Dial Up Motive
Presented by: Dan Hollinger, Nick Sullivan
Originally aired on January 2 @ 9:30 AM - 10:00 AM EST
Human-interest segment asking Cloudflare employees what their first Internet experience was and how it informed them joining Cloudflare. Dial-up modems, bulletin boards, punch-cards, Twitch, Twitter and more.
English
Transcript (Beta)
Welcome to Dial Up Motive on Cloudflare TV. Hello, hello everyone.
Welcome to episode 33 of Dial Up Motive. This is the show on Cloudflare TV where we have an amazing chat with a Cloudflare employee, learn about their early history with either the Internet or technology and some of the experiences that drove them to pursue their interest in technology and ultimately end up at Cloudflare and the job they have today.
Hey, I'm your host Dan Hollinger dialing in from the Bavarian coast.
And with me today we have Nick Sullivan, the head of research here at Cloudflare and Nick, I'm happy to have you on the show and look forward to, you know, what really drove you into the world of tech.
Great, well thanks Dan I'm really excited to be here, and I'm excited to share, I don't think I've shared any of these stories publicly, but, uh, you know, it's everyone has a backstory everyone gets involved with technology for a reason, and I'm glad you're doing this and some great stories from all sorts of different folks you've talked to.
Awesome.
And with that in mind, would you mind giving a quick introduction of who you are where you're dialing in from and what kind of work you're doing at Cloudflare today.
Sure. Um, so I'm Nick Sullivan I've been at Cloudflare since we were a wee baby company around, I guess 10 years ago now is when I joined my current role is that I'm head of research.
Cloudflare research is a department within Cloudflare that does academic research into computer science and helps bring that into production systems and it really focuses on how to bring Cloudflare's mission which is to help build a better Internet into reality, whether it's by new technologies working with standards bodies to establish new Internet protocols, working with researchers on what the latest technology is or working with our engineering and product teams to help build new, new insights new technologies into into Cloudflare systems, and so I'm dialing in from Brooklyn, New York.
And here I am happy to, you know, it's a 10am here happy to have this chat.
Awesome and maybe a good summary is, you know, our product team is typically looking at what we can deliver in the next year or two years.
We have an emerging emerging technologies team which is looking kind of three to five maybe three to seven years out, really the the area you're looking in is the 10 and 20 year span like really what's coming up on the horizon from a compliance or protocol perspective, what what are things that we need to be aware of or inform you know really that that far future, so the cutting edge of the cutting edge.
Would that be a fair, fair. Yeah, I mean I think everybody at Cloudflare is involved in innovation in one way or another.
And there's feedback loops from the product team to the engineering team ETI team and research and we try to get involved the whole way down but yes we're focused on on really long term trends like how is the Internet going to change over time and how can Cloudflare be prepared for that.
And how can Cloudflare design and systems to be resilient for our customers in as the Internet changes.
Makes sense and to kind of kick off the show and to draw things back, you know, I really love to really get that kind of first story you might have on, you know, your introduction to technology.
As you can see behind me I was in the Packard Bell era the dial up motive days and hence with our soundtrack.
I'm kicking off the show, you know, really love to dig into when was that first moment we're like oh my gosh this is magic.
I need to learn more. Um, well, let's see I was interested in computers before they were really interconnected on a broad scale.
So, my Internet was exchanging floppies with people at school and software exchange events, things like freeware shareware, mostly games, different types of technologies on, you know, the big fat floppies that we had five and a quarter inch on these old IBM machines.
And what kind of games were these were these all kind of text based era or was this starting to get into some of the the graphical kind of scorched earth levels type of gaming.
I remember scorcher I had forgotten about that one but that was definitely a fun one but it started with text.
And I had an older cousin at the time, who was able to program his own little games, and he refused to teach me how to do it I was maybe seven years old at the time.
And, and so, you know, kids don't usually have patients that they're better at learning than teaching.
So, as, as like a bright seven year old I got QBasic on my 8080 8088 I think it was it was like a hand me down computer from my dad's computer lab he was a teacher at a high school so I think it was an Olivetti m 24, which was, if you look it up, it's a really cool elegant looking kind of futuristic machine all those from the 80s.
But I got QBasic and started writing some different games I remember I had a had a had a game where I would draw with ASCII art, like a caterpillar and you had in the question, the question was, guess what this is or tell me what this is, and you know I'm go to statements were really big part of that you would just number the different lines and so I still have go to statements I know that's like controversial thing in computer science and, you know, my C code at Cloudflare is still full of them I think they're, they're very efficient.
But, you know, it's it started there.
You know, I also got into expanding the technology I was very curious about a lot of different things as a kid.
Astronomy was something that I really loved and astronomy magazine would print programs in the magazine that you could type up and and sort of run on whatever your, your basic based programming language was so I remember one time wanting to know whether Jupiter would be in the air so I could use my little telescope to look at it.
And there was this whole algorithm that would kind of calculate where in the sky, it was based on your latitude and longitude and these sort of things.
So, I remember that being really fun.
This is like, again pre Internet, but, um, and yeah that's that's fascinating that era where to even get information about learning how to code or learning what programs to run was in the form of books you had to go to the library, and, you know, depending on what city you were in, or get it from a magazine.
You know I've had early other conversations about magazines that were built for for programmers that literally had snippets of code here, go run this, you know that was the distribution system back then, before all of this was was dialed up or connected together.
Yeah, yeah I remember typing in transcribing from like a page of a magazine, this program and being really excited when it, you know, give me a result.
And, and when was that moment when you went from oh my gosh this is a fun toy to, oh I can program this I can make this game, I can make it do what I want as opposed to simply acting as a consumer of a new toy.
I think it happens slowly.
Part of the part of the joy of figuring out computers was getting it to do things that you wanted it to do.
So, being able to run programs that were too expensive to run on the computer so learning how to do things like edit the.
I don't even remember what those, those, those, it was like a command .bat or run.exe, whatever these were these sort of DOS things to help optimize which memory, memory settings you had on these old devices and.
And so, you know, once, once I could harness the power of, you know, the hardware, by, by touching different parts of the configuration that that's sort of when it clicked for me that this isn't just, you know, something you can do go to statements or ask queries of folks that there's there's actually, you know, a hardware, software synergy in place.
And as hardware was evolving pretty quickly back then so I wasn't able to get my hands on the latest computers so you know you had to kind of retrofit what you had to make things run better.
Mm hmm. And how did this kind of early dabbling and early code writing, kind of evolve when when the Internet did so they come either to your school or into your home.
Yeah, um, I think my first experience online wasn't the Internet, per se, I well I guess it was the Internet but it was something called go for net, which was a network that was operated out of universities I would dial in to the university and you'd get like a text based interface, there would be news and and things like that and honestly I wasn't really sold.
I don't think the global interconnected nature of the Internet. through, you know, making posts and then waiting for a while for them to be propagated among different machines really was that exciting for me so I did, as with a lot of folks in my generation get involved in bulletin boards, so I would dial into bulletin board systems all the time.
I know gaming seems to be a theme here but I was a kid, but I would play, there's a game called Legend of the Red Dragon, this like Lord Lord so I was a Lord Lord, trying to level up my paladin or whatever it was, which was really fun but um, we had high speed Internet relatively early where I grew up in Canada in Alberta.
And so I started getting into like web hosting.
And, you know, there's a lot of really creative things that people would do on the Internet in those days so you know I got involved in like a pyramid scheme or two.
There's a sort of like ad based software as you get all your friends to install and we're past the statute of limitations now right we're absolutely okay.
This is, it's been over 20 years, so not only 20. Yeah, over 25 years whatever whatever the statute of limitation is it's over.
And I was going to ask whether the Canadian Internet was more polite, you know, back in the day, relative to the American one or just the global one that was developing.
Well it was fun there's different business models right like at one point a company launched free Internet that was ad supported, where a third of your screen was really annoying loud ads but that sort of allowed you to go through it but um, yeah I think it was, it was, it was faster to get high speed than the US.
But there was, it became sort of monopolistic in Canada, in a while, only a few providers.
But it was fun I did a lot of web hosting sort of exploring things.
I got, you know, wrote fake viruses started a few viral email chains or two just to see how far they would go and embedded trackers in them and things like that's things I wouldn't do now.
I mean is there anything you particularly miss about kind of that Wild West era that was was the early Internet you know security was was on no one's mind, it was really about, you know, we can connect these computers, great.
None of the underlying software had security in top of center.
You know, some of these experiments you were running anger in, I guess, the, the motions of a nerdy teen saying like oh what can I what can happen if I do this.
You know what what are there parts that you miss from that era relative to how the Internet is today or parts that are stories that you really enjoyed during that era.
I don't know I'm. I like the, I like the HTML elements you could add to make like blink was recently deprecated that was a sad one to see go.
But, um, I mean it was fun just having like a little thing that you could do and that your friends can do I think, strangely enough, the constraints that were in place due to like memory and storage and bandwidth cause people to try to be more creative.
So, you did have. You did have some, some different restrictions that allowed you to come up with newer ideas and have to fit things into smaller spaces so I when I was in junior high, I did I started my own video game company, and I remember I did.
I did a clone of Legend of Zelda I built this clone of Legend of Zelda and for like storage and space safety.
I made the maps into basically based on the text file, and then each of the different characters was like a ICO file and icon file.
And so it was just like extremely compact and fast and, you know, built my rendering engine from that.
But, you know, I didn't get at the time, the impact of, you know, rapidly increasingly increasing clock speeds and being able to tune things so if you try to run that now it just incredibly fast.
But, but that was really fun I think there's like business aspects to that I, when I started my video game company one of my friends started his own competing video game company and then he like licensed some Wolfenstein 3d style engine to, you know, build his little world and mine and, you know, got really into into that sort of thing for a while.
Fascinating. Did you release more than just that that one game or how long did that last.
Um, I mean it wasn't really a commercial endeavor it was, it was more for for friends and family but I'm probably had around eight or nine different games puzzle games, games to help games to help with dexterity with mice I mean this was a time where older people were just getting used to computers.
And so I would have these different games to like help my mother get more used to using a mouse and more accurate and moving it around.
Yeah, let's see.
Yeah, like kind of shoot them up type side scrolling games, things like that, but I did eventually.
Yeah, kind of give up on give up on this but, um, you know, once we once the portfolio grew to maybe three or four games I started recruiting people in the neighborhood like my parents friends kids became like my employees.
So like, it's, I was a natural manager from the, from the early days, I would say child labor laws and Canada must have been a little lax, we're like okay I need you to go code.
Here's what we're doing. Yeah, yeah, it was like child labor child manager child everything was, it's kind of a wacky situation but a lot of those kids end up getting into computer science and getting careers in that area so just, just from the exposure but I didn't put lock them in a room and make them program, that's, yeah, I mean to some degree that sounds like exactly what most kids want is like, you know, you want to learn to code so you can make games.
And I had plenty of classmates in my kind of computer science undergrad that that was their intent and when they were when they realized like writing code wasn't all about making games or there were so many steps in between.
Suddenly it was a less fascinating career path.
They wanted to go closer to, you know, design or closer to game design as opposed to raw programming and so to some degree I'm sure all those kids enjoyed having that opportunity of like oh my gosh, I can go start to learn how to write code I can make some of these games and then really decide whether I want to pursue that or not.
Yeah, yeah, it was a, it was fun. You know, I think I'm just thinking back to one of one of those kids I haven't thought about him for a while he's yeah he ended up getting a PhD in mathematics and becoming a professor so there's a lot of different areas for folks who are interested in purely abstract ideas.
Yeah. And, and so where did you go after this early era in game design and game creation what was what was the next kind of technical leap for you.
So, um, you know, just when I was playing in high school or whatever, I kind of slowed down on the building the games and whatnot but I would, I got in trouble in chemistry class for like programming on my TI 85 graphing calculator, doing stuff there but I kind of took a break from computers for a while.
I was a little bit of an intellectual brat.
From that perspective, I, you know, decided, you know, what can I do in university that is going to be hard and challenging and I don't want to be involved with computers or technology so I ended up doing a degree in in pure mathematics and writing proofs with pen and paper and chalkboards and I really didn't get back into programming at all or or computer things until.
Well I took. There's like one class where I got into fractals there was like a fractal class and so I started doing some simulations and cool visualizations and things like that but I, I really didn't get back into programming in a real sense until I started doing cryptography, and that was, you know, at the very end of my undergrad and then in my master's degree.
So, I actually don't have an undergrad in computer science, but I do have a master's in computer science.
Now do you feel that mathematics degree provided just a broader foundation for then what feels like an application layer of computer science so the using of math.
The structuring of that logic of data structures.
So, in my mind I see that only as additive. How did that look like kind of in your experience.
Yeah, it's a good question. Um, so I think it comes out most in the design aspect of building systems.
People laugh when I tell them this but my favorite part of programming is refactoring.
So, you know, taking something that's messy and then compacting it down to abstractions that makes sense and I think that's really driven from my experience.
Doing rigorous mathematics is, you know, theorems have to be shaped a certain way, things, things can fit together a lot of different ways but the logic has to be sound, and there shouldn't be too many loose loose threads.
And so that that's been helpful for things like API design, and for, you know, making programs that are comprehensible in in this security cryptography world as I mentioned I did a master's in cryptography.
It becomes even more important because you're really dealing with mathematical concepts and constructs all this stuff is just discrete mathematics sort of upper level undergraduate math, and it's really easy to.
If you come at programming from a more kind of organic history, where it's more like a language to you and more like you're trying to build things and put things together and squeeze things together.
You're going to end up with hitting some really bad security vulnerabilities.
When you touch things like cryptography or, or really a lot of security sensitive code.
So I think this focus on, you know, clean abstractions tightly described functionality was sort of a natural lead in for getting into computer security, because, you know, the messier things are the less secure they are.
And I mean cryptography alone as to some degree the language of the Internet or the language that secures the Internet.
So, the math that goes into that and it makes perfect sense that makes perfect sense to me.
Yeah, yeah. And it's still growing. It's still growing. You know, when I studied cryptography.
A lot of the things that we use today on the Internet were were kind of just in textbooks, in a way, or they were just starting to be implemented for the first time and implemented poorly.
So, we've learned a lot in that science over the last 20 years or so.
And in our last few minutes like I guess what then caused you to take a chance on, you know, company called Cloudflare when it was in its earliest days.
And what are you most excited about for the Internet both what Cloudflare is doing and what some of the other companies are doing out there.
Yeah. So, how did I get to Cloudflare that's, I mean it's a good question.
After, after school I ended up getting into the security industry. I worked at Symantec, and I wanted to get my hands a lot dirtier on actually implementing things so I ended up going to Apple worked in the iTunes org on security software.
Also, well, also known as DRM, so you may be questioning your why we couldn't copy our music.
Is that it, uh, there's certain aspects of it. Yeah. I can't roll your clock back and play a rental that we can iTunes rental.
After you, you've had it for a week or something like that.
A lot of these different services.
Um, yeah, I think we also worked on like the iBooks store. The, the app store itself like a lot of the security around those applications came down to some of the work I did there, which was great.
It was very empowering and impactful and people hated me for it.
It was great party trick to tell people that the reason you can't share your music is because I wrote some magic code in there like it's hard to reverse engineer.
But, you know, eventually, I think I was looking for a way to have an impact on the broader Internet, and specifically on things like open source software and standards and I was very enthralled with the idea of how from coming from a place where it's very like at Apple where things are very tightly knit, where all the devices are meant to work together and all the protocols are sort of custom and everything's, you have to kind of upgrade them together to some something like the Internet where it's just thousands of different players, working on different things connecting to each other with shared languages and how those shared languages evolve was always very fascinating.
And so, loosely coupled highly aligned. Yeah, exactly. And, and that method of working was very appealing to me.
I think, working in the open with open source software and then also being able to work across different organizations.
Doing interesting things and furthering the mission of making the Internet more secure was was something that I was really passionate about.
And when I was looking around for for a potential new role after the, the reverse engineer that was hacking Apple ended up quit quitting.
After five years we had sort of a back and forth between, you know, building a new DRM system and then they would come out with a reverse engineer of it that would be able to bypass the protection and we had this like five year back and forth about it.
And eventually they retired because their kids were growing up and they, they weren't old enough for them for the hacker parent to teach them about what social responsibility was, but they're also too young to like, say hey daddy what are you doing on the computer like they're, they're old enough to know that, you know, hacking iTunes or whatever.
But in any case, Cloudflare, I was introduced to some of the folks at Cloudflare from a friend, and it just seemed like a place that had a really big impact with a small number of people and people really passionate about, you know, making something big and making something impactful and it was sort of aligned with, you know, making the Internet more secure so that's kind of how I ended up at Cloudflare.
And are you a little surprised like what's become of everything 10 years later, like, just the where we've gone and that amount of time.
Um, you know, I'm surprised.
I'm a little bit surprised it's the growth, and the like progress has been so consistent.
I, I've seen so many different companies who've had massive ups and downs and, you know, really kind of revolutions within their area and disruptions and competitors and, and all the all these sort of things and, and, and Cloudflare's really just, it's been a, it's been a company that has evolved and has grown in a very consistent way.
There's definitely been parts of the journey that were bumpy, but, you know, I'm, I'm really, I, to be honest with myself I sort of foresaw where we are now, when I started because it just seemed like this, this market of, you know, helping every website, be secure was something that could grow significantly over time, and if it was done right, then it could be a really big part of how the Internet works for a long time.
And I'll use that as the segue kind of the final question what, what are you most excited about for either what Cloudflare is doing for the future of the Internet or, you know, some of the latest research papers you've been reading what post quantum cryptography, you know, really what's most exciting to you about where the Internet is going next.
Yeah, um, there's a lot of areas. I think, you know, in the research team, Cloudflare research works with new paradigms for distributed systems, privacy, enhancing technology, Internet measurements, Internet security and more.
And so, some of the, some of the areas that are coming up that are really exciting is, well, the TLS which is the handshake the encryption layer for pretty much all web browsing has had this kind of security privacy gap for a long time called SNI.
I'm sure our technical listeners will be aware of it, but we're just now rolling out encryption for that layer.
So it's something that's been around for over a decade, and has been like a big privacy gap and we've managed to collaborate over the last five years to get different browsers and parties together and work in specs on the ITF to, to get this out so this is, this is something that's really exciting.
I think generally the privacy space is really interesting.
Cloudflare is in a great position to provide products for folks who are putting their business on the Internet and and help make their relationship with their clients, more private, and to provide privacy tools for them.
And there's some advanced cryptography that leads to doing that and so keep your keep your eye on new privacy technologies I think they're really really exciting.
Awesome.
Well, Nick, thank you for your time and thank you for taking your dial up motive.
All right, Dan. Thanks very much. All right. Have a great one. Bye bye.