Live data from Hacker News

Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

news.ycombinator.com

91–100 of 147 posts

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#91
I had a similar work path, most side projects ended up as READMEs, some I would finish because of interviews, but nothing satisfying.

Right now I work with in a restaurant, and the dude manages sales by hand. So naturally I tried making a stupid simple order intake thing in Vue.

- Having a 'real' use case with a potentially real user gives a lot more motivation. And at the same time I can learn and fail on my own.

- It grew from simple form to nested editor to multiple object stores and pages

- It's a bit above the usual frontend tutorial (todomvc and similar) without being too ambitious

Maybe that can inspire you a bit. Good luck

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#92
If what you are doing isnt working how doing something different?

For example how about working from the outside in? Start with a basic UI made in a graphics program and work your way to implement it. Or just do the smallest API call you can make.

The key thing is to do something you are interested in, not what WE are interested in. Do it because you are liking it and do not do it to post on Show HN.

And we don't care what technologies you use, really, it makes no difference.

Follow your passions.

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#93

I know how you feel. It takes time to design interesting work. People often say follow your passions but sometimes the passion requires a brief period of tedious exploration while you set things up. Through trial and error, you can slowly sketch out a landscape of interesting problems, like glossy monuments rising above the horizon. That’s how it feels anyways when you finally start making some progress; when you fin…

Very, very good comment! The past two years I've worked on three different pet projects: first one that was fun and interesting but where I was dependent on another developer, who ran out of time, so that project stalled.Then a project I did simply because I saw an opening in the market, but which wasn't fun or interesting enough - this meant I wasn't motivated enough to go the extra mile, or to pivot slightly when necessary. Project alive but going nowhere. For my 3rd attempt I sat down and tried to think of something that was fun and interesting, that I also thought MAY have market potential. It's still too early to say how it's going, but I notice I have no trouble at all fixing all the tiny, boring little details that need fixing here. It's just so much more rewarding.

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#95
Just as a suggestion - try to focus on your ideas, not on the technical knowledge that seems necessary for the implementation. To roll out an MVP, a Django app on PythonAnywhere may be enough.

Only after you have tested your app with some real users and when you start feeling that PythonAnywhere is limiting you, start learning about more advanced technical stacks or infrastructure - you will feel more motivated to learn it if there is a need to use it.

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#96

If you are wanting to pick up Vue, choose node.js and start with the basic tutorial on setting up your project and project organization. Then do a tutorial on state management in Vue. At that point you should just get started on a project and stop doing tutorials until you run into something you don't understand. To get moving I like to pick a piece of functionality. In this case, login could be a good place to start…

the problem i have with this approach is it's reinventing the wheel. continuous integration, testing infra, rolling upgrades, deployment, logging, analytics on that logging, should all get solved problems I get out of the box in 2019

if your job is about making wheels, you kinda have to reinvent the wheel. Just driving the car teaches you nothing about them

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#97

Earlier quoted context omitted.

the problem i have with this approach is it's reinventing the wheel. continuous integration, testing infra, rolling upgrades, deployment, logging, analytics on that logging, should all get solved problems I get out of the box in 2019

not surprised on the downvotes. Many devs will choose reinventing the wheel over using existing solutions. Their competitors will be on to release 7.9 while they're still trying to figure out how to ship version 0.2 without losing all the data from 0.1 Batteries included systems are a force multiplier. Engineers like writing code more than shipping. Go to any game jam and watch the teams using Unreal or Unity. Their…

> Go to any game jam and watch the teams using Unreal or Unity. Their progress dwarfs those that start from zero.

Pretty sure I would come out on top with my CoD:MW download then.

The point of game jam is having fun while challenging yourself doing what you like. If you can't animate a sprite or draw a wire mesh without game libs then by all means use Unity, Unreal or even Scratch. Nobody's judging. Just don't read too much into it.

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#98
You sit back down to start with the coding, then reality strikes!

There is a lot to do. There's market validation, finding customers, talking with people, creating landing page and coding.

You feel so overwhelmed about all the things to do that you decide you'll start the side project later, and it bites the dust.

So my advice is : Just start.

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#99
I can't exactly tell where you're at, as far as the need for work.

From what I can see, it looks like you are able to keep the lights on, and you don't need the work.

If so, this is where I'm at. It's both good and bad.

I may be coming from a very different place, so YMMV.

I worked as a "ship" programmer and manager for over thirty years. I've been writing shrink-wrapped software (actual deliverable) for my entire career. It's taught me a great deal.

The last decade or so, my management responsibilities meant that I couldn't code on the job, so I created a number of open-source projects. I didn't have a "We own the ideas that you come up with in the shower" clause at work, so I was free to do this.

They have been fairly successful, in a rather limited scope. Since all the work was initially done with a "bus factor" of 1, the projects are not massive; but they are very, very important to a certain demographic. They needed to be done at a professional level; not at a "hobby" level.

I am now at a place where I never need to work again, if I don't want, but I want to work.

During the week, I get up every morning at 5AM, do my exercise, then roll up my sleeves, and start coding. Weekends, I let myself sleep in until 6:30AM. I've been doing this for the last couple of years. It is now Christmas morning. I am taking my time doing my exercise, and will get back to coding the the Bluetooth OBD driver I'm making in an hour or so.

Yeah, I'm obsessive.

I write every line of code as shipping code; even if I never plan to ship. My test harnesses are fully-qualified apps that could be released to the app store, with localization and full-fat documentation.

Will I work again? Maybe. I certainly have skills and experience that could make folks with established corporations a lot of money, but I've found they are reluctant to hire folks my age; regardless of what I'm capable of doing. I've set up a consulting corporation, and we'll see where that goes. If nothing else, it gives me a structure to buy the toys I want.

No matter. I work as if I am writing shipping code. The satisfaction of finishing a deliverable-level project can't be replaced. I do everything as open-source. My portfolio is gigantic, and is not consistent of dozens of forked projects. Just about every line of code in it was my own. My GitHub ID is solid green. I have over a decade of commit history in dozens of repos.

I'd suggest that you consider establishing a portfolio. Write code that is meant to be shared and re-used. Write developer utilities, drivers and tools. Test the code well. Document it well. Craft it like you are a master goldsmith, creating royal treasuries. Establish a personal/professional brand. Decide on what that brand means, and make all your work reinforce that.

The world needs high-quality software. Write as a vocation; not a hobby, and whether or not anyone else values or pays for your work. If nothing else, it will be useful as an example to others.

Maybe find an NPO/NGO that doesn't have a pot to piss in, but a cause you can support. Write pro-level code for them that they could never afford on their own. That's what I did. I can definitely say that the code I've written has saved lives, and restored hope.

It also feels great to write beautiful code. Every project I do is completely shippable, and highly-polished, with strong branding, rich UX and a great deal of documentation. I'm a reasonably good artist. I was trained as one, many moons ago.

For me, doing the work is its own reward. I haven't been paid to write code in over twenty years. I don't care. I love coding. I've been writing open-source stuff for most of those decades. When they stopped paying me to write code, I didn't miss a beat, and kept writing.

I also like to write articles on my personal sites and Medium. I don't particularly care if anyone reads them or not. They give me practice doing another thing that I like to do: talk about myself ;).

Just my $0.02.

Have a great holiday season!

Re: Ask HN: Haven't worked for a while, best guide/advice to start a hobby project?

#100
post #89
post #85

Earlier quoted context omitted.

> Is there a better term for "hipster stack"? Javascript eco system.

You made me laugh, but it is unfair. vue, (p)react and angular all have a better frontend development experience than you'd have with Django/rails. It just increases your tech stack from one component by at least 3 (webpack, nodejs, frontend-lib like vue/(p)react/angular ) And that is sometimes more than it's worth.

> vue, (p)react and angular all have a better frontend development experience than you'd have with Django/rails.

I don’t know much about RoR, but I think Django has a pretty great front-end experience. In general I think the classic MVC approaches are highly underrated. Well maybe not so much now that React, Vue and Angular are all doing components that you can incorporate in your standard MVC, but I’ve always found the SPAs and it’s MvvM architecture needlessly complicated unless you desperately needed offline functionality.

It’s gotten quite popular of course, and great things can be build with it, but it’s so complicated that it makes a one-person project into something you often need at least two specialists to do well.

Post reply on HN