Learning Addiction Keeps Programmers in Chains
Vložit
- čas přidán 1. 06. 2024
- Learning new technologies, frameworks, and processes as a programmer gives you a feeling of accomplishment. But unchecked, learning can become a dangerous addiction that damages your software development career.
In this episode, I share how software engineers can cap their earning potential if they fall into common traps when learning is the absolute wrong thing to do. If you want to have a long career in software and get the rewards, recognition, and success you deserve - knowing when to (and not to) learn is essential.
Support me on Patreon:
/ healthysoftwaredeveloper
TechRolepedia, a wiki about the top 25 roles in tech:
healthysoftwaredeveloper.com/...
The Healthy Software Development career guide:
healthysoftwaredeveloper.com/...
Learn about one-on-one career coaching with me:
healthysoftwaredeveloper.com/...
CHAPTER MARKERS
0:00 Introduction
1:22 1. How Learning Addiction Harms Tech Careers
1:28 1.1 Procrastination
3:03 1.2 Sunk Costs
3:57 1.3 Declining Value
5:41 1.4 Social Avoidance
7:15 1.5 Work/Life Imbalance
8:48 2. How To Know When Learning is a Trap
8:54 2.1 Self-Delusion Your Project Requires It
10:01 2.2 Self-Delusion You Need It To Get a New Job
11:20 2.3 Influencer Hype and Vanity Metrics
13:03 2.4 Tech Industry Pushes Fear of Missing Out (FOMO)
15:10 2.5 Current Challenge Looks Easier in New Tech
#programming #learning #addictionrecovery - Věda a technologie
Have you struggled with the temptation to learn when you should be doing something else? How did you know? What did you do?
Just stopped learning AI recently, so I could make a game ive been wanting to make for a while. Can confirm its a learning addiction.
How did I know? Deliver, ship, deliver and ship more is my new mantra. If I learn something it has to be as a means to an end and not a solution searching for a problem.
"We can do this with jsdoc, but in Typescript it seems slightly cleaner." Did the improvements we hoped for (fewer bugs) actually manifest? We checked that. Result: no. Same amount of bugs in many different ways we used to measure them. Did the risks we worried about (quite a few, search for "Materialized typescript risks") actually manifest? Yes. Almost all of them. But we cannot get out again, because that would cost too much. And people are now invested in it.
I was thinking of learning Godot, Unreal, or Unity until I typed in the salary for Unreal and even in SF the range is $100-150K, it's some crazy "AI" game and it seems infinitely more complicated than just continuing on with mobile development. "Move button on admin screen to center", a little boring, but at least I can get my head around it. Until I get my PR in, then I realize I forgot to add unit tests, I didn't use the proper constant, I didn't put the description in the PR, and I didn't set the state of my work item to the proper thing, I only did it for iOS and not Android, the unit tests and build are failing, I didn't click the tiny triangle in Azure to actually read the description and acceptance criteria (there's a lot more to the work item) and I am humbled once again.
PHP? It's 2024 and Wordpress still runs the majority of the web. There is still so many legacy .NET framework / web form applications out there too. Shit, even old school jQuery is still in the mix.
@@kevinb1594 which gives a perspective on the most effective development of programming languages. Besides making Javascript fast, it may have been to make PHP multiple times faster. Thanks to Facebooks Hack which proved that PHP can be faster - and got overtaken by people actually getting PHP itself faster.
PHP 5.6 was almost 4x faster than PHP 4.4, and PHP 7 and 8 are again almost 4x faster than PHP 5.6.
That’s a 16x reduction of energy requirement for the same task. For most of the web.
As a dev with ADHD, this shit is REAL. I love learning new things, but I need to settle down and just focus.
If you learn because you are curious, continue. If you learn because of fear, stop - that's FOMO - and re-assess your priorities.
I too think this is a good way to approach it. As one gets more experienced the curiosity to learn every new thing tends to temper; and that is a Good Thing.
I realized that about myself at some point when curiosity gave way to fear. It was a sad realization.
Pause and reassess. Sometimes disconnect completely and the reassess.
Digital detox
My hack for reassessment
I completely recognize the overall mindset in the industry to keep jumping on new frameworks and technologies, which can be exhausting at times. Now that I have been working as a consultant for about 16 years, I taught myself to focus on mastering the stuff that I need today and only spend an hour here or there every week to read about the basics of some new technology or framework. The goal for me here is that I learn just enough about stuff that I don’t personally use on a day to day basis to engage in conversation about it or to know when I may want to use it and learn it for real. Note that you’ll never master a subject or technology anyway if you don’t use it in your work from day to day anyway, so up-front learning is typically wasted time, unless you really love a soecific subject and want your career to move in that direction.
2nd this approach. I usually limit myself to the first 1 or 2 chapters of a book on "X" technology-enough to understand what it is and what problem it solves. Also enough to create a mental reference that I can jump back to if I ever need to know more.
Truth. The bootcamp and training industry is probably booming now due to all the recently unemployed developers trying to figure out next steps. The AI hype-train in particular is marketing using fear and FOMO.
I agree with all your other point, but for me, in my opinion as a newbie programmer, with my current coding skills, am fearful because I am for sure replaceable by AI to be honest. That's why I am trying my best to upskill myself as much as possible.
@@julschamp5330
You're misunderstanding it.
We didn't just stop doing math because the calculator was invented. Certainly many professional calculators, called “computers,” have lost their jobs, but at the same time many new jobs have been created.
Every programming monkey will be replaced sooner rather than later. Ai writes better code than most programmers with the right prompts. But that assumes that you can distinguish between garbage code and usable code. There will be a paradigm shift in the next few years and I think programmers will be the first on the chopping block. Anyone who has just started actually has a huge advantage in my opinion. It will certainly be easier for them to adapt to completely different ways of working than people with 10 years of experience.
Don't be afraid of Ai. Embrace it. It's a tool, and tools always need someone who knows how to use them, otherwise they're useless...
@@peteerlustig8577are you saying we should all become prompt engineers?
@boratsagdiyev522 That would be a part of it as well, but it also highly depends.
If you can get better results in a faster time with just a prompt, why wouldn't you?
It's like you would say. "You really want to press just a button to calculate this complex algorithm function?"
That's the whole point. No one would say you need to calculate it in your head. Otherwise, it's not really math.
So yeah. Prompt engineer will be part of your job as well, and you may not gonna like it, but there are currently many Job listing which want exactly that...
In the industry for 7 years.
AI is no where close to replacing real devs.
At best, it will write getters, setters, and does a pretty good job documenting.
That's essentially it. You still need a mind behind it, implementing and architecting it
I personally took this to the next level, Im currently trapped into "Investigate to then Learn" addiction, all days I read docs, threads, blog posts, etc to determine "the best language/framework" to do x or y and I don't event get to the learning/implementation phase
Analysis paralysis. My cold, dark enemy (who visits often).
Usually in engineering there's no "best", it always depends on context.
In my years of the profession I inoculated myself against anything popular, if I smell popularity I'm immediately against the thing until the technical merits are proven. The more users something has, the less I care about it.
That's how I solve that investigate to learn phase.
I already have a set of tools and languages complete for a lot of problems, if I'm going to add new tools to my "belt", then they need to be proven on technical merits, I don't care X or Y is using it, I'm not X or Y.
Specially frameworks, frameworks are horrible because they constrain the domain of the solution, you want libraries not frameworks, that way you can pick and match what you need. And there's a reason why most programming languages are general purpose and multi-paradigm, that's so you can do the most without having to mix and match different environments, that creates accidental complexity that has nothing to do with what you're solving.
I'm way past the phase of trying new languages.
What I'm learning now is Rust because over the years it proved itself on the technical merit, its way better than all the shit about RAII, and prvalues and xvalues I had to learn to use C++.
Older things might not be popular anymore, they might not be sexy, but they're proven and get the job done. I'm still using C++.
The thing that I decided to master in the last 4 years was business management, which I needed because I'm now an independent consultant. Even then I just learn what I need when I need, like a JIT compiler, its insanity to just do "whole program optimization" on the internet.
This channel really helps me stay focused.
PL/SQL has kept me pretty damn busy for the last several years. There's a lot of value in knowing how to connect the boring old backend stuff to the shiny new tech.
🤣 same here
same with me too. if things get too complex i use my existing perl and python knowledge and use pl/perl or pl/python which is rare
The first language I really mastered was C. Still use it from time to time.
Networks haven't changed significantly over the past 2 decades either.
Just discovered you. 25 years in tech here. You make a lot of sense. Good job
Thank you! Welcome to the channel. 🙌
I hate learning to code but it always feels like I never know enough. One more video, one more book, one more reddit post.
I found out that best thing to do is follow few information channels (like yours) and don’t learn anything currently new or flashy.
Know fundamentals and best practices and you are good
The feeling of being in chains.. Yep. That is true. I feel that.
I totally agree with this, Im in University studying CS and im already feeling this way after learning a few programming languages, LOL. Learning so many technologies, it can become really easy to forget syntax, rather than knowing one good technology and being master in it (and focussing on the Real problem). focussing on a few can be a better way. Because,
"the more you learn, The less you Use" because you keep learning as an excuse that you will never be able to actually solve a problem
This is gold! Thank you for addressing this major fallacy that all learning is good.
Your videos hit so hard home for me, I can barely stand watching to the end. Your videos match so much with what I feel about software engineering. So many rabbitholes to avoid.
I will say, it always pays to revisit the basics. Investing in time to better understand the fundamental data structures and common performance / data modeling trade-offs can come in handy well into the future as they help you form a wide mental landscape with which you can approach problems with whatever tools you happen to have in your hands at the time.
Learning what's out their in terms of approaches, provided you don't dive into each rabbit hole you find, can make you a better problem solver over all. All that said, yeah, it's a waste of time to try and learn each new framework or tool. Pick a few to work with for some number of years and revisit that decision sparingly. Don't forget the Cobol lesson, it's often when the language is on the decline that you can make the most working on it (provided you can land one of the few jobs that is "stuck" with the language. There is a practicality point)
Cobol developers don't make a lot of money tho, average wages are high because average devs have 20 years of experience, accounting for that it's rather middle of the pack, especially given that you'll mostly be working at financial institutions where wsges tend to be higher. Lots of people were pushed out of the cobol job market when it started declining too.
Dude, you keep speaking straight through to me with things I find myself struggling with. Thank you so much for sharing your understanding in these videos!
It really encourages me to hear this stuff is helping you. Thanks so much!
Thanks alot for the info, really alot of scams out there making the industry look so overwhelming when it comes to learning new stuff, but thanks for clearly explaining how everything works. Really needed to listen to this because I really struggled to master languages hoping from one to the next big trend in the industry. Thanks for this really good info mate.
Yeah this checks out. I've always resented this aspect of the job: the implicit assumption that devs have to learn new tech in their free time in order to stay up to date. Unless you genuinely enjoy doing it (and even then with some heavy caveats as it indeed turns into procrastination) i find it an insane proposition. There's so much more to life than the 3.5 m2 in front of your eyes when facing the screen. Another thing: I've been focusing on the things that typically don't come naturally to devs for the last 5ish years: the soft skills. I've been evolving more into a "guy who gets shit done" instead of the guy who writes perfect code that no-one is going to care about. Fun part is that those soft skills are a boon to life in general, learning new stacks is not. Way more useful than new tech skills, especially if you are in a fairly static field with regards to languages (backend C# dev). Now I can spent my free time developing skills that actually work in the physical world, like woodworking, construction and fitness. Unless you are hell bent in being the best dev out there in some narrow field, learning adaptability is way more future proof in my estimation!
Great observation!! I think the best return on formal education is mastering how to learn most of the things that you might have to learn in pursuit of your career goals by yourself. Solid college level STEM education can be a good foundation for this but solid education in humanities and social studies up to high school level can also be a great asset to have for effective communication and empathy needed to work effectively with others. Most of the learning in IT is just learning the tools that you need to solve problems when the new tools can really make a big difference. We need to clearly understand that the internet as it exists now is a machine that is feeding on our data to influence us to sell things 24/7. It will become even worse with the adoption of AI. I think the best skill in this age is your ability to discern marketing from what is really important for your development. Thanks!
You nailed it - discernment. I've struggled with this my entire career.
You told me about all traps I fell into for 15 years. I will follow this channel for more awareness.
Welcome to the channel!
FOMO: Fear of Missing Out (on all of the good things you percieve others to be experiencing)
Instead, I've been trying to adopt LODGT:Love of Doing Great Things
Focus on yourself.
Be motivated by love, not fear.
Love computing, don't love frameworks, they come and go. The base of algorithms and techniques change at a slower pace.
I have been in a software development niche for a couple of years (e-com websites using Salesforce) and I thought I could be all right focusing only on mastering the craft around this platform, enjoy life after work. Then I saw a colleague from this e-com world saying that he thinks this niche is in trouble because most of the devs are not exploring new tech and tools to improve our skills so we can stay competitive. These kind of comments make me wonder if I just have to accept that keeping up the pace is part of the role as developer so no life whatsoever lol. Sometimes is hard to find balance. Great content sir, thanks!
Good points discussed here. Addiction to learning? Too many people stopped learning upon reaching adulthood or earlier. Lifelong learning is essential in ALL fields today, however, in the contexts which you speak of in your video, I agree with your sage advice.
Not the video I wanted today, but probably the one I needed. Thank you.
This is why I focused on the C languages (C/C++/C#) and left it there. I got sick and tired of the continuous cycle because of burnout and constantly changing gears to new technology stack for the sake of changing to it. Ridiculous - this leads to bloat and maintenance headaches. This is why I focus on the fundamentals as those will never change. There's a huge amount of type out there with "influencers" and social media, and you have to spend time sorting out the crap.
This is the same mentality I have now with Rust, Solidity and C. Honestly screw web dev and JS
@@dezly-macauley Agreed. I did web development once, that was during the "browser wars". Ugh. Never again! I can't even imagine doing web dev now as the ecosystem is a total dumpster fire.
@@Erik_The_Viking My version of "browser wars" was the "If it exists it will be written in Typescript" arc.
I realized that the opportunity cost of going down that road = less time to learn actual programming fundamentals, low-level concepts, and things like DSA.
C and Vim have been around fifty years or so and will likely be relevant for another fifty years, that will see me out.
Great points! Another one that we haven't touched is that learning also tends to crystalise the approach to solve problems with the ones used in the learning material, often taking away your opportunity to discover better approaches using programming fundamentals and creativity.
Amazing video. Just what I needed to hear.
One of your best video so far, this is an untapped topic.
The worst form of procrastination is ricing your Neo Vim config. and updating your plugins. Which is why I'm switching to Helix so I can learn that editor...
I'm prone to that, which is why I use Visual Studio and Windows, those are definitively closed and can't be easily modified, so I don't fall into temptation.
It doesn't really matter much because we spend most of the time reading code, not writing it. Just get a good monitor, a good chair and a good monospaced font, the rest really doesn't matter.
@@monad_tcpHard disagree with your OS choice. Unless you're coding in a lang created by Microsoft, it's a struggle to install and set up a dev environment for that lang. It's a struggle for simplicity while maintaining control. That control allows us to slip into procrastination.
one of the valuable channels in CZcams for SWE, Thank you for your content
Glad you're enjoying it! Thanks for the feedback.
I focus my learning on general CS topics: DSA, operating systems, OOP and compilers. These skills are easily transferable to all coding jobs.
This is why I discourage beginners from checking out developer roadmaps. It's a trap.
I learned programming for 4 yrs, and got nothing out of it cause I was not implementing any of it.
I was coding and coding, but none of it mattered at my job or in my current role in Help Desk IT, cause they want networking/people-management knowledge, not coding knowledge.
My mentor also got the boot this week for focusing on programming, cause it provided no meaningful impact to our department.
I’m not sure how proficient you are in coding. But if coding is what you want to do. Then pursue it. It’s been a very rewarding journey for me. Now… junior devs make basically fuck-all money. But your growth as a dev is determined by you and you alone. If you’re open to learning and determined to improve yourself, it’s possible to progress quickly as a dev. I know of guys who became seniors 4 years after their internships. Those guys aren’t the super-geniuses either. It’s usually hard-working people who take criticism well, and learn quickly from their mistakes.
Personally, I’ve slacked off quite a bit earlier in my career, that’s why I’m an intermediate dev with like 6 years experience. But that’s really just my own fault and no one else’s.
You are complaining that your non coding job won't let you use your coding skills? Maybe get a coding job and do that if that's what you want to do?
@@brandongregori995 For some reason I got fired from my programming job because I was focusing on my cooking skills
@@brandongregori995it sounds like they don’t feel they’re skilled enough to get a dev job and looking for an opportunity in their current role. Totally understandable.
I started out working at the help desk and used my programming knowledge to build various tools that helped my team members troubleshoot network issues and other related problems. Programming to me, is finding a solution to a problem and writing lasting code to solve that problem. Business desires value and if you are creating desktop games when you are supposed to be working well, to me, that is not a good investment, unless you work at a game developing company.
Great video! I am a beginner hobbyist. Plenty of your wisdom goes hand in hand with some of what @Techlead says about how it is really not that much about the technology you use but the impact you can create with your code.
This video came at the right time, thanks.
The best skill for programmer to learn is reading code, not writing code.
Reading code is so much harder than writing it! Agree completely time learning to read better is well spent.
@@HealthyDev I don’t agree with that. Writing good code is much harder. The same when you learn to speak a new language. It is much harder to produce than to consume.
I agree with the content of the video and I love your insights, but don’t agree with this specific idea (template pattern 😂😂😂😂😂😂😂😂)
Agree
@andreienache6290 I don't know. I think there is a difference between reading and understanding. Anyone who can read does not necessarily understand what is written. Many of my colleagues write good code themselves, but are super idiots when it comes to debugging (understanding) or joining a new project. Even if they write better code, I'm much more productive because it takes me a lot less time to figure out where things are going wrong and it doesn't take me as long to onboard either.
@@HealthyDevI would also add understanding. Like for example. Simply reading a log message and understanding the message does not mean that you truly understand what is happening. Unless you have checked all the lines between the entry point and the message, I would argue that you don't really know for sure and we'll just make an educated guess.....
Its greate video. i think you are one among very few to put your thoughts in reality with respect to your experience. I have encountered some similar situtations while doing the feasibility study for the best fitment of tool in Software testing. For Eg:"Manager comes and says i need the similar kind of behaviour in the existing tool/framework(comparing the features in the other tool ) which would take time and but they demand to get it done in less time." Usually we compare the tools (existing & new) with the POC and explain them about benefits to convince them.
Oh! Couldn’t agree more! Thanks for saying that ❤
This reminds me of my friend who learned Rust "just because", then considered forcing his team to write in Rust instead of C++ they're proficient with. That'a a way to introduce risk and delays to a project.
Yeah, happens to the best of us. I was on a project 2 years ago where I was hired to create a skeleton web application project that would be used to upgrade 13 apps from an old version of .NET to a more modern stack. I would have preferred to use something I preferred like angular, react, or maybe even rails - but the team maintaining the apps were long standing employees and deep in Microsoft technologies. So the skeleton I created was for the latest .NET technologies that weren't too different from what they were already used to. When I left, several of their apps were already converted and they were working on more - so it was a success. Key to this was I met with their staff and asked them questions about their learning ambitions and resistances. When you're a consultant, it's about getting the client results - not your preferences.
Good lord, Rust is the worst language to choose for that experiment. Definitely has a higher learning curve than average.
If you learn a language / framework / tool "just because" you are curious of whatever, that is OK. If you then force your team to do the same, that is a very poor move.
I don't understand what's Wrong with C++? Is the tool more important or solving the problem??, if you have someone who is a master in existing tool he could solve the problem better isn't it? Learning new Language definitely doesn't make you a better problem solver. Unless the the tool is different from a approach (not coding)
@@a.v7998 Nothing is wrong with C++. He just wanted a challenge or maybe something to put on CV.
Chat GPT is actually one of the things this is great at solving. Specifically the "is there a way to do this in my language?" part. Instead of spending an eternity going through docs etc, ask someone who already has gone through everything. Then once it has pointed out what it is, you can go read that doc specifically.
Learnt a lot from this, thanks
Thank you for this video
I'm hearing this as a form of "Begin with the end in mind". Learning is fine when you have a practical application for the knowledge. I'll do "routine maintenance" on my skills by skimming the release notes in newer versions of a language I'm already using. However I won't just study something like AI because it's currently buzz-worthy. I heard about a practical use for it, decided its benefits might be worth exploring and am learning just the portions necessary for me to build my prototype. Don't waste time on theories you may never apply.
thanks for saying those words. I can relate to these issues and found myself in that. I started at a big tech company 6 months ago and thought I have to learn new stuff everyday.
I want to use the 4 days off now to go into myself and focus on what is important. This video just helped me to get fresh ideas.
Thank you again!
Just one question:
how do you cope with the fear of doing wrong as a developer, or doing too little for becoming senior?
Well I’m a Christian, so I trust God has a better path for me or something to learn if I make mistakes. It’s just a regular part of being a flawed human being.
ORMs and things like TailwindCSS are going to be the large chunk of your programming sunk cost. The fashion changes and you have to use something else: all those specific quirks you had to learn are now all gone. Learning basics however will get you a ton of leverage; learn SQL and CSS well, gain in-depth understanding of the current programming language you work with and so on. Technologies come and go.
TWCSS is crazy, once I found a class that was a different name to the CSS property, I was out.
Atomic CSS is a great concept, but TWCSS accomplishes it very poorly imho. If you're new, stick with the basics, then understand what TWCSS is trying to do.
PHP dev for 10 years. Trying management and product management now. I am addicted to learning, but there’s so much truth to how there’s diminishing returns to being addicted to learning technology. Learning how to build teams and deliver value to the business is way harder to learn, but way more fun if you’re up to the challenge and has way more earning potential.
Good! You recognize you’re in a different role and need to learn new things. Definitely not trying to set the expectation that’s inappropriate in this episode.
Thanks for this. I have also found myself looking for work and struggling. The fact is it is difficult for me to be social (my fatal flaw!) and I've been getting a lot of advice to focus on that.
Sorry you're struggling. Hey at least you know what to focus on though!
I would say that there is a difference between "being social" and "improving communication skills" and I think the latter is both more pointed and more important. You may be struggling to communicate certain topics you are actually skilled at. Distilling a complex idea into something succinct/consumable is a great skill to develop
Bitterly true! Though learning hype in programming is more obvious, but is not limited to it. If you live 80 years, you have to learn things for 60 and only 20 left to act. Billions of "how to s", from on how to unplug your toilet to on how to end a relationship are on the media. It's made our life too complicated.
Your point with the temptation to use another languague in which you aleady know how to do it, it is indeed a huge trap in many aspects.
For example i once saw a guy convert a filtering block of 3~5 Java lines into sql code that was 50-100 lines. There was no practical reason to use sql since that query would not be used frequently to justify the perfomance gains. My point is we always need to remind ourselves that 3 lines of code will take 5-10 minutes to debug/maintain in the future, where 50-100 lines of a hard-to-debug-language will take DAYS !!!
I love learning, but I’ve found that what works for me is learning a new tech if I actually need it. For example, I never really learned React Native until I had a client with a React Native app that needed to be brought up to being user-ready. I feel like the thing that enables me to avoid endless learning of new tech for the sake of not falling behind is I trust my ability to learn what I need when I need it.
I did this to put off learning OOP design patterns. I do terrible at work because I can't understand what's happening in the code. Now I'm learning OOP. It's going really slow, but I know if I keep at it, I will eventually make sense.
You learn OOP in order to forget it ASAP, I hope.
Hmm I don't know that learning OOP design patterns is going to help you understand existing code better, it will just offer you a name for a certain pattern (if they are actually using a known pattern). You must read the code to better understand the code.
Learning OOP is a requirement for reading OOP code.
OO code can be very hard to decipher. It is very easy to create a totally incomprehensible OO hell and programmers that are a little too clever for their own good _will_ do that. Learning how to use a debugger efficiently will probably be the second prerequisite because lots of OO code can only be understood by stepping through it in a debugger.
@@HealthyDev False. I can understand code no matter how some guy in the 90s decided to call a thing
Much apprechiated, always on point, as you know NODDERS
There is good advice lurking here. There is also a risk. I love his descriptions of the traps; and I generally agree with his primary point. Yes, you can spend too much time learning. You can also spend too little. As you watch this video it is important to keep in mind that the end goal is to find the right balance.
I agree. I can include every subtlety and contextual positioning statement under the sun in my episodes, but at the end of the day each person requires discernment to see if it applies in each situation.
There are some things in programming that are timeless e. g. SOLID principles, proper approach to debugging and so on.
The question for me is. Should I learn also off the work hours? I'm a junior. I work as a Full-Stack and it's close to a year since I started. I always feel the pressure to be better and although I think I'm doing good job at my work (I adapted quite fast, learn quickly and when given feedback I try to improve based on it) I still feel that I should do better and learn in my free time. On the other hand I still want to live my personal life and I actually do but still I have this bad feeling in me that I'm not good enough and I should be learning more. What do you think about this? What stance should I take in this situation? How can I tell that I'm doing enough based on my experience level and don't worry about it?
I think the point of this is to focus your learning and make it relevant, not do learning as a goal in itself. The things you work on that are relevant to your work, you will get lots of reps and hours doing, so learning new stuff that is relevant makes sense. Learning what is in front of you deeply has benefits as well. If you do Typescript development, learn everything there is to know about JavaScript, Typescript, and the frameworks you use normally and become the expert. This, in preference to learning a lot of stuff at a shallow level. I am a senior programmer one the other end of the career, and I still devote time every day to learning something that is relevant to my job (currently looking into Data Science to be able to better support the Pythonistas in my company).
5:28 Watched this on a break from re-writing some visual basic code from the 90s. LOL!
There is a famous mathematician whom published “The Bargaining Problem". Which showed that for any finite game, all the players can arrive at an optimal outcome. So in theory, by playing the programming game we all can arrive in an optimal outcome. The point I am trying to derive is that instead of following the masses, especially influencers, do your own thing and success will follow.
One of your most level-head pieces of recent. I'm a bit in this trap right now. Mostly learning new languages. On the flip side, I refuse to buy into editor hype.
I'm a bit conflicted about this.
I agree that you shouldn't follow every trend and learn every new framework. On the other hand, I've been in situations where I've used the same stack for years at a time, and then found I'm not qualified for the interesting jobs. It's not that there aren't jobs in the "old" stack, but I don't want to do the same old thing forever. I want to do new and interesting stuff, and that usually comes with new technologies.
If you watch the episode again, I think you'll find I agree with you. I mentioned that there definitely are times you need to learn and can't rely on the same stack forever. I think the problem is we have these external forces like marketing pressure and influencers that convince us that's the case when it really isn't. The important point (I'm at least trying to make) is to try and make those decisions yourself - not based on outside influences. Hope that makes sense.
@@HealthyDevyou have to have some outside influence to know if you are on the right track, or need to learn something. This topic is super nuance.
Might be good to know when and how to evaluate when to learn new stuff so you dont feel burned out or left in the dust.
Good example is the cloud. There were so many saying it is “just someone elses computer” but more and more I see the need to pick up the skill.
You want to keep your ear to the ground on some level, especially for updating your platform or when looking for a new job. But who do you ask? You could use those influencers for a guiding posts. Could also look at relevant job postings and see what they are requiring. I think that would be problematic as employers want everything and only use a little of those skills.
I think covering when is a good time to upskill being a great topic. Also, how to learn that skill years after formal education or no education at all.
As a dev manger I see a lot of learning anxiety among devs regarding the cloud and cloud tools. This has become much more prevalent since cloud adoption has balooned and I don't see it stopping anytime soon. Still, your underlying message is valid.
I almost became homeless from learning haha 😅😅 I wasted at least two years "getting better" I was avoiding picking up the phone and selling my skills. This learning problem is so real I am glad I am not the only one who noticed...
Yikes! Are you in a better place now?
@@HealthyDev Yes but the key was to stop learning and start applying the knowledge. Any trip to LinkedIn will validate this video. In fact I am about to share it haha.
@@topdev_tech9156glad to hear! Hang in there.
A lot of careers can be "ruined" by constantly shifting from tech to tech and company to company. You can have a better career building on the tech you know and there is a lot of people needed in legacy tech.
99.9 percent viewers of this channel are not content with what they do , have FOMO or unemployed. He is just riding out those emotions and cashing it out. We all have some insecurities.
I am a "lifelong learner". I never considered that I was compelled by dopamine. Interesting.
I think if you really do enjoy learning new frameworks and stuff, then you can follow that. But you have to realise that it is a hobby you are doing in your spare time and will not improve your career.
There is programming and learning for work, at which you should strive to be an expert in a small number of technologies, and then there's programming for fun, which has to take a back seat and should not encroach on your deployments to production until you can be sure it's the correct move long term.
So true. I’m employed and make decent money but I spend almost every weekend studying programming just cause I feel like I need to keep up with the young ‘uns.
I’ve been wanting to spend $7K in my discretionary budget on a hobby but find I never have the time to just chill and do fun shit.
I’m always chasing it and hustling and my mind can’t just switch to have fun mode.
The other day I talked myself out of buying a guitar and learning guitar cause I reasoned I would never have time to learn guitar anyway.
As an agile coach.. with add, just signed up for a 9 month weekend course..
There's never a need to pay for programming training, not with CZcams and easily available documentation and guides everywhere.
excellent remainder to me
Listening to you while learning Go. 😅
Ha! Hey if you need to learn it, awesome! If you were hyped by someone else, well... :)
@@HealthyDev 😊 I neither need to learn nor I was following any hype (the second one I never do). I am just exploring new things, as a hobby let's say.
I just found it super funny when I was listening to you while reading the docs. 😂
But I totally got your point there. Your videos have been helping me a lot to become a healthier dev.
Thanks and keep the good work. 🏆🥇
@ you're very welcome. Nothing wrong with hobbies. I personally find working with my hands a great break from coding. To each their own though!!
I think I agree in principle, but we should distinguish between learning a technology and learning a concept. It seems reasonable to me to learn technologies lazily, on-demand (e.g., when joining a project that already uses the technology), but learn concepts eagerly, before it is needed. During a programming interview, nobody would ask about history of software engineering, architecture of CPUs, or type systems - but some understanding of these have a huge impact on the code one writes.
If someone knows only imperative languages, it would be a beneficial for them to learn the basics of a functional, or logic, or a stack-based language. But it is likely not worth it to switch from C++ to C# to Java, because it will not give them any new insight, and the "accidental features" that one needs to master to be proficient in that language are not transferable.
Software industry is extremely conservative, but we think it is not - because every day somebody comes up with a new technology, with only a minor conceptual improvement but huge differences in the "accidental features".
This thing, it’s all about me 😂
what's to learn? most of the stuff are mostly language specific frameworks and stuff, rehash of existing concepts, none of which isn't covered by things I learned during my college days in CS. The new stuff are mostly language syntax and specific uses in them like promises, await async etc. (oops, it shows how old i am lol)
Knowledge degradation also has a second angle separate from the further development of the technologies you used to use. The knowledge in my head also degrades even compared to the state it was in when I stopped using the according technologies. I used Java the last time over ten years ago, and I know very little from that time, because of all the stuff that happened in my life inbetween. That does not mean I have to start from scratch when I start using it again, but I feel it is important to also acknowledge that our minds are not perfect information storage facilities.
Absolutely agree.
The most future-proof programming language? I can tell you that with no problem at all. That is C. If you learned C 30 years ago, you're still golden. And in 30 year's time, you will still be golden. It is still used for all kinds of stuff and it is not going away. There is some evolution in the language, but not a whole lot.
C++ will also still be important for a very long time, but there is a lot of evolution in C++. There is a lot more maintenance in C++ knowledge than in C knowledge.
Other candidates: COBOL and Fortran. These have been around for a very long time and they will stay around for a long time. If you want a safe job with little competition, learn COBOL. It will not be hip, not cutting edge, not super exciting and will mainly consist of maintenance on ancient legacy code, but you will not spend a whole lot of time in classrooms.
I think C++ has gone off the rails. They're trying so hard to make the enhancements as flexible and generic as possible that they're making a language that was already a bit obtuse even more so.
@@loganmedia1142
I think C++ is fine. I like it a lot. Many new features make things simpler.
Smart pointers are essentially RAII out of the box. With a custom deleter, you can use them for any resource that requires a close function. Like java try with resources only done right. Speaking of java, in java all variables behave like shared_ptr and that is often not what you need. In most cases unique_ptr is better. Or value semantics, no pointers at all.
Lambda's are great. They synergize very well with standard library algorithms and the [](){} syntax makes them instantly recognisable. Also you have the choice what to capture and if you want to capture by value of by reference. In java, you always capture all local variables by reference, there is simply no choice. That is often not what you want or need.
Optional and variant are great. Absolute life-savers.
Structured bindings make returning multiple values simple.
Constexpr makes metaprogramming accessible for non-gurus.
Modules holds great promise, once it is finally properly implemented across compilers.
Concepts are great for template code.
I think contracts will be very nice too. You can have sanity checking on function parameters without exceptions or other convoluted constructions.
The parallel algorithms are really the simplest possible way of doing concurrency. If all operations on a range are independent, just slap on an execution policy and it is concurrent. Can't be easier.
Move semantics has some dark corners, but in 90% of situations it's easy. A lot better than those languages that have immutable strings that have to create a new string, including heap allocation and a copy of the data for every concatenation. And as in java, clumsy workarounds like StringBuilder that still won't do a thing if you for example want to change all e's for a's or something like that.
I really wonder why not more languages have adopted a concept as simple as it is brilliant like the word const, that can make any variable mutable or immutable whatever is desired. Much better than like in python have mutable and immutable versions of the same type.
The namespace system of C++ also is brilliant in its simplicity.
On the other hand, coroutines as they stand now are barely usable. They really need more library support and it seems like that has not great priority with the committee. And all the ranges stuff with the piping and all is a neat concept, but it is a matter of taste I think. I rarely use it because I think it is hard to debug. A lot of functionality comes in a single line and that is right or not. I prefer a more step by step approach that I can step through in the debugger rather than writing something that does a load of stuff and it will work correctly, do something unexpected or not compile for some strange reason. I think that stuff is harder to use than some authors make out to be. And some features have a rather specialised usecase (weak_ptr for example). Also quite a few defaults should really be the other way around. That is regrettable, but we will have to live with it.
I learn because I want to improve myself. Money, recognition and influence are whatevers for me in that case unless those can enable improving my skills. Currently, I am learning the C Programming language to get into the embedded systems field in light of disappointments in a customer support job after 1.5 years since graduating college. Am I deluding myself in this instance?
No, you're doing fine. C is the most fundamental language - short of Assembly - and a good stepping stone to object orientated languages such C++, C# and Java. For embedded programming C++ is the natural choice.
If you ever feel the need to know more about embedded programming, then learning Assembly will allow you to take a look under the hood of C. I found the Z80 tutorials, using WinAPE, on ChibiAkumas a nice introduction.
I keep jumping from one lang/framework to another, but never get past creating the login API. Can't help it, they're shiny. And then there's also the state of the job market, where the vast majority of jobs are looking for "frameworkers".
especially in the 1990's, a 40 hour salaried job, was an 80 hour week.
you are such a great actor
On the one hand I feel the truth of this, but on the other hand I'm in the process of leaving a job where I couldn't fit in because there was just too much stuff I didn't know. I though I could pick it up but I couldn't do it. Also from job interviews I've had I've been having my face rubbed in the fact that stuff I've done for multiple decades is just not of interest anymore. Things I spent a lot of time and effort mastering is not valued any more. I'm considering that it's time for me to bail out of the industry entirely. I'm sick of the treadmill. At the same time I really really really hate the fact that companies like it better when somebody does a lot of cut-and-paste and hack at random until something looks like it works, and don't like it and penalize somebody like me who takes a lot of time and effort to actually understand things. I despise that with every fiber of my being. I'm disgusted by the way the industry works.
I can relate. I’ve had to accept some of the big shortcomings of the industry and focus only on the things I can actually control. If you truly are unable to work because the technologies you’re working with don’t have enough jobs available, if you want to keep developing you may need to learn some new stuff - which is within your control.
That's true. I recently convinced my self that golang is way better than python. so stared to learn golang for backend when I realize I couldn't really get deeper in django.
It is easy better though.
This is so true
For embedded, the most future proof programming language is C/C++. I've been using C/C++ for embedded systems for 20 years. Maybe in another 20 years it will be Rust. But for now, the compilers don't support anything else. I'm not sitting on my hands. I'm always learning new stuff. But I'm not learning a new javascript framework every 5 minutes. Any webpage my device serves is probably going to be KB in size, so I cannot afford to use that stuff.
I have the opposite of a learning addiction, I could use a good learning addiction for the next 3 months.
Maybe you could explore what the motivation you have for working in tech is, and see if there’s something in conflict with you personally? Procrastination can be a sign of being out of alignment with your purpose, or an attempt to self sabotage.
Thank you very much for this video. I am not a software developer, but was in similar position in a different field. The things i was going thru were similar... endless learning, but maybe 5% knowledge needed and retained. This is broader problem in our society and economics, hope you will make videos similar to this in the future, cheers
Thanks for the feedback. Good to know this was helpful outside of software!
Tbh I do really just have a sense of curiosity around technologies, but I'm really just trying to keep focused on the principles so I can transfer my learning between languages. For now, however, I work in python
Begin between jobs, I'm learning Rust and Bevy (game engine) right now. It's fun and keeps me from vegging out all the time.
I would not pay anyone for it, though, the available free resources are too good.
“That makes it possible to meet the requirements” - reminder: these are all turing complete languages.
This is a gem! Thx boss
A lot of these scenarios have happened to me.
Great advice! I remember 10 years ago people told me dont learn .NET, its archaic blah blah blah. Here I am in 2024 writing .NET full time
I’ve done tons of work in .NET. It’s come a long way and really is excellent for many projects these days, despite not fitting the hipster aesthetic.
I somewhat disagree with this. I would say it’s very useful to learn the fundamentals and I mean true computer science and software engineering fundamentals. I don’t mean learning html, css, and js. I do agree with you when it comes to learning technologies those will come and go and unless you can understand the patterns used in technologies you will always spend most of your time learning. I also agree that learning when you should be practicing is bad and actually procrastinating
That you should learn fundamentals is kind of a given. I'm talking more about once you're on the job.
5:23 "[...] it is not gonna be worth the same amount over time." The exception proofs this rule. In The Netherlands COBOL-programmers are highly sought after and can charge any amount they want.
I see what you're saying but that's arguing the outliners of the bell curve. You have to ask yourself how many positions are available, and for how long. The answer to that question isn't so simple in the case of COBOL.
> I can't solve it in this, I HAVE to use this other technology
Then you don't know the language you are considering abandoning. In this case I agree having a bunch of shallow knowledge on a bunch of topics leads to frustration and no valuable skills.
Going deep and being able to do anything in a bunch of different technologies is a very good thing for moving up - but I'm referring to senior or principal level roles, not junior to mid level.
The more senior the more you have to learn on your own outside work to stay relevant but the majority of devs just need to do their jobs well and concentrate like you said on other skills.
I think you're making the last point I did in the video!
jokes on them. I never paid to learn tech skills. I learn it on a project basis. meaning I learn it if I have a requirement not because someone tells me it is the future.
Bang on. This is why I tell people who really love learning to consider consulting!
If you are learning just different frameworks/APIs, I can agree.
At the same time for many people who just went through bootcamps like me, there is never really the time to learn assembly/how exactly and why is the type system implemented. What is the threading model for your language/runtime. And these things are really important and you are blind if you are not aware of them.
Every web developer needs to know assembly? Is that what you’re saying?
@@HealthyDev well, at some point, yes. It opens your eyes to everything and you can read hex as a bonus. You realize everything is bytes moving around.
I would argue you can realize that without learning assembly, but to each his own.
@@HealthyDev I am not talking hard-core assembly - equivalent of a CS course 101. Didn't you study this in college?
@@ethanr0x I did an associated degree. We did get into assembly. I didn't find any use for it over my entire career. I don't think that's going to be true for everyone, but this is why I ask the question whether it's truly a requirement of all programmers today.
Interesting, how do you think this relates to the AI boom?
I have another episode a few weeks back on AI if you check the channel's video list.
I think the job market and recruiters are a big part of the problem, just one look at a job description for a beginner front end dev and the list is ridiculous
Hmm, I was going to learn Zig, but maybe I should forego it and instead just go build something...
I'm sorry. I can't work on this right now. I'm learning Chinese to be able to read the datasheet of this microcontroller.
How are we supposed to advance our career in a world of remote work? I have zero connection to my coworkers and find that trust issues are way worse with remote coworkers.
It's very hard. One of my personal mentors said recently "It's a lot easier to fire someone you only know through a webcam". I don't think we think about the cost of remote work enough on our career progression.
You are a free agent. Remote or not, you should search for good mentors inside and outside whatever company you are currently working with.
It is hard to get a job these days if you are not a code monkey in the specific library they are using.