Untangling business in the ISP industry with Elliot Noss

Avi Freedman: Hi, and welcome to Network AF. In this episode, I talk with my friend, Elliott Noss, CEO of Tucows, someone with a business background who came into the ISP industry and then has been, for over 20 years at Tucows, has done some revolutionary early CDN- ish work, really shook up the domain name space, started a mobile virtual network operator, an MVNO, and is now, in 2020,` is getting into and expanding the ISP business. We'll talk about that, a little bit about life and culture in companies. So please join us and watch the episode. Hi, and welcome to Network AF. I'm here with my friend of a few decades, Elliot Noss, CEO of Tucows. Elliot, could you give a quick personal and professional intro?

Elliot Noss: Sure. First of all, nice to see you.

Avi Freedman: Good to see you.

Elliot Noss: Second of all, I've been at Tucows, kind of running Tucows for, now I'm in my 25th year. Prior to that, a couple few years in the ISP business, and from my perspective, whatever happened before the first browser launched really doesn't matter. We've been in a number of businesses over that time. You've seen them all and we'll talk about them. But all of that has really had me kind of deeply inside of the ISP space specifically, therefore the networking space in general.

Avi Freedman: No, absolutely. And I really applaud, it's pretty amazing doing that in public, as a public company, is interesting. But you've always been pretty open about what's going on, which I really appreciate. So Tucows, the name.

Elliot Noss: Yes.

Avi Freedman: Were the founders cow herders or crosstalk?

Elliot Noss: No, it was the ultimate collection of Winsock software. It's a, I like to call it an anachronistic acronym. It was actually, this isn't talked about, it was originally Ucows. And somebody was smart enough to add the Tu, so crosstalk. It was the original formulation kind of in the room, just ultimate collection of Winsock software, and then really smartly added a Tu.

Avi Freedman: Oh, interesting. And so for those youths in the audience who do not remember, operating systems didn't used to come with TCP/ IP stacks. So we actually, at my ISP, as you said, like when I started Net Access in 1992, no one thought that everyone would have IP addresses at home. Of course, now, sometimes there's not, right, but we gave a slip disc with Trumpet Winsock, which I think we licensed, and Mosaic, and that was what got people on the net. And of course, Eudora, some people called it Endora when they called up for tech support and all that.

Elliot Noss: Yeah, we had a brisk business in creating ISP starter discs.

Avi Freedman: Oh, yeah.

Elliot Noss: Selling physical ISP starter discs that we'd burn and send out.

Avi Freedman: Yeah, no, absolutely. We did it just for ourselves. So how did you get connected with Tucows?

Elliot Noss: I was in the ISP space in Toronto, in one of the very early competitors, and Tucows, the original website was owned by an ISP. So the largest ISP at the time in Toronto, they well understood the importance of software downloads as a primary need of all the very early geeks who were buying accounts from them, and they were just smart enough to basically buy the website from the guy who originally put it together, a librarian in the Flint, Michigan system. So I came in, they hired me from the ISP that I was at, and I couldn't believe that they owned this website. In fact, I said to them, " Oh, I see, what is the Tucows thing that you have? Are you the Canadian mirror?" And they said, " No, we own it." And I couldn't believe it. Now, that may sound strange to anybody outside of'95, '6, '7,'8. But back then, that was a remarkable piece of internet landscape. And so they just let me build the business because nobody else was really doing anything with it.

Avi Freedman: Well, ISP nerds, not always passionate about software, unless you come from the BBS days, and downloading, and Kermit, and Xmodem, and all. It's funny, because in the later, mid'96,'97, I don't know if you ever ran into Andrew Koo, but he brought me to Australia, and we were doing satellite bandwidth augmentation, but we jokingly, we put inaudible caches in, and we called it the Social Engineering Caching Protocol, where they would look at what objects were in the cache and email their users saying, " Interesting content," so that people could keep the traffic locally back when bandwidth was somewhat more expensive than two cents per megabit per second.

Elliot Noss: Well, and Australia was, of course, a unique case. We were huge in Australia for exactly the reason you describe. I mean, imagine, when you were downloading, you glossed over it, but if you were Australian and you were downloading something from-

Avi Freedman: Cheaper to FedEx a hard disk.

Elliot Noss: Right. It was cheaper to FedEx a hard disk than to download anything from off the general internet, so anything you could cache was huge. And I mean, it was true really everywhere but North America, in the beginning.

Avi Freedman: Mm-hmm (affirmative), yeah, yeah. Well, and the business model for that was advertising?

Elliot Noss: It was advertising, yeah. So we partnered there, we became one of the two or three biggest in the world with download. com by partnering with ISPs all over the world, it was over 1000 ISPs in over 100 countries kind of at peak, and it was advertising, but I mean, this is very early days of advertising, so this was inbound. It started with all of the casinos, crosstalk-

Avi Freedman: Oh, so they just came to you directly? You were one of the crosstalk-

Elliot Noss: Oh yeah. They were, " Can we advertise?" And somebody was smart enough to take their money, and then that kind of evolved into all kinds of things. We probably had a performance- based marketing piece with WinZip for years and years and years. I mean, so if you wanted to get your trialware or shareware out there, we were one of the two or three best places to advertise in the world, so...

Avi Freedman: Awesome. So I was at Akamai for 10 years, and often when people are saying, talking about the first CDN, where I point out actually Sandpiper was a real CDN, it just wasn't as fast as Akamai inaudible, but I actually think of Tucows as in some ways the first CDN. Do you agree, disagree?

Elliot Noss: So it was, that was the way it was described to me. So I didn't think about it natively like that, you're, and have always been much smarter about networking than I am. I got the utility of caching, but I didn't see the bigger picture there. Yossi Vardi, who was an old friend, because we were the original distribution point for ICQ, he beat me over the head and face with that, as you guys at Akamai would, were just blowing it out in the public markets, he was hounding me with my failures for being first but not taking advantage of it. And I don't know if you remember, but I remember, definitely, us meeting at ISPCON well before Akamai, I still tell people, now that I'm back in the ISP space, I still tell people about the original My ISP Sucks Less tee shirt when I'm trying to help newcomers understand the industry. So you'd be happy to know that that history is being propagated.

Avi Freedman: Thank you very much, Elliot.

Elliot Noss: What, particularly in the ISP space, the reason that I think that is still so deeply apt, and we could talk about crosstalk companies and inaudible, but in this context is that ISPs today, the ISP service today overwhelmingly, over 90% of the accounts in the US are being provided by people who are not ISP people. They are telecom people or they're cable people or they're copper people, they are not ISP people, and there's just a different headspace and mode. I think the reason the dial- up ISPs, independent ones dominated over telecoms is because they rightly put the customer first, and there's just this ethos and this essence in it that just doesn't exist in the industry except when it's old ISP people.

Avi Freedman: Too disconnected from... I mean it's interesting, because if you look at the network as a service companies, or the interconnection as a service, they are people that did interconnection management, they were more recently, fewer levels of indirection, " the customer," which is why I started Kentik, because I was like, " Wow, when I left Akamai the networking sucked, I had to build my own," and people are like, " Oh yeah, sort of, we need that." So come back to networking, where I guess a little diversion, domain names. I want to thank you for OpenSRS, and actually, I had a lot of nerd friends that were suddenly starting little businesses doing domain registry, and how did that happen, and why did you decide to go for high volume, low margin and not milk the market at a time of huge explosion?

Elliot Noss: Yeah, so there's a couple things in there. How we came to it, people today just don't have a good sense of history, where around, prior to 2000 there was no web hosting industry. Web hosting was provided by ISPs, because there was this vibrant world of ISPs. They were kind of the nuclear events of'99, 2000, 2001, the switch from dial- up to now crosstalk-

Avi Freedman: crosstalk buying everybody.

Elliot Noss: ...cable and DSL. No, even before people buying everybody. It was really, once it went to DSL and cable, it wasn't freely competitive anymore. There wasn't that sort of regulated, dry copper you could buy, or just the copper phone lines that we could all buy to plug into modems. And so all of those ISPs essentially got crushed, went one of three places: out of business, into the very corners of the world, where you still see the WISPs of today, or they became web hosting companies, because they could go to the other side of the pipe, which wasn't regulated, didn't require the same amount of capital. So all of that's to say, we were the largest vendor of domain names, the largest reseller of network solutions. People also won't remember that you used to buy a domain name before the introduction of competition, that meant faxing and mailing things to network solutions. So as competition was coming, this is now pretty much with the millennium, January 2000 was kind of our launch, we knew that it was ISPs and web hosting companies that were the people who actually sold domain names. It wasn't this artificial construct called registrar. So our view was just to facilitate all these people, just basically instantiate the business processes that we wanted as resellers of these domain names for everybody else. So that thought was really simple. And then in terms of price, we went out and we said, " Well, this is a commodity that is completely competitive, it's naturally going to get priced down to a very low margin." We watched people coming out, we were a little bit longer because we didn't have to just stick up a website, we were doing wholesale, so we had to set up APIs, it was a bit more complex. And we kept seeing people come out at the same$ 35, $35. We couldn't believe it. I mean, I remember a phone call where I'm like, " Are we crazy or are they crazy?" And so we came out at$ 10. It took six months for that price to go to eight, like for us to be undercut, and today that's obviously very, very thin margin business. So it was just kind of looking at the market and seeing obviously where it was. Ironically, we priced it at$ 10, but eight if you were a Tucows mirror. Because it was still the case when we launched that that business was the bigger business, we were still three, four months away from the dotcom bust, and we were trying to feed the other engine.

Avi Freedman: Well, and take the ad revenue, yeah.

Elliot Noss: Yeah. So that's... And that business is still around 20 years later, we're the second- largest in the world, we're by far the largest wholesaler. It's now no longer... Well, it's now what I would call modern web hosting companies, so Shopify and Wix, et cetera. So you see that version. What is amazing, though, is I bet you that some chunk, some portion, 30, 40% of your friends from back then who started businesses? Those businesses are still around. The half- life of internet companies is crazy. They may have peaked at 1000 customers, and they might be sitting on just cashing checks on 400 now all these years later. So we have thousands of customers who are just these almost, it's almost like light from a distant star or something. But there's still millions of domain names like that kicking around.

Avi Freedman: Yeah, no, and it's fascinating, because I was going to contest the crushing of the independent ISPs, because I know a number of people that really, they built a business, they made a couple hundred thousand a year net, and it's morphed, but they're still doing it. And it's got some hosting, some consulting, and they're people like maybe me and my grandfather that maybe aren't as happy working for other people as they are working for themselves, they like making decisions, and a lot of them are HAM operators, or you get a lot of... We can talk about subtext of persona some other time, but-

Elliot Noss: Well, so that is, I said web hosting. That could be web hosting, that could be IT services, that could be a whole range of technical-

Avi Freedman: Well, and WISP, as you said, which we'll talk about.

Elliot Noss: WISP, absolutely, and by the way, we're going to see the re- emergence of that segment. Look, we're now doing it again at scale, it doesn't change. I mean, that's, crosstalk market.

Avi Freedman: People have a lot of patience if you have humility, if you have-

Elliot Noss: Yeah, that's right.

Avi Freedman: ...if they feel that they know the people, then they're like, " Okay, I could wait, my Facebook can wait for a minute." But when there's not personal treatment, then people get really, really antsy, really. I had a carrier issue yesterday where I walked into the store from one store from another in a mall, and he's like, I showed him the phone, and he's like, " Oh, okay, I'll help you out." It's like, " I'm sorry." It's like, " Hey, it's not you, it's not you. Someone in the back room made the decision, you're a nice nerd."

Elliot Noss: You protested earlier about removed from it. I think you cake the boy out of the ISP, but you can't take the ISP out of the boy. You're still a natural. You could pivot again today.

Avi Freedman: I could if I ... Well, in 2003, the ISP that bought the ISP that bought the ... It's a chad gadya, it's like the thing that ate the thing that ate the thing that happens... We talk about it at Passover in the Jewish tradition, it finally shut down Shell Services. And in 2003, 80 people were like, " Avi, you want to start that access again, Shell? You know for$ 5 a month, you could have a VPS, and that's better." It's like, "No, no, no, no, no, it's better. It's curated, we can chat with each other, we can chat with each other. It's your problem when Emacs sucks all the memory." I'm like, " No, no, no, Emacs is fast now, it's fine, it's fine, you don't have to worry about that."

Elliot Noss: We going to bring up VI now?

Avi Freedman: I use VI. Actually, it's interesting, because the network hipsters and the coding hipsters love Go and Python. But Go and Python are not well- tuned to my brain, because I don't think white space should be syntax, because it's UCL, which means I don't like Python, and I don't think uppercase should be scope, because the beautiful C is all lowercase. But back in the days of Fox Space and Turbo C, it was IDE, and I didn't mind. So I'm actually thinking, maybe if I use an IDE, which ironically apparently VS Code comes from Visual Studio, then maybe I won't be typing all the syntax, and maybe I will be able to get with it. But all I can write is running specifications inaudible could, so it doesn't really matter. But we'll see. I said Q4, maybe in Q4.

Elliot Noss: Love it.

Avi Freedman: So yeah, reminds me of early in my ISP days, because there was a while when I was the only ISP in Philly.

Elliot Noss: Yep.

Avi Freedman: A competing ISP would come from the BBS, so his first ISP was PCBoard with dial- out to a terminal server to get to the internet, said, " Avi, it disturbs me that you're pricing at," I think we were 20, or 12. 50, I forget what we got to by that point. He's like, because he was, it was funny, exactly at 35. He's like, " You need to raise your price to match mine, or I'll reduce my price and crush you." I'm like, " So you're going to drive everyone out of business and be$5 a month half the time, then raise the price to 50, and then do it again? I don't think that's the way the world is going." But I have to applaud you, you're prescient at starting it low, because I think we went 20, 12.50, and then 10, so-

Elliot Noss: So, I've got to tell you, this is a thing now where we sell a gig, like a fiber gig for$90 a month. We have no... My problem is building networks and connecting people, not signing people up. And I'll say to investors, " Look, we think long- term there's downward pressure on price." And they'll say to me, " But that's crazy. Why would that be the case. I've read every cable analyst report which says they're going to be able to raise the price by two to 5% a year for the next five years, why would you not be able to as well?" And when I try and describe the inevitable sort of outcome of these markets, they just can't grok it. They just-

Avi Freedman: You feel like Cassandra.

Elliot Noss: Well, what, now I-

Avi Freedman: Except it's not negative for the consumer.

Elliot Noss: Yeah, I'm just old and crotchety. It's sort of like, " Yeah, whatever, I'm just going to tell you what's going to happen, and-"

Avi Freedman: I have the rule of three now, I'll explain it three times, or ask three times, or offer to pay three times, and then I'm just like, " Okay my time" ... I don't want to annoy people. If they don't agree, it's fine. Everyone can move on, hae a fun life.

Elliot Noss: Well, and in this case, it's a statement against interest, right? Obviously investors would like it better if I was telling them we have pricing power. But I still think we're going to see, five years out, when we have a settling in this market, you will see... Because we're in the middle of a reformation, we can crosstalk whenever you like. You will see that price of internet access start to come down. It's inevitable.

Avi Freedman: Interesting. Okay. Even with... Do you think that's related to, not related to the disaggregation of content from access on the cable side?

Elliot Noss: No, not related to.

Avi Freedman: Okay.

Elliot Noss: It's 100% a function of... And we're jumping to crosstalk in the story, and we can work back as you like. But just, go to the end, the end is a US... And by the way, it's a different story in every country in the world, and I can tell that country's story, I'll just, the one I'm telling now is the US story. You now have fiber everywhere. What you're going to have is a fiber product that is superior to a coax product, you're going to have a DSL product that won't matter, you're going to have things around the edges in the remote places where there's no fiber-

Avi Freedman: Starlink.

Elliot Noss: Like Starlink and WISPs, and in the vast majority of the country, I'd say 120 of 140 million homes, it will essentially be fiber, which will be a better service, and coax, which will be an inferior service. The inferior service will have to charge less. And so then what you'll have is people who are willing to pay a premium for fiber, again, that premium is going to be some number south of what they're paying today, and you'll have a coax product priced underneath that. And it'll be some new stasis in market share there.

Avi Freedman: You're a dangerous person.

Elliot Noss: It's just inevitable to me. But there's going to be, I talked about 120 million homes being built. If we can build one million of them, I've done an amazing job.

Avi Freedman: Well, and-

Elliot Noss: My wildest dreams are three, four million. And by the way, never before, as the CEO of a public company, do I talk about my hopes and dreams, in the context of fiber I do, because the numbers are so big that it's crosstalk you can do what you want with them, " Here's my hopes and dreams."

Avi Freedman: Yeah. And there can be a lot of diverse successes making it happen, which is cool.

Elliot Noss: Absolutely.

Avi Freedman: So we'll come back and then come forward. Networking. How did you see it, find it, get interested?

Elliot Noss: Well, I think for me, I've always been a businessperson first and a geek second. I would let myself take one fun course every year in university, and my first year that was computer programming, which meant filling out punch cards with a pencil and putting them into a card reader. And then getting in the ISP space, you naturally just have to learn about networking. So I think it's deep osmosis. Where we are today, which is kind of interesting in this context, when we're trying to describe some of the success we've had, we describe it in terms of being multilingual. We can speak finance and speak ISP and networking, and speak culture and business, and it's really being able to understand natively each of those things that... And, therefore, the interrelationship. It's kind of like, if you're not bilingual, you'll just lose something in the translation between the things.

Avi Freedman: People don't-

Elliot Noss: I'm sure you've had great frustrations in your life with finance people who just didn't understand networking, for instance.

Avi Freedman: Yeah, or at Akamai, our CFO would help because, well, I mean, he had a PhD in electrical engineering, did design video games, and... But it's interesting, because sometimes people look at me weirdly when I say I business nerd or I finance nerd, people don't think of that as like a nerding thing like software, but it sort of is, right? When we hire people, we don't necessarily care that they're yet a nerd about the thing they're going to work on, but if they can't nerd about something, take a position, show that they're passionate, have paid attention to it. And it's sort of like the XKCD diagram, right? If you sit and wait for the received knowledge to come into your brain, it doesn't work. You need to be comfortable with confusion, like, " I don't understand, I don't understand, I don't understand ... Oh." And then the best is you teach, because everyone else is confused. If you were confused, then probably the teaching needs to be better.

Elliot Noss: Oh, I love that. One of the things that's most important to me when I'm interviewing people is just to try and figure out what the thing they love is. And I think way too often people conflate what they're good at with what they love.

Avi Freedman: Yeah, it's-

Elliot Noss: And what they love is that thing that they're going to do at the end of the day. It's that thing they're going to do on the weekend. And it's, really, it's the intersection between those two things and what needs to get done where the magic lies. So I think that's exactly right.

Avi Freedman: I can't believe how fortunate I am that the things which I enjoy doing are valuable.

Elliot Noss: Yeah.

Avi Freedman: That's awesome. So coming forward, so CDN that wasn't marketed as a CDN, or monetized as a CDN, domains. Which came first? Cell, Ting?

Elliot Noss: Yeah, yeah.

Avi Freedman: Okay, that's what I thought. So that's not IP, I guess there's IP in there. So why don't you describe Ting? And I'm curious, I haven't heard the whole origin story, there, I just crosstalk pop up.

Elliot Noss: Yeah, so I mean, that's Ting Mobile, which is now-

Avi Freedman: Mobile.

Elliot Noss: ...a brand that's owned by Dish, and will likely come back at some point. Then there's Ting the ISP.

Avi Freedman: Ting Mobile, yeah.

Elliot Noss: So crosstalk Ting will be the ISP just, but Ting Mobile, when we launched in 2010,'11, this was just, at its simplest, taking the things we were really good at in the domain name space, so sort of, there was great customer service, there was a great billing and provisioning platform, there was an understanding of the value of these simplified business processes. And so we looked at... But domains were too small. When I say too small, total addressable market that we could attack was tiny, and I needed growth. So at the time, mobile phone service in the US was the, it is still the single largest technology market in the world. ISP service in the US is number two.

Avi Freedman: Wow, I didn't... I should've known that, I guess,

Elliot Noss: Well, because people don't think about them as technology, right? It's like they're in this weird bucket of telecom, but it's all tech. Anyway. And the net operating margins in that, you want to business nerd a little, were in the, like industry- wide, were in the high 40s, Verizon was in the mid- 50s. So there were clearly excess economic profits, it was ridiculous. crosstalk-

Avi Freedman: Your margin is my business opportunity.

Elliot Noss: Absolutely. And all I did was I poked around and poked around until I could find somebody who could share with me what MVNO rates might look like, there were these things called MVNOs, mobile virtual network operators who buy capacity from the big telecoms. At the time, we could easily sell for way more than we could buy while still being well under the market.

Avi Freedman: And you had the provisioning technology that someone could be a 15 or$ 20 a month customer and make money crosstalk support.

Elliot Noss: I mean, the most profitable accounts we had were the, " Stick a phone in the drawer for somebody coming from out of town at$ 6 a month." We just, we approached it as if we were customers, and we kind of redesigned the service from the ground up. And there was a dirty little secret, there is a dirty little secret in telecom, which is over half of the customer service interactions are about billing. So if you just have simple billing, it's not, inaudible calls it the Confusatorium, which I love. Their business model was to screw you-

Avi Freedman: It's mandated crosstalk-

Elliot Noss: ... eveningsand weekends, calling circles, the way that roaming worked, in market, out of market, in country, out of country. The limits on things, and then if you go over those limits, you get punitively smacked about the head and face, so you overbuy your limits, and just all of it, right?

Avi Freedman: Right.

Elliot Noss: And we just eliminated all of that, and that was incredibly novel at the time, and quite profitable. We made good margins and good dollars right from the start.

Avi Freedman: So did you have an eye towards Ting broader ISP, or was it...

Elliot Noss: What I would say is I always hung around the fiber crowd in the US. It was a lot of fellow travelers. It was really about time and place. What happened for us... Fiber's very expensive.

Avi Freedman: Yes.

Elliot Noss: You need a lot of capital. Two or three years after we launched Ting Mobile, it was the intersection between I now had two businesses generating a shit ton of cash and didn't require any capital, and the market was ripe enough in a lot of different ways that we could start to build fiber or start to do fiber. So we bought a business in 2014, we partnered with a small city in Maryland in 2015, and we started building our first network in October of'16. And it was just, Avi, if you want to, which you'll appreciate sort of the thinking inaudible element of this, this was like, we had a set of core premises, and if you ran the numbers, it was like, this is crazy profitable, I'm not sure it can really be this good. And everybody else thinks we're crazy. So you know what, I'll-

Avi Freedman: That's a good combination.

Elliot Noss: ...push a few chips in the put, let me test my premises. Let me push a few more in, let me test my premises further. Okay, this looks good, let's start to really push some money in.

Avi Freedman: Well, it's funny, because Google, I always viewed sort of Fiber and Loon and stuff like that as, " It's unfair someone isn't able to click on our ads, it's unfair, we must make it so everyone can click on our ads." But you're coming at it from a different perspective. So it's not altruistic, people need better, it's that, plus we can, and you have the backend technology, which I probably would rather pull fingernails out than work on building systems all the time, but-

Elliot Noss: Yeah, so that's right. I mean, the very first thing in the story I tell, I walked into the first ISP I worked at, this was late'94, early'95, I come in, the day starts with two women going to a filing cabinet. One starts at one end, one starts at the other end, and they're pulling contracts. And I said, " What's going on, what are you guys doing?" " Oh, we go through all of the customers, I start at A, she starts at Z, and-"

Avi Freedman: Oh my God, not to see who's expiring.

Elliot Noss: ... "Andwe look to see," right, "We look to see whose day of the month is this, and then we bill them." So I wrote a piece of billing software, thankfully, that was rewritten by somebody else six or 12 months later. But it's exactly that.

Avi Freedman: It's funny, in the interstitial before Kentik, when I was at Akamai but I was, really thought we should do cloud, but they didn't want me to, but they didn't want me to leave, so... I had a Usenet company, which I still dabble in Usenet. So I bought a Usenet company, and I'm looking for the billing system. And I'm like, but I know that things are... And it was all in Windows, which is not my jam. And it turns out, there was no billing system. It was recurring with PayPal and their credit card processor, and what would happen is emails came in, and the owner's, former owner's wife would go in and manually set the thing. And so after a week, I'm like, " Wait," and I wrote a procmail to do the thing based on the... And she's like, " I have to go kill my husband, because I did this for eight years. I manually went through, I looked at the emails and set the users," because great programmers are lazy. Right, that's the-

Elliot Noss: What's true is, it's impossible to deliver a good customer experience, a good web experience, if you don't have sort of a solid backend billing and provisioning system, integrated with customer care, managing data in motion. Really, and the things you can do today with modern software... So I think there's a whole revolution there. I talk about it as, if we bill, there's going to be 70 million homes built in the US, fiber, in the next five years. If we can build a million, we've done amazing. If we can build three million, we've fulfilled my dreams. I want to sell software to the other 67 million.

Avi Freedman: Well, that was going to be my question. I know you're a public company CEO, so you have to be careful with projections. If you wave your magic wand for the 10 years fro now, is that where you'd like to be? You'd like to have high- mid single digit millions and be fulfilling and helping the growth of the rest?

Elliot Noss: I think that's fair. I think that my first love will always be the ISP business. I'll be... That business is still, luckily, anachronistic enough that I'm pretty fresh and current.

Avi Freedman: Yes.

Elliot Noss: Whereas when it comes to software, when it comes to data, when it comes to a few other elements, I keep myself well- read, but I am not an expert by any stretch anymore. So there, I think, over time I'll play a little bit less and less of a role, the other is more native to me.

Avi Freedman: Interesting. So it's interesting, one of the things you just said. If I could go back in time, I would have bootstrapped Kentik for probably another year. And in the SaaS business, software business, there's a move called product- led growth, where it's not just about shooting flaming balls of marketing money in the air, it's about being understandable, so we have a learning center we're doing, and even just the, when you start and you have people wanting functionality, billing, support, there's enough money in the venture market that if you start getting high growth you say, " Well, okay, we'll have some people do that." And with apologies to some of our own customer success people... Now, I will say at Kentik it wasn't really user- visible, but if I could go back and do it again, those foundational, how do we make it super simple, because there's real understanding that that breeds virality, which is the cheapest kind. It's exponential... Well, I guess you saw that with Ting, but the virality of the-

Elliot Noss: Well, I mean, with OpenSRS first. Well, I shouldn't even say that, I mean, with the download business first. But with OpenSRS, I mean, we went to an ISPCON in November of 1999, it was all about content and mirrors and software downloads, there was a little bit at the edge of the booth, " Yeah, we're doing this domain name thing." 700 signed contracts at the show.

Avi Freedman: Wow.

Elliot Noss: I mean, so yes, that, virality is... I mean, I think I just suck as a marketer, and so let me just try and grab a baton and get in front of the virality parade. One of the things that I'm wrestling with in telecom, well, that we think we're going to do... No, sorry. We're going to do, we don't know if it's going to be successful or not, is we don't think anybody has ever built or sold to the geeks in telecom. And you get inside of these big telecoms, and they've got some good geeks. They really do.

Avi Freedman: Oh, absolutely.

Elliot Noss: And, but everything there is sold to the CIO. So this is going to be the first real effort of, as you've described, doing that kind of really developer- focused, telecom developer- focused platform. And-

Avi Freedman: That's product- led growth.

Elliot Noss: It is. But-

Avi Freedman: It's Twilio. Ask your developer, right? It's-

Elliot Noss: It's Twilio, exactly, it's Mongo, it's a million of them, Slack. But the thing that I don't know is can developers kind of jump the blood- brain barrier to the CIO? Because it's still the CIO who's got the budget, it's still the CIO who has to write the check, et cetera. So it'll be interesting to see.

Avi Freedman: Yes, I will, we'll have you on in a couple years and we'll hear how it's going.

Elliot Noss: Yeah, that would be about right, too. But we are thinking a ton about exactly that, how to build it for them.

Avi Freedman: Okay, interesting.

Elliot Noss: Yeah.

Avi Freedman: So OpenView Labs, OpenView in Boston, and actually Georgian Partners in Canada write a lot about product- led growth. And it's funny, because they really are more focused on bottoms- up, but I think that, for example, even if you have a complex enterprise sales process, reducing the number of SE meetings by making onboarding easy gives business value. So traditional product- led growth is more like start$ 50 a month, or$6 a month, in your case, and then mine the big end as you grow. But a lot of those principles, I think, apply and it's demystify and it's understand, so I think it's an interesting area. It can be Talmudic when... You don't have to agree, but even the discussion is interesting. It's like, I call him the Great Lemkoni, Jason Lemkin is a VC and writes about SaaS, and was one of the first to help me realize, whether you're doing legal SaaS, inaudible billing SaaS, network SaaS. But he's very repetitive. But one of the things he's repetitive about saying, as CEO, as you know, you need to be repetitive and say the same thing. So even when I disagree with him, it would be, we've talked a couple times, it would be an interesting decision, right? Always hire two salespeople, because if you hire one for any role, you can't compare. It's like, well, you can't always do that. How do you support VPs? There's just lots of interesting conversation there.

Elliot Noss: Yeah. I feel now like I can go and get a map somewhere-

Avi Freedman: We should talk more, we should talk more.

Elliot Noss: Yeah, I'm just wandering around in the wilderness, so it's inaudible great.

Avi Freedman: So you talk about, you got into it through business. I love the business side.

Elliot Noss: Yeah.

Avi Freedman: But my ADD comes in when it comes to policy. Is policy fascinating, is it something you need to do? Because politics and policy and all that.

Elliot Noss: Yeah, that's a great question. So for me, I was an abject failure until I was in my mid- 30s and the internet came along. And I loved the online world from the second I touched it, which would've been kind of late'80s. And so I've always felt like I owed a debt to the internet, let me say it like that. It's also always been clear to me, when I say that, in the sense that it's always so easy to know what's going to happen eventually, but have no idea when. It's always more important crosstalk too many people are just like, " It's great, I can tell you what's going to happen in 10 years, now pay me some money." No. It doesn't work like that. And it was clear to me from the onset that the effects of the internet were to pull from national to global up and from national to local down. So it allowed you to do stuff at a local level that you needed more scale for previously, and that, I mean, I'll, sometimes have used a bunch of different taglines for this, but I have much more in common with a Iranian blogger than with a US or Canadian statist. And so it's kind of who's your community, who's your tribe. And it's clear to me, also, that democracy is sclerotic at this point, and deeply ineffective. And so just, you kind of combine all those things, and what you have is something else is coming, right? Anybody who says they know what that is, we're now on a decades kind of time scale. But I feel like the stuff we're seeing, I've spent a lot of time in ICANN which is where most people who kind of crosstalk with the policy think about that, and I've been around since the beginning, again, just because we happen to be in the registrar space, it was regulated. But what it is is, I use big, lofty phrases like, " It's a progenitor of global governance." And by that I mean, the countries of the world have a torn ownership of a piece of policy governing the single authoritative root, which is the one bit of centralization, tiny kernel that everything else in the open internet is built on. And so I just think there's so much... I feel like, what, we're in the middle ages, democracy is inevitable, the collapse of monarchy and empire is inevitable, and it's like, I get to kind of be part of the very proto, very early thinking. And that's just so compelling intellectually.

Avi Freedman: So would it be accurate to say sort of the ability to be part of that and where it's going helps assuage the short- term talking in circles and talking about how to talk about talking about talking about things in committees crosstalk-

Elliot Noss: Yeah, look, I can't-

Avi Freedman: I can't do that. I can't do crosstalk-

Elliot Noss: No, and my capacity over time has reduced, and I try to sit above it more and more and just lob in bombs. So really just try and sort of-

Avi Freedman: The disruptor.

Elliot Noss: ...crosstalk thing-

Avi Freedman: Yes.

Elliot Noss: ... ina slightly different direction, or make a different kind of black hole or magnetic force. We'll be doing some stuff in a couple weeks around content, like I can give you a simple example of this. Think of, I mean, you have a deep appreciation of the fact that cybercrime today is just completely asymmetrical. And it's asymmetrical primarily because you can't solve global problems using national frames. It's just, I can do a long version of that, I won't for time, but it's, the problem is these national things, frames. And so we all know where the bad guys are. So often, I had to spend well over a decade sort of dealing with my frustration around somebody doing something terrible, knowing exactly who they are and where they were, you can't do anything about it.

Avi Freedman: No, it's not legal. I mean, you can't BGP hijack back someone, because then you'd be committing a crime.

Elliot Noss: Right. So all the frames are wrong, and so we're going to try and do some stuff in the next little bit dealing with really easy, low- hanging fruit. Phishing, farming. When I say the worst elements of spam. But there's so many that are virulent, right? " Avi, thank you, we've processed your Norton Antivirus," and it's a cat and mouse game, and we can stop all of that at the level of the DNS. And so that's going to become something that I'm going to-

Avi Freedman: Interesting. Because Spamhaus effectively does that with reputation and whatever, just-

Elliot Noss: When you said effectively, you used it in the sense of...

Avi Freedman: Oh, not globally, but they're effective by not being sort of, by being hard to sue, by saying, " It's not us, the ISPs choose to use this," they sell the piece of it, I guess, but-

Elliot Noss: Inside of ICANN, some of this stuff can be formalized and should be.

Avi Freedman: Ah, interesting.

Elliot Noss: Yeah.

Avi Freedman: Interesting.

Elliot Noss: And so that's going to be... But all of that is just an example of, okay, cool. Here's this really low- hanging fruit problem around cybercrime. Now, maybe if we can just solve it in this one place where I can be part of it, maybe that'll inspire other people to take on bigger and broader... And I mean, it's amazing to me, look, we still live in a world... Let's talk about NSO, right? We still live in a world where there are these parallel universes between statecraft and crime. And until we all recognize that just because a state does something that's criminal, that doesn't make it right.

Avi Freedman: I still do have to be thankful that the criminals don't want to shit where they eat, because the internet infrastructure itself is still very fragile. And so whoever it is, at whatever level, I'm glad they're not attacking the internet infrastructure, because there are ways to do that that we know but don't want to do.

Elliot Noss: Yeah. Well, and I would say that's true today, right? I could lay out a set of facts for you where some threat actors, that might not be true for.

Avi Freedman: Right, okay.

Elliot Noss: And so I think that we've got to find a way to globally combat some of this stuff, or everything just kind of goes to shit.

Avi Freedman: Okay, well, we'll do a separate dedicated on this, because I'm fascinated. But I'll just say that I was on the inaudible advisory council from when it first, it became clear to me that I liked hanging out with my friends, and I applaud all my friends and non- friends that are on it. I think the breaking point for me was when like, " People aren't deploying IPv6 because it's too hard to get." It was like, no, that's not the problem. And then Jim Fleming, and IPv8, and jumping through a wormhole and peering with Uranus, and I was just like, " No, that's too much, I can't deal with this." And to see John Kern do so well, as super legal dude but inaudible and the patience, and then I actually, Paul Toomey was working with us at Akamai to do consulting and stuff, and he called me, he's like, " I'm going to be, they're interested in CEO of ICANN." It's like, " Oh my God, I think I'd rather clean high rise windows with my tongue and no ropes," or something. I mean, my ADD, so I guess I'm not power- hungry enough or patient enough, or altruistic enough and visionary enough.

Elliot Noss: We, sadly, again, like politics, which is the reason that democracy is so sclerotic and ineffective right now. And by the way, that's not me saying, " Therefore dictatorship" at all, that's a longer discussion. But there are certain people who can just engage in intellectual masturbation, and the system rewards it. Or engage in the sort of, the levers of power for power's sake.

Avi Freedman: Yeah. We'll be careful about politics, but I also think you want to have some friction in the system.

Elliot Noss: Part of the problem in the US, so I described this as it was like the fighting of the last war. Why is the US in, generally speaking, globally, terrible shape? 2016, China connected more homes to fiber than are homes in the US.

Avi Freedman: On the infrastructure side, right, yeah.

Elliot Noss: And the reason is, because the infrastructure that people were delivering internet on was cable, so it felt like entertainment. And then you had a whole swathe of companies, big telecom, big media, who wanted to see it as entertainment, right, that that was its primary purpose. And so it was not appreciated as infrastructure. The whole regulatory framework is that which is amenable to entertainment, not heat, water, power-

Avi Freedman: crosstalk all the way over to human right, yeah, like those things, yeah.

Elliot Noss: I mean, I don't think you need to get to human right. It's just, what, you get into semantic issues there. But I do clearly think it's key infrastructure.

Avi Freedman: Well, in the after times. In the before times, we could argue about it. In the after times of COVID, or, which hopefully we'll get to soon, or after- ish.

Elliot Noss: Yes.

Avi Freedman: It's critically, it's clearly required, and if you don't have enough bandwidth to educate your children and do your work, you're going to have problems.

Elliot Noss: If you knew some of the data in the US, I mean, you may, but I would expect you don't. It's unbelievable. And it's the intersection of a whole bunch of historical problems, that if there's one thing at a policy level on that side of the business that I am desperately hoping for is as much money as possible to go to subsidy, and as little money as possible... Like, just if you've got, this is your ball, how should the ball be split-

Avi Freedman: Right. Well, there's hundreds of millions, I think, as part of this, or billions as part of the infrastructure bill.

Elliot Noss: It's still a small percentage of the total.

Avi Freedman: Right.

Elliot Noss: And what people don't get is you build a network for one or two years, you operate it for 100. And in one case, you're subsidizing companies, and in the other, you're subsidizing people. And it's just-

Avi Freedman: Oh, I see. So sort of like... Although, I can make the counter, what was it called, the educational subsidy? I think it launched in the'90s, there was something rate.

Elliot Noss: Yeah, E- rate.

Avi Freedman: E- rate, yeah. That was also subsidizing the buy side, not the sell side. But in aggregate, it effectively was subsidizing the sell side. There was a little bit of choice. But it was-

Elliot Noss: Yes. I would say that's true about E- rate, and I don't question, look, I will get higher penetration rates the higher that subsidy. There's no question. But the reason that I'll get higher penetration rates is because somebody won't have to rely on this for their home internet access. And this won't be the screen that they consume it on, and it won't be the bandwidth limitations they experience. So yes, that is true, but it's kind of a second or even a third- order effect. There's just such a problem to address on a number of levels. But what all the momentum now is so positive and so inevitable.

Avi Freedman: Yeah, no, it's encouraging.

Elliot Noss: Yeah.

Avi Freedman: So I think we have a lot to talk about on a second show at some point, but I'll end with a couple questions.

Elliot Noss: Yes.

Avi Freedman: High level answer to this one, because I'm sure this is very inaudible. I was really impressed when I went to the website, which I hadn't been to for a while, and you ask people, " Go to Glassdoor, check us out."

Elliot Noss: Yeah.

Avi Freedman: And feature Glassdoor reviews. What is it that has worked well for you at Tucows that I guess seems to make people happy?

Elliot Noss: Culture is always in the little things. It is the things you do when no one's looking. So I really, it's amazing to me that people... It's just all Golden Rule, and that people complicate it beyond that. If you just do what you feel is... I'm sure that there are network geeks who love working for you more than anybody else they've worked for, and the simple reason is probably going to be that you're treating them how you would want to be treated. It's going to be the person who didn't do this for you, and you're going to do it for them. And that's all, and then the only other sort of big bucket there is filtering. I think that there are certain people for whom we will be the best work experience they'll ever have in their life, and then there are other people who really, it won't matter that much to. So the more money- driven you are, probably not going to do as well here. The more status- driven you are, probably not going to do as well here. The more, kind of you're looking to be upwardly- mobile, " I want a new job title every 18 months or 24 months." People sit at the same job title sometimes at Tucows for eight, 10 years. Their job, their responsibilities increase every year, the scope and scale increases every year, but it looks like they've been doing the same thing at the same title for eight years. Do you love the internet?

Avi Freedman: Yep.

Elliot Noss: And it's kind of those filters, and I'm unhesitating, and I think that trickles down through the ranks, of just, " Yeah, okay, I'm not going to chase this person. This is not a fit."

Avi Freedman: Right. No, that makes sense. And you have to, in order to do great support, you need to have great people that are happy in their jobs.

Elliot Noss: Oh, our fights are on the reverse. I'll give you one that you'll love. We originally, we let people pick their own router. Because that's a nice thing to do. But there are enough people that pick the cheapest router, and then it's like, " Well, why didn't I get a gig?" Because you've got inaudible on the back of the thing. It's not like I'm ripping you off. And so as we try to push in better communication if you're going to own your own router, our customer service people were resistant to that. It's like, " But this is our values, this is what we do. How can you restrict them?" And so I love that, when-

Avi Freedman: crosstalk that tension, we had lonely people that would call support, and we had to say, " Two hours a week." We actually said, " We feel bad for them, if they just want to talk to support, two hours a week."

Elliot Noss: That's great, that's great.

Avi Freedman: Yeah. So I guess last question, sounds like you've been having fun, and thank you for all the impact you've had and options you've given people. Any career advice you'd give your younger self?

Elliot Noss: Well, I think I'll repeat two things that I tell my children.

Avi Freedman: Okay.

Elliot Noss: One, make sure you do what you love, that was important for me growing up. But in today's hyper- competitive world, I just had to compete with people in Toronto, or maybe Ontario. You're competing with people from all over the world. If you don't love what you do, you won't work hard. If you don't work hard, you won't succeed. Full stop. Second, this is a story I tell, my daughter hates when I tell it like this. " Dad, I'm starting my first real job. I don't want people to hate me. How can I make sure people like me?" And I said, " This is really simple. If you make the people who you work with and work around you, if you make their jobs easier, they're going to like you, no matter how big a bitch you are. And if you make their jobs harder, they're going to dislike you, no matter how kind and sweet you are." Real simple, so...

Avi Freedman: Okay. That's good lessons. Well, thank you very much.

Elliot Noss: Thank you.

Avi Freedman: I look forward to a part two, I think we have plenty of things to continue, and best of luck with Tucows and the grand ISP dream.

Elliot Noss: Yes. More soon, more soon. I enjoyed myself, thanks.