The International Telecommunication Union is a one hundred and sixty year old organization which became an agency of the United Nations in nineteen forty nine. It coordinates radio spectrum, deals with figuring out communication mechanisms across international boundaries, helps to develop technical standards for communications, and today, it even helps coordinate the use of satellite orbits.
Now today, our special guest, Walid Mathlouthi, is the head of the infrastructure division of the ITU, specifically future networks and spectrum management.
Walid's experience is remarkable. It really is, as is his academic journey and his journey in the private sector before coming to the ITU. So in this episode, we'll hear Walid's story, some very interesting stories of his time with Google, with AT&T, and now with the ITU. I'm Philip Gervasi, and this is Telemetry Now.
Walid and Doug, thank you both for joining today. And, Walid, this is a a really great opportunity for me to get to meet you. I've been doing so much research for this episode today.
So fascinating, your background and some of the work that you and your organization is doing. So thank you thank you very much for, for joining. Appreciate it.
Thank you very much, Phil, for having me. It's a pleasure.
It's really an honor and a pleasure to be here, with, my friend Doug here.
I mean, it's it's really a great pleasure to be with such, such a star, in in in the telco world, in the in the, in the communication world, telecommunication world.
Oh, yeah. Doug's a treasure. He is the man who sees the Internet. Right? Isn't that your nickname, Doug, I believe?
Yeah.
He's shy.
He's shy today. He's being humble. So I do wanna know a little bit about, how you two are connected.
But before we get into that, Walid, I'd like for my own sake, because all I know about you for the most part is what I've read online in our couple small conversations that we've had. But I'm very interested in your background, especially in academia. I mean, you do have a a terminal degree. Your work in private industry, having worked for Google, and then now working for an agency of the United Nations.
This is a very interesting, journey that you've been on. I'd love to hear it. If you don't mind, and as much as you'd like to share, personal, professional, I'll leave that up to you. But starting, I guess, with your background and and even in your journey in academia.
Yeah. I I could I could talk for hours, so there's not enough limit there.
Thank you. Thank you for the opportunity.
So, yeah, I'm actually, a physics PhD.
I did, I did a PhD in the, in lasers, actually in semiconductor optical amplifiers Okay.
Back in, back in the the, I would say, like, in the early two thousands.
I started my career as, as a physics researcher.
I mean, back in the day, back in the days, there was still, in the in in the early two thousand, the discussion about, like, the Moore's law and how how long we can push it and, like, how much integration we could do.
And there was, similar to what's happening today, actually, a power crunch, at the chip level. And, Intel was really investigating, kind of like a forefront research project doing, like, silicon photonics and try to use photons instead of electrons for switching.
And so I added my PhD in Canada, in University de Laval.
And then I moved into the research laboratory, Philip photonics research laboratory, with, Mario Panitcha at that time, leading that, lab. And I remember my first time in two thousand seven, coming into Silicon Valley and my first day, at work.
There was this huge press conference. It's full of, like, journalists. I was coming in as a, you know, new grad. There was, like, these, umbrellas they put, like, for shooting and stuff.
Right. Yeah.
And I was coming right off out of Canada just finishing my PhD.
It's a different reality to join kind of an industry research laboratory, very different from university. But I did, kind of a couple of years of, research in silicon photonics. Very early on, I worked on the, first, Raman laser. So this is kind of like you excite the material, and then you get, like, a light emission.
And, Intel was able to do it in silica, which is a very, interesting part because you could leverage all the electronics industry and and fab to, you know, to use photonics instead of, electronics.
So that was very, very interesting.
But then quickly, I, I was not really sure what direction to take and what journey in my in in in my career. And I was still, like, very, attached to academia. I wanted to become, like, a professor, have my own lab. And Right. It's kind of like a dream of every student at that time.
So I I took back a position, a postdoctoral position at the, micro photonics laboratory in, in, in Toronto, in University of Toronto. Mhmm. I was very excited about it. It was very interesting opportunity.
I got the opportunity to meet great people such as, like, the later on to become Nobel Prize, professor, the the the French professor that got the Nobel Prize recently on laser manipulation of materials and stuff. It was it was really cool to to be there.
But then my my, my my my my short time in industry got me, like, a taste of Silicon Valley, and I was really, I I got the hang of it. And, you know, once you're in in the valley, like, it's very it's really hard to get yourself extracted out of it.
Very quickly, I was reached out, to by AT&T Labs.
And there's a small part of the lab, that does the, applied research that is in in, in San Ramon, in California.
And it worked very closely with the, the more famous part of the lab, which is in Holmdale in New Jersey. So that's, like, the kind of the research, and we take what what they do, and we try to apply it in the network. And, and when I joined AT&T, they just got the exclusivity of the iPhone at that time, and it was, like, only AT&T that was doing that. So they turned almost everybody into wireless engineer.
So my boss, I remember my boss coming into my office and was like, okay. We're not gonna do any fiber business. We're not gonna do, like, any photonics. We're done.
Everything is wireless now. We need to make sure that this device, this new device works.
It had an internal code name at that time when AT&T used, they called it ACME. And then we had to everybody was on ACME.
One of the things, that were was very high priority. I mean, at that time, people were very excited about, like, this new innovation, this new device, but it was very difficult to make calls. Like, all the call calls were failing because the network capacity to support all this hype was not there. Mhmm. Right. And AT&T had to, like, reinvest in its network and, and and do all the necessary.
So I became like a wireless engineer trying that kind of, like, tiger team trying to, help, the executive, you know, steer the investment and put more towers here.
It was a very a few fun facts there.
You know, in the during those years, like, the, the, San Francisco Giants, they were, like, winning the World Series. So, like, the ballpark in San Francisco was very crowded.
All the executives will was were going to watch all these games, and they couldn't even, like, share a photo on text message or something.
Oh, wow.
So they sent me there with, like, the team to, to test the network and try to build it up. And, like, there's a lot of, things talk at that time about DAS, distributed antenna systems and stuff. I did a lot of, like, really hands on cool deployment.
The first couple of years, it worked pretty well because I didn't know to I didn't know much, about baseball.
I had a hard time managing the team because everybody was trying to go there because we had, like, full access to the to the ballpark.
But after a couple of years, I was as distracted as the rest of the team. Right.
I I understood the game very quickly, and I picked it up.
So yeah. So I did a lot of, like, wireless work from there. You know, three g, 4g, circuit switch fallback, had full access to, all the logs in the network. I worked with a various bunch of, vendors, to kind of look into optimizing the network, big data, Hadoop. We had, like, all these things coming from, the devices into, like, a massive cluster, and trying to optimize and, like, find out pair events, like, such as call drops and stuff like that. It it was really fun.
I worked very closely with a startup at that time that ATT was a that's a plan that you could acquire, to put, like, a a small, you know, a small piece of software in all our devices and try to extract, radio information from the devices. Because it's good to have the network view, but it's nothing compared, understanding in what was going on at the customer's side.
The client's side. Right.
It had a lot of public implication on privacy. Mhmm.
The project got suspended at a certain time for investigation because, we had full access, and there's no limit of what we could source including the payload on the devices.
Oh, wow.
You had to trust the engineers that they were really looking only at improving the customer experience and I mean, being able to understand, like, the radio, the radio parameters there. It was a very interesting, play because, like, I had to deal with the, device manufacturers with the, the chipset, the application chipset, the radio chipset. It was a very yet nascent industry at that time, in the two thousand ten, two thousand twelve.
I mean, it's ridiculous that I'm talking about it right now. It's only ten years ago. But, like, when you see where it was ten years ago and where it is now, it's a huge massive I mean, it looks like I'm talking about prehistoric area.
It is amazing how technology, especially in in recent years, changes so fast that it does feel like ten years ago was ancient history. It it really is. I agree with you there.
And then all of a sudden, there was this disruption coming along.
Google Fiber came to town and, they were deploying fiber and, they had, like, an amazing, product. It was, like, one gigabit symmetrical.
So upload, download, one gigabit unheard of.
That was a time where, I think the national average, for, you know, for throughput was, like, around, like, one point six megabit per second, and then it was offering, like, unrestrained one gigabit upload and download.
Yeah. This is around twenty twelve or so?
Twenty twelve, twenty thirteen. Yeah.
Okay.
It was Provo, Utah, and then they extended to Kansas. And, when I joined, we were deploying Kansas, and then, we decided somehow mysteriously to jump the to Austin. That was a big mistake. But I I jumped, ship at that point, because I was really, you know, my background was optics. I was, really long into my original, you know, background of physics and lasers and and fiber, and I really like the project of Google Fiber.
So I I jumped around two thousand thirteen, and I, kind of, like, led the, network architecture piece, Google Fiber, design and actually the architecture, the, the outside plant, but also, like, the, the network architecture as well. So both at the IP level, but also at the, construction level as well. We had a lot of challenges, digging and, you know, I was talking about Austin. Like, it was after Kansas, we went to Austin, and it was, like, kind of like a flagship city.
But somehow, it was the decision was made by marketing and and business, not really consulting engineers. And Austin sits on a bedrock. It's like probably the that you would choose for, like, digging fiber. Yeah. And we had to to try to dig and, like, deploy this product that everybody wanted. Mhmm.
But it was a very interesting challenge for a network architect because I had to come up with alternative solutions. So we were investigating solutions like, free space optics. So shooting lasers from rooftops, to kind of, to condominiums, so to to a set of, like, big apartments, big number of apartments. And it made sense from a capacity and a product perspective because it was a short distance, laser shot, so reliable.
And there was enough business because we were driving we were trying to offer all this capacity at the end, that it would make sense to aggregate that through laser links.
Solutions like that, we built our own, you know, horizontal drilling, in the lab, and then we were trying to deploy it to accelerate the the digging and automate some of it. We were trying to well, a big challenge at that time that I that I took on with the team was to try to understand the infrastructure that has existed in in the underground, and map it. Right?
Because I don't know if you remember those days. Like, every every other day, there was, like, a water main hit by our construction crew, and Google Print was in the news every other day. Oh, look at what they're doing. And all these, like, you know, legacy or I would say, like, incumbent, incumbent ISPs, they they would jump right on on that event and say like, oh, you know, we were telling you, they don't know what they're doing. Mhmm. They're just a bunch of kids trying to, like, offer you Internet, but they they they don't know what they're doing. So we developed it like this kind of, artificial intelligence or machine learning, I would say, automation using different radars, to to try to sense what what was there, while the crew was going, to understand where's where are the pipes, where's, the water pipes, the gas pipes.
It was very interesting time also because, like, we had, a lot of, interesting, arguments with the incumbents when we were trying to put, like, our fiber aerial deployment in in in polls.
And there was, like, this, regulations where you had to be, like, you know, one foot under the the main, like, a couple of foot under, the main utilities and then one foot under the incumbent. And you had to wait for, like, AT&T engineers to come and sign you off so you could deploy your fiber. So we're Right. Right. Playing on the regulation with local, with local municipalities and stuff. Very, very exciting times.
You know, but unfortunately, with the evolution of, with with the deployment, that was getting slowed because of, like, different various reasons. I mean, the project, was scaled down.
I mean, in in hindsight, I don't know if it's scaled down or not because, like, the national average, for throughput was bumped by, like, I think, ten times.
And now AT&T was catching up. The incumbents were catching up. Comcast, deploying DOCSIS, very quickly. They had, like, a similar product.
They upped the they upped their game, and they were offering one gigabit per second, as well. In Austin, for example, our crew was going to, like, high end, multi dwelling units, and the AT&T crew would be would be waiting right at the door when we have a crew finishes to come in and put their own fiber. So in hindsight, like, I think it was very beneficial to Google, and to the, what they call them, like, the over the top players or something that very, very beneficial, because at that time, the Google to simplify the the the Google, business, business case was directly proportional to the amount of traffic.
The more traffic, the more ads printing, and the more revenue. Right? So it had, like, a strategic goal in really increasing that number, and that's why they created Google Fiber, I think. I mean, this is a personal opinion.
Alright?
Yeah. Understand.
For example, This is, like, few years back, like, understanding that and looking back at the business.
So it was a massive win for Google because, like, after a couple of years of existence of Google Fiber, we had, like, the national average goal, like, ten times up, and that was bringing massive benefits for, players like Google, Netflix, etcetera, for almost free.
I think the overall budget, for our deployment was around eight billion dollars.
But if you think about it, like, eight billion dollars is just a a drop in the bucket, for Billy, like, having this national Right. Relatively stable.
You know, ISP level and try to push incumbents to put more CapEx in their in their network.
AT&T was deploying fiber.
Comcast went on with DOCSIS and then DOCSIS three point one, etcetera, etcetera.
So it it it was a huge success, I think. Mhmm. Probably not for the engineers that wanted to really, like, build a full on ISP with, like, many, many other, aspects.
But there's there's still a lot of work to do. Like, believe me. I think till the day today I remember in those days, I was hanging around, like, San Francisco, like, some very touristic areas such as, Haight Ashbury, very known neighborhood, I was not able to get more than, four megabit per second. The old you know, like, it's kind of Yeah. Weird valleys and and dips in in San Francisco. And the only available, infrastructure there was ADSL, like, old copper lines. I think till the day of today, it's the same.
And then if you're in that dip and you don't have very good coverage, so your cell phone doesn't have really good connectivity.
So still even in San Francisco today, there are areas where connectivity needs major, updates, let alone, like, second tier, third tier cities all across the Yeah. The country where, I think back in the days, we were exploring things with Google, like, things like 5g related fixed wireless access Mhmm. Like, direct, millimeter wave links and stuff.
But there's still a lot of work to do, in, at at the country level.
Yeah. Yeah.
Yeah. But you were at Google for some time, for some years. What was it that, sparked an interest or at least the opportunity to move to the ITU, the International Telecommunication Unit? I mean, you went from, you know, academia to, huge, you know, company like AT&T to even you know, I don't know how big, Google is compared to AT&T necessarily, and then on to the United Nations itself. So I'd love to hear how that transition happened.
So, interestingly enough, when I joined Google, it was very small at that time. It was not big yet. And Google Cloud was really it was really like a smart small startup. I witnessed Google go, like, ten times.
I mean, at the time when I joined, it was around, fifty thousand. And then when I left there, it was around, like, a hundred fifty thousand people working for Google. It was, a massive it's pre alphabet. I lived through the alphabet transformation.
Post alphabet, crazy stuff crazy stuff.
So, yeah, the the the, Google Fiber had to put, like, a break on the, on its expansion and its, plans. So I took on a different role at the the next billing user unit in Google. I've taken more of an international role trying to find a solution how to improve connectivity worldwide and working on the global scale for Google network. And one of the ideas was to leverage the revenue that we were getting from ads to deploy actual infrastructure in underserved areas.
So subsidized Wi Fi. So in the same way you look at YouTube and, like, make the YouTube business case work by look, watch, and trading your attention for a few seconds to watch ads to generate that revenue so you can have all these infrastructure send a video to you. We will try to do the same for Wi Fi. We will trade your attention to look at an ad every now and then, and then that will hopefully pay for deploying the, access point and, like, the back hole and etcetera.
And I took on a role as, like, an architecture lead for that global network. It was also fun. We deployed fiber in Ghana, in Uganda.
We deploy I deployed Wi Fi networks in, in India on the train stations.
We were responsible for four percent of the global Internet traffic in India. We Mhmm. Scaled that to Nigeria, took that down to South, South Africa. I was deploying networks in, very dangerous slums of Africa, Guguletto, Kyle Leecha. I don't mean it sounds familiar to anyone.
I mean, there was active shooting in some places where I was deploying networks, Wi Fi networks.
And then that also got scaled down because it didn't really work.
That assumption that you could, like, subsidize, you know, take some of the revenue from the application layer, bring it down to the infrastructure, and try to make it sustainable Right. Didn't really pan out because, you needed to deploy in very dense areas for the business case to make sense.
Yeah.
You can print more ads because the ad price was going down. And that actually went against the premises that you need to go in less dense area to expand the network connectivity. That was kind of the goal.
So it it didn't really work, so they scaled down. So I had to took on take on another role, at with Google five with Google Cloud, optimizing Google data centers, the Google global network.
Did that a couple of years, but it didn't feel right. It didn't feel like I was really passionate about, like, building networks, expanding networks, have an impact to, like, the bringing Internet, to underserved areas. And working on the cloud services felt like, it was a very mature project that was going to mature markets, that and their need to compete with AWS, etcetera. It was not really my thing, even though it was interesting from, like, a technical perspective.
Sure.
I was like, hey.
How can I do how can I find impact, at at this point after all these years, in the valley, after all these different experiences?
And, one of the ideas I was, like, contemplating was, like, I I looked at what the UN was doing, and it looked really interesting. I thought, like, through the UN, we could really work with developing countries, and build skills in these countries, and build network engineers and and help them come up to speed and understand, like, how you can manage a network and what kind of regulation you need to put in place Right. Right. And how you could put it together so you can really build your own, while working with external partners.
And because, like, there was no other way to expand Internet in the underserved areas if it's not done by the local communities.
That was kind of my assumption. And, you know, naively, I assumed, okay, like, if you jump in a UN role, you have impact because you have, like, really direct access to these decision makers.
Right.
That influence the regulations, and then bring in my kind of, unusual background for the UN, like, with technical know how and being in Silicon Valley and haven't seen how these decisions are made and how the technical, aspects of the deployments, etcetera.
What is your what is your role specifically now?
So now, within ITU so ITU is the, as you mentioned, earlier, the International Data Communication Union was founded in the eighteen sixty, actually, eighteen sixty five Mhmm. When Telegram was invented.
Yeah.
They felt they needed because, they felt they needed an institution that will work between countries to kind of, you know, harmonize because Telegram was going across countries.
So they needed some sort of international bodies to, coordinate that. So starting with the Telegram, it evolved into this role, with the radio communications. So when the the radio waves were invented, Marconi in nineteen hundred.
So naturally, it found itself as the organizations that coordinated also, like, the radio, telecommunication Okay. And managing the interference, making sure that, at the international level, like, the communications are coordinated, correctly.
And then it became a UN specialized agencies in the nineteen fifty Mhmm.
Around the nineteen fifty.
And then it evolved into this role and to manage space orbits because in the sixties, the only communication from satellites was using radio frequencies and, radio transporters.
So, ITU found itself in that spot to manage the interference levels, and it inherited by that the management of the space orbits. So today, ITU, if you wanna send a satellite up in the sky, up in space, you need to file for ITU, you know, orbits and their studies, and they will look into the interviews. So it found itself as the organization that, coordinates that that as well as well. And then in the eighties, you know, they started to add layers to that. So that's the historical radio sector that manages radio frequencies and space orbits.
Then it evolved into, like, their they added another sector, for the standardization because they needed they felt like with all these now vendors and operators, they needed, like, sort of like standards to, you know, make sure that the communications are going, are are coordinated correctly across borders. Yeah. So they added the standardization, sector on top of the radio sector, the historical one. And the standardization, sector did such a great job. It it is actually, the, the body that is behind, the standards for the WDM standards, wave division multiplexing, which is the frequency grids, that that that is defined for the lasers that go in the subsea cables so that all the traffic around the world is carried in these modulated lasers. And there's a a bunch of them that go in a fiber. And the grid the exact grid on where you place those lasers and how you modulate them is defined by an ITU standard.
There's a bunch of also, like, fiber to the home standards defined by ITU. I think the standardization, sector did really, really well in being like instrumental and critical actor in, the expansion and making the Internet really, develop and grow.
Very, very interesting work that has been done that. And they're very interested in work that is still ongoing, even though now it expanded completely into, like, everything, which is I I don't know if it's really beneficial to the organization anyways. And then the later the the latest one that came, the latest sector, was at a certain time, I think, in the in the nineties with the Internet boom.
Many developing countries were witnessing this explosion of Internet and technologies that they felt left out. So they're like, hey. We need the development sector where we could, like, the ITU, from the other sectors, from the regulation and from the standardization, would come and build capacity in these developing countries and help out the developing countries make the right choices on how to leverage these technical advances to really, have economic economical growth in these countries because we need these technologies to be able to build our economies. And we need to build capacities, and we need to build, you know, personnel and technical expertise, to be able to catch up.
So that's how the development sector or ITU T or PDT, I think it's coming from, like, the French historical background. Many names in ITU, are in French. Actually, in these international negotiations where they do, French name is actually the default one for prioritizing the alphabetical order of things, which is something very strange. It takes up a couple of years to get.
Anyways, so they built this BTT sector. And so the BTT sector is the role is to take these technologies, these, ICT technologies and try to help developing countries based on their requests, understand them, grasp them, and implement them in, in their territories in order to drive economical growth, you know, and build development, and, you know, create, create, jobs, etcetera.
So I'm part of that development sector. And in the development sector, I lead a very unique, division, called, future networks and spectrum management spectrum management and future networks.
So my role is to help these countries and these decision makers and the regulators understand, what is spectrum management, how you can manage your spectrum bands, how you by being by, and this by being, in line with the international regulations, how you can build capacities.
I build tools for them that I make available for free, almost for free, to be able to understand, like, and manage, the complexity of, like, different spectrum bands, leverage them for different technologies, IMT technologies, 5g, 4g, where to allocate things, have an international role in coordinating between countries. For example, recently, in the African continent, we got fifty four countries to agree on a treaty for cross border interference levels and implement that using software that all the administration can take and use the same calculation method to understand the level of emissions.
And the future technology parts, involve, you know, understanding where's the Internet going, what what are the next technologies, how to leverage, you know, how to understand, fiber deployment in your country, how to, what best, last mile connectivity solutions you could use, how you can integrate new technologies such as direct to cell, satellite, these LEO constellations.
It could go also all the way to cover, quantum communication, quantum encryption, things like, RPKI RPKI, how to secure your network, really at the infrastructure level. I I won't I'm not venturing too much into the application layer of the OSI model. Right. We do a little bit of TPI right now, digital public infrastructure, which is a big thing as part of, like, the push at the high UN level, to have this kind of, digital public infrastructure for countries, that could you could put, like, building blocks on top of it for, like, digital identity, digital health, digital, digital services for for the governments.
So I had that division that is really at the cross border of, like, the different sectors because I need to interact a lot with the historical radio sector, with the standardization sector, distill all that, bring it to the member states, and try to advise them and build products for them and services for them in a way that they can, that they can leverage technological advances in the ICT technologies and build, and build growth. And a lot of that is also is building yeah. A lot of that is building, capacity building curriculum. So I do a lot of, like, now a little bit going closing the circle, going back to my academia days. I build courses, about infrastructure mapping, like, how does the Internet work, about space connectivity, where is, how does this things how does do this thing one on one, satellite connectivity, And I put that on the, IT academy, and I tried to make them available, for everyone, who's interested in taking these courses.
I wanted to ask a question. You brought up a you brought up a number of topics, but one that, you mentioned the spectrum management as it relates to satellite communications.
So this is a, a topic that is top of mind a lot. A lot of people think about SpaceX and Starlink is a a very famous technology now. And, even though satellite Internet has been around for decades, you know, there are reasons why the the low Earth orbit, constellation, technology is really, cool.
One thing that I I encounter, maybe every other day is somebody saying, well, like, why can't Starlink work in country x? And and there's and and there's really there's not a technical reason.
The satellites are essentially covering the whole, you know, surface of the Earth as far as where any people are, and, and then it becomes an issue of, the spectrum, as managed through the ITU.
And then, again, depending on the country, there may be a license that's required through the telecom regulator of that country. So you could have two, like, each there's different configurations here, but I I find that people are not familiar with this obstacle of, that they can't just turn the service on whether it's Starlink or Eutelsat or Viasat or any other, you know, satellite operator, they while they while technically, they could radiate to a ground terminal, anywhere in a footprint of one of their satellites, in some locations, they haven't satisfied these other, requirements of the spectrum management or the license, and therefore, to honor that, they're supposed to not. Sometimes that doesn't completely happen, you know.
But, anyway, do you and do you count the people don't appreciate the role here? Because I I feel like, I'm doing my part, in trying to let people know that there's there's all these other steps that have to, occur to connect.
It's a huge can of worms, Ben. Like, you started, like, a a huge discussion here. Like, there are multiple layers to it.
Many, many, many layers to it.
Trying to kind of, like, take a step back to where where where it could start. So, the demand is massive.
Like you, every other day, I have questions from member states, from regulators in countries who has who have no clue about the different implications of all this.
It's really disruptive technology, coming up.
Broadband satellite, with all with this massive, kind of constellation, is really a new thing. And, very few, regulators and very few, actually, operators are equipped, with the right understanding and the right skills to to really make up make the right decisions.
And now, Technica, as you mentioned, like, you can bracket, the earth and then you can have service anywhere you want. But, like, there are many applications. We can't, like, offer telecommunication services anywhere. Like, we can't I mean, I think I remember back in the days when I was, doing testing with AT&T, I used to have, like, a Yagi antenna.
And if I wanted to break down a a a a cell site, I would like shoot, like, a a two point 5gigahertz from, like, a little generator, shoot it straight up at the antenna, and then bring it down, bring the whole service down. No one would be able to make a single call. So interference is a major thing. And then there are operators in all these markets that invested a lot of CapEx and a lot of, like, money based on the business case to apply for licenses and, like, be able to operate at these frequencies.
So I can't just show up and, like, start, like, my service and say, like, hey. I can give you, like, a better deal. That's not how it works.
So the construct of this is that at the international level, the ITU sets the rule of, like, frequency bands and which which, like, which services are used for which bands.
Now it's at the local level, like the government or the regulator and the national regulator decides how to allocate those those bands in accordance with international radios. So there are different different ways. So if you wanna run, like, I think for SpaceX, for example, if you wanna run a service in the United States, you have to apply to the FCC. The FCC will make sure that, whatever you applied for is aligned with the IT regulation or will do, some application for you at the FCC level if you wanna, like, own some orbits or also have some resources or, like, operate certain frequencies.
So that's kind of the play between the local and and and the, and the international role. So applying for ITU does not give you the right to go and, like, provide service at at a at a certain territory. You have to also work out your way through the national administration as well as, like, the, the jurisdiction to manage its own resources as well, in accordance with international regulations.
So we have instances, for example, where, some countries, like, for example, Iran went to the radio the radio board, IRB, with an ITU to complain about, like, Starlink provided access there.
And then I was gonna ask about this case.
And, of course, like, you know, for the regulations and the international coordination rule, ITU notified the national administration. You you have to understand. ITU came from, like, the sixties where, like, you know, in the space industry, it was like countries. There was no commercial operators. So the administrations were responsible for the damage that you could do by sending your satellite, if it creates, like, physical damage or, like, also, like, interference. So they notify the administration, and the administration needs to make sure that it's abiding by the international treaties.
Let me can I on this on this case, if I could back up? So what for people who don't know what we're talking about here, this goes back to the fall of twenty twenty two, when there were the protests in Iran about the young woman that was killed in in police custody for wearing her hijab incorrectly.
There were many days of Internet shutdowns, Internet curfew. We were involved in reporting on this as well. And as happens with every one of these cases on social media, especially on Twitter x, there's people tweeting at Elon Musk saying, use Starlink to turn the Internet back on in this country or this other country.
And I think he loves that. And, and he was like, well, maybe we would just might do that. You know, he wants to respond to this. And in this case, and I've seen this because I cover this, so I've probably seen I'd seen it about twenty times already, where some you know, there's a big campaign to kinda, enlist, Starlink, its return of service on. Well, they supposedly did it in Iran, where they actually turned it on without, going through the ITU to get the speech spectrum, authorization without going through the Iranian government, to get, you know, a license or whatever is required to provide telecommunication services in that country. They turned it on so that, you know, technically is flouting whether we call this international law, this treaty based, you know, system.
So Iran then takes, you know, find like, on their own, they collect evidence. I guess they get their own terminal. They they test that it works. They take the evidence and bring it to the ITU to complain.
And the ITU essentially said, correct me if I've got any of these details wrong. I think they kinda just said, okay. We'll, we'll accept your evidence, and we're just gonna convey the message on to I think it's Norway and the United States, are the two governments that represent, SpaceX internationally, that you should you should have them turn the service off in Iran. I don't know that they've done anything. It's, there is no enforcement mechanism, in this. We don't have a one world government, you know.
I guess the one thing that SpaceX Starlink I would think from my perspective, again, I have a very simple understanding of this, was that, you know, the spectrum allocations that they have are valuable to them, and there are those allocations only exist as long as the ITU's spectrum management is the rule of law. So the more they wanna upset that apple cart, they've got a lot to lose if everybody just decides, we'll just all do the all do our own thing.
So it's a risky thing, and I don't know how I under I I, you know, you it's public day public, publicly written, like, what the decision was. I'm not sure what action Starlink, took or didn't take.
Do you know? Did they did they turn turn the service off?
I I think it changed their terminals, and they were not offering the service anymore.
Okay. So they're they're complying with the decision of the ITU. That's interesting.
You know, like, that's so I wanna make a message very clear here. Like, ITU is not enforcing, tracking down or a ITU does not have a position of, like, supporting, you know, Internet tracking down bringing down Internet in in countries. We are the ITU, it plays a role just to make sure that things stay coordinated at the international level. The Internet or the international connectivity, is not based on one operator. It requires, like, a bunch of operators similar to how the Internet was built. So we need that coordination, and we need to work out through the also the local administration.
And we need to have things, you know, coordinated internationally. And that's what the ITU role is.
But, like, we're not ITU is not responsible for the decision of a country of, like, to turn off our a service or turn off, like, a certain, you know, connectivity.
We're we're not part of that discussion. We're at the technical level.
Yeah. I've seen I've seen some, criticism along those lines, and the you know, but I mean, I guess I guess the the concern I see is if you were to then start dictating to, you either have to kinda represent, faithfully represent the the sovereign government of Iran or you're you're in a position where you're telling them they they can't do something. And the more you start telling these countries they can't do something, then they can just leave. You know, they don't have to like, you you you risk the thing just coming apart. You'd rather have them at the table. It's like the UN.
North Korea has a table at the UN. It's better that they're there, than they're not there, so that we can have a discussion of the to resolve something, or, you know, this is, I think our understanding of international relations. But, yeah, the more you you you have you have a balance you have to strike there where you want them to be represented, so that they, stay engaged.
That's very, very political. Because this is there's like a mechanism. This goes like a radio resolution board, the RB. The radio resolution board is voted. So it's in certain countries. They they vote, and there are certain countries there that sit there to take these decisions.
And then it gets, you know, we voted every four years, I think.
So there's a whole, you know, process and administrative process based on, like, coordination between all these countries. And it's very, very political, very, very, very, very heavy, very slow. And as you mentioned, there's no enforcement mechanism.
You know? We did that. There's a decision, but, like, we the ITU does not enforce it. ITU gives, like, hey. This is a decision at the international level, and this is the what has been agreed by the member states.
And then you have to look into ways of how how to, implement that, but they don't force anyone to implement it.
But there's a cost as you mentioned. If you don't, then that means that you're not part of the treaty or you don't really recognize that treaty and then it gets really complicated.
One thing that I really wanted to finish on that, that you brought up. I think you're also involved in that, quite a bit, Doug.
The space the satellite connectivity is all about terrestrial network.
And the challenge for also for STARLINK is that, it is not clear how the traffic gets routed, at the terrestrial level and where where are the, you know, the the terrestrial stations where the traffic goes to to space.
In many of the developing countries, there is no, there's no, you know, terrestrial station per country. So it goes across borders.
So one so taking a step back, one of the challenges when I was deploying, like, these Wi Fi networks, with Google, and they were open Wi Fi networks. You have you all you had to do to connect is to watch this, this ad and then you you're connected.
You would be surprised how many malicious actors would take on this connectivity and try to do bad things.
One of the main driver for the traffic in India, and sorry for being this, this direct was like, pornography.
And some of it was juvenile pornography. Like, some people will come to, you know, the the, to, the train platforms and will upload content and will start doing, like, these things. And we needed to track these things because in in India and in most of the countries around the world, that was not okay. And it should not be okay.
We should be able to. So we needed to implement at the network level a way of, enforcing, legal intercept. So there was, let's say, subpoena from the judge that was making sense. We're not here to enter in the judicial part, but, like, if there there's a judge who judge that this activity is not legal, then with the subpoena, we need to be able to, like, track the the the logs and give them to law enforcement.
And in that year, we were doing, like, a dump of all the Nat Biden records and, the stuff, and we were giving it to the government. I mean, India is a particular country to operate, like, networks in. But, like, we were doing our part to, like, give all the logs that we're able to help you to stitch in in in in a case of a malicious thing to stitch the connectivity and retrace it back to the device or to to the MAC address or etcetera. Right?
Now if we take that same, challenge and we, we extrapolate that now to, like, the LEO connectivity.
Now the problem a lot of the concerns in these countries and even in many countries, they don't even are not even aware of these challenges is that when something bad happens with these networks, and let's say, like, terrorists or, like, malicious actors are leveraging these networks for certain activities, the local administration should be able to kind of, like, retrace back, based on, like, their judicial, you know, process, etcetera. The problem is that this spans over borders. And in most of the cases, there's not even a terrestrial station where the traffic is exchanged. So it goes immediately, like, I don't know, traffic in Botswana will go and be exchanged in Kenya. So there's no way no no way for you to really be able to so it's not a traditional thing.
So there are discussions right now with the regulators to see how the best how what what is the best practice to regulate this? Should we enforce these operators to have, like, local, terrestrial station in country where the network is where the traffic is exchanged so we can at least have some sort of control in case we need to?
I mean, the problem is not only in developing countries. It's more exacerbated in developing countries because they don't have, like, the technical skills to look into this. But even in the United States, I mean, recently, I I I'm not sure, like, how the FCC is enforcing, whatever is starting connectivity because, like, the traffic could be routed, I don't know, in Colombia or another place for, like, national traffic school. And and now when I was having discussions with a regulator, the FCC, they're not really aware. There was no clear mechanism how to enforce these things.
Yeah. You know?
Well, the the US, lawful intercept system was completely hacked recently, so maybe it's for the best. I think I they probably assume it's all coming down back in the United States and probably you know, that's that's probably a good assumption for but it's not guaranteed. It's not guaranteed.
It's not guaranteed. And and I I know some of this because, like, the terrestrial network of Google Cloud and Google is the one actually powering up Stonic.
So Stonic is taking a ride on Google Cloud and Google infrastructure. That's not public information. I'm not sure, if I'm allowed actually to share it, but probably if you dig in the Internet here, you'll find some information about it.
The problem is that the footprint of the Google Cloud, it's like a cloud operator. So it's it's designed not to have, like, coverage in every country. It's not an ISP.
Right? But that that information is hidden. So if you present a request for stunning, and this based on my knowledge five years ago, so I I don't know if it's still the the case. But if you present a a a a a request for Starlink to understand how the traffic was routed, Starlink would go to Google and ask for Google logs.
So you you are supposed to know that you're presented a request not to Starlink, but actually to Google who's a Tiresio operator of, like, empowering the, the Starlink.
So you see, like, there's a lot of complications in this, and it's not really very well regulated.
It's still, like, the fastest. And we need really I mean, network engineers should come in and, like, sit together and, like, try to find the best way how to, kind of have, get this sorted out, like, in the old days of the Internet where we're trying to, like, route traffic through BGP and, like, have, like, these things exchange and the information exchange from switches to switches, etcetera.
I have a look at it, from a a technical background. I don't wanna add the political layer where it's in a case of a war and stuff like that. That that part, I don't wanna get involved in.
Yeah. And I'll I'll I'll I'll I'll stop here. The last part also, like, it's the commercial part. Right? I still don't see a direct sustainability path for operators like Starlink, because when you come in in countries, the areas where the ARPU is making sense are the dense areas where already operators are invested massively CapEx to offer 5gs and stuff like that and buy very expensive, licenses.
Now in the less denser areas, the business case is really hard to make sense. And that's why we have three billion people that are not connected today because it's really hard to make money when you build the the infrastructure there.
Right?
I mean, the saddle the LEO constellation is even more expensive from a from an infrastructure.
And they have to replace the satellites every couple of years. I mean, they they don't they don't have long lifelines. Not nothing lifetime. You can't have something in low Earth orbit that lasts that long. Eventually, it comes back in.
Well, it's not clear to me that till today, maybe I'm not part of, like, this hype, but it's not clear to me how it's gonna make sense. I mean, like, deep pockets could make it sense. They're probably at scale when you have, like, I don't know, maybe hundred million users in the network that will start making sense.
But I I'm still struggling to see it. I I would like to see the numbers. And I run the numbers multiple times. And, I mean, when the regulators come to me and ask me, I'm like, this is not a silver bullet. We can build, like, maybe a resilient network, a combination between terrestrial wireless, you know, and satellite connectivity where everything comes together, and then you can have a fallback mechanism between the different technologies to make it work seamlessly. I think the iPhone recently announced that they will be supporting the satellite direct to satellite connectivity even though it's still not standardized at the ITU level. It's not regulated.
It's coming up, I think, in WRC Right.
Understood.
Two thousand thirty.
But eventually, it may make sense. But for now, it's like just tee pockets. Yep. Right. You know?
Yeah. That that must be my personal idea. Okay? It's not an IT opening. I don't wanna get into trouble.
So as we're getting closer to wrapping up, Walid, I'd I'd like an update on the progress, maybe even some of the challenges that you're encountering with the giga project, which is a little bit of a departure from what we're talking about with, you know, low earth orbit satellites and things, and we're talking about how the ITU is involved in, enabling and serving underserved areas with Internet connectivity. So, yeah, give us an update on the, the Giga project, if you would.
Alright. So the Giga project is a venture between two UN institutions, UN agencies, ITU, which is the Mhmm. Telco, international telecommunication, and UNICEF.
So we noticed that, in many countries, like, I think the official, recent number that was published by IT was two point four billion people that were not connected, absolutely not connected, so they're not online.
And part of that, like, a big part of that is, actually children.
So it's it's a huge problem to, understand, how we can bring connectivity to these schools, mostly in the rural areas. And, UNICEF and ITU combined efforts and came up with this, Kinga project to try to bring connectivity, to, schools all around the world.
The first part of that is to end and to understand, and, quantify the problem.
So Mhmm. We started by trying to understand where are the schools.
And as surprisingly enough, it's a question as simple as that. It's very complicated to answer. Many of the countries do not have a grasp of the location and the exact whereabouts of the schools. So we started by, UNICEF, started working by, like, identifying the lat long and the location of the schools and if the school was still operating or not.
Once we we are in a good understanding of that part, we need you to understand if the school is connected or not, which is even a more complicated question to, answer.
We were working with, like, some, you know, maybe Doug is familiar with these guys, mLab, and, like, some sort of, like, trying to collect telemetry and understand the state of connectivity in these schools. So it's a very hard question to answer to.
Is the school connected or not? What is the level of connectivity if the school is connected? Is there electricity or not? And now we're slowly getting into, like, trying to understand, like, the infrastructure that is available near those schools. And that's the role of IT. And that's the role of my team, as we are responsible for mapping the infrastructure, the ICD infrastructure around the world. And we collected a lot of information over the years, over the past ten years about, infrastructure availability, all around the world, terrestrial fiber, wireless connectivity, the towers, etcetera, etcetera.
So then ITU comes in and try to understand how far these schools are from the existing infrastructure.
Let's assess the the location, the level of connectivity, and then what infrastructure is available. We map all that, and then we run through, like, a standard, exercise of network planning, network design, and, business planning around that. So my team is making tools available into, like, to, actually design the gap based on the understanding. If we know the location, we know the existing infrastructure, we know the, streets, the street map and the street infrastructure, and we try to route some sort of, like, the best network and the less, the less expensive CapEx.
Should we do, like, direct, direct wireless connectivity? Should we do fiber connectivity? And we collect data to put, like, kind of a business plan around it.
And then we try to find the investors and the government, procurement players to support them. And UNICEF plays mostly that role because they're massive, in their presence with the countries, and they're massive in helping countries procure solutions, including vaccine for kids. In this case, we help them, like, procure network equipment, with working with countries and investors.
So the role of ITU and my team is to bring derisk that barrier and make it more grounded into, like, more of a network engineering, exercise by doing design and planning and, business, business estimations.
This is the stuff I used to do with Google and with Google Fiber as well, county and greenfield and stuff. Well, I mean, it's very challenging, to really get this to expand.
You know, IT was a hundred sixty years institution.
It's really entrenched in its inertia and its massive processes.
You add to that UNICEF, which has, like, another layer of really crazy administrative processes, and you try to bring those two together into this agile kind of, like, project that's which is giga, man.
Yeah. It's a noble endeavor, but the, but frustrating to actually carry out with all of the bureaucracy and, like you said, the inertia. These two behemoths coming together and the inertia of that is very difficult to overcome. But, but certainly, Doug, Doug, I can't speak for you, but I'm gonna speak for you. We're we're both very glad that you're there and doing the work that you're doing.
I'm I'm trying, man. I'm trying.
Well, Walid, it's been a it's been a pleasure, speaking with you today, getting to know you a lot more. Learning about the ITU and your work there, of course, very fascinating, very informative. Doug, do you have any, closing comments or questions for Walid?
No. Thanks for coming on. It was a good discussion.
Yeah. And, we rely on your help, Doug and, guys.
I mean, there's no way, I mean, we could do this alone. I mean, we need, like, private, players in private sector, and we desperately need more, you know, partnerships and collaboration.
Yeah. The UN is just the tip of the iceberg of this. I think for me, I I remain that a firm believer that it's the private sector, with the right incentives and, with the UN trying to reduce the barriers to access in, these markets and, you know, deregulate safely, most of it, to allow, kind of a marketplace, a vibrant marketplace. That's the only way we could solve the problems, and, you know, have universal, meaningful connectivity.
Eventually, that's kind of the goal for, like, twenty thirty.
So yeah. Yeah. I appreciate, the help and the partnership, guys. Really, Doug.
Mhmm. Our pleasure.
So, for our audience, if you have an idea for, an episode of Telemetry Now or you'd like to be a guest on the show, love to hear from you. You can reach out at telemetrynow@kentik.com. So for now, thanks so much for listening. Bye bye.