Eddy Barker and the Birth of DSL

When we started looking at this, we didn't initially have the vision that this really changed the world. We started out trying to solve a problem and realized the solution would change the way that we live and work.

This is the Techsploder podcast, episode two for Friday, May 10th, 2024. Eddy Barker and the creation of DSL. This episode of the Techsploder podcast could not happen without the financial support of our wonderful patrons at patreon.com/JasonHowell. If you like what you hear, head on over and support us directly. And thank you for making independent podcasting possible. Hello, and welcome to the Techsploder podcast.

I'm Jason Howell. It is so good to be back. It is so good to hear from folks who watched or listened to last week's episode with Tom Merritt. Man, I'm so happy that it was received so well. People really connected with the conversation.

That's really what I'm looking for here. The Techsploder podcast is really meant to be a chat, like a casual conversation with people who live and breathe technology. It's also an opportunity for all of us to find some common ground about what we find makes technology so magnetic to us. In the case of today's guest, what that magnetism to technology can actually lead us to create and to innovate. We've got an amazing guest coming up, but I'm going to get to that in a moment.

Before we get into the show, a few bits of housekeeping. First of all, like I said, this is, of course, episode two, which means we're very early into the Techsploder podcast experiment. So at this stage, especially, your help goes a long way to make sure people discover the podcast, and then hopefully they subscribe to it. So please, if you are enjoying the podcast so far, give us a review on Apple Podcasts, share it out to your friends, or review it really anywhere. It doesn't have to just be Apple Podcasts.

If there's a review mechanism wherever you happen to get the podcast, drop us a review. It really does help us out. If you want to support us directly, you certainly can. Patreon.com/JasonHowell, that actually supports the production of this show. It also gets your name read at the top, like John Garrison, who's been incredibly supportive in my transition into independent content creation.

So thank you, John, and thank you to everyone who contributes at Patreon.com/JasonHowell. Okay, let's get into it. Today's guest is not likely to be someone you've been listening to in the podcast world or even reading their news reports on the tech news sites.

Now, this is a little bit different today's episode. Sometimes I have the opportunity to take a look into the past and talk with people who literally shaped the technology that many of us probably take for granted today. And I'm certain that there are tech fans who have wondered how we got to this moment in time when high-speed internet is incredibly abundant and very speedy, actually, as well. In order to get here, I don't know if you had this experience.

I certainly did. We had to muck our way through the dial-up years. And on the other side of that muck was a little technology you may have had some 30 years ago called DSL, or digital subscriber line. Now, it really took the paradigm that people were used to, that paradigm being the use of the phone line for transferring data from one network to another, be it BBSes or the internet, whatever.

And it added speed and flexibility to it. Now, I was a child who had his parents often picking up the phone while I was downloading a game from a BBS, a bulletin board system with my Commodore 64, using a dial-up modem. And so DSL technology actually also prevented that. It prevented anyone from having to hear those dial-up tones screeching in their ears anymore. Well, that is obviously until now, because there you go.

I don't know. There's something that's pretty nostalgic about those noises now with some distance. Anyway, today, I am thrilled to have the opportunity to talk with someone who is critical to the development of broadband technology back in the mid-90s. Eddy Barker, AT&T's AVP of Mobility and Access Architecture, is here with me to chat all about his involvement in the development of DSL technology and so much more. Eddy, it is a pleasure having you here today.

Thank you, Jason. Thanks for having me and for having me on Techsploder.

So let's talk a little bit about technology history. I know I have my own personal memories of DSL back when I was younger. Let's start a little bit before that. Tell me a little bit about how you started your time with AT&T. Was it actually developing DSL? Was that your first project or was there something prior? How did that all start?

So really, I started as a child just mystified by communications, Jason. So telephone and radio and television, and I just wanted to know how did these things work? And I'd kind of done some building crystal radios when I was young and audio filters in my teens. So I went to college and majored in electrical engineering. And specifically, a portion of that was analog and digital communications.

But while I was in college, I also started to work with ARPANET. And that's what ultimately became commercialized as the internet using TCP/IP. So at that time, though, it was entirely text-based.

So it wasn't glamorous like we see today with all the graphics and videos and things like that. So I was interested at that time in working for a telephone company. And due to the rich history of AT&T and its innovations, and I'll mention a couple of those, like Claude Shannon developed digital information theory, and Doug Ring developed the concept of cellular communications, and William Shockley invented the transistor. And all those things were invented, interestingly, in 1947 in sort of the post-World War II technology renaissance.

So a lot of rich history there. So I started with Southwestern Bell in 91. But little did I know at the time of the massive changes that were about to happen. And that was based on many different technology developments that were just about ready to collide as this came together. So those developments were things like inexpensive personal computers in people's homes for the first time.

Windows 3.1 was developed. And you look at Netscape, that was sort of this awesome browser of its day that made graphics appear on your screen and not text. And ultimately, digital subscriber line, which allowed us to have a better-than-dial-up experience. And then all the applications that really took advantage of those new ways to communicate and change people's lives.

Yeah. It really was a moment of innovation in all of these different directions. And when that happens, it's exciting, but it's also very complicated to kind of organize around that. Can the technology keep up from the perspective of the networks that are supporting it? And like you said, graphical internet coming on board had a higher demand for bandwidth.

So many things happening all at once. And I actually really recall how slow dial-up could be. Then the internet came along with Netscape and that graphical interface and experience.

That all needs a better infrastructure to support it. So I know as a user that I experienced pain points, but I'm sure AT&T being on the other side of that equation experienced its own pain points as well. It's AT&T's phone system being used for all this data. Talk a little bit about the challenges from AT&T's perspective with dial-up that really, I'm sure, led to the development of DSL.

And you really nailed that because when we started looking at this, we didn't initially have the vision that this really changed the world. We started out trying to solve a problem and realized the solution would change the way that we live and work. So our initial challenge was really associated with the long durations that people would connect to the internet over dial-up modems. And our problems were that our voice switches that were designed for connecting people's phone conversations for short conversations with the busy times of the year being Mother's Day and things like that.

So we had a problem that we were having people connect and staying connected for days and nights, and we had to make some decisions on how to deal with that. So DSL came about and we worked on commercializing it and standardizing it. And so what digital subscriber line allowed us to do was to nail up an always-on internet connection so you didn't have to connect and wait for long periods of times. And that internet connection worked on the same phone line at the time.

It was already mass-deployed for many, many years. And then it worked at frequencies over top of your voice conversation. So on the same line now, you could have always-on internet, and then you could pick it up and have dial-tone as well.

So we did field trials with large corporations back in the mid-90s, and those trials were looking at this high-speed internet access and our first virtual private network technology for work at home using secure clients that we could put on windows at the time. And really, the rest is history on sort of how that evolved.

Where did the idea of using the existing telephone lines for that separate stream of data come from? I mean, granted, at that point, we were using our existing lines for dial-up.

So that was kind of a paradigm that I suppose we're kind of already enjoying. Was there ever a talk of building out a separate connected network? When I think about that, that's a pretty gargantuan task, especially at that time in history. But where did that idea come from?

Well, that's interesting because if you look at the cost to build a utility, most of the cost isn't the networks connecting the United States and internationally. Most of your investment's that last mile to connect to the customers' homes and businesses.

And interestingly, so you're incented economically to try to use what you have if it will meet your needs and requirements. So about the same time that we were exploring DSL, we were also testing some of the first fiber-to-the-prim technology that was available at that time. It was very, very early, but we looked at it and that would be a huge investment to replace all the copper that we had connected to all our customer base.

And these are millions and millions of customer connections. So we put that on the back burner, still wanting to look at that in the future as really the next revolution that we had to go through. But the fact that this breakthrough in technology that the gentleman, John Cioffi, that his team first developed some of the capability to make DSL work through a new modulation capability, he was AT&T in prior years. And so this technology was maturing. And ultimately, I think when we finally commercialized that and made the investment, the decision, we were going to cover about 70 million lines using that existing copper in the ground to get this technology out between like 99 in 2002. So the fact it was already built allowed us to really go quickly and to roll that out amazingly fast. Yeah.

And when faced with that versus the idea of building out separate new infrastructure, it kind of seems like it's a no-brainer at that point. Did it take long to validate the concept? Did you face any sort of resistance along the way to the idea of this working and being, I guess, accepted and adopted at a wide scale? Or was AT&T just eager to kind of get this out as fast as possible, knowing that people on the other end were going to understand what the value of that is and be able to enjoy it?

We did have challenges at first just on scaling the technology like anything, right? The first things we got were on breadboards with soldered connections, and we had to make them work. And we did field trials with it, and it worked great. But obviously, what we were able to deploy then was just there were a lot of different proprietary implementations that didn't work with one another with different suppliers, different chipsets. So we standardized it in ANSI and ITU for a global standard, and that helps drive down the cost. And then we had several companies that we worked with to really scale it.

And at the time, we actually partnered it with ATM, Asynchronous Transfer Mode, is sort of the switching and transport side of it. And we mass deployed it. And when we did trials, we knew this was going to be popular. We knew it solved our problem. But as we said, about that time, you now had eBay, you had Amazon, you had all of these new technologies that were cropping up.

Ultimately, when you have something that sort of sells itself, so you have word of mouth marketing, it's a great product project to work on. And even if you look at today, and we've evolved a lot, back then, people when they moved even wanted to make sure they could get these things where they were moving to because it allowed them to have so much more flexibility in their lives.

Yeah, it's interesting when you were listing kind of some of the companies and the technologies that would immediately benefit from this eBay, some of those other companies of the time, the name that appeared in my head near the top of the list, possibly because I was the age that I was when it happened was Napster. And this whole element of the internet that suddenly, this trading of media, and I guess, peer-to-peer sharing, peer-to-peer file sharing, just in general, it couldn't have happened nearly at the scale that it did if we were locked to 56k dial-up internet, for better or for worse. I guess it depends on the side of that perspective that you said, but that's interesting.

Yeah. And even if you look back now, look at Microsoft when they would do a new OS download, think how long that would take over, or if it would even be possible over dial-up. So it just changed everything.

And I think the velocity of technology adoption for the industry. So it was an exciting time. And we've certainly now evolved beyond that. I don't think we had that revolution and then a lot of evolution. And then obviously, we added internet connectivity to smartphone adoption and that started happening. So we've certainly continued to evolve, at least until the next revolution.

Well, yeah. And you've kind of hinted a little bit at that as well. All of that work really led to an ecosystem where speeds are just crazy fast and capable by comparison, right? So specifically, you work closely with AT&T's fiber deployment now.

Maybe talk about some of the common challenges. Something that keeps coming up for me in the last handful of months as I think about technology is the more things change, the more things stay the same, right? And the technology here with fiber, no doubt, it's so far and away more advanced than what we were seeing with DSL or dial-up or those other technologies that led to it. But I'm sure there are some similar challenges that you face between these technologies some 30 years later. What might those be?

That's a great question because when I look at DSL, other than just the challenges to get it deployed and economics and the technology, one of the little issues that I recall very clearly is when we first started deploying it and going into customers' homes, what we would literally do at the time is we would, first of all, have to put in an Ethernet card. They didn't have them at the time, if you think about that sort of standard today, right? So we would put in an Ethernet card. It's probably 10 megabits per second, maybe 100 later on megabits per second. And then one of the things that we would test when we were there was we would do some speed downloads. And a lot of times, at least in the early years, the bottleneck was their PC. The hard drives did not have a fast enough read and write to download video and images.

So carry that forward to today. And as we've gotten into faster and faster broadband and specifically fiber to the prem, we are going to continue to hit those bottleneck points, whether it's all of the huge pace of Wi-Fi development in terms of as we go from Wi-Fi 5, 6, and now we're at Wi-Fi 7 with these faster channels. The Wi-Fi, if you're not too far off or going through too many obstacles, is as fast as the Ethernet connectivity. But we're also now going from one gig. We've gotten to a gigabit using Cat 5 cabling. Right now, as we go higher than that, we're getting into newer types of Ethernet in addition to Wi-Fi, and that's driving new types of cabling.

And that can be confusing for customers, right? Well, what is Cat 6? What is Cat 6a? What is Cat 7?

What do I have that I'm wiring with? And so I think as we go through these points of time, we go through these stair steps of how technology affects people and what's in place at the time, not just in our plant, but even in customers' homes. So we have to look at that really carefully to see if it's ready and to see what the adoption is and what the application. But I think anytime we do this, though, it just opens up a world of new content creativity that allows them to be creative, essentially, and create new businesses.

Yeah. And especially at a time right now where maybe the buzzword has cooled a little bit, but the metaverse, this idea of these virtual worlds. Obviously, artificial intelligence is huge right now. I kind of see those technologies at some point converging to a degree to where these kinds of experiences become much more desired and much more pursued by people on the other end, which does require fatter pipes, the ability to transfer all of this data in order to support that. And kind of along those lines, where if you had to use your crystal ball and look forward in the next 10 years, obviously not revealing anything that you can't reveal because you still work with AT&T. But from a broad spectrum, what do you see all of these technologies enabling? And what are you excited about for the next 10 years from this perspective?

So when I step back, and fortunately, I'm in a position where I get to work on a lot of new things, right? And in a lot of that, I get to work on things that may not even see adoption or development for years out. And these are things like we do at the international standards bodies, where we work with engineers from all over the world. And everybody brings in their ideas. And we want to, as I mentioned earlier, we want to all adopt something similar so that we can get economies of scale, right? And make what we deploy inexpensive, where we can get it to all walks of life and geographies and income levels.

That's really the goal. But when I look at technology, I'll say speeds are important. And today, we've just had a huge growth in the speeds we offer. But really more critical than speed is always being connected, latency, you know, optimizing that for all the applications that we need to use and capacity is the big one.

Because speed is important. But really, what's most critical is to make sure that during those busy times of the day, you know, the Mother's Day that we talked about in the old voice network, well, now we have to do that for the last mile connections into our network, the internet, the cloud environments that we host all of our applications on, that has to be able to scale. So when we look at new technologies, whether they're for just consumers at home, or whether they're for businesses, we have to make sure that we've got the capacity to give them a very good user experience. So really, those three attributes enable, you know, future technology development across many areas. So entertainment, and medical, finance, education, employment. So, you know, we're today we're seeing a massive investment really in fiber from, you know, public and private investment. And in even our own companies, you know, we're building 30 million living units, which is a big part of our footprint. You know, we're going to have that done by 2025.

We're on schedule. But if we looked at the future services that these new technologies will support with all this additional bandwidth, you know, I would say that the types of things that are going to be driving that bandwidth are really, you know, things like personal, personalized virtual worlds, you mentioned the metaverse, real time gaming, you know, which is different than what we have today, holographic imaging is going to take a tremendous amount of bandwidth and some of that really medical based. One of my favorites is really connected wearables and enhanced reality and that could even include implants. So imagine that, you know, you walk into a room with, you know, your maybe something a little more stylish than the goggles that we have today, but you could look around and you might see somebody that you met a few times, you should know their name, but you forgot it. But what if you had the ability through your glasses, they would kind of help you, they would give you that advantage to, it's got people's names, it does facial recognition on it, right? What if you're giving a speech and you got your glasses, you could have your speaker notes, you know, where you didn't have to have like a separate, you know, podium pushing those, the notes out, a super connected home. And one of my favorites, and I think it's really, it's close, is personal robotic control. It's going to have AI, but think about a real life digital twin.

Think about now you have to, you've got some thing, kind of like a someone that you need to keep busy while you're working during the day, and you need to sort of keep tabs and make sure they're functional, they're doing things correctly. That's connectivity. And, you know, it goes back to the things that I grew up with, you know, that were just dreams between Star Trek and the Jetsons and Lost in Space, right? And these are the things that we're seeing realized today. And I think, you know, as they continue forward, you know, they're going to keep driving new applications. So optical fiber and wireless communications, and that cellular and Wi-Fi, and now satellite to fill in the spaces that you don't have connectivity are the current technology enablers. And I think these will be the things we use for the next 30 years. What happens after that, I think will be another technology revolution. And I think that's where we'll see quantum communications, which further add capacity and latency improvement over distance. And that'll improve the lives of humankind tremendously at that time. And, you know, and I wish my crystal ball was a little clearer than that. But I think there's some amazing, you know, things on the horizon that we haven't dreamt of yet.

I think you're absolutely spot on. I love thinking into the future to kind of see how these things, you know, make past desires or, you know, things of science fiction, become a reality. And actually, you kind of alluded a little bit to this fight, if you don't mind, to kind of round this out in the spirit of Techsploder, which Techsploder really is about kind of the human element within this soup of technology that we all live within. We're so surrounded by technology and bits and everything like that.

But there's also the human perspective and the human element and experience within that. I'd love to ask you a question that's along that line, which is, you know, probably a little different than some of the questions you've been getting as you've talked about this stuff in the past. But you've alluded a little bit to being younger and some of the influences that you had when you were younger that really that you were exposed to that I imagine kind of set the pace and laid the foundation for your career in technology. Can you reflect a little bit on like a piece of technology that you could look at and say, that really shaped my love for what I do now at a younger age? And what was so different about it? What was so magical about that?

You know, to me, the technology that influenced, you know, me was probably more than anything was television. And, you know, I started life. We had a few channels. Right. And, you know, I remember growing up and we were so extremely excited when, you know, when we could get cable.

Right. And I'll never forget when MTV and HBO, you know, came out. And, you know, that was that was, you know, a changing point in life when you could get, you know, we could probably get 20 or 30 channels right at that point in time. But for me, it was kind of like, you know, as I mentioned earlier, how did this work?

How do you send these images across the cable or across the air and the audio when, you know, get color and get stereo audio? And, you know, and I so I, you know, tried to learn what I could at the time and, you know, built some electronics and things like that just in trying to learn. And I was just fascinated by that.

And I think that's what changed me was, you know, seeing these things that seemed supernatural to me. Right. How do you make this thing work to send this stuff? But, you know, from nowhere, you can't see it. And I was just always enamored by that and challenged to want to understand it better, to see how these people created these things, these ideas. It was just amazing.

I love that answer. And what what occurs to me listening to you answer that is I'm certain that there are people that look at technologies like fiber, like DSL back in the day, and thought have thought and ruminated on the same fact, which is how the heck do we take all of this, you know, all this data, all these bits, and send it from one point of the world to another one, all seemingly instantaneously. And you were a part of that development. So that's pretty, pretty awesome how that all works, Eddy. Eddy Barker, it was a pleasure chatting with you today, talking about some really important historical moments in the world of technology and things that I think we all benefit from and enjoy on a daily basis. So, Eddy, thank you for doing the important work. And thank you for talking to me about it today. I appreciate it.

Jason, thanks so much for having me. It was a real privilege. Thank you for inviting me to Techsploder.

All right. A huge thanks to our guest, Eddy Barker, also the folks at AT&T for helping set this chat up. What a cool look at a technology that I've used a lot in my lifetime. Listen up, everybody. We could not do this podcast without your support. And right now, the most direct way for you to support us is at our Patreon. That's patreon.com/JasonHowell.

There, we offer ad-free shows, early access to videos, a Discord community, regular Hangouts and Q &As with me and the extended Techsploder family, and a whole lot more. We also offer the chance to be an executive producer of this show, just like this week's executive producers, Geoffrey Marraccini, John Kuehne, Catie Lake, and our newest executive producer, Bill Rudder. Thank you all so much for supporting this independent podcast.

Literally could not do it without you. Techsploder podcast publishes every Friday on the Techsploder YouTube channel. This week's episode was a pre-record, but when there is a live recording, you will find that event on the YouTube channel at youtube.com slash at Techsploder. I'll put up a thing there that you can follow it so that you're notified when it goes live. Don't forget to like, rate, review, and subscribe wherever you happen to be.

It really helps us out a lot. And you can find everything you need to know about the show at Techsploder.com. Thank you so much for watching and listening. I'm Jason Howell. We'll see you next week on another episode of the Techsploder podcast. Bye everybody.