Programming @programming.dev jawa21 @lemmy.sdf.org 2 mo. ago

I need help getting back into development

A little background: Through my teens in the 90's I did a lot of the things you may expect. I was a script kiddie on mIRC, made a tank game in Unreal Engine, and did some Quake modding. From 2002-2004 I landed a job doing Java web dev, SQL, and overall database administration because my father's friend needed someone that could do that. I was ok at the job, but not great. Being young, my hobby that turned into a 9-5 made me want to stab my eyes out and I quit.

With that said, I can understand a lot of what's going on, but it doesn't "click" anymore. I spent 20 years as a career machinist, but I physically can't do that anymore. Here's the rub - my twin brother is a brittle diabetic and can't work (lots of other stuff going on as well), and our mother is getting old (father passed this year). The only reasonable way forward that I can see in order to be able to support my brother is trying to get back into development.

When I stopped, subversion was what we used. I'm trying to understand Git, but it's a giant conceptual leap. I guess, what I'd like to hear from you all is a way to jump back in as quickly as possible in such a way that it may be a career.

Thanks

51 comments

For getting a grip in current web development, I would recommend fullstackopen.com - it's a free online course by the University of Helsinki and starts from basically zero, and gives you a lot of insights into web dev.

It may take some time, but it's really worth it and an overall great course.
- This is great. I'm going to try it out.
  
  FSO is good, but it does assume some JS knowledge to start. I would argue that App Academy's is more comprehensive (over 200hrs of material).
  
  https://www.appacademy.io/course/app-academy-open
When I stopped, subversion was what we used. I’m trying to understand Git, but it’s a giant conceptual leap.

It's probably not 'that much of a leap' as you imagine. If you're looking at Git tutorials, they're usually covering all kinda complex scenarios of how to 'properly use Git'. But a lot of people barely care about 'properly using Git' and they just kinda use it as a substitute for SVN... You create branches, you merge them back and forth, and that's about it.

Like if you want to contribute to an open source project, all you have to do is create a fork (your own branch in SVN terms) - commit some stuff to it, and create a pull request (request to have your changes merged) back to the original branch. git pull is just svn update - getting someone elses commits

Not saying there aren't more complex features in git, or that learning git properly isn't worth it, just saying, I don't think you have to see it as a 'giant conceptual leap' that's preventing you from jumping back into programming. Easiest approach just to get started would be probably to just download a GUI like Sourcetree or Fork, and you just kinda pretend you're still using SVN - approach wise
Someone will probably shoot me down for this but I actually find ChatGPT good for explaining concepts to me. Especially when I just want a high level understanding of a concept as I try to understand another one without getting too bogged down. A lot of Google results go into way too much detail.
- It’s also great for solving issues when you’re stuck. Not because of its superior reasoning skills but it can solve beginner issues and write you a list of things to try when it doesn’t know the answer right away. It’s like a rubber duck that will talk back.
- A few months back I would have said the same. Lately, however, ChatGPT very often returned very incorrect information on very basic topics. Each wrong answer erodes my trust a little bit more. Lately to a point where I consider googling first instead of asking ChatGPT
- It's great for explaining entry level information on a wide variety of topics. More advanced / obscure topics are more prone to hallucinations. I used it to learn React and it was a great introduction.
- I’m doing this now as I learn Unity and come back to C# after a couple decades away. It’s great for helping with syntax and reminding you of libraries, also with debugging steps as you mention. I haven’t had it full on hallucinate, but it has given me suggestions that fail to do the thing I specifically asked it for. I also went down a couple rabbit holes following its suggestions to later realize there was a much simpler answer that it didn’t tell me about. All in all, I’d still highly recommend it for my situation and OPs. If you’re able to follow the logic and point out the flaws, it’s a hell of a lot faster than googling or following tutorials.
Just remember: imposter syndrome is real. Everything you learn exposes you to ten things you don't yet know. Successful devs are comfortable with this reality - the job is one of constant learning. Best of luck!
- You said it. I spent the first six months at a FAANG convinced that everyone could look at me and instantly tell that I didn't belong there. Glad I was able to flip my thinking on that one.
First of all, you can totally do it! The field is massive, but also full of very bad programmers, and seeing how you were able to write a coherent text of three paragraphs, that already puts you ahead of the curve. Determination and perseverance is key.

I would suggest to play to your strengths. Java is still Java. Most of the progress since the 1990s was in the libraries and tooling, which only recently have become passable. The language itself also evolved somewhat, but there's nothing that you won't pick up in a couple of days of working with it.

Start with [1], work through all the boxes that are unfamiliar to you, practice a little on a pet project, or an open source project, and you'll land a job in no time.

[1] https://roadmap.sh/java
- Also, if you remember enjoying hacking, maybe pick that up again? There's a massive shortage of security engineers at all times.
Best of luck to you!

I’m trying to understand Git, but it’s a giant conceptual leap.

Git is not that different from svn (I mean, the biggest hurdle is going from a shared folder to any version control system)... I'd say the main difference is that branches live in a different namespace than files (ie. you don't have trunk/src/whatever but just src/whatever in the main branch). On top of that there's that commit and push are two different things (and the same with fetch and checkout) and that merges are way easier than in svn (where you had to merge stuff manually).

If you create a repo locally and clone it twice in two different directories, you can easily simulate what would happen when you and a coworker collaborate via a centralized repo (say, github) - do a few experiments and you'll see it's not as complicated as it seems (I'd recommend using the CLI instead of some GUI client: it's way easier to figure things out without the overhead of learning to differentiate between git concepts and how the GUI tries to help).
- To piggyback, the true main difference is svn stores commits as snapshots and git commits are deltas from previous. This is why git is depicted as a tree since it's inherently a node-based structure.
  
  A git commit is a snapshot. The node-based tree structure is an artifact of recording pointers to other snapshots and labeling snapshots with a branch name.
Just use a good GUI tool for git and Google how to do stuff with command line when needed, you'll eventually pick it up. Chatgpt and it's ilk can also be helpful with figuring out things when Google fails.
My personal recommendation:

Pick up Python, it's easy to learn and highly productive. If you also learn fastAPI, you can benefit from highly validated, declarative models to build REST APIs in the backend, well fast. It will yield quick results, you won't become demotivated and you can pick up a paid project soon.

Pick up Rust. It's "in" right now and I get requests from marketing people that know nothing about programming, asking if their project could be implemented in Rust

Go with memorizing the shell commands first, try to understand git later. Get productive, try to get where you were with e. g. svn or cvs. If you are comfortable, look at something in depth if you have a problem that you can't solve with the knowledge you have.

Fuck Java, seriously.

You have commercial interests, so it is probably wise to look into becoming a fullstack dev to maximize the kind of projects you can do. Look into React, vue.js, svelte. React is probably still the most widely used framework, you'll quickly do a project with vue.js and svelte is a super interesting look into things to come.
- Fuck Java, seriously
  
  Toxic...
  
  You can find a lot of good work doing Java.
  
  I was going to say this. I hate Java as much as the next dev, but everything runs on Java. If it’s web -> Spring, DevOps -> Jenkins, Event Streams -> Kafka, Big Data -> Spark, Logging & SRE -> Flink. All of these are built on JVM based languages. I am fortunate enough to program Rust daily at my job, but my options for getting another Rust job are severely limited. Everybody always wants Java or Go. They always ask for C++ , but I’m convinced that day 1 they would have you switch to Java.
  
  Java have had very bad press lately (since the log4j fiasco I guess? maybe since before).
  
  IDK why people blame Java for any issues with any library/project written in it... it's as dumb as blaming C/C++ for all the windows fuckups, and nobody blames php for the various cpanel vulnerabilities or python for all the shit people write in it.
  
  Side note: please don't abuse the word "toxic" until it becomes absolutely meaningless. Let's keep that to a more fitting context, having a
  
  I was expressing an obviously personal opinion about the language itself, which is objectively a dull, barren wasteland that sucks out your soul while you walk it. That is precisely the reason why it's so widespread and loved by business entities and managers - there is no excitement, no surprises, just an everlasting monotony of keys clicking produced by a horde of clones wearing button-down shirts while sitting in absolute identical cubicles, creating yet another instance of FactoryProducer. It's very easy to plan and schedule for, while at the same time being unnecessarily verbose and mildly unproductive (compared to other languages).
  
  Look, the JVM is fine, just pick another language. There is plenty of work doing Kotlin. But yes, if you're doing this only for the money, go ahead. I've always been unable to separate my job fromy personal life and my other interests, I couldn't imagine being cursed by Java again.
  
  If you can sit somewhere for 8 to 10 hours each day, doing something that isn't fun and separate yourself from it, not going insane, all the power to you. I also get that not everyone has the luxury of picking their favorite toy and making it their job, but I firmly believe there are options that are not Java.
  
  Now, if you're one of the rare types that actually enjoys Java, meet me in the closest Denny's parking lot, I need your cranial measurements.
  
  Please note: this post contains hyperbole and humor. I don't hate any of you, I just hate Java
One of the best tutorials on really "grokking" git concepts, and it's online and interactive: https://learngitbranching.js.org

For programming, start with buildings things for yourself. Be practical, start small, and iterate, regardless if you consider the previous iteration was a success or failure. I've heard good things about https://automatetheboringstuff.com/ (in Python) in this regard.
Best advice I have:

https://www.youtube.com/watch?v=0FPO4fm4nxc

Keep at it. Do actual projects. Actually use the tools for a while. It will eventually make sense.
- I want you to understand how surreal this particular link is for me. Video edits are my hobby, and I've just recently started yet another play-through of Morrowind. I get the whole "do actual projects" thing and that's valid, but I've no clue where to start here. Should I finally learn C#? Is Rust the only way forward? Should I just try to catch back up on Java? I guess what I'm asking here is should I just try to find a FLOSS/OSS project and try to contribute or think of something new?
  
  All the cool kids right now are mostly on JavaScript/type script. It’s probably a good place to get started since you’ll need it on the front end.
  
  I’m guessing most work on the U.S. side is react. You can probably pick up a bit of node/react native as you go for the backend and for native dev.
  
  Java is the new COBOL. Big corps luuuuuuuuurve it. It’s not what I would expect a single dev to use. C# is similar but smaller, if you forget unity game dev. And you should forget unity game dev.
  
  Rust is basically a less bad C. If you do things you used to do in C, learn rust. If not, don’t.
  
  Python is also still a thing. If you like it you can probably use it for most things, but if not there are other options.
  
  You can either decide by what is currently in demand in the industry and then pick a project that you can exercise that language with or you can think of a project you'd like to do and then go by what the best language is for a given project.
  
  In the end, languages are just like different wrenches. First you have to learn how to use a wrench, size or features don't matter much at this point (unless you already know that you want to become an expert with one particular wrench).
  
  I think starting a new project is way easier than contributing to an existing one.
I'm trying to understand Git, but it's a giant conceptual leap.

To start with, start with just using git locally. Don't worry about GitHub or similar. Then git and SVN will work very similarly. The main difference is that you need to git add files with changes inside before you commit them.

Once you're comfortable with using it by yourself, then I suggest running something like forgejo locally to be your own code server. Then you can play and learn how the two parts work together.

Generally, you need to give yourself a little time. You need to do the work. Be efficient...sure, but don't try to force it to be quicker than the time you need to learn.
Did you use Spring back then? It would be a useful thing to learn. Might as well see if you like to do frontend things such as React / Vue
- Yeah Spring wasn’t 1.0 until 2004. We had XML files upon XML files just to describe one single Java “Bean”. I did java programming from 2001-2002 and the again from 2011-today. Things dramatically changed (framework-wise) in that short decade I was away from it.
  
  I would agree, Spring Boot and Spring are very useful to learn. React, despite having its origins in Facebook and still with Meta’s hands on it, is a good web framework especially if you use it with Typescript.
- No, we didn't use that. I did both front end and back end in a sense. I developed the web UI for the customers and managed the PHP and SQL back end since it was purely a web interface. Yeah, yeah, I know that's not true back end but that's where we were at that point.
As for git, many basic concepts (e.g. staging area) clicked for me after reading some articles that Atlassian (people behind BitBucket) wrote. Other than that, I'd recommend adamj.eu 's book "Git DX" which is on gumroad. Haven't read it, tho. But I read his Django DX and like 90% of it was stuff I had to learn on my own, and thought: oh, how come I didn't find this book earlier...
I guess, what I’d like to hear from you all is a way to jump back in as quickly as possible in such a way that it may be a career.

You might check job listings in your region/country to figure out what languages and technology stacks are in high demand, as where you are seems to matter a lot.
Get a book on Spring Boot and jump back inro Java web development.

With git, you don't need to master it, just find some subversion to git howto and start using it. It's half a page of text. Once you learn the basics you can learn more as you go. There are many otherwise competent programmers who don't know git too well, I know because I support them from time to time.
So the big important part of git is that it's a collection of commits. A branch is just a labeled commit and each commit is a list of what changed from the parent. Rebasing (the most confusing one for people) is when you fiddle with a commit from underneath yourself. Or in even more simple terms editing a parent commit. Rebasing is extremely powerful but most useful for when you notice a bug you wrote a couple commits ago. Fixing such issues via rebase (or !fixup commits you auto squash at the end) keeps your history clean. It's as though you never wrote the bug. The other thing you do a lot with rebasing is moving your branch up in the history cause somebody updated the remote.
- I mean, this could be useful, but I need a much higher level of understanding. When I was a dev, all we used was basically version control. There was no concept of commit, pulls, etc. I'm having to understand this from a standpoint of no understanding whatsoever. For most devs, they've done this for at least a decade with Git. Git is so much more complicated than a simple version control. I understand why it's necessary for large teams, but I still see it lke this:
  
  You likely have a mental model of Subversion, so what I would suggest is to try to forget as much of it as possible first, as Git is very different.
  
  Take a tutorial that is aimed at Git beginners, e.g. [1]. This will help you start building a new mental model as well as get first practical experience.
  
  After that, read the official docs starting wtih the object model[2]. The reason why many people struggle to get into git, especially after using other VCS, is that it was built from the ground up, without much regard of the established conventions and terminology. Linus Torvalds once mentioned that he used his experience designing file systems when developing git.
  
  So the object model of git is very simple, but also not intuitive. However, once you understand it, everything will start making sense, including the xkcd you've linked.
  
  [1] https://gitimmersion.com/ [2] https://git-scm.com/book/en/v2/Git-Internals-Git-Objects
Start with a git GUI application. I use git extensions on windows.

You'll be able to get a feel for how it all works and it actually shows you the commands it's running.

I also use bingai a lot when learning. The more specific your question the more likely you are to run into hallucinations and that. But for explaining basic concepts or query things that are well documented it's really good.

It also sources it's answers so you can follow the link for further reading if needed.

Just don't trust it to generate large amounts of code.
If it's a motivation issue, you will need to figure it out on your own. I could give you some advice but motivation depends on a specific use case. If you want I can share what motivates me but there's no guarantee it will work for you.

As for getting back on a tech horse, there's a trick I've been using every time I was learning a new thing and it worked every single time. I start with a project idea. I write down the major goals I want to achieve and start working on it. At first I do the "quick start" adjusted to fit my project. It usually ends up with a working proof of concept and a list of things I don't know/understand. Next, I learn about the those unknowns and update the project with what I have learned. This raises more unknowns which leads to more things to learn about. After few iterations, when I'm happy about the project, I start a new, more complex project. And so on.
"Toxic" was the correct word choice. Thanks.
It sounds like you already have a foundation to build on. If suggest using chatgpt or something similar to explore concepts you struggle with. "explain git to me as if I knew SVN 10 years ago but have forgotten a bit"
I’m trying to understand Git, but it’s a giant conceptual leap.

In that case, I suggest learning Mercurial first. Its underlying design is very similar to Git, but the interface is more consistent and does a much better job of presenting the concepts to humans (especially those already familiar with traditional version control). Then, once you're comfortable with the concepts and commands in Mercurial, learn Git, since it's everywhere nowadays.

Consider learning at least one new language, if not immediately (since you're in a hurry) then in your free time. Java is only one of many (and not even a particularly nice one IMHO). Try to find a language that you enjoy using; your programming life will be better for it. If you want something flexible and productive, consider Python. For more structure than that, maybe Go. If mobile apps interest you, there are Kotlin (Android), Swift (iOS), and Dart (both). If web development appeals to you, along with lots of job openings and lots of competition for those jobs, JavaScript. If you fancy the esoteric (and well-paying), Elixir and Erlang are worth a look. Lower level languages are in demand as well (e.g. C++, Rust) but they come with various kinds of pain that I wouldn't recommend to someone in your situation.

Above all, please make sure you're reserving time for things that make you happy, sleep, and (if still possible) some kind of exercise. Your own physical and mental health are important. If you endure a bad working environment or wear yourself thin for too long, you will burn out, which won't help you provide for yourself or anyone else.

Take care, OP.

You've viewed 51 comments.