Leon Adato - IT Origins
Leon Adato: So I've been in IT for 30 years. I focused on monitoring for about 20 of those 30 years. My degree, full disclosure I have no actual training or background in anything remotely IT. My degree is in theater and NYU undergrad drama Tisch School the arts Greenwich Village 1985 to 89. Go Bobcats, fighting violence, fighting violence. It was very 85 to 89 Greenwich Village. So that was my degree. And coming out of college I discovered that the world didn't need one more short Jewish nebbishy looking actor who wasn't nearly attractive enough to be an ingenue and not nearly as buff and buff enough to be an action hero. So after being a temp secretary for about two and a half years I sort of liked tinkering with things even though my training is in theater. I spend a lot of time in the technical side of it doing lighting and light boards and computerized light boards were just coming out at the time. So that along with the temp work that I was doing showed me that, hey, computers are kind of a thing. This is-
Rich Stroffolino: This is going to catch on!
Leon Adato: This is 89 and so I got my first job at a training company which at the time to be a software trainer a computer trainer really required two things breathing and a suit and one was optional. Actually a lot of the classes were taught by what I affectionately called green sock blue sock people you know they don't know that they're wearing two different socks and it doesn't really matter. They can talk bits and bytes but they can't exactly put a sentence together. So I was the flying pink unicorn of training because I was pretty much coming at it from the same perspective as the people in the class. I didn't know Paradox, the week before I taught it. I didn't know you know DOS basics or WordPerfect or whatever. And so my process for five and a half years was show up on Monday install the software play around with it Wednesday start writing the manual get the first draft of the manual kind of down by Friday come in Monday teach the beginner class of that class revise the manual on Tuesday and then just throw it into the pile and once you do the beginner class and you did it like five or six times you had questions in class to say oh that should be in the intermediate class. And then you'd build the intermediate class. I did that for about five and a half years for various training companies around the Cleveland area. And then I was ready to stop being a train-o-matic because the reality was at that time we're talking now mid 90s I could have taught a four hour windows basics class twice a day every day seven days a week and never satisfied the demand. It just you know there was something unintuitive about a mouse or whatever. And people were nervous about it. So it just got tiring after a while. And I had learned so many really cool things from the questions people asked in class and I wasn't, again I had no airs about you know, well I have to pretend I know this like I didn't know anything and I was happy to say I didn't know anything. So people would say hey how do you do this thing mail merge from a database or whatever. I have no idea, but that sounds like an awesome question.
Rich Stroffolino: Yeah, let's find out.
Leon Adato: Let me find out about that, I'll get back after a break or whatever. So after five and half years of that I was ready to start doing more so I got into desktop support and worked my way up the IT food chain in the larger organizations in Cleveland from desktop support to server support to, you know, networking. And then I fell into desktop standardization. There was a big project I was at Nestle at the time and they were standardizing all the machines to the same operating system, same install sort of the beginning of roaming profiles like those didn't exist before. All this stuff. So I was figuring out what software they were using and how to get it installed on a server and how to make it you know server based instead of you know each machine had to be installed as a custom work of interpretive dance.
Rich Stroffolino: And what kind of scale was involved here?
Leon Adato: [00:04:12] That was about 300 machines in the Cleveland office. But the US headquarters was about 10000 machines and I wasn't part of the whole group but I was working with them. So I did that and that got me into a little monitoring tool called Tivoli. Tivoli had just been bought out by IBM and they were still doing their own little show, planet Tivoli, and they were still their own thing. And so Nestle had bought that tool but not trained anybody about it. They were sold as so many modern tools are sold. Somebody told them, oh no you just install it and it runs by itself. You never need to do any support or any maintenance or anything like that. Making matters worse they had three different teams using the tool one for software distribution one for inventory and one for actual systems monitoring and each team was making changes to the core framework without telling the other team. So after two years and the investment just going nowhere and not working particularly well they said Hey Leon here's the agreement we're going to on our dollar let you learn how to use this stuff and you're not going to leave. How's that sound. OK so you know I learned a lot about it. Got the U.S. installation working with a team of about six people and they said hey this was so much fun let's do it let's take this global. Let's go for the worldwide tour. And so they moved my family and I to Switzerland and we were there for about a year. And we set up Tivoli for 250000 systems in five thousand locations around the globe.
Rich Stroffolino: Wow.
Leon Adato: So there was that. And that was the beginning of my monitoring career. And at that point I was already realizing that monitoring itself is best a heterogeneous not a homogenous solution. You need multiple tools no single tool is going to do everything you need round about 2003 I started using Tivoli for a different company. They wanted to build a monitoring as a service. And so I started doing that. And we built it out of [00:06:31] SolarWinds [0.1] and you know as I moved from company to company the typical two to four year migratory pattern that IT people have I would learn other tools whether it was Patrol or OpenView or Nagios or whatever it was. And you know but Tivoli always held a special place in my heart. It was you know it was at the lower end of terms of cost. It was at the higher end in terms of usability and it had this insanely active user community. You go on the IBM forums and it's like it's broke helped me fix it. Here's the fix. Thanks bye. That's the entire you know and on thwack.com I found a community of people who not only talked about it's broke, help me fix it but also hey I'm setting up a router I'm not really a router guy and what are these ACLs. Can someone oh here's a tutorial here let me help you and it's not the forum leaders. It's just other people. And then there was the conversation about you know who's the greatest starship captain. You know what's your favorite color- blue. You know like all these-
Rich Stroffolino: The universal forum questions.
Leon Adato: Yes it was. It was amazing. So I got really involved in that and coincidentally at that time I was working for another company that was creating a monitoring as a service offering and I was leading that project and it was built out of Tivoli- built out of SolarWinds. And so I had found some rather creative ways around some of the shortcomings that SolarWinds had, all software has it. And so I posted it and that got a lot of attention. And then around the same time SolarWinds started playing with the idea of doing a yearly convention. Originally they wanted to an in-face you know come in convention and that idea didn't take off. So they said we're going to do an online one. We're going to do it completely online free no travel no sign up no registration convention, who wants to do a talk? Oh I'm an attention whore. I like, yeah I'll talk about anything. So I put in a couple of ideas. And they said Sure do them all. All right. So they flew me down there and I did these two talks pre-recorded and what I didn't know is it was my interview before my interview it was like they were saying. This guy writes He's got his own blog a couple of them. He seems to be fairly outgoing. He's not afraid to be up in front of people. We have you know this head geek thing. Let's think about him. And so Patrick Hubbard and Lawrence Garvin were the head geeks at the time. And Tom LaRock. And so they talked to me about it and I talked to some of the other staff. And you know they said Okay great. So it honestly when I told my wife about the job she said and do they know that you'll do this for free. Do they understand-.
Rich Stroffolino: Those fools!
Leon Adato: Wait let me make sure I get this straight. You get to play with software before anyone else gets to you then get to stand on stage in public and brag about how you get to play with software before anyone else gets you. You get to write. You know as much as you want whenever you want. You get to make goofy videos like did they just make a, did they say what's a Leon Job. So it really is like perfect.
Leon Adato: Well that's sort of the functional like how I see it. Really, the head geek role, technical evangelist is probably more accurate term you know technical marketing whatever. But really it serves three main functions. The first one is externally facing cheerleader evangelist, right. And the great part about it for me is that I don't have to sell a thing you know the shtick is hey monitoring is amazing and you can do incredible things for monitoring. It's good for your company and oh by the way the company I work for makes some really good stuff but that's not important. What's important is that you understand that monitoring is really good for you and you should do some. So that's the you know the raw raw piece the internal pieces I turn around back inside and I'm also the voice of the customer both because I was a customer and because I get to interact with customers and say as new features are being proposed or new ideas are being developed or new workflows without having to worry about NDAs or tipping our hand or whatever it is the head geeks serve as that internal let me look at that with my customer hat on that smells kinda BS-y. Let's try something a little deeper or oh my gosh take my money now. You know you've got this really those are going to binary options right there's no in between.
Rich Stroffolino: There's no spectrum.
Leon Adato: No no no. So we can do that. We're sort of a safe space for developers. And then the third piece is what I affectionately call the edutainment whether it's giving conference talks or doing SolarWinds lab or any of those things getting up and solving a real problem. So it's not hey we have this piece of software and let me show you how to use it. It's Don't you hate it when your router won't help with net flow when you can configure it. Let's talk about how to do that. And I always tell people that when you watch This Old House you're not thinking about I need to buy some Craftsman tools. And yet every tool on this old houses Craftsman tools but that's just what happens to be in Bob Vila's toolbox. You know that's just happenstance so I'm talking about net flow and the tool I happen to have is SolarWinds' NTA. Okay great but it's not because NTA is the only tool it just happens to be the tool I have in my toolbox. So we get to do that also. So like I said it edutainment piece.
Leon Adato: It existed, it clearly existed before I started in 1989.
Rich Stroffolino: OK.
Leon Adato: You had old, old tools like Forest for the Trees which was on a single three and a half inch floppy and was supposed to look at all the data on your hard drive whether it was spreadsheet data or database data or whatever it was and sort of put the pieces together and give you a single pane of glass. You know that idea has been around. I'd have to do some internet research but I really do believe that it was long ago, so almost as soon as I came on you know Sidekick which was a wonderful tool. I still consider loading DOS 4.11 And WordPerfect 5.0 And Lotus 123 just to see how fast it would run on my current machine and Sidekick and QEMM because you need memory management. Sidekick, the idea was that you could flip between multiple programs simultaneously. It was a task switcher and the term came in even there.
Rich Stroffolino: Do you feel like there's been any inflation in the use of that. I mean from my perspective, the reason I bring it up is it seems like that term engenders a very visceral response from the Tech Field Day audience when they hear it. So I was just wondering from your perspective if there's been any kind of inflation in the use of that term or is it just always going to be present whenever someone must talk in grandiose terms about a software monitoring or any kind of software dashboard.
Leon Adato: Right. Well the challenge is that it's incredibly descriptive of what people are trying to get at that they're trying to give you a single view and a single pane glass just seems to roll off the tongue. It's great sales speak whatever I think the reason why Tech Field Day has such a visceral reaction is because they don't have to deal with object oriented as a visual thing to react from or you know tabs versus spaces or.
But the reality is that it's hard to get away from because it's so darn descriptive, it's so darn perfect for what vendors like SolarWinds and everyone and it doesn't need to be a monitoring company it can be any other software solution. But when I'm giving you everything that you want to see right now what else do you call it.
Leon Adato: Right. Right. Yeah. And my first computer which had 720 K of RAM, 360 Bill Gates said that's all you need.
Rich Stroffolino: You're really just being selfish at that point.
Leon Adato: Really, now you're just showing off anything more than a 5 megabyte hard drive. You know that's just because you need to put all your porn somewhere. So the biggest change so I'm going to answer it twice because why answer a question once if you can answer multiple times. I think the first one from a very technical standpoint is the convergence of functionality whether or not it's on your cell phone or it's on a server or it's in the cloud or whatever it is. We've talked about conversions for a really long time. I do remember the [00:17:16] Compaq American Tourister gorilla luggables, [2.4] they call them portables. And the joke was they're portable if you are Arnold Schwarzenegger or the [00:17:24] American Tourister gorilla, you could carry it around. [0.8] Which came with two five and a quarter inch floppies and a keyboard that was on a telephone cord and red gas plasma screen and a phone they put a phone on there because like the idea of convergence has been there forever that I have these separate functions but they're all transistors they're all circuit boards and what if I could get all the circuit boards into one place that's been around for a while. But getting that to happen and also getting it to happen from a conceptual basis. There was a great part of a keynote that Martin [00:18:52] Casado [0.1] gave a couple of years back where he talked about the idea of convergence and the idea of function's moving from hardware to software and it was in the context of networking and context of like the big metal box that you would buy and you'd stick it on a rack and you do things and now it was becoming software. So network is code was his point. But he said you can see that in going from [00:19:30] Garmin Nuvi, [0.1] and he holds up his cell phone. Once upon a time you had to buy a specific specialized box with a specialized circuit board and specialized software that had to be updated in its own way. And it used closed market technology to tell you when to turn left and to tell you how to get places and lots and lots of people had those little boxes in their cars to tell them where to go. Now all that functionality is in that little rectangle we keep in our pocket that does all these other things too that it didn't just become miniaturized it became completely subverted into software. So that from a technical standpoint the idea of convergence and the idea of functional convergence into software where everything eventually becomes just code and doesn't need to have a discrete hardware component I think is the biggest technical shift that I've seen but there's been this other shift which is not done by any stretch of the imagination. You know you want to call it the democratization of IT. You want to call it the rise of dev ops culture you just want to call it people learning to get along better. That's something else that I saw when when I started in you know 89 the Crystal Palace which is what a lot of people called the Data Center because it had this big glass screen and you saw people in the fishbowl and that was the, it was a Crystal Palace because all the things you needed had to be requested you had to go up on high to the data center and ask the people in there if you could please please please have a report. And in a couple of weeks you would get it and you know then it was you know don't bother me anymore unless you need something else and everything basically was a software project you know a report was a software project and you know a set of data was a software project etc. I was coming in at the very tail end of that era. PCs were already on the rise and that was becoming a great again democratizer because all of a sudden people were saying oh no no no I have a PC on my desk, don't touch it just leave it alone. I know I have a terminal but I can do some stuff myself and I'll do it. And some of the real you know cutting edge folks were doing that and then PCs were moving into the workplace en masse and they became a IT supportable thing and so on and so forth I don't need to repeat 35 years of IT history. The point though is that even in those early days where we're coming off of the data center mainframe [00:22:05] VAX [0.1] environment into PCs and sort of servers as we sort of knew them then with [00:22:11] Novell [0.4] and things like that there were still silos you had the network people and you had the data people and you had the developers and now we're seeing those lines really blur. We're seeing a lot in monitoring. That's my perspective on it because once upon a time you had the network and the network had its monitoring stuff and you had the server folks in the server and they had their monitoring stuff and the storage had their monitoring stuff and so on and so forth. And if there was a problem it was not me not me not me finger pointing, finger pointing until finally someone, it was a game of musical chairs and one chair would you know what someone is left standing and they had to figure it out. Now again with the rise of the dev ops mentality with the rise of just people I think shifting from you know I started off a network but now I'm in application design or I start going off in database and now I'm in you know networking or server or whatever it is you know people have moved enough that they have a sensitivity to know problems or multi modal you know it's not the network or the database (it's never the network). It's not you know there's a piece of this that is contributing. I think also cloud because again everything is code the network is code and the database is code and the server is code and you know the only thing that's not code is the data and sometimes the data is also very code-y, but anyway. The idea that you know the problem is the problem and the user is having the problem and the user doesn't care whether it's you know the storage or the network or whatever it is they just want to get back to work again and more and more people are taking ownership of that concept. They mean you know there still maybe silos in the company but they're more semi permeable than they once were. So again the two things that I've seen change is the convergence both on a physical technical standpoint and also the philosophical standpoint. And actually I guess that's the same thing convergence in terms of work function that more people are willing to say yeah I'm in the network team but I'm going to troubleshoot this database issue or I'm going to go troubleshoot this you know storage latency problem and see if I can use some of the skills I've learned to bring to bear. So yeah convergence.
Rich Stroffolino: Well that's very much at the heart of kind of the philosophy behind Gestalt IT is that all of these you know silos are kind of coming together and if nothing else people need to know what's going on from a news and media about these industries if not need to know how to work in them at some point or another or that their work you know they're no longer the storage admin they're the admin. You know stuff like that. But you know that was a very high minded very aspirational shall we say. So let's throw some shade. What is the current worst trend in IT?
Leon Adato: Single pane of glass! Ok no. There you go. Boy there's so many bad things. I think the hype cycle has always been there but I think it's worse. So just the hype cycle in general is the root cause of a lot of this but the current victims of the hype cycle are bitcoin and block chain cloud all the things you know. Yes cloud is incredibly useful. There are certain things that are not going to are not reasonably clouded and actually SolarWinds did a survey last year and found that of the projects that were moving to the cloud 20 percent were moving back either because the cost saving. The cost wasn't what they thought it was going to be. Meaning it was much more expensive. The security wasn't what they thought they thought it was and they couldn't handle it. The functionality the performance wasn't what they thought it was they're bringing it back. So cloud all the things which we see in a lot of companies is another one. AI and machine learning first of all the complete ignorance of what those words mean. The difference between those where AI and ML. You know it's almost like you know one of those ships you know Hollywood ships right. So it's like you know like Bennifer right like AIML it's just one word.
Rich Stroffolino: And, powered by block chain in the cloud. Exactly.
Leon Adato: And there we go we hit it.
Rich Stroffolino: Bingo!
Leon Adato: It's the quadffecta! So I think that the hype cycle. But again the current victims AIML which is not not a thing but I think that most businesses have no use for it. They need just a really good algorithm and be done with it. And you know so you see but you see a lot of executives saying you know we're going to do this magic thing and it's going to fix all of our stuff. But you know we saw it with SOA the software oriented architecture you know I'm just gonna buy a box of SOA one and then sprinkle it all over her,just sprinkle that crap everywhere, this is gonna be great. So that's I think the the worst trend again the hype is at the root cause of it.
Rich Stroffolino: Now it's curious you use the word victim. Is it my reading of that is that these are interesting perhaps worthwhile technologies but that people are going to get so burned out on these that then they just turn off their ears whenever they hear block chain on anything.
Leon Adato: And when you say people we mean IT people.
Rich Stroffolino: Yes. Yes.
Leon Adato: Yeah. Because the execs are all hot on it because it is the new hotness. IT people are tired of trying to A, figure out what the exec is talking about B, parse it out for them in functional terms what it means to our business then be told no no you don't understand. This is going to be awesome. Somebody on an infomercial told me so you know or you know the other part of the worst trends is the we want to be just like.
Rich Stroffolino: Oh yeah...
Leon Adato: Apple Google Netflix Hulu whatever.
Rich Stroffolino: The Hyperscaler effect yeah.
Leon Adato: Right we want to be just like them but we don't do movies we don't do anything like what they're doing. But yeah but you know we sell widgets but we want to be the Netflix of widgets. What does that even mean. How can you say that.
So and I do mean victims. I do mean that these are good blockchain is a good technology. It does have worthwhile use cases. The same thing for AI I just heard this week that IBM ran an AI that had a debate with a human debate either team or individual and it was judged by journalists and the AI won one and lost one. So we're you know it does have whether or not that is the perfect use case for it is immaterial. The point is it's not nothing it's not vaporware but it's not everything. And again hyperscale or your SMB. I run a bakery. I don't exactly know how this is going to apply to me.
Rich Stroffolino: Just need that point of sale work. Yeah exactly.
Leon Adato: Exactly. I need a website that people can find and a point of sale that doesn't throw up on me every three transactions.
Rich Stroffolino: We had an interesting talk on a couple of episodes ago with Ted Dunning of MapR. He's their chief application scientist and his perspective on AI I thought was very interesting is that it's constantly either aspirational or blasé in that what is the groundbreaking what's the IBM debater of today becomes the boring feature that Google implemented tomorrow and that there is no. The problem is everyone kind of wants to be in the middle of that and it moves so fast that it's very very tough to be there.
Leon Adato: Whiplash commoditization.
Rich Stroffolino: Oh wow.
Leon Adato: It goes straight from you know oh my gosh we just discovered it to two dollars and 35 cents per minute. Like in seconds it's like oh code. Got it. Let's go install a new widget. Here we go.
Rich Stroffolino: Let's spin it up. All right so let's be happy now. We've thrown the shade. What's the best current trend in IT?
Leon Adato: I actually think I'm gonna go back to the things that have changed the most. I think that the convergence both the workplace convergence. I absolutely believe that the more empathy that we can build for our IT cohorts and beyond. I'll talk about that in a second. The better IT is in general and that the more things converge down to again a little rectangle in our pocket or the cloud or whatever it is I think the more exciting things we're able to do. I think that commoditization that fast commoditization is also good. I don't think anybody really enjoyed the six months it took to learn XML and Salt and all that only to find that it didn't actually do what they wanted it to do or in the language that they had to use. That was not fun. And I think that you know quickly going from teaching code to recognize objects to you know getting a little Amazon box that can tell that I'm not a hot dog or that [00:32:18] Thomas LaRock [0.5] is not bacon. Everyone on my team just got you know they went to AWS Reinvent and they all got the Amazon box and they're all talking about you know hey look I can tell this whatever it like the fact that that happens so quickly means that you can do some really interesting things. The only analog I can think of off the top of my head is stereos. I remember my dad's stereo with the really interesting glowing tubes inside the metal frame and if one of those tubes didn't glow then you had to take it apart and know it. My dad's a musician so he's pretty you know savvy when it comes to that stuff. And there were approximately 8242 components to his stereo. And then for my bar mitzvah I got like an all in one turntable radio whatever, he's like there it is you know and that was empowering because it did exactly what I needed it to do and I was able to take it places and then there's the Walkman and you know again we're talking about not only the miniaturization but just the ability to take something with you to use in all these different ways. I think that's probably the best trend now. The convergence in terms of not just IT people having empathy but also having empathy for parts outside IT. One of the things that I talk about a lot and I push a lot is a need for IT folks really to learn about the business. I'm not talking about becoming a salesman. I'm not talking about becoming a manager. Nobody wants that. If you do want that that's great you do you. But most of us love being technical and really are afraid of that moment when we will either because we no longer have the chops which never actually happens or because we have been left out of the conversation or whatever suddenly we won't be able to keep up. We're afraid of that. So we fight against it by remaining as technical as possible. Lot of us do learning the business is learning another language you know just like you can speak French and Spanish and Hebrew and sign language and Hungarian and Esperanto and Klingon and elvish which you know very geeky things lots of us learn to do that. It's kind of cool Shakespearean you know we go to a ren-fair we want to fit in we learn to you know business is no different than that business is actually easier than learning to converse with folks at the ren-fair. I think we need to do we've always needed to do that. I have said for a long time there is no such thing as an IT project at all. There are only business projects that have an IT component to them and if you do understand that you're constantly going to be at cross purpose with the people who are actually approving the things that you want to do you need to learn that when you go into Mahogany Row Oak Avenue the C Suite whatever you want to call it that nobody cares about the technical anymore at all really it drives down to three things. Can you describe what you want to do in terms of increasing revenue reducing cost or avoiding risk to the business? The end. All the other stuff is cute. Nobody wants to hear it.
Rich Stroffolino: Yeah that that's been something I have I've enjoyed reading the work of what Keith Townsend does with CTO adviser where I think he does a really good job of bridging those two domains and saying like business has a mission. IT can either enable or inhibit that. It needs to enable that. But it's not a purpose unto itself unless your business your mission is to deliver IT services in which case you are then that becomes a different a little bit of different conversation. But I enjoy, that's a vast oversimplification, but I've enjoyed that. That kind of conversation that's been going on. So I have seen that at least you know in a little bit of the Field Day community side. I've seen the kind of that perspective getting some traction for sure.
Leon Adato: And I'll add a third point you can either enable the business you can inhibit the business or you can be irrelevant to the business and that's the kiss of death right. And certain technologies will become irrelevant to the business they will find a way to outsource it. They will find a way to commoditize it. They will find a way, that's okay. Again dad stereo my single you know my you know boom box right. You know certain parts of that process became irrelevant and that's OK. Nobody mourns the loss of the 8000 component you know [00:36:49] Bang and Olufsen [0.4] they really like our MP3s. I'm alright with that. Technology can become irrelevant. But once you as an IT professional you and I know that that word I think has certain, inspires certain emotions. But you as a person who has worked in IT with and have experience in it if you are doing things to forcibly make the business see you as irrelevant. You have quintessentially failed in your overall mission which is to do really cool stuff.
Rich Stroffolino: Where do you see IT, and maybe the monitoring field specifically going in the next three to five years?
Leon Adato: AIML! In the cloud, hyper converged! Actually. OK so joking aside-.
Rich Stroffolino: Ultra converged.
Leon Adato: Ultra converged right. All joking aside I actually think that in addition to monitoring in general always needing to address the new stuff you know monitoring in the cloud. And I'm a big believer again heterogeneous environments right not homogenous it's not all the cloud it's not all on premises said it right,.
Rich Stroffolino: Thank you.
Leon Adato: No problem. It's a some mix of them. Very few companies are 100 percent one or the other. Almost all the people we surveyed 8000 respondents all had some not only blend of on prem and in the cloud but on prem and in multiple clouds you know some the average is 2. Okay. And we know which two they are but many of them 3. Sorry lost my train of thought, what was the quesiton I was answering?
Rich Stroffolino: Where it's going the next three to five years?
Leon Adato: Thank you, so moniotoring specifically needs to recognize that you know all of that as a single environment from the perspective of the people supporting it that yep you're right you've got sales force and that is part of your IT environment. And the three ISPs between you and sales force are also part of your IT team. Yes that means you have to fix the internet.
Rich Stroffolino: Okay that's actually what I feel is really interesting that I've seen from SolarWinds and other companies like ThousandEyes or something like that is viewing that path of transit, now that we're not everything is going over and MPLS and you kind of give up ownership of large you know what you're going to assess you know kind of based application or something like that that that kind of monitoring is kind of blowing my mind of what I'm seeing in terms of being able to kind of even though you don't own that chain necessarily you can have visibility and can have actionable insights into that.
Leon Adato: Now just to blow your mind a little larger you know a little further out there whatever. You can see your path to that service that's fantastic right. What if you could also see what other people on the same path or similar paths what their experience was at that moment. At the same time what if that information could be aggregated you know you know we're starting to build a picture of literally you know of the Internet not just of your view of it from your point of origin to where you want to go. So that's an interesting that's an interesting idea that we keep playing around with. But in monitoring where we're going is recognizing that you have this multi you know location you know hybrid environment et cetera. And at the same time the data is all you know not the data that you're using not the customer order or whatever it is but the monitoring data all starts to become more and more related and finding those trends finding the what I first heard [00:40:38] Charity Majors [0.4] call the high cardinality event. You know I don't want to know the thing that happens every five minutes that's that's old news. Like yeah hard drive fail or whatever I want one of the things that almost never happened. But are notable or whatever how do I find those so looking at all the incoming data and and that is where machine learning maybe artificial intelligence maybe not the dumpster diving through vast quantities of data that are coming at you in near real time and beginning to assemble you know hey this is interesting we should look at this. You know those kinds of things. It's not an alert you know. It's not down. We have plenty of tools that will tell us. The thing is up, the user's experience of the thing is good. That's really all we care. And the thing I'm trying to accomplish is happening you know the user's experience of my shoes website is good. The website is up and they are buying shoes at the correct rate that I was expecting them to buy at. I don't care about anything else. I'm selling shoes and they're buying them and it's all good. But there are things happening within the data that may be notable that may either lead me to say oh I didn't realize that that usage pattern was happening. I can capitalize on that. Or oh this is an anti pattern but it hasn't hurt me yet but I can immediately start to respond and start to address that. That's where I think monitoring is going is the ability to apply algorithmic understandings there. I avoided the buzzwords, algorithmic understandings to the incoming data. Then it becomes a game of can I get all the data I need and can I get it all in the right place, and can I process it fast enough? And the answer is obviously yes. The doing it is kind of tricky is like yes it's simple. That doesn't mean it's easy. It's just simple.
Rich Stroffolino: Yeah that's one of the other things that I've really become aware of is not just you know data is the new oil drink your buzzword bingo game but that it's not that that idea is old now. That's conventional wisdom even though it was said a year ago or whatever. What is interesting to me is that not just that it has value or that it's super valuable or that businesses need to you know find a way to use it is that there is a timeliness to that to that value as well. And so knowing when it has value is now kind of becoming you know it's not dead is the new oil. It's knowing when your data is valuable is the new oil is the new oil know like that to me is very interesting to me that kind of feeds into being able to process it and being able to make it actionable within you know that's an IT talent certainly.
Leon Adato: Right. And if we wanted to beat that that metaphor into the ground until it was indistinguishable from the ground around it. You know data is the crude. OK. You know now we have to learn how to refine that crude oil effectively and then distribute it to where it needs to go and then use it in the right. Are we talking about a fuel injection system or are we talking about jet fuel are we talking about like how are we using our crude resources? And do we have the distribution tools i.e. whether it's data flows or application front ends or back end processing or whatever or monitoring you know whatever it is to get that that processed refined crude data to where it needs to be.
Leon Adato: So recommendations for books. It's interesting. I just a few months ago took a bunch of guys who had never really been in IT at all the most complicated technical thing they used was a flip phone. And in six months brought them generations ahead into IT. And you know help them understand what the IT world was in general and then they picked a path whether it was system administration, network engineering or programming was the three [00:45:08] choices. DA [1.5] But those were the few things I thought were achievable, very particular group of people who have the ability to learn large quantities of, learn and assimilate large quantities of data quickly but just had no experience in technology at all. So I took six folks six guys through that. And so I had to tell them like here's required reading for you who've never been in IT and really you know really out of the game. So what I told them was Accidental Empires by Robert X. Cringely, which gives a fantastic view of how we got here. Still it's a great history lesson in IT, lot of fun to read. You know one chapter is called Chairman Bill leads the happy workers in song. The chapter that talks about Apple is written like an episode of Bonanza with Steve Wozniak as Hoss. He is the big strong guy and Steve Jobs is Little Joe Cartwright. He's not big Joe. He's always like in the shadow trying to remember is amazing.
Rich Stroffolino: That sounds amazing.
Leon Adato: It was written in the late 80s early 90s. So it had a particular view of those organizations. But a good book. So that was the first one and then the Phoenix Project. I thought that was important and those are probably the two besides the subject specific. I'm currently working my way through Satya Nadella's Hit Refresh. I was a longtime Microsoft hater. I feel that I have earned my hatred. I used Windows 286 when it came on 12 5 and a quarter inch floppies for free. When you got a copy of Excel 1.0. So I, I've used it for a long, long time and it has always been. I remember when Microsoft willfully destroyed GeoWorks you know which ran on the 1886 but just got buried under the garbage and never but it was in my opinion a better operating system at the time. I'm old, really old. I remember [00:47:37] OS/2 Warp. [0.1] I remember Next. I remember a lot of operating systems and Windows was not any of those. It wasn't sexy it wasn't exciting the way that those was. It was the VHS format of computing. It wasn't the best format. It just is the one that won and I lived through a lot of years when Microsoft was making bad choices for the users for the customers. Making things more difficult. Whether you want to laugh about Microsoft Bob or Vista or Windows 8 you know I mean it was it was really hard. There is in my to do list. I'm sorry and I take it all back. Is the name of the essay. I was live blogging the Windows 10 announcement, the Windows 10 desktop announcement four years ago and with my usual Microsoft skepticism and like ah this doesn't mean anything and even then when I was the first time I ever heard Satya Nadella say anything and I was like OK, so he kind of gets it. He kind of understands they'll never execute on it but he kind of gets it and being at Microsoft Ignite last year and seeing the changes and it's not just that they're you know embracing Linux it's not just that they're putting Sequel on Linux not just that they're you know-.
Rich Stroffolino: Which is insane in and of itself.
Leon Adato: Which is insane and wonderful. Understanding that I've run Linux on my desktop for the last 12 years but it just the culture shift the willingness to embrace true you know innovative ideas little things and big things you know really have caused me to turn around and say alright alright this this is a Microsoft that I can be comfortable hanging out with. You know this is a Microsoft that I want to have drinks with Microsoft that I want to have on my like you know friends list on social media that kind of thing.
So that's you know that's what I'm reading and let's see everything else I'm learning is obscure and bizarre with this religiously based for those people listening I'm currently learning the Maseket Beitzah. So like you know if you know what those words mean. There you go. If you don't know what those words mean, it doesn't matter.
Rich Stroffolino: All right. First computer you owned, I heard reference of 720 K of memory.
Leon Adato: Yeah.
Rich Stroffolino: But was there, was it any particular box?
Leon Adato: I was an Amstrad. I don't remember the model it was a 286 and that was and I remember getting on to customer support first customer support call because it would do the memory check and it would come up OK but I couldn't run some of the programs I wanted that required a meg of memory. And it's all OK, I don't understand, it's OK but I can't run this and the tech support person said that's zero K, zero K high memory, zero K extended memories zero K expanded memory like you get. You have none. Oh and that was like first tech support call ever.
Let me take a step back. That was my first PC. I do want to point out that my first compute device was a Atari 400 not the pong thing. The Atari 400 that used cartridges and I had to write my own printer drivers and it used five and quarter inch single sided single density floppy disks. So yeah that was it. And I quickly graduated up to Atari 800 but I never got the 1600. That was my dad's. He had that one and then I moved to PCs from there and so that was my.
Rich Stroffolino: Writing your own print drivers. Yeah sounds like it probably works better than most printers today.
Leon Adato: It wasn't hard it was an Epson dot matrix printer. The funny part looking back on it like if I knew today what I, if I knew then what I know today I would say you can't write a printer driver that's insane, you're a theater major. Like what makes you believe that writing a printer driver. But you know I was on Cleveland Freenet and somebody said yeah you just put it in this. I mean it actually wasn't complicated. It was basically a basic program but I had no idea what I was doing. I was I was you know effectively cutting and pasting. And I was going out of the manual and the manual is like yeah you can write your own printer driver and here you go. In the days when the printer manual had instructions on writing a printer driver not knowing what computer or OS you were using. It was incredible, it was a heady incredible time.
Rich Stroffolino: What do you do when you're not working in IT and writing your own printer drivers?
Leon Adato: Right. So work takes up a lot of time. I love my job which is amazing. I love my job which means I do too much at it. As I said recently you know knowing when to walk away from work is possibly the best skill one can develop. And it's one that I've utterly never done. So in my almost non-existent amount of spare time you know I've got four kids and two grandkids and so I spend time with them. My daughter runs a bakery so I help her. Now that's IT stuff because it's her Web site and it's all that. But I also you know just help out. I can't cook at all. You will never taste my cooking and you should be happy about that. But I wash a mean dish. So I help out there and then because I'm Orthodox Jewish and part of the community a lot of time is spent just learning texts from a religious standpoint like I said when I said those work, Maseket Beitzah is part of Talmud and there are whole buildings of married guys like me who show up at odd hours of the day and spend a couple of hours learning this or learning that. And they do it every day of the week.
Rich Stroffolino: Reading in the margins.
Leon Adato: Reading, yeah reading in the margins exactly. So that's what I spent my time doing.
Rich Stroffolino: And how do you caffeine, if you do?
Leon Adato: I do caffeine. So I met my wife a really long time ago. I met her when I was 16 and she had two brothers adopted from Colombia. And so her parents would make trips down to Colombia every year or so. Colombia has no tariff on coffee. You can bring back as much as you want as much as you can carry. Don't worry they'll make more. So my girlfriend at the time said oh my parents are coming back from Colombia. My back would begin to hurt. You know. You know. In anticipation of the four hundred pounds of bags of Colombian coffee beans they bring back. So I learned to caffeinate very early on real, like from Colombia Colombia coffee beans ground and brewed right there. So it is coffee all the way. I don't really drink soda pop. No Red Bull. No. As you can tell I'm a really sedate.
Rich Stroffolino: I was going to say you're soft spoken so you probably need the caffeine.
Leon Adato: No. I've worked in jobs where they were actually barred me from the coffee machine.
Rich Stroffolino: Now do you have. Are you particularly snobbish I guess is the word in the way you prepare it? Are you a drip coffee maker, french press, aeropress?.
Leon Adato: No, I mean we have
Rich Stroffolino: Whichever way you can get it.
Leon Adato: We have, We have a grind and brew because we have the coffee beans.
Rich Stroffolino: Yeah. The fresh beans.
Leon Adato: So I prefer it that way. The one thing I draw the line is instant. You can not do it while it's a thing again in the Orthodox community when you have no Shabbat or the Sabbath and you can't touch anything with an on off switch. Brewing coffee is impossible and straining coffee and all that stuff. So you have to like instant as a way of life for a lot of people and I just refuse I brew a giant Yeti thermos of coffee right before sundown like you know in the minutes before we time it. And I get that coffee into the thermos as fast as I can preheat it with boiling water and everything and that will give me reasonably hot water for the next 24 hours.
Leon Adato: Few Answers. I'm gonna give you some choices. The first one is any and all of the head geeks other head geeks. So that would be Patrick Hubbard @FerventGeek on Twitter. Destiny Bertucci @DesSays on Twitter. Tom Lorac @SQLRockstar. [0.0] So anyone of those folks would be great. I'd also love to see the product manager for the network monitoring side. Chris O'Brien who has the greatest Twitter handle ever at @PacketMagic. I would love to see him answer those. Those are some of my top choices.
Rich Stroffolino: Excellent, excellent we'll have to definitely invite them on. And finally any career advice you'd like to pass on to our listeners?
Leon Adato: If you're just getting started I would say understand that being a generalist which a lot of us start off as a generalist is great but know that in the near future you're going to need to choose. You're going to need to pick that thing that you want to focus on that you know if you are this amazing do all be all to everybody thing. You're going to be limited in where you can go. It's just IT is too broad. But right now use the generalist to learn everything about all those other areas and then pick that thing and then do it. IT is wonderful and amazing in that it changes so rapidly that if you just wait a little bit you actually will know more because you just got on the you got on the trolley at this point and now you're riding the trolley so it doesn't require everyone to have years of experience or a degree certainly or anything like that. If you're not new in your career if you are experienced in your career. Share everything, give it all away. There is nothing that you know that is going to make you more valuable for not having told someone else. In fact everything that you know is more valuable if you tell someone else or another way to think of it is irreplaceable is unpromotable. You will never leave that thing that you're doing if you make yourself irreplaceable and there's going to come a point when you don't want to be doing that thing anymore you're going to want to do something else. So share it all. It will only make you more valuable. Empathy. Sprinkle that all over the place.