Ask Slashdot: What Should I Study?

A fellow Slashdot reader is seeking advice on a new field of study: After many years at the same company, I'm now thinking of a change. At my current place of work, I have worked on many different projects, from server side development, to UI development, and most recently, a lot of data science work. If I were to rate myself, I consider myself to be a good developer, thorough, conscientious and always willing to learn new things. Even my recent foray into data science (though not entirely new, since my graduate studies specialized in machine learning) has had reasonable success, and ideally, I'd really like to continue working in this space.

But, I'm starting to feel in a rut and I'm looking for a change. And looking outside my company, I'm not sure how to begin. Should I hit the books again? Should I focus on any specific technologies? I haven't particularly kept up with new technology -- after working for so long, I tend to think of that as something I can learn, when I need to. Any advice on how I should go about preparing for interviews? I'm quite willing to put in a few months of work into prep, so all suggestions are welcome!

Comments

Business Administration or MBA

[WillAffleckUW]

They also have MBIT if you want to stay in IT.

Money is the root of all Google.

FP

[reanjr]

If you haven't delved into it, try learning more about functional programming. In my experience, most devs haven't really built anything using FP, and it can be illuminating. react.js is a pretty good example of a system that works well with FP.

Re:FP

[ShanghaiBill]

try learning more about functional programming. In my experience, most devs haven't really built anything using FP

Functional programming is a hot topic ... yet very little has actually been built using it. That indicates to me that it is just hype. Programming with "pure" functions sounds nice in theory, but the real world has "state", and you don't get far by pretending that it doesn't.

Re: FP

[Anonymous Coward]

Functional programming is a hot topic ... yet very little has actually been built using it. That indicates to me that it is just hype. Programming with "pure" functions sounds nice in theory, but the real world has "state", and you don't get far by pretending that it doesn't.

Huh? In functional programming, state is passed in the arguments to the functions.

Seems like you don't really understand FP.

Re: FP

[ShanghaiBill]

Huh? In functional programming, state is passed in the arguments to the functions.

Which means it doesn't scale. At all. Please show me a device driver written in FP. A backtracking parser? A deep-learning engine? Are you going to pass around 200MB tensors with every function call?

Seems like you don't really understand FP.

If you understand it so well, show us your useful applications.

Functional programming was invented by mathematicians who want chicks to think they earn as much money as real coders.

Re: FP

[phantomfive]

Programming with pure functions is a great idea that reduces bugs. I do it even when I am using oop . It is not good in all situations of course, so it is a case of using the right tool for the job.

Re:FP

[LynnwoodRooster]

Ever done real-time embedded firmware? It's pretty much all functional because it's highly deterministic, stable, and easy to debug with limited UI access. EVERY paradigm has its place; only a rather ignorant engineer would claim "X is a useless paradigm"...

Re:FP

[110010001000]

"most devs haven't really built anything using FP"

Gee, I wonder why that is?

Re: FP

[ShanghaiBill]

Programming with pure functions is a great idea that reduces bugs.

Functional programming makes sense for simple things that are inherently stateless. Sqrt(), sin() and cos() have been "pure" functions for as long as there have been compilers. But it does not make sense for things that are inherently stateful, which happens in almost any useful application.

I do it even when I am using oop.

So does everyone else. Many oop apps use simple stateless functions. But if you are actually passing around objects by-value by shoving them onto the call stack, you are taking it way too far.

It is not good in all situations of course, so it is a case of using the right tool for the job.

The whole point of functional programming is to wedge the use of pure functions into places where it is the wrong tool. Otherwise it would be nothing new.

Can you give a single example of "functional programming" providing a better solution than traditional methods?

Re: FP

[phantomfive]

Java strings are an API that is functional even though it is in an oop language. Each time you make a transformation on a string you get a new string, the strings themselves are immutable.

Re: FP

[nowwith25percentmore]

If you want to delve into functional programming, try Erlang. It's a practical distributed functional programming language that's been in production use in telecom systems for quite some time. It's also a mind-screw to learn.

Re:FP

[mnemotronic]

Functional programming is shit. It's complex, it's difficult to learn and use, it's slow to develop, and it's just plain inefficient. It's nothing more than a thought exercise except to those who want to masturbate publicly to demonstrate their egos for knowing how to use a useless language paradigm.

oh, it feels so good ;)

Sounds kinda like it's the programming version of a major in Art History.

Re:FP

[mikael]

Every language is like this - they are wonderful at small-scale academic proof-of-concept experiments with a few hundred lines of code. Start going to the millions of lines of code for things like large scale GUI applications and then every language starts to have problems: C - too low level, requiring a whole set of functions to be written for a simple ADT's to represent a new list structure. script-based GUI's - impossible to debug and step through. C++ - too many ways of doing things and everyone starts trying to demonstrate their extreme cleverness by trying to find the most insanely complex way of doing something using STL and Boost. The "slightly slower" speeds of languages of just-in-time compiled languages also start to show themselves in terms of performance - the spinning wheel of fate.

Re:FP

[ShanghaiBill]

Every language is like this

... but not equally. Some are much worse than others. If you were starting a big new project, and expecting to write tens of thousands of lines of code, then Java or C++ would be reasonable choices. Haskell or some other FP language would not. Here is a list of Haskell programs. The list is short, is mostly small programs, and many of them are mathematical utilities, which is a narrow field were FP actually makes sense.

Re:FP

[serviscope_minor]

Functional programming is a hot topic ... yet very little has actually been built using it.

Try: half of C++. The C++ templating mechanism is a pure funtional functional language with an um interesting syntax.

Programming with "pure" functions sounds nice in theory, but the real world has "state", and you don't get far by pretending that it doesn't.

Doesn't mean the techniques aren't useful. It's become popular in C++ for instance to take those ideas from functional programming, like pure functions. The modren style is to have as many functions as possible be pure, and keep the state well contained. That makes the program much easier to reason about. And, have constexpr functions which are a limited form of pure functions now.

Re: FP

[serviscope_minor]

Functional programming makes sense for simple things that are inherently stateless. Sqrt(), sin() and cos() have been "pure" functions for as long as there have been compilers.

They have not been. All of those will set errno. This in fact has turned out over the years to be a right pain in the arse for threading and optimization.

But if you are actually passing around objects by-value by shoving them onto the call stack, you are taking it way too far.

Going too far towards making your program fast...

C++ is very hot on value semantics as a style and the passing by value thing on the stack also means the compiler can reason perfectly about the scope of varibles for conforming programs. This gives it an advantage over GC languages like Java, since they have to rely on escape analysis which is always less good than perfect knowledge.

Re:FP

[Hognoxious]

That's a totally unfair comparison.

A major in Art History could land you a job as a museum guide.

Re:FP

[serviscope_minor]

Unless it's the other kind which is a fuck off massive state machine.

Re: FP

[angel'o'sphere]

A backtracking parser?
This is a prime example for functional programming.

Re: FP

[angel'o'sphere]

You don't know much about functional programming, first of all it is not what you think, your functions here like sin() have nothing to do with it.

Secondly, real world functional languages allow to modify state via monads. https://en.wikipedia.org/wiki/...

In a functional language "functions are first class citizens", that means they behave like objects. e.g. the following code:

      a = sum(sqrt(3.7), sqrt(4.2)

Creates three function objects, one instance of sum and two of sqrt, then it evaluates the "expression tree"
Behaving like objects means, you can store them, pass them around as parameters, create new functions by combining existing ones.

Regarding your answers: FP is not a hype, it is ages old, most likely older than you.

All modern languages support features of FP (C++ with functor objects, and now with lambdas), and nearly all modern languages either have lambdas or closures.

Depending on project and Java Version I use lambdas every day. If you use the STL in C++ you can not help yourself doing FP (albeit a bit limited)

Most prominent is probably Erlang. They used to write all the router and switching software in Ericssons gear in Erlang. https://www.erlang.org/about

That FP is only limited used is because the modern implementations e.g. in Java are so watered down and only support a subset of usages (e.g. lambdas/closures and the Streams API: http://www.oracle.com/technetw... )

Re: FP

[serviscope_minor]

Kind of sounds like you're replying to the GP, not me.

Re:FP

[epine]

C++ - too many ways of doing things and everyone starts trying to demonstrate their extreme cleverness by trying to find the most insanely complex way of doing something using STL and Boost.

You think that's bad?

Someday you should count up the number of different ways God invented to compute pi.

pi = 16 arctan(1/5) - 4 arctan(1/239)

Seriously, any computation of pi more clever than that is pure showboat.

Re: FP

[angel'o'sphere]

True, how could that happen ... I answered to ShanghaiBill.

Re: FP

[Hognoxious]

But if you are actually passing around objects by-value by shoving them onto the call stack, you are taking it way too far.

Nobody would be that stupid.

I just put everything in a massive struct of doom and pass a pointer to it everywhere.

Re:FP

[Hognoxious]

Somebody, somewhere, is thinking that simplifies to 12 arctan (1/234).

The worrying thing is that he's probably somebody's boss. The other worrying things is that apart from "it just obviously doesn't" I can't really explain why.

Re: FP

[reanjr]

I would guess it's because you don't need nearly as much FP code to do the work of comparatively large amounts of OOP code. All the good FP is being handled by a relatively small group of talented coders, while we keep shoveling more bad coders out of university who are more comfortable with OOP.

For example, erlang is used in absolutely massive telecom systems. Meanwhile, OOP developers are trying to figure design patterns that work well with parallel processing, and designing complicated build and failover systems to overcome the inability to live patch.

Re: FP

[reanjr]

"Object oriented programming is shit. It's complex, it's difficult to learn and use "correctly", it's slow to develop, and it's just plain inefficient. It's nothing more than a thought exercise except to those who want to masturbate publicly to demonstrate their egos for knowing how to use a useless language paradigm."

Fixed that for you.

What is your goal?

[guga.hagenbeck]

Why do you want to study a new field? What are your expectations from it, your goals?

Re: What is your goal?

[Anonymous Coward]

I'm in a similar situation to OP and for myself, the reason is because most employers expect you to be able to hit the ground running day 1. If you want to transfer you better find your exact position somewhere else or you have to start adapting to whatever specific tools, technologies and approaches you want to get into. A bit ridiculous but hey that's the landscape these days.

Complete change of pace may be required.

[Anonymous Coward]

Become a plumber.

Re:Complete change of pace may be required.

[rogoshen1]

Well, that's removing one of the middle men at least -- perhaps try proctology ?

Re:Complete change of pace may be required.

[The Fat Bastard]

I knew a guy who retired from IT to start his own roofing company in his mid-50's. Makes more money in a summer than he does all year in IT.

What do you find interesting?

[Anonymous Coward]

Studying things that other people want you to sounds like a recipe for boredom.

Become a Renaissance Person

[al0ha]

Look at the opportunity out there and become skilled at something completely different. There's a crapload to be made in many skilled trades now that Baby Boomers are retiring out. Some trades like plumbing and electrician can't find enough people, and the opportunity to become very successful is wide open. Be a long time before robots take the job of a plummer, electrician and other skilled laborer.

This is what I'd do if I were in my 30s even.

Re:Become a Renaissance Person

[b0s0z0ku]

What about teaching/academia? Neither is going anywhere quickly either.

Re:Become a Renaissance Person

[geoskd]

Look at the opportunity out there and become skilled at something completely different. There's a crapload to be made in many skilled trades now that Baby Boomers are retiring out. Some trades like plumbing and electrician can't find enough people, and the opportunity to become very successful is wide open. Be a long time before robots take the job of a plumber, electrician and other skilled laborer.

It is also painfully difficult to break into either of those trades from the outside. Most places require you to be licensed to practice, and even if they don't, no insurance will touch you if you don't have the requisite certifications and/or licenses.

If that doesn't sound too bad, look at what is required to get those credentials. Almost all accreditation programs and licensing rules require you to have at least a year or two as an apprentice under a master. The problem with that, is that there is absolutely no reason in the world that any master is going to want to take on a random apprentice. For the first two years of apprenticeships, on average the apprentice has negative value to the master (They cost more time and effort to look after than they produce in useful output). That is why the few that do offer apprenticeships, do so at minimum wage for two years. Most of the rest, even if it were of net value to the master to take on an apprentice, they would still be wary because every new apprentice that you support now will be a competitor in 5 years, and in any given middle size city (50,000 - 100,0000 metro area population), can have as few as a few dozen actual plumbers or electricians. For each new master that gets added, every single existing master will take an approximately 3% paycut; why would they willingly do that to themselves?

Most times when you see an apprenticeship, it is a son or daughter following in the family trade, and the only reason that elder master takes on the apprentice is because they are kin.

Re:Become a Renaissance Person

[Strider-]

It really depends on where you are. In a lot of places, the trades are absolutely starved for people, and will readly accept anyone who is willing to put in the effort to actually work hard and learn the trade.

If you're interested in electrical, always a good idea to contact your local IBEW branch, many of them will have all the information you need to get into the trade, and get better wages/benefits than you would otherwise.

Re:Become a Renaissance Person

[geoskd]

10 years ago, I asked at the local IBEW about it. I already have a degree that would allow me to skip half of the requisite experience to get my license, and I know what I am doing thanks to a family history (Grandfather was a lineman for a utility in California before he retired). Because I would be coming in at low seniority, I could expect two years of hit and miss work (maybe one week a month of paid work, and when they call I must show up or get dropped). After that, they said once I got a permanent position somewhere, I could begin my actual apprenticeship and could expect 2 more years of full time employment at $15 per hour before I could test for my license. Once I got the license I would have more options, but until then I was effectively a slave.

Even at that time, I was making $25 as a line supervisor. Granted it was a dead end job at what can only be described as the shittiest company I have ever heard of, but to get that license I would have to basically go back to living in my mothers basement for 5 years, and wait with my fingers crossed that I got lucky and got the shorter end of the waiting period. Even then, I could expect a cap of around $80k per year unless I was willing to put in 80 hour weeks for years to start my own company and handle all of the business side of that deal.

I elected to go a third route. I fought tooth and nail to get back into my principle field of study and now I make that same 80k, working 40 hour weeks, and can expect to go management sometime within the next few years to get into 6 figures.

Going into the trades only makes sense when you still have the option of living at home to cover the rough years, and even then it only makes sense if you don't have the means to get that higher education and the degree that goes with it.

Re: Become a Renaissance Person

[nowwith25percentmore]

Is there really that much demand for new electricians? I've been seriously considering switching to that field, but the demand & pay stats I was finding online didn't seem horribly motivating.

Re:Become a Renaissance Person

[RKThoadan]

You can rise to the top *real* quick though. My brother and brother-in-law are both master electricians. Many of the people who try to fall-back on the trades are people who are barely able to succeed at tying their shoes every morning. If you show up on time (sober), work hard, can follow instructions and aren't a total jerk you're doing better than 80% of the people and will quickly be become a guy the master electrician wants with him on every job - that can be way before any official apprenticeship starts. If you happen to understand how electricity works that's nice as well. I doubt they pay more than $15/hr, but that's not a bad wage for somebody without a degree. If you already have a degree, probably not the best idea, but far from the worst.

Re:Become a Renaissance Person

[GLMDesigns]

Disagree totally. Academia is going to be rocked sooner rather than later.

Right now people are saying that it's real use is making connections. Ha!!! Learning. No. Connections. (As a parent of a teenager I hear this constantly.)
"Video killed the radio star" comes to the current university system.

ALL lecture courses will be gone. Lab work is a different story.

Re:Become a Renaissance Person

[b0s0z0ku]

That's what was said 100 years ago when audio movie projectors were invented and 50 years ago when VCRs were invented. I'll believe it when I see it.

Re:Become a Renaissance Person

[Hognoxious]

Almost all accreditation programs and licensing rules require you to have at least a year or two as an apprentice under a master.

I doubt that's true for everything everywhere. My brother was looking at getting his CORGI (gas fitting) cert some years back and it was possible to do an intensive course in a couple of months if you already had the hands-on skills - which as a plumber, he did.

In the end he was put off by the classroom side of it, but I think he could have done it if he tried.

AI/Networking/Security/Law

[sdinfoserv]

I'm in Seattle area... People with Cisco CNE's, Security CISSP's are constantly getting poached. Good security people bring $200k-$1M salaries out here. Network engineers make in the $100K range (as do programmers out here).
AI is really growing an high paid, but you need a Phd to grab a top salary in AI. If you have that, you can start at the same wages (or more) of a neurosurgeon.
If I were 21 today and starting over... seriously.. I would spend 4 years in the military. Get out and get a job as a fire fighter. They start out here at $80K. Some work 10 days on, 20 days off..(those 10 days you live in the house). Retire at 53 or 54 with a full pension and health care and spend the next 30-40 years fishing, hunting, playing with grand kids, traveling... what ever.

Re:AI/Networking/Security/Law

[jon3k]

Came here to post this. Security, security, security. It is an absolute gold rush right now and the problem is getting worse not better. I don't know how many people outside of the Bay Area or NYC are making $200k-$1M in security, but $100-$200k with only a couple years experience and a CISSP can probably get you 100-200k in pretty much any tier 1 or tier 2 city. Alternatively you can get into government contract work, get a TS(/SCI) and bounce around contractors with insanely good insurance and pretty much guaranteed work.

Re:AI/Networking/Security/Law

[Alypius]

This is hugely reassuring. I live on Whidbey Island, halfway through my MS in infosec, and about to start on a second career. I know enough to keep my 4.0 and my current day job is risk management. Can't get the CISSP because of the employment requirement, but I can get close enough and I enjoy the subject matter. Thanks!

Re:AI/Networking/Security/Law

[b0s0z0ku]

Thing about the military is that you may be required to kill or do harm for causes you might not believe in. Is it really worth having this on your conscience?

Re:AI/Networking/Security/Law

[sdinfoserv]

You're conscience, don't speak for me.
Humanity is a tiny thread from chaos and mass killings. Every single event in animal history - humans included - where resource contention happens, survival turns us into vicious killing machines. Civility today in America (most of it), is a result of a rare time in human history where most have "plenty". That has never endured in the course of human existence so don't assume it will continue. ...... but it's nice that you're life, and the lives of those you care about is so sheltered, blissful and bountiful you never have to consider strife.

Re:AI/Networking/Security/Law

[b0s0z0ku]

Why be a part of adding to strife worldwide, though?

Re:AI/Networking/Security/Law

[RPI Geek]

Getting a full-time firefighting job is harder than you might think.

My brother-in-law was in the military, was deployed in a combat zone, and upon discharge (honorable and with multiple letters of recommendation) spent ~5 years trying to get into a decent firefighting academy. He went up and down the east coast and despite being in the top 10% of the test-takers, both academic and physical, he was frequently turned away because of the rampant nepotism and one person just "not liking him" (the ones who DID like him there told him he could appeal the decision or sue, but it was basically the same person who ran the appeals board and he'd just be wasting his time - but would he like another letter of recommendation?).

He did end up with his dream job as a firefighter and is happy and secure now, but the competition is fierce and not even close to fair.

Re:AI/Networking/Security/Law

[david_thornley]

That's not really relevant to someone enlisting. The strife will happen, and one more or less recruit will have approximately no effect.

Re:AI/Networking/Security/Law

[Hognoxious]

That's like saying that if I throw one empty bottle on the beach it doesn't matter. It matters if everyone takes the same attitude.

Since you apparently aren't a millennial I'm surprised you haven't heard the expression "Suppose they gave a war and nobody came.

Re:AI/Networking/Security/Law

[david_thornley]

I have heard that expression. I sang songs on that theme when I was in college. Realistically, however, large numbers of people will enlist.

Also, in the real world, that depends on nobody from either side showing up, and that's a lot harder to arrange. It's a good thing people from all over the world showed up in WWII.

Re:AI/Networking/Security/Law

[b0s0z0ku]

It's not a good thing that people showed up for WW2. It's a shame there weren't more people like Stauffenberg -- one of them would have been a true hero if they were successful.

What's the question?

[DogDude]

This is a terrible submission. What the hell kind of question is "what should I study" with zero context? How fucking arbitrary is this?

Re:What's the question?

[RonLillycrop]

Agreed. The answer is "whatever makes you most happy" which could be "what will make me the most money" for some and "will get me to travel the most" for others and "give me lots of social interaction" for others and "let me get my elbows-deep in pig shit" for others. It's anybody's guess what will float this guys boat.

Re:What's the question?

[wvmarle]

It also makes the answer simple.

Study what you find interesting! If only because personal interest tends to be the strongest, longest lasting motivator.

Re:What's the question?

[serviscope_minor]

What the hell kind of question is "what should I study" with zero context?

You know, what would be really neat is if slashdot would provide a longer summary to go along with the very brief headline. They could even have links in there. Honestly though I don't think that would work because no one would read the summary let alone the links.

Clarification from the original poster

[Anonymous Coward]

Original AC here. I should clarify - I don't want to entirely change my field of work. I still want to stay in programming, and possibly data science. I'm just really nervous about interviewing after a *very* long time, and I'm wondering how to go about it. I also have a very varied set of experiences, not specializing in any one thing - just really a matter of doing what was needed, when it was needed. I'm not sure how this will go down in interviews, and how to best portray it.

Re:Clarification from the original poster

[Arzaboa]

Just go try it. Failing is how you learn to succeed. If you're nervous about a few interviews, go interview at a few places you don't want to work for practice. You'll figure it out, just takes some practice learning to talk the talk.

--
"Thanks All Folks" - P. Pig

Re:Clarification from the original poster

[Hognoxious]

If that's the best you can do when asking a question I'd suggest you try to work your way up to ditch digger or road-sweeper.

Re: Clarification from the original poster

[Camembert]

In general, try not to interview first for your dream job, best to have a little experience handling interviews.

Re:Clarification from the original poster

[LostMyBeaver]

You're making the mistake of thinking that a person in your age group should interview. I'll imagine that you're at least 35+ possibly even as high as 50. We don't interview at that age anymore. We figure out what we want to do, then we think about who we'd like to work for. Then we make the friends through social networking that would present us with the opportunity to meet the right people to get us on payroll.

I don't even hire people I have to interview. I sometimes have lunch with someone that's recommended to me. But to be fair... as soon as I see a resume and I hear job interview, that's over with. I sure as hell don't want to hire anyone who is over 30 who is going to send me resumes.

I have often talked with people who I find on Github and Gitter. If I like their code and they play well with others and display a good work ethic and they make the comment they're looking for a job or make hints they're interested in moving, that's a great way to meet people. Meetups are more for desperate people. It's like speed dating for people who were splashed with acid.

If you're interested in something, invest the time in making sure people know you're a smart guy and willing to move. This way you draw jobs to you not the other way around.

Interviewing is something you do as a college grad trying to get that first job. Or it could be something you do if you're trying to pimp yourself off to Microsoft, Amazon or Google for example. But even with those companies, I'd just make friends with senior level developers and mention that I think it would be interesting to work for a behemoth from the inside for once... but I wouldn't want to be just another badge number. I would recommend in that case that you don't express interest in their vest and rest plans.

Re: Clarification from the original poster

[Anonymous Coward]

Actual data scientist here. Want to up your game? Learn the rarest skill among those who bill themselves as data scientists: statistics. Not R, not SPSS, for ghu's sake not SAS. Pencil and paper statistics. It's incredible how many people in data science simply cannot reason statistically, or solve elementary problems without software.

Re:Clarification from the original poster

[tigersha]

I did that a few months ago when it looked like our company was going down down down.

I put my name up on a online job matching agency and got one Skype interview after another. The first was a bit touch and go but is went better very quickly and I got a good feel for what the local market (Germany) is looking for. I also got a few nos but one or two interesting offers. I am 48 years old btw, with experience in SE since before Windows came out.

In the end the company I have worked at split off our division. Into a small sort-of-startup with a research lab backing it with money so I decided to stay.

Re:Clarification from the original poster

[ranton]

You're making the mistake of thinking that a person in your age group should interview. I'll imagine that you're at least 35+ possibly even as high as 50. We don't interview at that age anymore.

That isn't my experience when job hunting. My last two roles came from recruiters / partners contacting me, but I still needed to interview. Last time I actively job searched it was a combination of companies whose recruiters contacted me and openings I found through my network. In every case if it went past a basic screening and I felt they had the necessary budget for my salary expectations, there were still interviews. For the four companies where it progressed to interviews, I interviewed with at least three people at each (one I interviewed with 10 people in 3 panel interviews). Even my previous company when one of the co-owners sought me out for the position, I was interviewed by his partner and one other senior staff member before being given the position.

I am in the 35-40 age range and all of these roles were for Solution Principal, Development Manager, or Technical Architect positions, so they weren't entry level. I'm sure there are plenty of employers who don't care about resumes and hire purely based on recommendations or Github work, but I sincerely doubt they are the norm. I haven't run into one of these companies yet anyway.

I would say that by your 30's you probably shouldn't be posting your resume to job boards and HR portals and just hoping for an email response or call. But that is far from saying you should expect to not interview for roles at any level of seniority. My current company just finished interviewing three candidates for our COO position (three that I know of, maybe more).

Re:Clarification from the original poster

[mikael]

I agree with this. Set up your website and blog documenting what you are interested in and have done. Then have companies come to you. Not sending your resume to startups and corporations begging for scraps.

As long as you remain task oriented

[bferrell]

You're gonna stay stuck doing tasks and getting bored.

If you can, find a principal you like/admire/challenges you.

Slow news day?

[OrangeTide]

I thought I accidentally went to Reddit.

Derp

[TimMD909]

Don't ask such questions.

Re:Derp

[UnknownSoldier]

That's partially crap advice by Mike Rowe.

The secret to happiness is to remove false expectations

Yes, some people absolutely SUCK at what they love. The deluded ones are the ones who definitely SHOULD follow Mike's advice. They suck and always will, and no amount of talent will save them.

The problem Mike is painting everyone with the same brush. That does NOT imply that they will NEVER get better.

When I first started programming I sucked -- like every other fucking newbie -- because that's what a beginner is. Someone who DOESN'T have the knowledge and skills. I kept at it because I _loved_ it. I invested the years to becoming great. Today it pays the bills and I have a job that doesn't suck.

One of the secrets to life is to find what you love, and what your talents are.

Chances are, that if you invest in yourself, you can find a way for it to make you money.

There is no guarantees in Life. That's what makes it frustrating. Life isn't a simple checkbox-follow-these-instructions-and-success-is-guaranteed. Life is what you make it. Sometimes you need to _try_ things in order to know what _not_ to like.

Invest in yourself -- because chances are, no one else will.

Re:Derp

[UnknownSoldier]

> It's only crap advice if you follow it blindly without trying to understand it

Exactly.

> You seem to be under the impression that you need to work really hard to be a good enough programmer

I didn't say "good", I said "great" -- sorry if I gave the wrong impression.

You k now that cliche:

It takes 10,000 hours to master something.

People seem to think that success is an over-night thing -- and forget that it takes years before you truly master something.

Programming has always been extremely easy for me. It has been my 1st love; nothing even comes close. I have invested most of life in it due to:

* I started around ~10,
* There is a LOT to know,
* I am always looking to be better,
* There is ALWAYS something to learn,
* I love tearing apart programs and reading what the assembly is doing, and.
* 50% of what makes a good programmer is only technical. The *other* 50% has nothing to do with programming. If you aren't responsible, can't communicate, can't estimate, can't prioritize, etc. then you will suck regardless of how great technically you are.

i.e.
If you are not working to improve your craft then you are in danger of being stagnant.

Programming was my passion. I just happened to be "lucky" that it was also my talent. Mike gives the false impression that being passionate about your talent is not possible -- which is simply false.

Re:Derp

[david_thornley]

I've got a friend who followed his passion and got a degree in medieval history. After a rather unpleasant attempt at a career, he went back and got his MBA, and things improved. Fortunately, my passion when young was mathematics, which is a heck of a lot more salable in the job market.

AWS Cert

[TheSync]

Get an AWS Cert, best study material is Udemy A Cloud Guru (Ryan Kroonenburg). I spent a few weeks on it, and passed my AWS cert, plus have a great introductory understanding of AWS cloud.

Innovation

[JaredOfEuropa]

I have worked on many different projects [...] I tend to think of that as something I can learn, when I need to.

Sounds like you're a bit of a generalist with the will and ability to dive into a specialism when needed. If you really feel you need to "pick a side" and specialize, then all advice I can offer is: find something you love doing and specialize in that. But if you enjoy the learning process itself, the experimenting and ground-breaking work with new tech, then maybe you can find a job working in an innovation team.

Innovation is a bit of a buzzword, but there is plenty of legit innovation work out there. Innovation teams often offer a chance to learn new tech or new ways of doing things, and require a lot of flexibility from their team members. Perhaps that will suit you... I've been involved in innovation for 20 years or so, and I not only enjoy the great variety of technologies I have to deal with, but also the fact that I often get to wear many different hats: from project manager, team lead, architect, to coder and business analyst. Sometimes you'll be a one man team, sometimes the team will need someone to write a couple of tests for tomorrow's experiment or prepare a short presentation for a visiting VC, and yes I am sticking up my hand to volunteer. If you think that doing something yourself is often faster than getting others to do it for you, and if you can actually deliver results that way, then innovation might be something for you.

Positions in innovative work are few and far between and are often sought after, so you need to position yourself well for that when preparing your CV. Your background in data science and your machine learning study will help, since those fields are currently firmly hanging ten at the top of the hype cycle. But also emphasize your versatility as it's a key quality in such roles: show that you have experience in adapting to circumstances, and in diving in when the project calls for it.

Art. (No joke)

[Qbertino]

Study art. Better yet: *Train* for an art.

Seriously. Is there an art (performing art in particular) where you say "OMG that is so awesome, I wish I could do that."? Study/train that. Obviously there are limits. If you're in a wheelchair doing ballet won't work. But perhaps music, singing, acting is something that would be an interesting challenge. I have a diploma in performing arts and even though I've never done anything remotely like that in the last 2 decades (except being quite good at social dancing (Argentine Tango)), the experience was like nothing else. It does help me do presentations, that's obvious, but I've also learned about styles and aesthetics, art history and how to move gracefully. It helps me with GUI design and understanding emotional aspects of the user experience.

Imagine getting a Chello and learning that. Your horizon will expand into a universe you couldn't dream of knowing doing IT/Software every day for the rest of your life. You probably have IT pretty much down and getting into some newfangled technology or PL is a walk in the park once you've got a broader perspective on life in general.

Art most likely won't earn you big bucks but from what I get that's not what you need right now anyway. Note that fine art is closer to programming as an art than performing arts, so I strongly suggest performing arts, but perhaps you do want to get into drawing or painting or illustraiont or - an intersection with IT - 3D/VR and stuff - then fine art might be a neat alternative.

But generally rest asured, if you move away from IT and into an art, your life in general will improve for the better. Especially with your life right now having you struggling for sense and meaning. If only art becomes an enriching addition to your life as an IT expert right now, that will spill over into your IT career and have measurable positive effects. Promise.

My 2 cents.

Re:Art. (No joke)

[tigersha]

In particular, if yiu have anything to do with UI work at least take a dedign course. Lots of them around. Start with reading David Kadavys book ‘Design for Hackers’

Re:Art. (No joke)

[nastyphil]

I second this. I am completing my masters in Art History this year, on the back of 25 years in IT. It has been a phenomenally broadening experience.

If you are already a mature expert in *anything*, then an interdisciplinary study is a significant multiplier to your ability to think abstractly and architecturally.

Computer Aided Productivity/Living Might Be Cool

[dryriver]

Imagine some kind of quasi-intelligent software technology that lets you manage all the scattered pieces of your digital life in one slick UI. Your photos/videos. Your emails. Your office documents and source code files. Your daily task list. Your games and apps and movies and music. Phonecalls you should make. People you should meet up with. Places you want to visit or dine at. Websites and blogs and social media accounts you should check for news or information or research on. Things you have to pay for. Things you have to file or send off to somewhere. Things you have to physically do. Things you are waiting to be delivered to you. Pretty much anything you have to do or want to do in any given week of your work and leisure time. But not a stupid "cloud service" where all this sits on somebody else's datacenter servers half a world away where it is datamined to death, or is accessed through some stupid smartspeaker device that is also tethered into the internet, peeking into your home and work and general life all the time. A fluid, intelligently designed visual user interface that sits firmly on your computing devices/home server of choice, does not constantly send collected info about you to some faceless corporation, and can keep track of pretty much anything in your life for you, and even launch 3rd party software for you when you click on a task, or intelligently automate many tasks for you, such as ordering flowers for your wife from a particular internet site on Tuesday, or waiting for a certain stock to hit a certain value level and then buying 500 Dollars worth of that stock. Your server side coding, UI design and data science experience would probably be perfect for creating something like that. A sort of Super-UI through which you can perform and keep track of a hundred different necessary tasks, but one that gives you peace of mind because all data is local - not sent to 3rd parties over the internet. If I had your particular skills, I'd probably build something like that.

Those who can't do...

[Lab Rat Jason]

teach.

Re:Those who can't do...

[b0s0z0ku]

Nothing wrong with teaching, often more enjoyable than doing, when the people who hire you to "do" typically treat you like disposable trash. OTOH, teachers in blue states and at university level are actually valued.

Re:How old are you?

[Anonymous Coward]

Why does it matter?

Stop being an ageist prick.

Re:How old are you?

[Memnos]

They mentioned "after many years at the same company" and "since my graduate studies", so you can probably set a lower bound just from those hints.

Re: What a loser!

[Memnos]

No, it's a version of Tay that got into the wild.

There are only 2 real answers:

[UnknownSoldier]

* Do what interests you, and/or
* Do what pays.

Next question.

Re: There are only 2 real answers:

[TJHook3r]

Pick one :(

Re:There are only 2 real answers:

[UnknownSoldier]

Did you catch that: "and/or" ? ;-)

AI, ML, automation

[Pfhorrest]

For long-term job security, either get into AI / machine learning / anything to do with automation, or else something as immune to automation as possible, because those will be the last jobs to go.

Re: AI, ML, automation

[TJHook3r]

Last job to go will be soldier, fighting the robots!

Re:AI, ML, automation

[Tablizer]

For long-term job security, either get into AI / machine learning / anything to do with automation

It's possible we are in an AI bubble*, so be careful, because in the shorter term you still have to pay the bills.

I would suggest you pick 3 areas that pique your interest and explore them deeper, including asking practitioners. After you know the 3 better, then select 1 to focus on in a formal career sense.

By the way, genetics/data-biology seems like it has a bright future: it doesn't smell as bubbly as AI, yet gene scanning is growing ever cheaper such that there will be tons of genetic data to analyze. We will soon get to the point where it's economical to scan the entire DNA of an individual patient.

* AI is not going away, but we may hit a wall or two where too much money is chasing too few practical results for a few years and investors lose patience, pulling the plug on tons of R&D, sending AI researchers flooding the resume boards.

Something that allows you to teach...

[b0s0z0ku]

Go back to school, get a master's or Ph D. If you can teach at a university level, the working conditions (hours, ability to have fun, ability to do one's own research, prestige) can't really be beat.

Medical school (even abroad), residency, and working as a physician or researcher is also a nice gig. Consider going abroad and staying -- steady pay from a public system + benefits + ability to help people are good things.

Try Psychology...

[Anonymous Coward]

... especially the Dunning-Kruger effect.

Re:AI?

[geoskd]

If you're tired of data science already, jump on the AI bandwagon!

I would have to agree, if you have an interest in big data (data science is the gateway to big data analytics), then by all means, pursue it. You're already ahead of the curve by quite a bit.

If you want something outside your comfort zone, but close enough to your experience to be very interesting, I would suggest playing with a Raspi, especially the hardware end of things. Understanding RS232/RS485, I2C and/or SPI communications can be very rewarding work, both intellectually speaking and financially. Embedded hardware is fascinating because it really forces you to start considering all the ways that things can go very very wrong. High level development has a lot of simplicity in that you can pretty much always count on a certain subset of fundamental operations always working as expected. In the embedded space, you can write data to an I2C bus, and what arrives at the far end isn't always what you sent. The only error correction is whatever you bring to the table.

You might even be set in a very enviable position of being able to bridge the divide between big data and embedded systems. In the near future, the IoT will start producing simply vast quantities of data, even by today's standards. All of that data will be worthless without data analytics to figure out how to make it actionable.

The cynical yet logical side of me says.

[AbRASiON]

Wilderness survival.
Cabin building.
Water cleansing.
Very Basic engineering / how to use your hands.
Woodworking.
AGRICULTURE.
Animal husbandry.
Basic medicine / biology.
Basic weapons and self defence training.
Sustainable energy generation.
How to build a basic windmill / watermill.
Oh and did I mention
AGRICULTURE?

How near to 40?

[TJHook3r]

Retirement age in IT!

Phrenology and Astrology

[goombah99]

Historically Phrenology has been a more reliable source of income than AI skills. AI comes and goes in popularity. Phrenology and Astrology are forever.

Biotechnology

[coastwalker]

Bioscience. We had digital technology in the 20th Century and we will have Biotechnology in the 21st. You can thank me later.

Girls and drinking

[Bite The Pillow]

Girls and drinking. Or boys and weed if that's legal. Or just social customs on websites or people.

Plumbing and HVAC

[Virtucon]

I'd avoid any career in IT or Software Development. Why? Companies 20 years ago viewed IT as a core competency that needed to be developed in house and fostered. Now they view it as something that they can buy or outsource. That's not good if you're starting out and looking to get 40+ years in the industry. Sure, there's always web development jobs but all these bootcamps and schools popping out web developers only plays into the hands of minimal wage growth.

My suggestion, apprentice as a Plumber or HVAC tech. Both are well paid, in-demand and everybody has plugged up a toilet from time to time.

easy

[e**(i pi)-1]

decision science with focus on cost-benefit analysis.

Re: How old are you?

[b0s0z0ku]

Funny, I know quite a few 35-40 year old men just starting med school. I know some 45 year old new Ph D.'s as well. Problem is that American white men are their own worst enemy -- they're expected to follow a career for life by society, like some 1950s nightmare. Society doesn't jugge a 40 year old woman going back to school OTOH.

Do what interests you

[guruevi]

If you're looking for variety, go into sysadmin/operations type jobs with a healthy dose of in-house development, perhaps a small business or startup.

Alternatively given your brief resume, I have a feeling you may be more into research/academics, if you want to do research yourself, get/finish/use a PhD but otherwise good institutions are always clamoring for good people (data/computational/research scientist) regardless of your degree. If you go more into the administration/operations of a research institution rather than the academia, you'll get to do a lot of stuff.

If you just want to make money, right now I'd say, find the recent buzzword and apply for jobs - hundreds of applicants are lining up for AI-related point-and-click programming and for some weird reason companies are paying up the wazoo for them. But real commercial programming is not very prevalent anymore, you're usually tying together some cloud accounts these days and a workflow in between them, it's boring.

The answer

[Provocateur]

Rocket science.
 
To boldly go where no man has gone before.

Re:AI?

[vtcodger]

AI is worth considering, although it may be too late to catch that particular wave in the sense that by the time one knows enough to be employable, interest may be fading. It'll come back in the long run of course, but like J M Keynes said -- In the long run, we are all dead. All depends on how far current capabilities can move beyond the parlor trick stage I think. I'm guessing not very far, but maybe I'm wrong.

Security OTOH looks like it will be a pressing issue fpr a **LONG** time, The population of exploitable bugs appears to be immense and most folks are operating on the decidedly dubious assumption that the tradeoffs between security and usability are not a big deal. That leads to the belief that magic like 2FA can somehow get around the fact that secure systems tend to be unusable and usable systems tend to be insecure. I'm guessing that people will be creating new security issues at least as fast as old ones are resolved for decades.

Re:How old are you?

[mnemotronic]

Does age matter? Let's say you're Baby-boomer age, say 55, and do some math ....

What's that you say?

... so all suggestions are welcome!

Well OK then. Let's really try thinking outside-the-box. Start studying medicine with a long-term plan on getting a job in Pediatric ICU or Pediatric cardiac OR.

• 4 years undergrad (unless you've already done it)
• 4 years Medical school
• 3 years Pediatric Residency
• 3 years Anesthesiology Residency
• 3 years Pediatric ICU Fellowship
• 1 year Pediatric Anesthesiology Fellowship
• 1 year Pediatric Cardiac Anesthesiology Fellowship

Roughly 19 years schooling, residency and fellowship.

That means you'll be about 75 by the time you're ready to start work. You might have racked up some enormous education bills to pay off. Just guessing that'll take 10 years to pay off. Then you can start saving for retirement. Another 30 years ought to do the trick. Assuming that Parkinson's or Alzheimer's hasn't set in by then, you can probably look forward to settling into a nice relaxing retirement at 105. Tee time's 5am. Be there!

Did I go for a worst-case scenario? Obviously. Just to make a point age can be a relevant factor.

Data analysis

[thePsychologist]

Every second job involving a little math is basically data analysis right now. If you already some data to play with, you can get hands-on experience as well.

Quantum Computing

[Darkling-MHCN]

Go to a seminar on Quantum Computing, one thing you'll quickly pick up in the audience is that there are a lot of people who have experience with quantum physics.
  There are also a lot of people who have great skills in computer programming. There's almost no one who has an understanding of both.

At some stage in the future quantum computing may be abstract enough for developers to not need an understanding of quantum physics in the same way most developers don't understand nand and nor gates transitors, multiplexors etc etc. However we're a long way from that and in the interim there'll be shortage of people who understand quantum computer design and programming

I was in a similar position.

[wertigon]

After five years at the same company doing the same soul-sucking work in web development, I decided to hit the books again.

For me, I decided to start fresh and go to university for a bachelor and master title in embedded systems, but if you already got those under your belt, then you can always apply for a Ph. D. position at any university in the world. While an MIT or Stanford Ph. D. does come with great bragging rights, many other smaller universities, especially abroad, are happy to have you. Like the one I went to, which is a small but focused university able to compete with many bigger universities in my specific field (Embedded).

If Academia isn't a suitable position then pick any field of interest and learn enough to land a job in that.

And I cannot overemphasize

[gargleblast]

Animal husbandry

Re:And I cannot overemphasize

[AbRASiON]

I'm not sure if you're kidding (because I certainly think the term is hilarious) but sadly, it is the right term, I believe, for animal rearing for breeding.

Christ I hope that's the right 'rearing' in context.

tl;dr

[zifn4b]

You need to get a piece of paper that demonstrates that you have mastered an in-demand discipline in the economic sense so that you can have a place to live and food to eat and hopefully be able to retire some day. The consequences of not doing this are likely to be poverty unless you come from a wealthy family.

You shouldn't

[OneHundredAndTen]

If you are even wondering about it, you clearly do not have what it takes. Do some menial work instead, or go into management - you can still make lots of many with either. With the former, at least, you will be useful.

Electrical Engineering

[midifarm]

Super versatile and not too specialized. It will always be in demand, even after the fall of society. Don't just stick to theory, learn practical applications and uses as well.

Re:Nursing

[b0s0z0ku]

Truck drivers are at risk from automation.

There's a lot of demand for nurse practitioners, but for the amount of education it requires to be an NP, might as well go to med school. 2 years post-collect vs 4 years for med school. Yeah, yeah, residency, but residency is already a paid "job."

Foreign language

[Not-a-Neg]

Learn the language of another country you have an interest in, move there and work with startups trying to get off the ground that have solid financial backing. If family makes that too difficult to pursue, then get your MBA and prepare for a management position before you get age-discriminated out of your field of work.

I want to say one word to you. Just one word.

[XxtraLarGe]

Are you listening? Plastics.

A warning before quitting your day job

[lyz]

As someone who just earned a couple of Masters degrees in business for an insane price tag, I recommend that you exercise caution when transitioning careers. Breaking into a new field that is more of a passion project than one that is in demand is just brutal. Unless you are financially set and don't need the income, a transition could wreck you financially. The absolutely insane cost of graduate school cannot be understated and needs to be weighed against any advancement you may experience through it. It is massively easier to have a career track inside of a company that supports you rather than doing it on your own.

Re:AI?

[SunTzuWarmaster]

No worries - you'll be good enough o catch the next wave!

https://en.wikipedia.org/wiki/...

The Wikipedia page is a stupidly-good document of the "AI winters" thus far - 1966, 1974, 1987, 1993, 2001... We're about ready for another winter, and if it takes you 5 years you'll be ready to catch the next summer!

Re:How old are you?

[mnemotronic]

No, you are still wrong. Your argument for being ageist is fundamentally: "don't advise people to do things they can't be guaranteed to complete".

There is a lot to be had by going on the journey, and if someone wants to start a medicine career at 55 then good on them. If they can't complete it because they die along the way, then so be it.

I assure you that the use of age as a qualifier for advice, will only ever result in prejudice and cannot be useful or constructive in any way.

Point taken. For many of us, the journey is the destination.

Re:How old are you?

[mnemotronic]

"Tee time's 5am."

Bit early for 18 holes, no?

It takes a while to get both the user and the walker out of, and back into, the cart 12 times per hole. Times foursome.

Re: AI?

[Monster_user]

Reading that wikipedia article, and looking at the tech that is available currently, and I would hazard to guess that we are at the end of "AI Winters". Siri, Cortana, Alexa, and Google's digital assistant would seem to be a realization of the dream which funded AI during the "AI Summers", so I expect the present reminder of actual results would mitigate or eliminate a full blown winter of AI funding.

AI in its current form is all about big data analytics. Thus there is work to do and value, regardless of public hype and popularity. Though, self-driving vehicles may experience a winter of their own, which may impact image recognition algorithms. One just has to make sure the job in AI is a boring field, not a trending one.

Break as good as a Change

[ElitistWhiner]

Write down your dilemma questions and solution set. Take a break; a sabbatical maybe travel year trip live in hostels around the world just off the technology planet you are consumed with every day.

Ugly ruts comforting familiarity are the bane of existence. Freedom is the emergency parachute out of the uglies onto planet reality.

Return in a year, return to your list. You'll have no problem answering the questions.

Study something interesting

[TJHook3r]

Go on Coursera and find something interesting... learn a foreign language, study geology or biology. Sounds a bit like you have learnt to use useful tools but don't have anything to do with them. Data analysis is boring when you have no interest in the data - find something you want to learn more about and worry about tools and frameworks another day.