage and, you know, economic activity that it's sort of, you know, compensating for some of the GDP related issues that, you know, ISPs that don't pay as much taxes as a M&O sort of influencing. But yeah, in general, like spectrum is probably the biggest barrier because that is the entry, the entry bar there is pretty high. Okay. And, you know, there are private LTE networks, I know are another area where, you know, people are using basically wireless technologies to replace running cables in plants and facilities. And again, it seems like as you lower the cost and the complexity and improve the capabilities of these tools, they're just going to be a lot more of that. There was a some news last week from one of our, the companies that we've known for a while is, or the people that we've known for a while, Boris Rinsky, he started a company last year called FreedomFi, which is kind of diving into this new world. And they announced a partnership with the Helium Network that gives you a FreedomFi gateway, which is pictured here with a smiley face on it. It's basically running magma. It connects to these kind of standardized radios that are now available and then is a way to offload traffic from using the Helium Network and their kind of like coin payment system that they have. You know, if you look at this headline, it has pretty much all of the buzzwords in it at once with cryptocurrency and 5G and mining and, you know, FreedomFi and all this kind of stuff. But I think that this is, you know, this is another example of something that that just probably 18 months ago people wouldn't have really been thinking about. And when we were chatting with Boris today and he said that, what did he say, Mark, in terms of the response? He said they had over 10,000 paid pre-orders for this box. And to me, it's just like really interesting as you think about like what, you know, Martin and Amar were saying a minute ago about sort of like the tax revenue and governments and like sort of this concentration of power and a small number of providers and in these different, you know, government systems and spectrum is this kind of like choke point, so to speak. And then you think about decentralization as kind of this big wave the other direction. And you have like this crypto thing and this, you know, distributed private 5G or 4G thing that, you know, FreedomFi is doing. So it's interesting to see these kind of forces and counterforces at work. But I'm curious what Bruce thinks about all this crazy stuff that we're talking about. I mean, I think it is a good example of how if you if you create a platform that doesn't sort of presuppose a particular outcome, then you open up opportunities for innovations. So like I guess I always like to think about how, you know, the internet didn't particularly optimize for any one application. And like I can remember before there was a World Wide Web there were lots of other ways of getting access to information on the internet, but none of them were particularly easy to use. And then, you know, one day the World Wide Web became the easy way that everybody got access to information and we stopped using things like Gopher and the other things before it was like, I'm sorry, I'm really old here. But the idea that you could build a network without having any idea what what the dominant application would be. And suddenly when I see something that combines like 5G and cryptocurrency, I'm like, OK, well, there's two things I didn't expect to be combined into a single offering. But it's like, well, you know, both of those are just, you know, building on a platform that doesn't presuppose a particular sort of optimization for one outcome. Even like the way the way broadband is provided. Like, I think it's it's very interesting to me that, you know, when I lived in the US, I had a choice of roughly two broadband providers. Whereas now that I live in Australia, connectivity is is kind of done as a regulated monopoly effectively. And then that you get layer two from the from the connectivity provider. But then there's dozens and dozens of ISPs that run on top of that. And so they've kind of made connectivity a basic offering. And then the ISPs compete on top of that. So there's like completely different way of slicing the market that it certainly led to more innovation in terms of how ISPs compete than what you see when it was sort of sliced in a different way. So it just does kind of show you how making certain policy choices will have impacts on how the technology rolls out. Yeah, I love that. And I think I tell you when we started OpenStack like a decade ago, we had no idea that telecoms would want to run it at all. And now here we are at this point and we see like four billion people in the world are actually connected to networks powered by OpenStack. It's insane. I mean, we had no ideas, sort of a general purpose technology, right? Pete storage networking. But the fact that phone calls might be routed through networks powered by OpenStack was the furthest thing from our mind at the beginning and now like nine out of ten of the top, you know, providers in the world are actually powered by OpenStack. So I'll get my quick plug in there for OpenStack. But it's just a really good example of, you know, when you design a technology or even just kind of enter a market with some some different tools that enable people to do things, they're going to do stuff that you just absolutely did not expect. And so it's a great point. I want to just say that I have no idea about the freedom five thing that you showed, but, you know, it sounds super buzzwordy, right? Like 5G and like crypto and yada, yada, yada. However, and again, I didn't even know about this going into this. It actually makes sense, right? Which is routing has always been a distributed thing, right? And connectivity has always been a distributed thing, right? And there's always been this dream that why do we have to have central infrastructure that's run by a telco when, you know, listen, I have real estate, you have real estate, I have a tall, you know, part of my building, you have a tall part of your building. And so if you can create a way to incent people to offer real estate bandwidth, which has always been a very important thing, and you had a way to monetize that securely, it all of a sudden makes sense. So sure, 5G is just kind of, yes, this is fast. You could imagine some incentive scheme on top of that to promote more bandwidth, more connectivity, more points for us an asset that is actually a fixed stash of which is real estate. So, again, I don't know if that's even what they're doing. I just think it's so cool that you have a platform now where people can marry these concepts and what seems to me as, you know, the uninitiated in a very obvious and cool way. So I think that was super cool. Yeah, I mean, they had that seems like they had a great response and and I can't wait. I mean, Boris is is is always going to make something happen. Boris is the best. He's a force of nature. We love Boris. I love Boris. He's awesome. So, you know, maybe this is I guess this could be any of y'all, but what, you know, if we are thinking about the technology pieces that are that are going to drive this, we've got we've got a few of them in open source already. What do you think are the pieces that that we we still need to be working on as as kind of communities overall to open up and to keep to keep enabling more access and more innovation in in in terms of the network. I know we've touched on radios a couple of times and there are things like O-Ran and and we've touched on obviously Magma with the core. And what what do you think are the the pieces that that bring it all together that might still be missing? I mean, I do think one of the things that's often neglected is how do you make it easy to operate these kind of services that network management is kind of always been the sort of the unfortunate stepchild of networking. And I think that's one of the areas where I really like the idea of being able to pick up ideas from other parts of the community. And so if you look at part of what's going on in Magma that I think is potentially very powerful is you don't necessarily build your own system for doing logging and alerts. You leverage something else from the open source community. And then you say, well, we can just get a best practice from there and bring it in as opposed to we're going to go and implement something that's completely optimised for a telco environment. So that to me like how do you make the operations as easy as possible? I think this is this sort of thing that I found very frustrating dealing with the telcos over many years. The sense of we're going to go do everything open source because it's going to drive down our costs without realising the level of expertise they would need to stand up and run open source software. And so, you know, we shouldn't sort of get over rotated on to thinking open source is the cure for everything. It only becomes a solution when you can operate it. And to me, that's the thing which we've got to be putting a lot of focus on. Couldn't agree more. So it's a really good point. So I have a slight it's great point, Bruce. I couldn't agree more. And I think leveraging a lot of the cloud infrastructure is pretty critical to build the next generation of computing and network. I think one thing that is sort of holding back the at least the connectivity space is the complexity of the standards. I think a lot of what has happened with 3GPP is that it's a one size fits all standard. So regardless of whether you're operating a small fire fixed wireless site at the end or like, you know, like, like what Martin's doing at NeuralNet or whether you're like running like a mobile network in downtown New York, like the specification and the requirements that the output out to the network are almost the same. And I think just being able to decouple the use case from the specification is actually a pretty big thing that we need to achieve. And showing that it's done in the open is a good way of saying, OK, hey, you really want to run this fixed wireless network here is 80% of the standard that you don't need to worry about. And then, you know, showing that as a reference implementation and trying to influence the industry to move that forward is actually a pretty big thing. And I think we should probably continue to push on that as well. Yeah, that's another one that I think is is a that that becomes one of those those barriers to kind of breaking into influencing the direction similar to to spectrum policy and other things like that where it if you want to try to come in and and do something innovative or or real really different than what this what has been happening then then there are some of these roadblocks. I think I think what's excited about this is if you zoom out just a little bit, which is there are massive technology trends. There is a market there are new applications. We've talked about all of them. But I mean, I almost feel like doing a call to arms, which is it's hard to find another sector or another area where the problem domain is this rich. If you actually look at like what Omar is doing with magma, it touches tech, economics, regulation, social issues, operational logistics issues. I mean, like that, like all of these things are required in order to like do deployments. And so, you know, often tech is like this kind of 2D chess game, you know, like with tech pieces and then maybe tech and economics is a 3D chess game and like whatever. I mean, I think that this is kind of as kind of complex and interesting as rich as problem spaces I've ever seen in my entire life. And so, you know, for those that are listening that are interested, I do recommend them actually looking into this broad problem of, you know, connectivity. What does it mean going forward? What are the convergence? What's happening? What is magma and so forth? Because again, like we're at this like, you know, inflection point now. And we're seeing, I think, the same type of evolution with, you know, last mile connectivity that we saw in the PC industry, you know, back in the 80s. And so it really is quite interesting. Yeah, you mentioned social issues and in addition to getting involved in kind of, you know, these technologies and open source communities, I know that several of you also work with some organizations that are trying to address some of that disparity and connectivity. Maybe Amar and Martine, if you want to shout out to some of those organizations that people can look into as well. Sure, I'll start very quickly on this. And then Amar, if you want to jump in. So so I co-founded an organization as a 501 3C. It's a nonprofit. It's called Miro Networks. And what we do is we bring cell connectivity out to indigenous areas in the United States, right? So in rural areas, less than 70% of tribal land has cell coverage. So if you actually like this, the population, from a population standpoint, in the more rural areas, less than 70% actually have connectivity. But it's, you know, like, listen, so to come and actually to provide solutions in these areas, if you just look at the social aspect which you brought up before, Jonathan, it's very complicated, right? I mean, you know, like this is, you know, new technologies that impacts people. There's a lot of benefits. You have to be sensitive to the needs. You can't just bring in and walk away, right? Kind of leave them with something that's non-functional. You have to have economic viability. You've got to do capability building. You've got, you know, they've got to be trained or learn to do what they're doing. I mean, you know, this is a very multifarious effort. And so, Mural Networks has been doing this for about five years. We've connected a lot of organizations, but there's a lot more to do. Then Amar, you should probably hop in. You've been doing a lot of work in... Well, as in Mural Networks, actually it was our first site that went up, right? So it's been our first since day one. Yeah, so I think, yeah, no, I do agree that, you know, like there are a lot of these community-centric like sort of ISPs out there that are actually, you know, bringing connectivity into like the underserved areas. And a good way to engage with that would be, you know, join the open-source movement around a lot of these efforts, including MAGMA, as well as help and evangelize a lot of what we're trying to do, which is, you know, bring connectivity and try to decentralize some of the connectivity initiatives, especially in the U.S. as well as in Latin America. All right, well, thank you all so much. This has been an awesome series of topics. Obviously, it's very, very important that we all get involved in open-source and policy in all these different ways and maybe throw some crypto in there. But before we go, I wanna just make sure that everybody knows how they can reach all of you. So I'm Sparky Collier on Twitter, but Martin, how can people connect with you if they wanna work more with you on this important stuff? I am Martin underscore Casado on Twitter. All right, and Amar, how about you? I'm reachable on the MAGMA course Slack, as well as on Twitter, I'm Amar Fed and both. Awesome, and Amar's our MAGMA expert, so you wanna get connected with him and talk a lot more about MAGMA. And Bruce, where can people find you to work with you all? Yeah, Twitter's probably the easy way or LinkedIn, but on Twitter, I'm underscore Dr. Bruce D. Awesome, and then last but not least, Jonathan, how can people find you? I'm Jay Bryce on Twitter, but I'm not on Twitter very much. So I'm so old I still use email, Jay Bryce and Jay Bryce. Oh, wow, electronic mail. Well, thank you all for coming on Open In For Live, and this has been another awesome episode. So thank you all so much. All right, well, we are back live. That was such a great conversation. I really enjoyed talking with Bruce and Martine and, of course, Amar. So maybe as we have a few minutes here, we've got some great questions coming from chat. Just to kick it off, looking back at all of the comments and what we discussed, maybe we can talk about some of the points that were raised. One of the things Bruce said that I thought was really interesting is that throughout the history of computing, you see these computing paradigm shifts happen and that always changes the network dramatically. We have to rearchitect, create all kinds of new technologies and we talked quite a bit about Magma. And so maybe Amar, you can talk a little bit about what Magma is and kind of how people can get involved in the community as an open source community that's very welcoming to contributors and what it's trying to accomplish. Yeah, definitely, thanks a lot, Mark. So yeah, Magma is a fully open source project. It's a neutral organization that has an open governance that you can find at the link below magma core.org. That's a great entry point for the conversation. Stepping back a little, Magma at the end of the day is trying to build like the universal network back end for this next generation of computing. And so as part of that, like, simplistically, you can think of it as a core that is access agnostic like it's Wi-Fi, LTE or 5G capable. But going forward, we're also starting to look at integrated access backhaul for enabling a lot of the satellite use cases and Brian Barrett who is on the technical advisory council of Magma is pretty influential in that space. And so we're trying to take a pretty clean slate approach of like, okay, what is it that the network of the next generation needs? Some of it is access agnostic. Some of it is like embracing cloud native technologies or like borrowing from the internet architectures and then other aspects are moving things to the edge. So we're trying to do all of these things. And so we'd love to get your participation as well as ideas and thoughts to sort of help move the program forward. Again, magma core.org is the best way to get engaged with the project. Oh, thanks, Samar. Jonathan, what did you think was a good thing to follow up on from the live comments or the discussion earlier? Well, we had a lot of good comments coming into the chat, especially from DRUK and he actually, I think, you know, picked up on one of the things that Bruce mentioned as well around operations being complex. And he also asked a question, which was any advice on choosing the right open platform? What stresses me out is technical debt, investing in technology, getting it embedded only to see the supporters or vendors ditch the platform. I think that's, you know, a really, really good question. Definitely something to think about, you know, whenever you're adopting these core technologies with, you know, we were talking about open stack and how that's nine of the 10 top major network operators are running open stack. And when they make those kinds of investments, those are decade or multi-decade type investments in technology. And so you definitely don't want to, want to be picking something that is, that's flashing the pan or going to disappear. And I think that the way that I think about it, there are several key elements to consider when you're looking at open technologies. For me, you know, one of the most important is the community around it, making sure that there is, you know, multiple contributors from different organizations, that there's diversity in the companies and the use cases, because that helps to ensure that there's truly a long-term market for it that where it won't just, it won't just disappear. And, you know, to Amar's point, Magma started out inside of a Facebook team that launched it and open sourced it. And in the last, you know, six months to a year, that team has spent a lot of work moving it into a fully open community like that and has garnered contribution and involvement from a number of different organizations now and expanded use cases into different geographies. So I think, you know, community is definitely one of the top elements to look for. And I think another key piece when you were thinking about these fundamental platforms and the open technology is to kind of check out how the development process itself is managed. You know, there are all kinds of ways to do open source and there are lots of, you know, lots of different approaches to how open source maintainers accept code and test code. In the Open Infrastructure Foundation, you know, we have found some approaches that we think are very robust and help to bring maturity into that software development process. And so when we are hosting projects within our foundation, we make sure that there is CI, we make sure that there's code reviews, we make sure that there are some kind of standard processes for how the code flows into the project to kind of introduce that maturity into it. And so I think, you know, the community and then the way the code gets built is really key for evaluating which projects you may or may not want to use. And you know, that's not to say there's, there are millions and millions of open source projects out there and it's very easy to pull in projects that are maintained by a single developer and sometimes those are great for specific use cases. But I'm talking about, you know, when we're thinking of this kind of multi-decade type platform, you definitely want to put a little more evaluation into it. Yeah, and I think, you know, to that point, you know, one of the things that Bruce mentioned was that, you know, let's not just blindly assume that throw open source at everything, it's gonna work out great. You know, we need to be thinking to your point about kind of how the open source has developed if you're gonna be relying on it long term. And I think, you know, kind of related to that within the Open Infra Foundation, we have a community called Open Infra Labs. If you go to openinfra-labs.org, you can learn more about it. But, you know, this is an effort to really run a production cloud, massive production cloud with a bunch of open source components, but also to share back the knowledge on telemetry. They have a telemetry working group that actually looks at what does all this software look like in a production environment? Where are the bottlenecks? What configurations work well? So other people can replicate that. Because a lot of times, I think, you know, people just grab a bunch of source code and roll it all together. They're not gonna have a great experience, right? So we have obviously a big ecosystem around some of these tools, you know, OpenStack. Obviously it has a big ecosystem around it. And now that we see, you know, projects like Magma's, you're starting to see an ecosystem grow around that. And actually on that point, I think Jonathan, you mentioned earlier that this Freedom 5 is one of the startups in the Magma space. And they announced that they're doing something cool that involves crypto and 5G and kind of like all the buzzwords. So, you know, do you wanna say another moment about the Freedom 5 announcement? Yeah, I think that this, probably one of the points that stuck with me from that was what Bruce and Martine were both saying about how it's a combination of things that you might not put together immediately. You know, you might not think, okay, blockchain and 5G and open source and you know, that's gonna create a whole. But each of those technologies is really shaped around how do we create platforms that you can arbitrarily build on top of for whatever kind of use case. And so it does make sense that, you know, we're gonna see a lot of surprising things happening in I think in that space. And I think the other piece that's exciting too is these are representing ways where we are taking technology and making it more accessible to more people for more use cases. You know, I think one of the things that's always excited me about Magma is it expands the possible use cases for this type of technology because it makes it simpler and lower cost to run that kind of network. So, you know, you wouldn't necessarily go out and try to run one of those top 10 networks in the world on Magma today. You know, maybe at some point in the future. But I think the interesting thing about it is all of the other kinds of networks that it's enabling and all of those new use cases that are going to be developed that we don't really even have any idea what they are. And I think that's part of why the FreedomFi one has really caught my attention because it's an example of that where that was not a thing that I was thinking about, not a thing that I had necessarily connected all those dots in my head. But that's what's great about platforms is, you know, you enable everybody to innovate on top of them. Yeah, and then Bruce kind of brought that up too is just like you create these platforms and when everybody has access to them and can influence them, then they end up getting used in ways you just absolutely had no ability to predict, which is actually the magic of it. And but on that point, I'm still going to put Amar on the spot and try to make a prediction here. So on Magma, I know that there's a lot of ongoing development work. And so what do you think is kind of the future of Magma? Are there any kind of features coming that are in development or exciting things that you see might come to life in Magma as we look a little bit into the future on the development side? Yeah, thanks for that, Mark. So one of the big pushes we're doing this year is to productionize the 5G essay, which is basically the standalone architecture. And one of the cool parts of the way we built Magma is that we're taking a lot of what we've built for 4G and reusing that for 5G as in because we've built the 4G stack roundup, we had the luxury of looking at 5G and building that once again. So the process of hardening is not going to be as long as it was for 4G. So folks can expect a release that is 5G essay capable, at least lab ready in July and then more production ready by September. The other thing that I just want to also plug a little bit is that there's a lot of exciting stuff happening in the space altogether. So spectrum is becoming easier. Germany has industrial spectrum, Japan has industrial spectrum, Latin is now loosening a lot of spectrum and shows the US with CBRS. The supply chain as Martin pointed out has gotten mature. So just get engaged, right? Like, so even if you're a user, you're a developer or even just an advocate, the equation of like connectivity and connected platforms, be it smart manufacturing or connecting to users is changing. And so this is a great time to get engaged in this new space because there's a lot of new and interesting things that are happening. So, well, thank you. Well, this has been a great chat. It was awesome earlier to get to talk to Bruce and Martin and get back on live with UMR. So I just wanted to say one last thing in terms of where everyone can get involved. There are lots of different projects going on and at the Open Infra Foundation, we put on Open Infra Live. And so we would encourage everybody out there to join the Open Infra Foundation. It is free for individuals. You can go to openinfra.dev slash join. We really are the foundation. We are a foundation that's focused on building communities who write software that runs a production, particularly for infrastructure and trying to work on the next decade of open infrastructure. So come do it with us, openinfra.dev slash join. And let's work on this stuff together and make the future a little bit better with open source. Thank you very much for joining us. Thank you. Thank you so much. Thanks, everybody.