Speaker 0 | 00:09.564

deck net sort of things battles in the mid to late 80s and um you know sort of knowing the money the smart money was on tcp because of the fact that it was open source and it could be going anywhere any vendor could implement it but But we could talk a little bit about that.

Speaker 1 | 00:27.293

Give me some history on that just real quick for, you know, because we’ve got people that are alive now that were born after the dawn of the internet and things that many people take for granted. So, you know,

Speaker 0 | 00:46.000

what was really interesting, you know, if you go back into the mid 80s is you had. you had this nascent work with TCP IP starting to build out in academia. And so, and at National Science Foundation in 85, 86, up until like the early 80s, you probably had less than 50 to 100 places connected to the internet. You know, some really big computer science schools that were there. You know, you go back to the… mid to late 70s, it’s four or five schools. By the early 80s, it might be 40 or 50.

Speaker 1 | 01:28.208

What were the first five? What were the first five schools that you can think of?

Speaker 0 | 01:31.351

Oh, it’s like UC Santa Barbara, University of Utah, UCLA. I forget who the fourth one is, but it was sort of in that group right there. And then you had BBN, which was out of Massachusetts. And it was really just sort of testing out the idea. with IP that, again, you know, one of the interesting things on the internet to me is that, you know, when you look and we say, well, geez, the internet, the internet doesn’t have good security. And that was because it wasn’t designed from the get-go with good security. It was designed to be able to survive a nuclear attack. You know, up until that time, the whole idea was that you could take out critical infrastructure when you were doing point-to-point attacks with phone. where if you took out key switching centers in the U.S., you would break down all communications, and so you wouldn’t have command and control capability. And so TCP IP came out to be, well, what we need to be able to do is to reroute our communication. So if most of our traffic flows through Atlanta, but Atlanta gets taken out, we can automatically adapt and route. through alternate mechanisms. And that’s what the internet is good at. That’s the whole idea, routing protocols and certain things go down, but the internet itself stays up because traffic gets rerouted, et cetera.

Speaker 1 | 03:02.396

Is that still, I often think of this having worked in fixed wireless as a kind of wholesale backhaul, internet backhaul. I was selling internet backhaul for years at this company called Airband, where we did these large three-foot to four-foot long line-of-sight shots and running backhaul for whether it be cell phone towers or private government networks, etc. And I still have this, I still remember when we had an earthquake in DC and everything came to a halt because… None of the,

Speaker 0 | 03:45.456

like,

Speaker 1 | 03:46.797

the back hall just couldn’t handle the amount of people that were on a, on the cellular network at one time, which led me to think of like, well, I wonder if like, you know, what if someone was smart and they kind of really knew where all the carrier hotels were, which is not that hard to gather that information. And they just like, you know, took out, you know, 14th Street and K, they took out that carrier hotel and then they hit Atlanta. And then let’s say they hit, you know, Dallas at the same time. Would the internet still survive? Could we still technically take down the internet?

Speaker 0 | 04:18.651

So that’s a really great question. By the way,

Speaker 1 | 04:23.974

I hit record just in case this turns out really good. I think this is good enough that we could just keep talking and just make the show now.

Speaker 0 | 04:31.118

Well, I’m a little worried about that. I’m willing to say certain things off the record.

Speaker 1 | 04:37.222

I will allow you to pre… We all allow you to pre-scrub this. And the good thing is, is no one will really know what we erased or not.

Speaker 0 | 04:45.471

So, so, so thinking about the, what you said, there, there was a, I don’t know if you remember, but there was somebody attacked a set of fiber links. Oh, probably about a decade ago in the Silicon Valley area. And, you know, it turned out that it had some implications on the internet. Now it, it was one where ultimately it rerouted. Um, and, and people were able to sort of get things set up, but you’re going to definitely feel it. It’s not like it could, you know, if you take out big chunks and, and to answer your question, what’s happened over the, um, the interim period is the internet per se, what we all use may not survive some of these sorts of things. You know, if you, let’s say you took down, um, you know, certain areas, east, west, coast, and people know where these are.

Speaker 1 | 05:42.115

And you mean like 1X in Northern Virginia? Yeah. That would be problematic, I would assume.

Speaker 0 | 05:47.718

Yeah, it will have some problematic implications. On the other side, what you’ve also seen, though, is that a lot of groups have built their own private networks through this mechanism. So you have… Certainly the military has built a whole set of private networks.

Speaker 1 | 06:06.769

I would imagine Facebook.

Speaker 0 | 06:10.292

Facebook, Google, all these places are building private networks to sort of also be able to be in a survivable thing where they don’t necessarily have the same sort of all points of presence. They usually have connections there, but they may not be that that is fundamental and they can’t route around it. So to answer your question, yeah, there’s some spots. And. If you were looking at one of the other areas, there’s some very limited transatlantic, trans-Pacific links. And so the reality is certain areas, if those went down, they’re going to take down traffic to either Asia or Europe. And you could probably route around one of the two going down. You probably couldn’t route around. all of them going down simultaneously. And so that’s some of the challenges that, you know, we all have to sort of deal with is, is that, you know, it’s, the internet is good, but it’s, it’s not designed right now, I think to be able to take any sort of hit.

Speaker 1 | 07:15.532

Surprisingly, if you do look at the maps of the, do we call them subterranean submarine cables? It’s actually quite amazing how many there are now.

Speaker 0 | 07:26.318

It’s a lot. They have, well, and you know, what really has happened is, is being able to have longer distance between repeaters is really the key. And they’ve, they’ve gotten it down to the point where they’re really good at how they lay the cable and stuff like that. So yeah, no, it’s, it’s really interesting. Some of the stuff that’s, that’s going on in through that.

Speaker 1 | 07:48.602

So historically we have a TCP, this battle, can you describe that for a layman? or a young college student nowadays. Like when I got into telecom years ago, I remember we had to take like a historic, like a telecom history, you know, and we had to look at like, what are these little brick buildings that they call, you know, COs and central offices? And you took like a brief history and you got through this history class in about like a week or two weeks. But like, what would that look like now?

Speaker 0 | 08:16.116

Well, that’s a really interesting question. So, you know, thinking back, you know, the first sort of, Packet-switched networks, I remember, were what were called X25 networks. And X25 networks, it wasn’t TCP IP-based. It was something that sort of came out of the late 60s, early 70s. It had a little bit of packetization, but it was more point-to-point. But the idea was you had some error correction by creating packets. And so… You could sort of resend something if you had an error that was going on in through that. And these X25 networks sort of came about early on. But where I was going to go is as you get into the mid 80s, you had this idea that TCP IP was sort of starting to happen in the academic education space. And in 85, 86, National Science Foundation decided to create some supercomputer centers. And they wanted to be linking more universities to these supercomputer centers for being able to do academic research. And so they greatly expanded the number of universities that were connected to the National Science Foundation network. And that very quickly over a four or five year period ballooned, you know, where you probably went from. Before that number, it was less than 100. After that, it was probably up around 500 to 1,000 by 88 timeframe. And you started to see this, though competing against that, were what I would call proprietary commercial networks. So IBM has its SNA network that is linking IBM mainframes in multiple locations. And you had a mechanism that… allowed you to have interconnections that were linking to run jobs from a remote location and things like that. DEC had this very powerful network called DECnet that allowed you to connect and transfer files and move things about multiple places geographically. Now, you had to set up as a company these point-to-point networks. And so, you were buying And what’s really funny is you were buying links that were 9,600 baud or 56,000 baud. And really, that’s bits per second. And so it’s just so funny to be thinking like the first network that we set up here from UMBC to connect to the Internet was in the very early 1981 timeframe. And it was the PDP-11 that connected from… the Baltimore campus down to the University of Maryland College Park, which is just outside of Washington, at 9600 baud or 9600 bits per second. And, you know, we could sit there and we could do an FTP, we could do a telnet. Now, the telnet, you know, it was all like a teletype type thing. So it really didn’t matter. Nothing was graphical, you know, in through that. But, you know, those early stages of the network. You had these sort of nascent things using TCP IP, and then you had these vendor-specific mechanisms where you had DECnet or SNA. And what sort of came out of that was there was a protocol that came out of, I think it was governments, and it may have been even between Europe and the US, but there was something called X500. So if you go and you Google X.500. It was this protocol that really had been designed. It was pretty significant in the sense that what it reminds me of is of a programming language that you’ve probably heard of but never used. And this would be Ada. Have you ever heard of Ada? ADA? No.

Speaker 1 | 12:31.899

Most of this is way above my head. I’ve opened up a can of worms here that I need to.

Speaker 0 | 12:35.422

Anyway, what I was going to say is Ada was this programming language that the government said, oh, okay, we have Fortran, we have Pascal, we have C, we need this higher level programming language. And so they got a group of people together. And of course, what happens when you get a group of people together is you never optimize for the minimal viable product. You throw everything in and the kitchen sink. And you say, well, we’ve got to meet everybody’s needs. Right. And at that point, Ada had all this, it was so big, it frankly never really, it produced code that was so memory hoggish, if you want to call it that, that it never really quite made it. A similar sort of thing happened with X500. X500, everybody sort of added their wishlist into it, and it became so bloated that it was very hard to implement well. It wasn’t something that was highly robust. robust. And whereas the TCP IP stack was sort of mean and mean, it could be implemented on lots of different things. You really couldn’t do X500 as well. And there was just this period though, where the governments were sort of promoting this idea of X500 and watching the fact that, yeah, but we can’t really implement it well. And then quickly, you know, you had, you know, sanity return where people said, but… If we extended TCP IP with this or that, it would make a difference. And frankly, what really changed the dynamic was once the World Wide Web came out that was based on TCP IP, that was the end of it. You know, I mean, people saw the benefits of using TCP IP because the web just changed everything once that was released in late 91, early 92.

Speaker 1 | 14:34.542

By the way, if we do turn this into an episode, we are speaking with Jack Seuss. Am I pronouncing your last name correctly?

Speaker 0 | 14:42.607

You did, actually.

Speaker 1 | 14:43.908

Excellent. I mean, a whopping 40 years as VP of IT. I don’t know if you were VP of IT for all 40 years.

Speaker 0 | 14:53.954

No, I was not.

Speaker 1 | 14:55.115

And CIO at University of Maryland, Baltimore County. But a math major, is that correct?

Speaker 0 | 15:07.323

I was a math major, yes.

Speaker 1 | 15:09.384

My father was a math major also. Never graduated. He did three years of math at… Bowdoin College and then went right to medical school. Yeah, he was going to be an engineer and he’s like, I don’t know why I’m doing that. And then he ended up going to become a doctor with no college degree. So you used to be able to do that back in the day.

Speaker 0 | 15:29.320

You can still do it today.

Speaker 1 | 15:30.761

Yeah,

Speaker 0 | 15:31.422

as long as you know you have to find a medical school that will accept you.

Speaker 1 | 15:34.784

He went to Michigan for medical school. No, no,

Speaker 0 | 15:37.327

no, no. It’s back then it was that was pretty common that you would start and you would apply and if smart people just got in.

Speaker 1 | 15:44.993

If you got in, that’s it. See ya. So never graduated college, but he’s a doctor, MD. So we’re doing a history lesson that’s, you know, involving bloated coding languages, which is, that might be the, maybe we title this bloated coding languages that never made it, which is interesting. People don’t have that. I don’t know if, would you say that the modern day, I don’t know. web developer, web developers, not the right thing, uh, software dev, software dev teams. Um, um, they must have some appreciation for this history.

Speaker 0 | 16:24.119

They don’t. Um, and you know, it’s, it’s really kind of funny because, um, you know, one of the interesting things that I looked at it, so, so I ended up doing applied math and, and one of the, I had, I grew up first generate first in my family to go to college. So I’ve never, I came to UMBC. I was at one point thinking I’d be an electrical engineer started and I was working I was commuting and I was working 20 hours a week and it was just too hard to do all the lab and other courses but I liked math and so I was doing mathematics and I was going to do applied math and one of my teachers said well you ought to take computer science you know at that time computer science was in the mathematics department so you could take There was a computer science track and there was an applied math track and there was a theoretical track. So I was taking the applied math track and I started taking computer science. I ended up finishing also the computer science track, but it was only six courses. So it wasn’t like it was 10 or 12 courses. But the thing about it was that it really sort of opened my eyes to that was the first time I’d ever used a computer was sitting there doing that. Now. The first course that you took, you actually, speaking of history, we had to take a course where you had to do your programs on card decks. And so, you type up punch cards and then you get these punch cards, they’d run the job. And, you know, if you made a typo, you’d have to, you know, all right, now I got to retype that card, resubmit my job. You might get four or five runs a day because it could often take one or two hours between when you submitted your card deck to when you get your printout back. And so if you made a stupid error, you really hated it. And so what you got good at in those days, which is interesting from today, is what you got really good at is, one, checking your code before you submitted anything. You didn’t want to just sort of not make a mistake. Right now, the model is… compile and let the checkers just tell me, and I just keep fixing them as I go through the lines of code that would be there. So one, no, you didn’t want to do that. You wanted to find all your syntax errors right from the get-go. The second thing that you got really good at is you didn’t want to make dumb logic errors. And so you’d sit there and you would literally sort of plot out in your mind the code flow and how data is being changed. You’d have these big printouts and you’d be. making tables and sort of tracking how it should be going through your code because you didn’t want to have to be trying to debug on the fly. And so the world today, in many ways, there’s a lot of wins to being able to have very quick turnaround and very quick compilations. But I remember something in the late 90s where one of our Unix systems started to get a little bit slow that the computer science students were compiling on. And I got an email from one of them who said, It took almost 90 seconds for my compilation to finish. How do you expect me to be an effective computer science student? My compilation takes 90 seconds. How can I get this through quick enough? And I just had to sort of laugh.

Speaker 1 | 19:53.341

Let me tell you about it.

Speaker 0 | 19:54.222

I did an environment when I was going through it that, you know. 90 seconds would have been your dream. Sometimes it was 90 minutes, you know, and, you know, you just sort of learn to not make dumb mistakes. But, you know, what we did is we told, we apologized to the student. We told him we’re working on adding more capacity, but it was, it’s interesting sort of the change in the world nowadays.

Speaker 1 | 20:18.843

So what do you do now? So, I mean, really, can you be just as effective now as Was it easier to be good back then? Was it easy? Were changes, did change come easier back then and big leaps and bounds come easier? Or can we still look forward to more of that in the future? Have we hit the wall, so to speak? Are we almost through this non-industrial revolution, whatever we call this, whatever this is going to be written down in the history books after a thousand years from now when we’re all dead, what is this? Have we hit the wall? Are we still growing?

Speaker 0 | 20:55.797

Oh, most definitely. I mean, because I think one, so I’ll give you sort of two distinct differences. So one of the interesting things as I look at my career and what I’ve done, that’s positive for me, but almost impossible for someone new to do. But it’s also a negative for me, and I’ll explain. So I started out, my first job as a student, I was doing basically. COBOL and Fortran programming on a mini computer for the university’s administrative things. So I got this, started working with some sort of nascent databases, but it was pre-SQL. It was mostly administrative applications, some screen design, things like that. Got this sort of two years of, after graduating, got hired to do that for about a year full time. And so I got this sort of… understanding of what the analyst application programmer really does. My second job at the university at that time was we had bought a big mainframe and they needed someone to be a systems programmer. And so, since I was the one who had a computer science degree, I said, okay, I’ll do it. I’ve taken assembly language. At that time, systems programming was done in a lot of assembly language, a lot of low-level code. So, I ended up… becoming a systems programmer. I did that for about three and a half years. Really got to learn the innards of operating systems and low-level things. We got rid of that mainframe, went to a more mini-computer environment that was using Unix and, at that time, another operating system called VMS from a company called Digital. So, I got to do a lot of work. that was at a little bit higher level where it wasn’t around assembly line.

Speaker 1 | 22:54.025

Let’s give a shout out to Digital real quick. I mean, Digital.

Speaker 0 | 22:56.627

It was great.

Speaker 1 | 22:57.327

Digital out of Maynard, Massachusetts, like right down the street from me. I have a customer that I’ve worked with for years, Acacia Communications, recently purchased by Cisco, but they’re in the old digital space. Oh,

Speaker 0 | 23:13.260

really? Yeah. Yeah, we’re like between Massachusetts and New Hampshire. They had a lot of stuff in through that.

Speaker 1 | 23:19.726

That was like a big deal. I remember my brother back in the Q, they were for digital or Bose. It was like, come on. My brother, he graduated from Bentley. He was a business major, but I just remember, I remember as a little kid, him talking about digital and being like, wow, I guess it’s like a big deal.

Speaker 0 | 23:37.636

It was actually a great company. But again, it fell quickly.

Speaker 1 | 23:43.720

Yeah, that’s a great, so that’s why. What’s the lesson learned there?

Speaker 0 | 23:48.606

So the interesting thing was Simple beats Complex. And so what was interesting, because I saw it happen and it took down another vendor, and I’ll explain into that one, was so Simple was, DEC had come out. So if you look at the original DEC VMS systems, mid-80s, Because I used to actually, because I taught a similar language, and I actually taught it on the deck side too for a while. And VMS architecture was what they called a complex instruction set system. And the idea was you had like hundreds of specialized instructions. And so it made writing compilers easy. But what it did is the chips ended up running multiple instructions. And so you could be a little slower. And around the late 80s, you had a number of companies start to come into the marketplace. The most… The one you might remember most is a company called Silicon Graphics. Silicon Graphics, they were based upon this chip from a company called MIPS, which used a reduced instruction set model. And the idea was to have less instructions in your processor, but do that in hardware so they could be faster. And so DEC came out. They were trying to compete with sort of MIPS, and they came out with this alpha chip in the early 90-ish timeframe. And the alpha CPU was really sort of tried to be an amalgam of both. And it was just much more expensive than what you could do if you were doing a smaller number of instructions. And so DEC… couldn’t quite compete at the price level that you could get from vendors like, you know, MIPS and Silicon Graphics and others who were using the reduced instruction set model of chips. And so ultimately, that and the fact that their version of Unix, Ultrix, just didn’t quite win out. DEC was trying to be sort of not adopting sort of the full BSD suite and they ended up really struggling and quickly lost favor in the marketplace within like three or four years.

Speaker 1 | 26:39.537

How many companies are still stuck running on, I don’t know, DEC servers, operation systems, things like this?

Speaker 0 | 26:50.301

I have no idea. I am sure. So where you see it, is, and it makes sense. I mean, I, you know, I, we’re not too far from NASA’s Greenbelt location. I’m sure there’s probably something running some, you know, a satellite or something that launched 20 plus 30 years, 25, 30 years ago that might be on an older hardware. Cause a lot of times when they would launch these sort of things or build systems that are going to last. 15, 20, 30 years, or they never planned for them to last 30 years. You know, you could be built on older hardware. By and large, I think, you know, most of the world has moved on. We got rid of our last VMS system. We ran it, we would have gotten rid of it in 94, 95, but we had one faculty member who had a set of applications that he was running that we kept running. that system until through the early 2000s, just so that we could support him. And that’s the way you find it with a lot of places is, you know, you get into those boats. But anyway, that’s a little bit of the history. But anyway, where I was going with the thing and we’re diverging is I started in applications. I did systems. In the mid 80s, I ended up getting very, we launched our first Ethernet network. And so we ended up doing the thick Ethernet cable, if you remember that, and what’s called vampire taps, where you had this big pin that tapped into the coax cable. And we started setting up servers in different places.

Speaker 1 | 28:33.669

That’s great.

Speaker 0 | 28:34.429

I’m going through some of this stuff. Vampire taps. And, you know, and at that point, you know, we were first starting to sort of build out the campus network on here. And so for probably six or seven years. I did a lot of work sort of in the networking space of, you know, how do you run cable? How do you, you know, link routers, switches, other things together as we’re starting to deploy the campus network. In the early 90s, we went really heavy with large scale set of silicon graphics machines when we opened the building that I’m in now, the engineering building in 92. And at that time, I… did a lot of work in the Unix systems administration sort of world, but I also got involved in the World Wide Web. So I was doing a lot of stuff and setting up web servers and thinking about setting up diverse websites and stuff like that here. And then in the mid-90s, I ended up doing some work where we linked a legacy mainframe with a web, you know, basically, so that you could create a web portal so that students could be doing their add, drop, or register for classes. Web, but not have this. And so getting involved in some of that. Then I got involved in some high-speed advanced network projects. in the 97 through what was high speed back then it’s yeah it’s funny so um so what we were setting up at that time was high speed was um 155 megabits per second so it wasn’t slow um you know but it we we had at that time um a 45 megabit connection to the internet you And then we were moving to 155. And our long-term goal was to try to move up towards…

Speaker 1 | 30:42.275

What year was this? This was in the

Speaker 0 | 30:43.835

90s? Yeah, this was 155. I was smoking fast. It was like 98, 97,

Speaker 1 | 30:51.078

98. Yeah, that was real fast.

Speaker 0 | 30:57.021

And then, you know, subsequently ended up getting involved in… a lot of identity management stuff, and some other projects. So where I was going with this is that I’ve had a chance to sort of touch on a lot of different areas of IT. And it’s really hard for people to do that nowadays because you come in and IT is so much more complicated. You sort of get pulled into a specialization. And it’s pretty rare that you get to bounce out of that area of specialization too much. And so the… good news is that because we have such levels of specialization, you’re seeing jumps happen more quickly, but they’re happening in individualized areas. What you don’t have, though, is people who are used to sort of understanding the end-to-end systems piece of all this stuff. And so I think one of the good things that I have is, well, I don’t know everything nowadays. And- You know, we’re doing a lot of stuff in Amazon and we’re doing things in other places. And our network has now gotten, you know, our network connection from here to the Internet’s 100 gig. We all of this stuff, I have a good sense of how it ultimately works. And so when things don’t seem to work well, you know, I can ask good questions to my staff and say, hey. you know, what’s going on with this? Or why is this happening? And if someone says, well, I think it’s because of this reason, I’ll say, well, why do you think of that? And why wouldn’t it be this? Or, you know, and so I think the skill that I have is I can sort of tease out and asking people good questions and help them think through when we have problems, or as we’re trying to design systems, where we’re going to get bottlenecks. And so that’s my unique superpower is because I’ve had so many different experiences. I have a pretty good sense of how things work.

Speaker 1 | 33:07.468

I was blown away that Facebook or any providers or Comcast even, for example, still has major outages. That Facebook outage was a couple months ago. I was like, it can’t be the network. There’s just absolute, like if I was playing your role, I was like, there’s no way it’s the network, is it? Like, how could it be the network? It has to be someone made a weird change in a firewall or someone, someone made some kind of change or it was completely on purpose.

Speaker 0 | 33:40.786

Uh, you know, I, I don’t have the, I don’t have the details on the Facebook. I can certainly say, um, you know, human error. Often goes a long way in a lot of these sorts of things where you should have checks and balances. You should have tested something out. But, you know, you are so confident that it could never not work that you don’t. And so we’ve seen some outages like we’ve had some weird outages because we have load balancers and other sorts of things that are managing internal traffic. And. Every once in a while, you know, something will go haywire and, you know, potentially it could be because of the load balancer.

Speaker 1 | 34:28.687

Load balancers have been problematic in my history. switching over the different IP address or something like that, things stop working, weird stuff like that. Is load balancing archaic anyways? Shouldn’t we be using SD-WAN for everything by now? Shouldn’t we be aggregating bandwidth?

Speaker 0 | 34:48.468

So I would say that in a lot of internal activities, load balancing actually makes sense more so than SD-WAN. Yeah. Yeah, you may want to be spreading things out if you’ve got remote, but if you’re really trying to be thinking about how you’re managing service within this. Now, in the future, what you will see is we’re trying to take a lot. We use more load balancers internally than with up in AWS. And as we move to AWS, they have their own sort of set of tools for balancing how you manage load. And that’s what’s really sort of interesting is sort of watching how stuff breaks. I’ll give you an interesting example of how something breaks. And so this will be a story that I can talk about. So you may not know UMBC, but you know about UMBC if you’re a college basketball fan. Okay. Because UMBC was the 16th seed that beat. University of Virginia, which was the number one seed. And that was the first time that that happened. Okay. So this is 2016 when this goes down. And while the game is going on, UMBC is up by like 10 or 15 points at halftime. And all of a sudden people are, who is UMBC? And so everybody’s going out to our website.

Speaker 1 | 36:23.254

It was like a self-proclaimed DDoS attack.

Speaker 0 | 36:26.315

An aisle of services. So what sort of happens is that we had actually provisioned a lot of our website out in AWS to be sort of thinking about some of these things, but we didn’t have everything out there. And so we had, what was funny was, is that, and this just sort of shows how things can go. awry is we had some content that was on the main website that was linking to this news website that one of our groups runs. And when people would click those pages or click the links, they were getting to this small server that was on campus that was running.

Speaker 1 | 37:19.233

Like posting another page.

Speaker 0 | 37:21.614

You know, if it gets 20 hits in an hour, that’s a…

Speaker 1 | 37:26.756

The third string basketball club.

Speaker 0 | 37:28.837

Right. Yeah. You know, and also all of a sudden, you know, everything in the web environment was hanging while it was waiting for this news website to sort of try to respond to things. And so we were quickly trying to fix stuff in the middle of the game, you know, because it was like, wow. And we won. And so we spent a lot of time that night. because we were going to be playing a day and a half later, completely re-architecting the website. But it just sort of goes to show you that, you know, if you haven’t sort of looked through end-to-end solutions of how things are going to flow, especially in strange situations that are not normal.

Speaker 1 | 38:09.531

That’s a really cool story, actually. That had to have been like a fun break. Pardon? Was that a fun break? Like a fun break.

Speaker 0 | 38:17.495

It’s a nice story to tell because we won the game and it was a positive reason for having a break.

Speaker 1 | 38:25.522

That’s just like, it’s just a really cool story. I don’t know what else to say. Everyone had to have been proud that we’re on the IT department and that this broke down for this reason.

Speaker 0 | 38:37.372

You know, it was nice. We were proud of our team. Let me just say that.

Speaker 1 | 38:40.975

How about the person that had that news server? Who was running that? Was it some kid?

Speaker 0 | 38:44.998

It was a news group. And since they do a lot of the marketing and communications, they were very helpful in helping us move the content to different servers. So it ended up working out fine the next day, but it was kind of a funny story. This is awesome.

Speaker 1 | 39:01.552

We’re getting killed.

Speaker 0 | 39:02.813

We got so many hits, we broke the web server, right?

Speaker 1 | 39:06.395

That can still happen. That can. I’m hoping that’ll happen with maybe that’ll happen with dissecting popular IT nerds. Maybe my web hosting will shut down and I don’t need, you know, I’ll need to, to beef it up some, someday something will happen.

Speaker 0 | 39:20.009

So what’s a topic that you’d like to, to jump into real quickly? And we can just spend a minute and we’ll see if this makes, makes sense. So pick a, pick a topic.

Speaker 1 | 39:30.164

among your four themes well how do we help new it leaders sell it as a business force the guys are stuck as a i mean because a lot of them are stuck in it is still as kind of a cost center we’re paying for point-to-point circuits we’re paying for server refreshes all matters little now in this world and people are struggling um making the the paradigm shift of it is no longer a bill it’s a it’s like oxygen we need it so there’s this fine balance between um being able to grow you know grow it and no matter what organization they’re in no matter you know, making the necessary changes, but having to sell those changes, right? So how, I guess, you know, with 40 years of experience and seeing how much, you know, the world has changed, right? People used to laugh at Jeff Bezos, right? Working something out of his garage or whatever, and he’s not going to sell, why is he trying to sell books on this internet thing? You know, Sears was laughing at him. No one would say that now, right? But people still are. So I don’t know what you have to say about that. It’s either that question or I really also really want to just ask you, what’s the future going to look like 10 years from now from your perspective? Because you have more experience than most people. I’d much rather say like, hey, is everything going to go wireless? And is that going to be safe? Or, I mean, you have so much perspective. I also want to ask you, are we screwed on security? Are we screwed? Because security seems like we’re screwed. It really does. From a security standpoint, it seems like it’s much more easy to be the criminal than it is to be the CISO thing. So those are the three. You pick one. You pick one and we’ll go from there.

Speaker 0 | 41:38.674

What was the first one again?

Speaker 1 | 41:41.136

The first one is kind of, it’s typical, right? Every IT manager, leader in the world.

Speaker 0 | 41:45.738

Let me answer the first one really quickly. So to me, at the end of the day, what I would say to any emerging IT leader is… You don’t sell IT for the sake of IT. You sell IT as delivering business value. And if after the pandemic, you can’t sell IT for delivering business value, you probably shouldn’t be the leader of IT. Because almost every business has had to rely upon virtual, online ordering, you name it. I look at the university. For almost a year, we went entirely remote, entirely online. We had never done that, but we did it. And that delivered intrinsic value that if you would have asked me four years ago, oh, if a pandemic comes up, will you be able to run everything online for a year without having to do… anything where you could come onto campus, I probably would have said no.

Speaker 1 | 42:59.454

But why? Why would you say no? Why would you have said no? Because like what things were not integrated? That what things were you able to do this year that you weren’t able to do back then?

Speaker 0 | 43:10.980

So I think the huge leap that came about in that five-year period was just much better desktop video conferencing. Because we started launching video conferencing in the early 90s with some of the early, you know, sort of bigger systems between campuses to do remote classes and things like that. And then in one of the groups I work with, Internet2, we started in 2008 or 2009 launching some remote desktop video conferencing tools. that some of the communities like the high energy physics community had some tools and you started to see some companies coming about with desktop video conferencing. But boy, it just didn’t work seamlessly. And even if you went to, we had bought a product in 2014, 15 that we were trying to deploy. But again, it just didn’t quite work perfectly. By the time you’re at, you know, 2019, You’ve got Zoom has been out. You’ve got Cisco WebEx was starting to be happening. Microsoft, Google Meet. But all of those vendors were able because they’re, and what was really the key is, and this is something to be thinking about. A lot of these other products were designed to be products where you ran software on the mini computer. You know, you downloaded something and you ran it. And so you couldn’t. innovate. at the speed of cloud. To me, what really changed the dynamic in the pandemic was we had gone to all of this web-based video conferencing. And so, Microsoft Teams, Google Meet, Zoom, WebEx, yeah, they may have a small client, but they’re really doing the innovation engine at the cloud scale. And so, they had built that out. And to me, that’s the secret thing that survived, that we were able to sort of move to video conferencing because the cloud could scale and enable it. And it was doing it at levels that no one ever anticipated. And so I think that’s, frankly, the secret thing that allowed a lot of stuff to sort of happen. That you would have had all these people go remote was really a shock to me. But basically, Zoom and other things allowed it.

Speaker 1 | 45:46.756

Okay. Question one or two, you get to choose the next one is, are we screwed on security aspect or where are we going to be 10, 20 years from now? Are we going to be all on the wireless? Are we going to be wireless network, more wireless network or, you know, from a, from an internet standard kind of protocol? What are we doing? Which one?

Speaker 0 | 46:08.648

I don’t want to guess that one. That isn’t necessarily my level of expertise. I know a lot of people think you know, 5G or successor 6G or whatever the heck they come in the future, they’re going to call it is going to be the way that everything is done. And I think that for most people in the, what I would call home sector, that’s always going to be the case. I think there is a benefit of wire. So at least in enterprises or for connecting enterprises to the backbone. let me say that it’s to me, I’d much rather have fiber connections than private networks,

Speaker 1 | 46:47.174

private networking, even if you’re going to be off the grid.

Speaker 0 | 46:50.056

Okay. So, so that one’s right. So the security piece, what I would say to you on the security piece is that I think in the short term, we are, I wouldn’t quite call us screwed, but we’re in a difficult spot. The reality is, is we have not built. all the things that we need to build to do it well. Where I see hope that says we’re not completely screwed is deploying multi-factor authentication makes a huge difference. As we’ve deployed multi-factor, that helps a lot. I think that what we have to be, one of the projects I got to be involved in, and this was where Perfect got in the way of of better is I was involved in a project that the government was trying to do called the National Strategy for Trusted Identity in Cyberspace, NSTIC. It got launched in 2013 under the Obama administration, and it got killed when the Trump administration came in in late 2016. And the challenge that you had was that we were trying to do too much at one time. And because we wanted not just security, but privacy and other sorts of pieces all built in at exactly the same time, it was really hard to be able to have everyone come to the table and say, we can deliver that. And so we weren’t able to deliver it. But the reality was there were services that we could have started to do. that would have been better. And we had a vision for things. And I think that you’ll be seeing a lot of those elements start to come back over the next few years. It was around, how do you take advantage of multi-factor? How do you take advantage of the fact that you can identity-proof people in a variety of ways? By identity-proofing, you’re able to say, I’m actually Jack Seuss. And Jack Seuss really isn’t my name. My real name is John Seuss, but I go by Jack Seuss. And this is my email address. And here’s other information that is going to make sure that if you want to be connecting with Jack Seuss, who really John Seuss, who goes by Jack Seuss, you’ve gotten the right person. We have the tools in place to do that. We just have to have. an understanding of how to grow that and how to make that happen. But I think coming out of this, you’re going to see things start to come into play that give me hope. So I don’t feel we’re completely hosed. I think if we can improve identity management. We get more multi-factor going. You’re seeing some of the elements with zero trust start to come into the marketplace. All of these things are really good elements that are going to help us a lot in the future.

Speaker 1 | 50:11.408

What’s your general, last question, general emotional thoughts on social media?

Speaker 0 | 50:20.460

I don’t really make heavy use of social media. I follow a lot of people on Twitter, but I don’t really tweet. And I will occasionally follow people on Facebook, but I don’t really post. And it’s mainly because I just don’t believe in giving more of my information out for people to better market to me.

Speaker 1 | 50:54.942

True. It is a marketing tool, ultimately. I’m always surprised when people are surprised that when something comes out, whether it be over the news, they’re like, they have your information. Yes, they’ve been pixeling you forever. I find it surprising that it’s not common knowledge that everyone knows this.

Speaker 0 | 51:18.272

Yeah, I think the data aggregators are really pretty amazing. And frankly, I don’t know that it helps me much by not posting much to these sorts of things. But I’m very judicious in what I do and use. And so that’s how I sort of manage things myself. But I don’t know that that’s a best practice. That’s probably just the fact that I’m an old person now.

Speaker 1 | 51:48.492

This has been a pleasure. This entire conversation has been a pleasure. I think there’s something to be said with, you should be teaching history class. You should be teaching history class and it’d be a lot of fun and people would ask a lot of questions.

Speaker 0 | 52:02.976

So the best class I ever got to teach part-time in was a class in the mid to late nineties where it was with a faculty, a professor who was a sociologist and anthropologist, and we taught cyberspace and society back in 95, 96, 97, 98. And that was just a really fun class because you were bringing in a lot of students who weren’t computer science students. These were sociology students, some grad students, some undergraduate, and trying to think about how, seeing how they were thinking about cyberspace, society, um, how they hoped it would go was just really enlightening. And, you know, for me, I just recommend that, you know, we’ll get past a lot of the problems. I mean, the good thing that I believe happens with technology is there’s an immense number of creative people who are out there, both with economic opportunity to make money, and because they care, who are trying to solve a lot of the problems that we see. that gives me some hope thank you sir