Keeping up with Technology: Drinking from the Firehose


Audio podcast:

Enjoy the Podcast?

Don’t miss an episode, subscribe via iTunes, Stitcher or RSS
Leave us a review in iTunes

Target Audience:

The technologist who is frustrated with the pace of technology and feels lost.

Abstract:

SQL Server releases every 2 years. SSMS releases every month. Power BI releases every week. It’s impossible to “keep up” with technology, and it’s foolish to try. Instead, you should realize you have a limited amount of TrainingBucks: time, energy and focus. And then you have to decide how to make the most of those TrainingBucks.

First, we’ll talk about how exposure and mastery are different goals with different paths. We’ll cover how exposure does not equal mastery. Second, we’ll talk about strategies for adding more learning with the time you already have available. Third, we’ll cover what types of training lead to deeper, stronger learning. Finally, we’ll talk about having a theme in your learning. We’ll talk about about how learning the wrong things is wasting your time.

During the presentation, I will use an interactive Power BI report to visualize how using these strategies will lead to more growth for less effort. This report will be publically available, so the viewers can follow along as they watch.

The goal of this talk is to give you an analytical framework and multiple dimensions to look at, instead of the simplistic “more is better”. More is not better, and if you try to drink straight from the firehose, you’ll drown. By the end of this talk, you’ll have a roadmap for how to focus your training and grow your career.

Why I Want to Present This Session:

Because I think this is a talk that would be helpful to anyone looking to keep up with technology. And, selfishly, if I present on it I can crystallize my thoughts and better understand it myself. This is Rubber Duck career development, and everyone wins.

Additional Resources:

Session Transcript:

Brent Ozar: In this next session at GroupBy, Eugene Meidinger will be talking about keeping up with technology, so how you learn and what you should learn. So take it away Eugene.

Eugene Meidinger: Alright, I’m very excited to do this presentation. It should be really interesting. Anyone who wants to contact me, I’ve got my Twitter and I’ve got my email up there on the slides, and then if you go to just SQLGene.com/keepingup, you can grab these slides, you can get a link to the GroupBy presentation whenever it gets updated and you can grab a link to my epic blog post, which is longer than this slide deck, which is about 43 slides because I just kept adding stuff during the day. So I’ve got two hours, should probably be fine.

So a little bit about me real quick, so I’m a business intelligence developer, consultant, whatever you want to call it. I worked for a company called All-Lines for about five years now, and I speak at our SQL User Group, I speak at any SQL Saturday I can drive within four hours. I help run our PowerBI user group in Pittsburgh, and hopefully, at some point, I should have a Quora site course coming out this year. But the important line here is that I started out as a SQL noob, and now I’m presenting to all of you, and I want to tell all of you watching that you can make that same transition.

When I started my job five years ago, I was not qualified for the job, and I love telling this story because the job said .NET slash SQL Developer, and I’d taken a database course in college, I TA’d a little bit, I could do basically everything that was in the last session, except for the backup piece. Maybe a little bit less than that, I don’t think I had HAVING down at that point. And I had to start working, and I had to look up my first month, what’s the difference between a stored procedure, a view and a function, because I didn’t know. So I remember Googling furiously for the answer, and then I found out that user groups were a thing and I was really excited, and I found out that SQL Saturdays were a thing, and I got really excited and went, and then I made a mistake of going to the after party. Do not go to the after party for a SQL Saturday because you’ll end up with a speaking gig. I went to the after party and I got cornered by Gina Walters on one side and Rick Heiges on the other. Gina was running the user group, and Rick was a former PASS Four member, and they were like you should present. I’m like, I’ve been doing this for a year, I’m not qualified, and they were like you should present anyway.

And so I did, and at the time we’ve been getting eight people to our user group and when I presented on Execution Plans, there were 30, so a little bit of pressure there, but I didn’t catch on fire and I really loved it, and it just kept growing and growing and growing. So presenting is a really great way to give back to the community, it’s a really great way to learn. I really encourage all of you to get involved. Local user groups are dying for speakers, so really do get involved.

Alright, enough about me. Let’s talk about what you came here to learn. So we’re going to answer some questions in this presentation, and the first one is, can you really keep up? Is that even a thing that can be accomplished? And second, why is this an issue to begin with? I think that IT is unique compared to a lot of other fields, with the changing pace and all this kind of stuff. And the next question is how do we get our hands around it? How do we really understand it? Is there some way that we can model in a way that’s going to click? Because I think for a lot of us, it’s the vague, nebulas anxiety, right? It makes me think of the term the Cloud. The Cloud is basically, with old visio-diagrams, for the internet, they just draw a Cloud because it didn’t have any actual meaning, and now we use it as a marketing term. I feel like keeping up with technology is like that. There’s not really a well-rounded definition for it, and I’m hoping to do that at least a little bit today. And then finally, some practical stuff about how we keep up, and what it’s going to boil down to is are you learning more things, learning more of the right things, and learning more things that last longer. So those are the three things that we’re going to focus on during this presentation.

So, one of my favorite aphorisms is Betteridge’s Law of Headlines, and it says, “Any headline that ends in a question mark can be answered by the word no.” So here’s the question: can we keep up with technology? No, you can’t. It’s going to be a very short presentation I swear. So why not? Why can we not keep up with technology? So the first thing that I would say is that it’s not well defined. In my mind, keeping up with technology is less of a technology problem and more of an emotional problem, right? It’s this vague, uncertain anxiety, and you’re never done and you’re never finished, and there are no real good criteria for what that means. Really I think it’s more of an emotional problem, so we’re not even going to talk about, air quotes, keeping up with technology, we’re going to talk about some more concrete things, and we’re going to give you a model to understand the prominent way that’s actually helpful.

But another reason you can’t keep up is that the data platform is broadening. If you’ve been to PASS in the last year, if you’ve been paying attention at all to any of the announcements, you’ve got all this stuff with data science, you’ve got all this stuff with big data, you’ve got all this stuff with NoSQL, they just turn document DB and the cosmos DB, which basically involves slapping four other NoSQL models into the same Azure platform as a service. There are more and more and more and more things, and there’s just too much for you to keep up with.

I also think that the rate of change is accelerating. SQL Server 2017, do I have to say anything more? If there’s a SQL Server 2018, I’m going to start pulling my hair out, because there used to be an idea of databases being one of the most conservative platforms in the data – in the programming world, in the IT world, whatever you want to call it. It used to be the tortoise of the IT world, compared to say JavaScript frameworks, which change every six months. Well, that’s changing too. The pace is going faster and faster, and Cloud I think is part of that.

So if there’s no point in trying to keep up with technology, what is the real question behind it? What’s the question behind the question? Why are we so worried about this stuff? And I think there’s two things. So the first question that whenever you say how do I keep up with technology that you’re actually asking is, how do I keep my job? If you didn’t need to worry about keeping your job, this probably wouldn’t be a problem. Maybe your personal sense of curiosity would still be involved, maybe there’d still be some frustrations here and there, but really, one of the biggest questions is how do I keep my job? But there’s another question that adds some tension to that, and the question is, how do I keep my friends? Right? Because we want to keep our job, but one way we can do that is if we just work 70 hours a week and we’re constantly studying and we take amphetamines so we don’t have to sleep so much. And unfortunately, that’s not a realistic choice. And so we want to have a social life, we want to have a personal life.

I’m married, and I can tell you that we’ve just determined through trial and error that nine to ten pm is sacrosanct. I’m not learning technology, I’m not playing Minecraft, I’m spending time with my wife. Now, whether that’s playing a video game or reading or learning sign language, it doesn’t matter. But there’s certain lines in the ground that you can’t cross if you want to have a healthy, stable life, and so there’s this tension. How do I keep my job while also keeping my friends? And that’s the real question here, and really they parallel something different. How do I learn effectively, that is learn in a way that causes me to advance and grow and keep my job, and how do I learn efficiently, in a way that requires less of my time and energy? So that’s also the same kind of tension. So those are the real questions that we’re trying to answer today. How do I keep my job and my friends, and how do I learn effectively and efficiently.

So again, if we can’t keep up with technology, what are our options? So one is maximizing out input. It’s the resources that allow us to learn. Second is being more efficient with those resources, and aiming them in the right direction. And the last thing is it’s increasing how long that knowledge stays relevant. In short, we need to make the most of what we’ve got, and this is something you can do, because this is less of some ambiguous fire flung goal, and more of just an optimization problem. It’s taking what you already have and making it more efficient or more effective.

So, why in the world do we have to learn new things to begin with? Why can’t we just get a four-year degree, get a job and just cruise on through? And there are two different reasons, and they’ve got different implications, which we’re going to go into. So the first one is pretty obvious. New stuff comes out. You’ve got SQL Server 2017, you’ve got JSON Support, you’ve got PowerBI, you’ve got all these new things that you have to learn at some point or another, and the parallel of that is that old stuff gets old. You’ve got SQL Server 2000, you’ve got data transformation services, you’ve got performance point. You’ve got all these things that lose value over time, that at a certain point aren’t relevant anymore, and so to understand better this dichotomy, let’s take a detour into an analogy.

Let’s talk a little bit about financial investing. So I’m hoping that everyone here watching has a retirement plan, and in that retirement plan, you have to make a decision about where you’re going to invest your money, and one of the simplest comparisons is between stocks an bonds. Now, I’m fairly young, and so most of my money is going into stocks, and the reason I do that is to deal with inflation. Inflation means that that hundred dollars that you had last year has the buying power of ninety dollars this year, or something along those lines. Your money that you have is worth less and less every year, and so to avoid inflation, we invest in stocks instead of bonds, because stocks have a higher rate of growth. They allow you to outrun inflation, they allow you to outrun the decaying value of your money year after year. But there’s a problem. If you’re smart, you’re not investing in single stocks. Anyone who has their wits about them doesn’t take their entire retirement fund and go, well Apple seems to be doing well, I’ll just pour my 100,000 dollars in that, because you never know when Apple is going to go out of business. So stocks are high growth, but high volatility. Volatility means that our investments can implode at any moment, and to avoid volatility, we diversify our investments. We say you know what, I’m going to pick a couple different things and spread my investments around them in case one of them just tanks.

And this is a really useful comparison to understand how we deal with old technology that’s getting older and new technology that’s coming up. So with financial risk, there are the two risks that we talked about. Inflation kills you slowly, right? Because if you got a 100,000 dollars that is sitting under your mattress, assuming that no one robs you, that money is not going to disappear overnight, right? And let’s – okay, you have a safe, instead of your mattress let’s say. But whatever the case, that 100,000 is going to slowly ebb away. You’re going to slowly lose value. Volatility kills you quickly. That’s when you invested that 100,000 into Lycos.com or Excite.com or any of the half of the websites that are on monopoly internet edition from the year 2000 that I used to play as a kid, that are just completely out of business now. So you’ve got these two very different risks going on when you talk about financials. You’ve got inflation which is a slow moving risk, kind of like climate change for the dinosaurs, and you’ve got volatility, which is a very quick moving risk, like a meteorite slamming into the earth.

And so let’s take this comparison and move it back to career stuff. So old technology kills you slowly. This is your SQL Server 2000 or whatnot. This is the technology that your value doesn’t disappear overnight in general, right? If you’ve got skills with SQL Server 2012 or Windows server 2012, those skills aren’t going to just atrophy one morning like you wake up on a Monday and you go, man, everything I know from that is useless. Like that doesn’t happen for the most part. Instead, it’s more, well I know how to do this thing, but I know that in 2012 they added some of these Windows functions or in 2016 they added JSON or whatever. So old technology is a slow moving threat. New technology is a fast moving threat, and so that is something you have to be careful about because you don’t know if that technology is still going to be around five years from now. Things are still being shaken up. So let’s look at a real example to understand it a bit better.

VHS is a slow moving technology, it’s a slow threat. VHS was around for a long time, and it disappeared very slowly. HD-DVD kills you quickly. That’s gone. There was a format war between Blu-ray and HD-DVD, it was going on for a couple years and then bam, HD-DVD didn’t make sense anymore. And so let’s take another leap into actual technology you might deal with. VB6 kills you slowly. Hadoop kills you quickly, right? So you’ve got different strategies for dealing with these things because VB6 is still being used today. I have to support an application in VB6 and it makes me weep every time I have to look at the code. But it’s a slow moving threat. It’s inflation, it’s the slow decay of the value of that knowledge. That knowledge of how to code in VB6 gets less and less valuable over time but very slowly. Hadoop is something that may not be around five years from now, and if it disappears, then all of that value, a lot of that value that’s invested in that new technology disappears with it. And so there’s a different type of risk. We’ve got the slow moving decay versus this high volatility.

So I want to move into a quote – wrong person – there’s a Buck Woody quote. He says, “In an immature market, you want to generalize.” Right, so this is Cloud. It’s an immature market, or big data, where there’s all these competitors, all these pieces that are going on. But in a mature market where the path is kind of laid out, and that I would say is some of the core like Windows, SQL DBA fundamentals; where there is a solid path, you want to specialize. So in an immature market, generalize. In a mature market, specialize. And part of that is immature markets are high volatility, right? So we talked about diversification. If you’re learning Cloud right now, you don’t want to make too heavy of an investment in any one single technology because you don’t know if they’re going to kill it or not. An example is SQL data sync. For a while there, it looked like that functionality was never going to leave preview and they were just going to kill it off. You really don’t know. Or just you look at some of the competitors to the big three when it comes to Cloud. I mean, you got AWS, Azure and Google Cloud, and those are the clear winners, but a while back, you had HP’s Hellion Cloud, or whatever IBM is doing these days, who knows. And so these immature markets are high volatility, which means you need to spread around that risk, whereas mature markets are low growth, and so you need to focus on something in particular. You need to specialize so that you’re still adding value. There’s less volatility in these mature markets. So going even deeper with this…

Brent Ozar: There’s a question from Ella. Ella asks, “How do you know which markets or technologies are risky versus which ones are safe? Does it come down to how mature or immature they are?”

Eugene Meidinger: That’s a great question. So there are a couple different things that you really need to think about if you’re looking to that. So one is how much of a history does this technology have. And just a simple example is you think about Go compared to C# or even better, like Google Dart compared to C#, right? C# has been around since what, at least 2000? So it’s been around for a good 16, 17 years, it’s backed by a company like Microsoft that has, for certain things, a decent strategy of putting their weight behind it. This is at least something that clearly made a lot of investment. So you look how long has it been around, you look at okay, does it have some sort of sponsor, and you look at are there a lot of competitors, or has it kind of solidified into a more consistent market. So there are still risks. I mean, anyone who’s invested time and energy into Silverlight is a little bit of a sad panda right now because you know, it really seemed like rich internet applications were taking over the world, and then Steve Jobs killed Flash and it just dominoed over. But I think yes, look at how long the technology’s been around, look at how long the cycles are.

JavaScript frameworks are a perfect example of this. You look at angular and I don’t know if this has changed, but I’m pretty sure they said they’re going to move to a six month like big number breaking release cycle. So that means every six months, you’ve got backwards incompatibilities going on. Even if you just look at the area in general, there’s a lot of upheaval. Or like I said, you look at the number of competitors. When you looked at big data a couple years ago, there was a lot more competitors. Stuff like Vertica, even now, there’s still a ton of technologies.

Cloud is starting to come together into just a few big providers. So there’s a bunch of different things that you can look at, but you’re always taking a risk. You’re never going to completely specialize or completely generalize. You’re going to have to find a balance of both of those things.

Brent Ozar: Cool, awesome.

Eugene Meidinger: Alright, so going a bit deeper, there’s this contradiction of learning because to add value, we have to specialize. Yes, there’s situations like if you’re an architect, where you want a lot of generalization, but you think about it, what things in IT would you be willing to hire a just off the street high school student for? Not some prodigy who’s been doing this in his basement off his TI89 calculator for years. But someone who literally does not have a lot of experience. How much are you willing to pay for someone who has one hour of work experience versus ten hours versus a hundred versus a thousand? And that’s not to say that hours correspond directly to relevant experience. There’s the saying, I think Scott Hanson has, that you can work at a job for 20 years, and a lot of those people had the same year of experience 20 times. But my point is, there’s a minimum bar. There’s a reason that a lot of teenagers are getting paid minimum wage is because they’re unskilled. There’s a reason that consultants like Brent gets paid gobs and gobs of money, is because they’ve taken the time to get a certain skill level, get a certain level of specialization, that they in an hour can save someone weeks of time, theoretically. Right? Like I had a customer that I did a health check for, and they didn’t have much of any SQL experience at all. They just had Windows admins, and so part of that check was looking at the backups, and they were doing – they had it backwards. They were doing transaction log backups every night and they were doing full backups every half an hour, and because I had spent much, much more time going deep with SQL, I was able to look at that and go, that’s not good. Right?

So people pay money for specialization, but again, there’s a contradiction here. To become – to avoid becoming irrelevant, you have to generalize, because nobody wants to be an expert in Cobalt. Unless you’re going to work for Nasa, no one wants to be an expert in Fortran, and there’s a reason that I don’t put VB6 on my resume. So we have to spread out enough that we’re able to at the very least, maybe hop onto the next big thing, but first, if you’re out of college, if you’re straight out of college, your biggest impediment is helpdesk, in the sense that you’re going to be stuck on helpdesk for all of your career if you don’t get some specialization. So that’s kind of the contradiction, is to get paid, you need to specialize. To get paid in a decade, you need to generalize, and both of these things require different resources and different learning styles, and you individually may have a different area where you’re more at risk right now, and you have to figure out what that is.

Generalization costs time and short-term risk. So that’s what we were talking about before, where to generalize, to learn a bajillion things, you need to put in a lot of time. And there’s the short-term risk that – at my last job, and a little bit in my current job, there’s a technology called Adobe Flex, which is now actually Apache Flex, because Apache is where technologies go to die, and it was based off of the Flash engine, and it’s something called Action Script, which was a derivative of JavaScript. And as a rich internet application tool, it was really cool, really intuitive, really easy to work with, and even better, they had support for iOS and Android. An amazing technology, but you know what? Adobe had to move away from Flash because again, Steve Jobs, the tyrant that he was, killed off Flash, and just made that technology useless. And it’s still being used in places, but it’s a perfect example of short-term risk. Can you imagine if you spent years specializing in Adobe Flex and becoming an expert in that and focusing very tightly on that technology and then you find out that it’s deprecated and it’s been handed off to Apache?

So generalization costs time and short-term risk. Specialization costs focus and money, because anyone can attend a GroupBy session or a SQL Saturday session about Hadoop and learn the basics and become conversational without a lot of focus. You can listen to a podcast, you can attend a session, and you can do that without a lot of focus. But if you want to go deep, you need focus because blogging is hard. Presenting is hard. Home labs are hard, but that’s how you go deep. And there’s a long-term risk because of the things like VB6 and Cobalt, because it’s like this slow moving iceberg that eventually is going to crash into something and you’re not going to have a job if you’re not willing to leap to a different technology when the time is right. So to summarize that, you have to go deep to pay the bills…

Pinal Dave: Just to break the flow and before you start the next slide, that is very interesting comment from Curtis, he says, “This is great material. I can read some word when I am dealing with the young immature developers who are always wanting to just use the latest things because they think it makes them cool.” I think it’s amazing comment. I often feel it, like one of the person I recently said, can I use DMV – can I use Query Stores, I hate DMVs, and I was like, I’m speechless. Both are very different things, and yes they do same things at one point, there is a little common area, and I think this comment about – I think this question I get all the time. I would like to see what you think about this Eugene, when people say the new people who come in and propose immediately a new technology. How do we handle them? I think it’s a fantastic – the contradiction of learning is good because the people who is experienced want to use something different, and people who are just brand new want to experiment and take a risk. I want to see what you think about it.

Eugene Meidinger: Yes, that’s a great comment, a great question. So one of the challenges, there’s this thing called the Dunning Kruger effect, and it’s this really fancy term for the fact that you don’t know what you don’t know. And so one of the challenges that I’ve seen is that it takes years of going deep with something to realize how that’s materially different. Like one of the things whenever you actually do specialize or you do get some real experience under your belt is you learn how to find pitfalls and anticipate them before you actually have to run head first into them. I think some of that friction’s always going to be there, and it’s pretty natural to want to use the latest, coolest thing, but you know, I don’t know if there would be a good way to communicate it to people, but I would definitely advise – if you’re someone young, you should not in any way be learning Docker or Hadoop or Query Store or any of this cool new stuff – like maybe a little bit in your free time, but you need to think, okay, how do I add value to the business so that my time is worth their money. And it’s amazing, but there are not that many silver bullets these days. There’s stuff that helps, but you really have to like – it can be hard to get out of your own skin and think from the perspective of a business owner, but if you can talk to someone who actually runs a business, being able to see okay, why are you paying me money and how am I adding value can really help with that. But I think that friction’s always going to be there because whenever you’re young, you get excited about these things or when you’re young, you don’t understand why having some maturity in your technology is so important. You’re just excited about being able to do stuff.

Brent Ozar: There’s – from the business side, there’s a really great quote, “Do risky things with safe technology, or do safe things with risk technology” but you don’t want to do risky things with risky technology.

Eugene Meidinger: Yes. I think there are times where it makes sense whenever you have somewhat experienced developers. Like I think the whole thing with like Trello where Joel Spolsky said, yes, I go ahead and use like all this cutting edge kind of technology, but I think he had a pretty good management system in place, he had experience with delivering software products and he had developers that had the experience, where it was safe for them to decide okay, everything’s going to be Mongo and hyper scale and whatever. But yes, there’s definitely this friction that’s always there, and it’s really, really hard whenever you’re fresh out of college because you think that your value is how many lines of code you can add, and measuring a program by lines of code is like measuring an airplane by how many pounds or kilos it weighs. A mature developer understands that they’re adding a lot of these qualitative things like stability and scalability and reliability and all these things that you can’t measure by how quickly your fingers are moving.

Brent Ozar: Ella asks, “Part of Agile methodology is encouraging generalization over specialization. Should I just follow Agile completely or are you suggesting that maybe Agile isn’t the right thing for my career?”

Eugene Meidinger: So there’s – so I’m definitely not an Agile expert, but in my mind, I think part of the value with that is being able to interact with the rest of the team, understand the project flow, all that kind of stuff, right? But I think I would say that it’s fine with this stuff to satisfice. And so satisfice is a combination of the word like – I want to say it’s like satisfy and sufficient, or something along those lines, it’s some fancy term, but basically, it means do what’s good enough. And what I think here is, you have to decide okay, what’s enough specialization, what’s enough depth that I know I’m adding value in a concrete, solid lasting sort of way? That doesn’t have to be super deep. Are you adding $15 an hour worth of value? $20 an hour? You don’t have to be like David Klee who’s the guy if you’ve got a virtualization problem for SQL, who’s probably billing $200 an hour. None of us have to be at that level. But you do need a certain amount, so that at a bare minimum, think about it, if you’re applying for a different job, what skills can you put down to show that you’ve actually done something? Do you really want 20 different technologies on your resume that you’ve got basic level experience with? Or do you want something where you can say, at a bare minimum, intermediate with DAX, intermediate with SQL, intermediate with report server, whatever. So I’d say at a bare minimum, you want to make sure that you’re going deep enough that you can put stuff on your resume with a level of confidence that they ask you about it, you’ve got something you can point to and you can say yes, here’s where I added value.

At that point, it’s okay to look at branching out a bit. I mean, you think about architects, like software architects. Their job is to generalize. They need to understand the full stack and every part of the system, but it’s really easy to think that you’re becoming and architect when really you’re just becoming a jack of all trades and a master of none. That’s the risk.

Pinal Dave: I just want to add one line to this thing. Last year I watched Brent’s presentation where he said something and which I really – exactly what you just said, you have to be very, very specialized and after that I changed even my business card. I mean, because – okay, I can’t present it properly, but he said, nobody will hire a founder, nobody hired a CEO for doing your performance tuning task. Call yourself what you wanted to do, and I think, exactly that day I changed my card saying Database Performance Tuning Expert. My wife asked me do you know really that thing, I said I will learn it from first customer. So the point is, I think we need to be master of something. We cannot just generalize. You cannot say I know everything, I can’t call myself Database Expert, I need to be a little bit deep, exactly what you just said, so thanks. I think Brent changed my life since that day, I’m making more money.

Brent Ozar: Alright good, as long as it worked. That’s the good part. Richie’s in now too.

Richie Rump: Alright, so since I’m the only score master here, I could mention about the Agile stuff. Agile isn’t for the individual. Agile is for the organization. So for the organization, it is better for them if they had more people knowing more things so that they can pull and plug different people in different places. It’s good for that. Agile is not for you, the individual. It is not going to help you in your career. Now maybe knowing about Agile and how to implement Agile and maybe you could move your way up to a score master or whatever the Agile layers are, that may be good, but it’s designed to pull projects from one place to another and encourage collaboration. So you need to make a difference between this is what I do in a group it work, and this is how I move my career forward because what I do in a group, it work may not be the best for you as an individual.

Eugene Meidinger: So Richie, I’m just curious, would you say then there’s kind of a small conflict of interest between the organizational needs and the individual needs in that sense?

Richie Rump: Absolutely, and there always will be because where the organization wants to go may not be where you want to go, and what you need to do is understand, hey, this is where I want to go in this organization, how far can I go, versus where do I want it to go, and understanding where they diverge and at what point in your career do you need to kind of go somewhere else that more closely aligns with what your career goals are.

Brent Ozar: As Richie’s boss, I feel like I need to say something in here. So like the typical company would say, I would like Richie to know everything, and I don’t want to pay him anything. You know, so many companies are just like that, yes I’d like you to generalize and know everything as deeply as necessary, I’m not going to pay you a dollar for your training budget. So it’s – yes, what works for the company can be very different.

Richie Rump: I could use a paycheck, Brent.

Brent Ozar: I’m sure you could, but back to the presentation.

Eugene Meidinger: Yes, so I guess one last thing with that before we move on. A personal example for me is, you know, you don’t have to find that specialization within your workplace. So for me, I’ve decided that, okay, PowerBI seems like this new hotness. I’m going to take a risk and plant my flag in that, and so I help run the local PowerBI user group, I do presentations on it, I’m hoping to start blogging more on it. All this kind of stuff, and at work it’s part of the direction we want to go but right now, there’s not necessarily a ton of stuff coming in for that, right? My work is a lot more spread out. I do administration, SSRS, all these other things. So you can find that specialization and dig into it really deep without having to you know – without that having to fit into your workplace 100%.

Okay, so let’s keep moving because there’s like another 30 slides or so – 20, 23. Alright, so kind of little bit what we were talking about. Again, you go deep to pay the bills and you go wide to keep your job. And it’s this cycle – when you’re in college you’re getting a broad brush education, you’re trying to get a little bit of everything. And as soon as you get out of college you need to make it past that helpdesk position so you need to go deep to have some specific value, and then you need to go wide again because the Cloud became a thing and you’re worried about your job. So there’s this never-ending cycle, and you have to figure out where in the cycle you’re at right now.

Now, what I’m going to say is specialization is always harder, so I’m going to encourage every one of you, if you’re not sure which way to go, pick specialization because that threat of specialization being the wrong one in general – if it’s with a solid established technology, that’s a slow moving threat. Okay, specializing in Hadoop, really risky. Specializing in SQL internals, not so risky. They’re not going to change that quickly, and you’ve got time to jump to something else. But part of the issue too is that you can generalize via passive learning, and this is one of the biggest points in this whole presentation, is all of you who are watching GroupBy today, learned a little bit about continuous integration, operational stability, and TSQL basics and you probably did it while you were on Twitter or pretending to work or drinking some coffee or whatever. But specialization, you can’t multitask that. It’s not going to work, because that requires a lot more focus and energy. You have to do active learning. You have to do something that’s going to challenge your understanding of the material, and so reading books isn’t going to do that. Reading blogs isn’t going to do that. Listening to a podcast, watching videos very rarely is going to challenge your understanding of these things.

Instead, you’re going to have to do active learning, which means presenting at things like this or user groups. It means writing a blog and double checking your stuff. It means doing home labs, it means writing code. Specialization is always harder because it’s always harder to find focus than it is to find time whether you believe it or not. And so if you don’t know which way to go, pick a specialization.

Okay, so now let’s take another detour into a different analogy. So we’re going to talk about radioactive decay. So this is my other analogy for again, trying to get your head around what does this learning thing looks like – what does it look like so I can kind of work with it, versus just this nebulas thing that’s floating around. So in nuclear physics, there’s this idea of a half-life, and all half life means is you know, if you’ve got X amount of something today, then once that half life duration is past, you’ve got half of it, and then you go another duration and you’ve got half of that. That’s all that it is. If you took calculus it’s an exponential curve. That’s all that it is, and I think that IT knowledge could make sense modeled as a half life because there’s this saying, how long does it take before half of what you know is useless? So to quote Alan White, he says that every five years you have to re-tool yourself. So that gives us a number, and I remember I was in Pittsburgh, the SQL Server data partners podcast was there and they did a panel thing and he had said that and then I looked at him and I said, well, I’ve been doing this for five years, does this mean I have to start over? And it was half a joke and half serious, because like, do you have to start over? And obviously a lot of the knowledge transfers, but I’m kind of doing that re-tooling, like what I said about trying to dive into PowerBI.

So I would joke in college the same thing. I would say every five years, half of what you know is obsolete. So this gives us a model. We can say okay, I’m going to solve for X. I’ve got a rate of decay, which is every year I only keep X amount of my knowledge in terms of being relevant. Maybe you still know DOS, but it’s probably not relevant to your job for the most part. So that rate of decay to the power of five, because it’s five years, it happens five times. We’re going to have one half of whatever our original knowledge was and so X to the power of five equals half, if we solve for X we get 0.87. So what does that mean in like lay terms? That means that every year, 87% of what you know is still relevant.

Now, you can tweak the numbers here, maybe it’s every ten years half of what you know is obsolete, 20. The exact number doesn’t matter. What matters is now you’ve got something that you can think about, because conversely, every year, 13% of what you know is irrelevant. Think about that, 13%. That means if you’ve got a hundred things that you know, 13 of them at the end of the year are no longer relevant. That means you’ve got a lot of learning to do. That means you need to learn 13% of your total scope of knowledge every year just to keep up. Let’s take a look at what this looks like, so again, this is assuming you’re completely stagnant. This is you came out of college with your fancy computer science degree, you knew a hundred things, you learned a hundred things in college, and then you look at this curve, in five years you know half of that, in ten years you know a quarter, one year and you know 87%. So this is a model that we can actually think about and try to understand how we need to deal with this.

Again, another way to think of this is we’ve got a 13% interest rate. That’s bonkers. That’s going from an A+ to a B, that’s terrible. It’s crazy, it’s scary, and the question is, how are we able to fix it? How do we take care of it?

Brent Ozar: There’s a couple great points, like Steve Jones for example, the 13% thing is getting really interesting in chat Slack and people are going, like the things you know build on each other but they lose their value. The value on them goes down.

Eugene Meidinger: Yes, and so it’s worth mentioning that you’ve got some stuff that’s going to be timeless, or has a very high half-life, which I’m going to talk about later. So like, how to write a good email, probably doesn’t decay very quickly. But how to get around the fact that SQL Server 2000 doesn’t have triggers is going to decay pretty quickly. So every piece of knowledge has a different value half-life, and so some of these things last really long, some of them last very short, but again, just as a rule of thumb to kind of get our figures around it, we’re looking at five years. So let’s assume that all that knowledge on aggregate is still valuable for on average, five years, or half of it. You get the idea. But piece by piece, it varies dramatically. Okay, anything else before I move on?

Brent Ozar: No, go ahead.

Eugene Meidinger: Okay, cool. So how do we fix this, because this is clearly a problem. So as I mentioned at the beginning, we’ve got three options. We can learn more things, we can learn more of the right things, or we can learn more things that last longer. And so we’re going to go through each of those three options to deal with this issue. So the first thing is learning more things. Well, what are the costs of learning? What are the inputs? Because one thing we can do is just try and brute force it, right? So you’re like okay, I know a hundred things, but I really want to know 120 things. And so instead of having to learn 13 new things a year, now I need to learn like 18 or something like that. So what are the inputs? So first one is time. So how do we get more time in a day? Again, I suggest things like caffeine and not sleeping, but at a certain point, you’re probably not going to be able to cut out as much as you would like.

One thing I would suggest considering is doing a time monitor of your day. There’s a great website called Toggle, which is just a time sheet application, but really I would consider taking a week and writing down where your time goes for that specific week, and just see where your time is and how you’re spending it on things because you’re going to be surprised at how it actually matches up or doesn’t, and that’s going to give you a better idea of where you can pull stuff out.

I can tell you for me personally, there’s not a lot of room because you know, during the school year I do some youth leadership, I told you have that nine to ten time with my wife, and 10 to 10:15 is cleaning because we’re giant nerds and everything has to be scheduled. I get groceries for my mom, I work a job, occasionally I like to actually have some fun. It’s hard, so I don’t think – there’s this cliché of just cut out watching TV and you’ll get four hours a week back and it’s like, I’ve already done that, no. So you know, there’s a limit to how much you can cut out time. So what I would – the next two things I think are a bit more useful.

So the first one is multitasking. This is huge. If you’ve got a commute and you’re not listening to a technical podcast, you need to rethink your choices because there is so much stuff you can learn during this dead time or this time that doesn’t require a lot of mental focus. So for me personally, I have a 20-minute commute which isn’t that bad, but I know I definitely don’t need to do a lot of thinking when I’m driving. Another example is exercise. Doesn’t require a lot of mental focus. Washing dishes, all these things. There are all these opportunities for things that aren’t mentally intensive where you can get back some of that time and learn some of this passive stuff, and this is great for some of the more generalization stuff because it’s a way for you to learn about these technologies that might be a hit or a miss, without making a huge investment.

The other thing is to utilize dead time. Now, you don’t want to go overboard with this, you don’t want to always be looking at a screen, but think about those weird periods of time, like 15 minutes before a meeting or 10 minutes at the doctor’s office, or you know, you just finished lunch but you’ve still for five minutes. Right? During that time, are you reading Twitter or playing Candy Crush? Or are you learning stuff that’s actually relevant? If you have a feed reader or something like that, you’re able to read stuff that you know has value because you’ve curated that list, and that’s a great way to learn, so I think with time, to get more time back, the two big things are finding times where you can listen to podcasts or watch videos while doing other stuff, and find ways to get a feed reader and get some of that time back. Because I suspect that a lot of us are doing meaningful things with our day to day stuff, and it’s not so easy just saying okay stop watching TV, you’ll get four hours back. Some of you have kids. A lot of us have significant others who are not cacti that need watered once a month but in fact need personal time. So those are some of the options with getting more time so that you can learn more.

I’ve got a list of a bunch of podcasts. Richie’s going to yell at me because his isn’t on here, but I don’t consider it a technical podcast, so – but there’s a lot of resources for listening and watching and learning and that sort of thing. Now, the next resource, so the first one is time, which I think is probably more plentiful than focus. I’ll say that. So focus is really, really important, and it’s a rare resource because if you work just a regular nine to five job, you’re probably a bit wiped out by the time you get home, right? And if you work as a DBA and you’re on call, your focus is even more scarce because you’re just burnt out. And so focus is really important to treasure and guard.

And so the first thing is I would consider making a learning space. Schedule some time, set aside time during the week, set aside a physical space and find ways to remove distractions. Literally set up a computer that doesn’t have access to the internet. There was a period of time – I’ve got a blog post about this, where my primary computer was a Raspberry Pi because I couldn’t install stupid apps on it. You know, the next thing is knowing your cycles. What time in the day, what time in the week do you have the most energy? And again, don’t be reading Twitter at eight in the morning on a Saturday when you actually have some energy. Read it when you’re brain dead after work. Know your cycles, and finally, for focus, take care of yourself. Eat well, exercise, sleep well. If focus out of the three of these, time, energy, money – if focus and energy is the most rare and the most valuable, then you really need to be making an investment in yourself so that you can have more of it, because this is how you get specialized. This is how you go deep, this is the constraining bottleneck resource on going deep is focus.

Some quick apps just to mention, Pomodoros, there are apps for it, but it’s basically just Italian for tomato, and it just means using a timer for 25 minutes and then taking a five-minute break. But it can be really helpful, I did it for a while. Did somebody have something?

Pinal Dave: No, no, it’s good. I think Brent is making us laugh, but I have one comment to make…

Brent Ozar: I had a glass on my desk.

Eugene Meidinger: Got it.

Pinal Dave: And that’s a secret weapon with which Brent makes maximum money, but coming to the point, I – one of the thing which really – I build SQL in 60 seconds videos okay, and my point was everybody said why 60 seconds, and I said, hey, while you were standing in a line of elevator you can do this while you can do this. I think the biggest – and I think I was very much proud. Somebody told me this and it made me pen put and point, said, hey, I’ve been always busy so when I’m standing in a line of elevator, when I’m making a water cooler run, when I am just climbing down the stairs, when I’m doing things, that’s my break time because after that I’m able to think more. And you are actually asking me to go intense and when I said, “Here the podcast of this and that,” I said when I’m driving it just gives me time to spend time with myself because after that I’m actually able to think more. So when I’m not planning, I plan the best. What do you want to say about it? I don’t want to trigger a discussion, but when I get this argument I kept quiet because I think I was waiting for the answer which you are going to give now.

Eugene Meidinger: Well, so let me make sure I understood what you’re saying. You’re saying that you know, for you, it doesn’t necessarily make a lot of sense to be using your commute to be listening to – to be doing more and to be doing podcasts and that kind of stuff because whenever you’ve got some unstructured time to actually think and not be as intense, then you come up with some of the best ideas or you just have a chance to organize yourself. Is that what you’re – did I get that right?

Brent Ozar: I hear that a lot.

Eugene Meidinger: Yes, I mean, so I think it goes back – you know, in the previous slide, one of the big things I had was know your bodily cycles, right? You need to understand how you learn best and what kind of resources that you have. So I’m just going to highlight that. Know your bodily cycle. So not everyone is the same. If you need some quiet time, that’s great. I think something important is you know, with the time and the focus, I don’t think it’s healthy if all of your time outside of work is spent learning. It really is not healthy. I know for me, whenever I’m driving to work, if I wasn’t listening to a podcast, I would be listening to music anyway, or something that really isn’t benefitting me in any significant manner. I’m still getting started in the morning, and whenever I’m coming home from work I’m definitely wiped out. Good.

Pinal Dave: You have a very valid point, I agree. Like Steve just said a few seconds ago, Steve Jones said, I go to yoga so I can eat a donut, and somebody also said I go to shower so I can think. So I think what you said makes sense totally.

Eugene Meidinger: Yes, but I think – I guess part of my point is I think there’s a lot of people who – they have this dead time and they’re not using it for anything right now, and so at a bare minimum, they might as well listen to Office Hours or SQL Server data partners or whatever, and at least be learning something, even if it’s passively, even if it’s half paying attention. Like there’s been a significant number of things that I learned where I wasn’t paying a ton of attention, and then I’d rewind and be like, wait, what was that about? JSON or SQL Server 2017 or something. So I think the overall message is be productive with your time. If you feel that you’re being productive with your commute just thinking about your day or planning or decompressing, awesome. But if you’re just listening to pop music and you don’t feel like you’re getting a lot of value out of that, maybe consider getting an iPod. I think that’s kind of the overall theme.

Pinal Dave: Yes, I totally agree. I think the message is very clear, that don’t find excuse to not to learn new thing.

Eugene Meidinger: Right. Going back to the apps real quick, again, the Pomodoros thing is just using a timer. The next three are browser plugins you can install, that will block you from time wasting websites, and you decide what that is and all that kind of stuff, and for some of them, you can specify specific times of the day or days of the week, and so maybe you say alright, whenever I’m supposed to actually be working, I’m going to block Facebook or something like that. Or you can even – some of them have like a nuclear option where you can say alright, for the next two hours, I’m going to do a blog post gosh darn it, and I’m going to block out any website except for stuff that’s on my white list. So those are really useful, and then the last one, people are going to think I’m crazy for suggesting, but I did it for a while, I thought it was really cool, there’s something called Beeminder, where you’re basically betting against yourself to get stuff done, and I think that’s an awesome idea but that’s because I’m a programmer so I don’t think quite right. But the whole idea is that literally like, let’s say you said I’m going to do 50 blog posts this year, or you said, I’m going to get a certain amount of points with CodeSchool.com or whatever. You make a contract thing on there and you say okay, if I don’t meet X goal, then I’m going to pay them $5 and you give them your credit card number, and then if you don’t meet that goal, it re-ups to twice that. So then it’s $10 and then it’s $30 and then it’s $90 and then it’s $270. And so basically, you’re paying them to take your money if you don’t meet the goals that you said, and there’s a lot of them where you can set up an automatic integration. So you can set up some custom ones by using their API or something like Zapier, but they’ve got stuff like GitHub. Let’s say you want to say I’m going to make a certain amount of contributions to this project on GitHub, or again with blog posts, there’s ways to integrate it.

There’s a bunch of different integrations, but it’s such a goofy thing, but if there’s something where you need a little bit of motivation or little bit better focus, I think it’s a really cool tool. So these are just some apps that can help you find some more of that focus, but I would honestly suggest like get a separate computer, don’t give it internet access, and make it so that when you’re sitting in that chair, you’re working on stuff that’s making your career go better, go farther.

Okay, so the last resource is money, and all I’m going to say is you have money, don’t tell me you don’t have money. If we look at the median pay in the United States for database administrators, it’s far and above the median pay for households in the United States. And I understand this is taking into account places like New York and Los Angeles and all that kind of stuff, but you know what, even in Pittsburgh, if you’re an entry level DBA, you’re probably still making at least $40,000 a year. Pittsburgh isn’t the best market for this kind of stuff, right? And you know, I understand that people have student debt and problems, but I think that one, don’t be afraid to spend money on learning. Because if we talk about your time is rare and your focus is even rarer, your money is a way to get some of that back because you lean on curation. I love Pluralsight because I pay them $30 month so I can get content that’s curated, and if you want to watch some of Pinal Dave’s videos you can go and see them on Pluralsight. And to me, it’s worth filtering out all the noise and all the junk that’s out there, or buying a book that you know someone’s put in a certain amount of time. And so money is going to be your most valuable resource, and again, even if you feel like you got to squeeze or something like that, think about this. If you’re getting paid $40,000 a year, that’s what, $15 an hour, $20 an hour, somewhere around there. If a book saves you $7 an hour’s worth of time, if it’s a $20 book and it saves you 3 hours, that’s worth the investment. And I don’t know how it is for everyone else, but I found that in my life, I’ve made this switch from having more time than money to going to college and not having much of either, to being employed and having way more money than time.

You know, we were joking about board games and stuff, and I buy way more board games than I can play and that’s a problem, but I’ll tell you what, I’ve got more money than time, and the further along you go in your career, the worse that’s going to get. Because when I was in high school, I probably would have gotten paid minimum wage if I did actually work, and so I played Everquest, which was a predecessor to WoW, because I had way more time than money. The opportunity cost was not a problem. But now, again, let’s go back a second. If you’re getting paid $40 an hour, your time is literally worth $40 an hour, and so again, as you advance in your career, you’re going to have more and more money and less and less time, and you need to start making some of that trade back. So don’t be afraid to spend some money on learning.

The second thing is get a budget. Just get a budget. Min.com is free, EveryDollar.com is free, I paid for YouNeedABudget, I think it has some nicer features that I like and just get a budget and keep track of your money. Because you’re going to find that if you’re managing your money, you’re going to have more than you thought you did once you’re actually keeping track of it instead of just fluttering out the window because you’re not paying attention. So those are the three inputs that we have to manage. Time, energy and money.

So let’s move on to the next thing is that’s just learning more stuff. We’ll just brute force the problem, throw things at the issue and we’re going to learn more things. Well, let’s see if we can learn the right things, right? So we maximize our inputs but it’s kind of like this funnel where you got a hundred raw resources, maybe a hundred training bucks or whatever you want to call it, and the problem is that now you only get 50 training coins or whatever unit of measure because you spend it on the wrong things. So to learn the right things, you want to increase the signal to noise ratio, you want to have a learning plan and you want to focus on depth and specialization. So let’s talk about each of those.

First, increase the signal to noise ratio. I know we all love Twitter, we all love our hacker news, but if you read those things, you’re going to learn about the last presidential election and you’re going to learn about Uber. You’re not going to have a very focused target. Whereas if you follow blogs that specialize in what you want to learn, if you use video content websites, if you read books, it’s going to have a much better signal to noise ratio. You’re going to have a much tighter focus with the content that you’re learning. You also want to learn things that are in your plan.

And the next thing is have a plan. Have a learning plan. So how do you get a learning plan because if you – like for me, I know that in three years, my three-year goal is I want to be working in data science or more likely data engineering. As I actually learn more about what data science is I realize I don’t want to do that, I realize I want to be a data engineer. But still, I want to learn R, I want to learn Python, I want to get on this bleeding edge. So part of that is reading job postings and seeing what companies are asking for. Because I’ll tell you what, when you read some of this stuff and you see the word data lake or Hadoop, you realize okay now I know there’s some backing behind this technology, I need to start learning some of this.

Consider certifications, and I know, certifications get a bad rap, they teach you stuff that’s impractical, they teach you the weird edge cases that no one ever uses. I have to wonder how many people here have used schema binding, how many people have used all four different types of XML output. I really have to wonder that, and those are the things – no, here’s the best one. Cube function and group sets, does anyone here use cube functions and group sets in T-SQL? I have to wonder. Those are the things that were on the T-SQL certification that probably have not helped. But a certification gets rid of unknown unknowns. It gives you a lay of the land, it tells you what’s out there. And I can tell you what, the admin certification for me was a life saver because I remember going through all the high availability options and learning about those and trying to remember the difference between mirroring and replication and thinking, we’ve got like three SQL Servers, when am I going to need this? And then suddenly I ended up doing consulting and it was really, really important for me to understand the difference between availability groups and availability failover cluster instances. And because I had taken the certification and read the content, I knew all of the stuff I needed to learn and I at least had some foundation. So if you don’t have a plan, a certification is at least a good starting point.

Part of that plan is have a one-year goal, have a three-year goal. My one-year goal right now is two things. One is to do the edX course on data science in a year, and two is to blog every other week. And I guess my three year goal is to get into data analytics or something like that, and once you have that plan, start working your plan, because once you have a plan, more of the stuff you learn is going to fit in it, or at least you have a rubric to say does this really make sense with what I’m doing?

Continuing on that, I said this before, I’m going to say it again. Focus on specialization. Exposure and generalization require, by definition, learning the wrong things because you don’t know a priority. You don’t know ahead of time which ones are the right things. That’s the whole point. If you don’t know anything about big data, can you tell me whether Spark or Hadoop or Flume or Pig or Scoop or Vertica or Greenplum makes sense for what you need? No. You have to get conversational about it before you can make the right choices. Exposure, passive learning is about dealing with those unknown unknowns. The things you don’t even know you don’t know. Also, huge tangent, but in Slack, go ahead and Google big data or Pokemon, you won’t be disappointed. Anyway, so by definition, exposure and generalization is going to have a high noise level because you don’t know ahead of time what stuff is wheat and what stuff is chaff. And by definition, you’re learning about things that aren’t going to be right for you because you’re getting the lay of the land.

But if you need to know a specific area, going deep is more likely to pay off. If you need to learn SQL Server, then reading SQL Server internals is far more likely to pay off. Mastery and expertise are about known unknowns. They’re about the things that you know you have to learn, and by definition, they’re going to have a higher chance of payoff. So that’s the second big thing. So we learned – we figured out how do we learn more things, we saw how to learn the right things, and now the last piece is how do we learn things that last the test of time? How do we learn things that last longer?

So the first one is avoid volatile skillsets. Angular, you know, JavaScript frameworks, certain Cloud technologies, a lot of this big data stuff. Don’t make too much of an investment in these things that you know five years from now are going to change a lot. Or you know, may not be around. I’m not saying big data is going away, but I don’t know if Spark is still going to be the go-to tool or Hadoop is still going to be the go-to tool or what.

Go deep, okay? Internals, fundamentals, theory. These all last the test of time much better than surface level functionality and features. So go deep, learn things that are timeless, learn things that are going to transfer. So one of the big things is soft skills are timeless. SQL Server changes versions every year. Generations of people change versions every 40 years. People have not changed that much in the past 100 years. Sure, we wear different clothes and our bikes are a lot smaller, but people don’t change that fast. Take a technical writing course. Learn how to write a good abstract, learn how to write a good email, learn how to communicate. Present for Pete’s sake, so that you learn how to talk to people. I’ve said this before, but if you can present to 50 people right in front of you or 300 internet strangers without sweating, I can promise you that if you’re a consultant, you can do it in front of two customers. Learn soft skills. Learn project management, learn how to keep things on the rail. So soft skills are going to have a lower rate of decay. They’re going to be more timeless.

Some resources really quick, find a Toastmasters, or find a user group. Like I said, user groups are always looking for speakers and they love to promote the local community, so find a user group and make a presentation and you don’t have to be an expert. You just have to be honest and you have to do your homework. You don’t have to BS people. You say I’m still learning this but here’s what I found out. That’s okay. My first presentation was execution plans, I had been working with SQL for a year. It’s okay.

Some books I’d recommend that have helped me a lot, How to Win Friends and Influence People is a standard by Dale Carnegie. A lot of commonsensical stuff but I needed common sense when I was in my teens. It was good. How to Have Confidence and Power Dealing with People. It’s a mouthful, it’s from like the fifties and again, it’s more commonsensical stuff, but it really changed my life in the business place, and one of the biggest things – well, two biggest things was one, understanding the value of small talk. Because I’m an introvert, and if you start having a conversation with me, it’s like a little taxi cab meter starts running, and every hour I just got dinged a dollar, and if you’re talking to me about your dog, that is a very expensive conversation to be having. But if we’re talking about something a bit deeper, I see the value in it, so it helped me understand the value of small talk and it helped me understand the value of paraphrasing things and speaking words back to someone to make sure you’re on the same page. The last book I would suggest is Crucial Conversations. It’s really good for dealing with those heated moments, and then for a podcast, there’s two by the same people – Career Tools and Manager Tools, and they teach in a practical way soft skills, people skills, and they give you literally like an actionable check list. So much better than fuzzy stuff about we’ll just be nice or be agreeable for be friendly or whatever.

Another thing is deep dives last longer. These give you a rich mental model that you can use to create analogies to other technologies. I can promise you that if you learn how a transaction log actually works, when you switch to Oracle or MySQL or Postgres, you’re going to be able to guess a lot of that kind of stuff. If you understand how a floating point data type is stored, then you’re going to understand a lot more about implicit inversions in these other sorts of things. So learn the fundamentals, learn the theory, learn the internals because those last so much longer than which Window function features were introduced in SQL Server 2012.

Alright, so to recap everything, one is use multitasking to get more of your time back. Take advantage of the time that you’re not using your brain, and consider using it in a way that’s not too intensive. Second is take care of yourself. Get a good sleep, get good exercise, get good food in your body. Third is schedule time. Set aside time for blogging, home labs, whatever, make sure you’re making time for this stuff because it’s so hard to make time for it and it’s so easy to waste time on Twitter and Reddit. Fourth is spend money on curation. Spend money, buy a book, buy video training, something. Your time as you get further in your career is going to get more and more and more valuable. Be willing to spend money to support that. Five is understand this tension that’s constantly there between specialization and generalization and understand how specialization presents slow-moving risks, and generalization presents fast moving risks. Building upon that, pick a specialization, you can change it later, it’s okay. Whenever I started my career in this, it was SQL and it was BI development and that kind of stuff, and now I’m making a pivot into more PowerBI and eventually data analytics and data science, and that’s okay, because right now I’ve gotten to the point – I’ve got enough specialization that I can stand in front of all of you and pretend like I know what I’m talking about, I can stand in front of a customer and tell them that they should have their full backups less frequently than every 30 minutes, and so I’ve gotten enough specialization that I’m adding value and that decay on that is slow enough that I have time to jump to something else and build up something else before I’m suddenly a Cobalt developer.

Seven, make a plan. Make a plan, make a plan, make a plan. Write it down, write down your goals and find some sort of accountability partner who’s going to follow up every quarter or follow up in a year and tell you hey, you said you were going to do this and you didn’t. And then the very last thing is learn things that will last the test of time. People skills never go out of style. True fundamentals never go out of style. I’m not saying you have to learn a bubble sort, but I’ll tell you what, if you understand some basics, if you understand how a transaction log really works and you understand how data types really works and you understand how a database – a relation database actually functions, you’re going to have so much more flexibility whenever you need to switch the big data or something like that, and you can compare and contrast and go, well how do they deal with this problem because here’s how SQL deals with it. Alright, that is my presentation.

Brent Ozar: Man, there was so much active discussion in Slack.

Pinal Dave: I have to say one statement before I let everybody speak. I think what you did Eugene today, the word I have, the statement I have for you is share your knowledge, it’s a way to achieve immorality. I think you just has amazing presentations, so much wisdom, you coach us so much, I think this was fantastic, and so much conversation out there from bath shower to eating donuts.

Eugene Meidinger: That’s very awesome. I’m excited too because this is like of those Evergreen things that you know, people are still going to be bugging me about five years from now, maybe we’ll see.

Brent Ozar: It’s true, it’s absolutely true. So much discussion around the percentage…

Eugene Meidinger: Awesome.

Brent Ozar: And I never thought about it as solve for X, five years in, what percentage decreases. So there was even things out on Twitter where people were saying well look, the stuff that I learned a long time ago is still valuable, well that’s kind of true but there’s less of a market for it. You know, there are less people who are interested in – so when you’re making your own decisions about what technologies you want to go bet on next, what made you change from say SQL Server over to PowerBI?

Eugene Meidinger: Yes, a couple different things. So one of them was you know, we had that conversation about okay, do you specialize in your workplace or that sort of thing, and for me, there was – it helps at least have some alignment, right? It doesn’t have to be 100%, but it’s really, really hard to become an expert in something that your workplace isn’t even doing. So it’s definitely something a direction my boss was pushing in, so that was one thing. Another thing is that you look at where Microsoft is dumping their money, and they’re dumping it really heavy on this.

Pinal Dave: [Crosstalk]

Eugene Meidinger: Go and become a Windows phone expert, yeah. Silverlight, Windows phone, XNA game development, no, not so much. But I mean, you look at like the Microsoft data insight summit and the stuff that they’re doing with that, and just I think that PowerBI, I joke, is a solution waiting for a problem. I feel like Microsoft went alright, we need to do BI, we need to do Cloud, and we need to do something that is in SSRS. Find me a solution, build it agile and we’ll find the market later. You know what I mean? And the third thing is you just have to pick something at a certain point. You just have to make a decision and hope it’s the right one.

Brent Ozar: Yes, I switched not – your deck was fantastic, but there’s so much question discussion in Slack that I might as well show it in here while…

Eugene Meidinger: I’ll post the link back in there so at least…

Pinal Dave: And you know Steve is saying national donut day right?

Brent Ozar: Is that what he’s talking about? Donut day? Yes, of course, he’s all about that. Steve makes a good point, he says, “Be careful, Microsoft just dumped a lot of money in some pretty bad holes.” That can’t be the only thing you look at. You have to look at other things too, like is there customer demand for it, you look at – Windows phone was such a good example because Microsoft set a ton of money on fire in there, but as long as you were in touch with consumers, you knew that you’re like yes, this isn’t going anywhere.

Eugene Meidinger: Well, another thing is you have to look at like okay, if we’re worried about these being ships and one of them hitting an iceberg, what other ships can I jump onto that are right nearby? Right? So a couple different things.If you – like for PowerBI, because I just know about that, if you take the visualization route, a lot of the technology and skills can transfer to Tablo or Clickview. It’s not perfect, but you can make that jump. Another thing is like okay yes, let’s say PowerBI does die. I don’t think they’re going to kill Tabular for SSAS. Again, they might, but that’s another pivot. So when you’re making this risk, figure out do I have any pivot points where I can say well this just got killed off but there’s somewhere else I can jump onto.

Richie Rump: Well I think there’s some other fundamental type stuff, especially a developer type hole where you’re understanding Git, you’re going to – whatever language you use, understanding Git, knowing how it works is going to be – and unit test. No matter what tool you use, understanding how unit tests works is going to be beneficial.

Brent Ozar: And I would say too, and not only knowing Git but leaving a public trail if you can. If checking things in a projects in GitHub, building a blog, I mean, you look at people who built blogs over time or built open source project, sometimes folks say I feel guilty for leaving this stuff behind that looks kind of elementary. It doesn’t look that hard. That’s your proof that you knew it then, that you’re building up a long trail of the things that you know.

Erik Darling: You feel cornered by the fact that you have SQL in your name?

Eugene Meidinger: I’ve heard people make that argument, I know Brent’s like use your full name. SQL’s older than me, I feel like it’s not going to disappear. I think it’s like – you know, yes I promise my middle name isn’t like SQL – it’s not Eugene SQL Meidinger, but yes, in the Twitter handle, that kind of stuff.

Drew Furgiuele: I get a ton of requests every day from everybody.

Brent Ozar: I’m in a mute Drew, hold on a second, there we go, okay.

Eugene Meidinger: A lot of people make that advice of you know, don’t put a technology in your handle or something like that, but I really don’t feel like SQL is going to be that fast of a moving target that it’s going to go away, so I’m not too worried.

Erik Darling: Yes, I had to get rid of all my side base Erik nicknames, it sucked.

Eugene Meidinger: With the Foxboro stuff you kept around.

Erik Darling: Foxboro Darling, still my first Gmail address.

Richie Rump: Calling myself [inaudible] just wasn’t a good idea.

Brent Ozar: Well thank you very much for Eugene, or SQL Gene presenting today. Excellent job, sparking tons of discussion in here.

The following two tabs change content below.
Eugene is a Business Intelligence developer in the Pittsburgh area. He helps run the local Power BI user group.
Previous Post
T-SQL for Beginners
Next Post
DBAs vs Developers: JSON in SQL Server 2016

6 Comments. Leave new

Andrew Notarian
February 7, 2017 9:12 am

Teach a man to fish, right? Will there be 0 technical content in this?

Reply

    Short answer: No code. No demos.

    Long answer: 0 technical content, but an assumed audience of SQL professionals. So I’ll have a list of SQL podcasts when I talk about multitasking, for example. So there is a target audience, but this could apply to almost anyone in the tech field. I’ll also assume an intuitive understanding of math, so things might get a bit mathy when I mention things like exponential decay. But I’m hoping to keep the content approachable overall.

    Reply

    Now that you’ve got me thinking, I could take the mathematical model I’m imagining and turn it into an interactive Power BI report. Hmmmmmm. I like this idea.

    Reply

You’ve grabbed my interest! One of the most common complaints I hear from our devs is there’s never enough time to learn all the tech they have to and get their jobs done. With all that SQL Server has to offer now, I sometimes feel overwhelmed as well.

Reply

I have to admit that I do sometimes (often!) feel like I’m behind the curve in keeping up with tech is concerned. It would be interesting and useful to hear about coping mechanisms and learning techniques and your point of view on things

Reply

I do feel the impossible pressure of staying up to date with so many things that are changing. I would love to get someone else’s perspective on this.

Reply

Leave a Reply to Andrew Notarian Cancel reply

Your email address will not be published. Required fields are marked *

Fill out this field
Fill out this field
Please enter a valid email address.

Menu