The biggest reason to make useless stuff instead of useful stuff is to avoid harassment. If you have something useful on GitHub then entitled users will come out of the woodwork to request features, report "bugs" (read: the user is doing something wrong), or otherwise expect free work out of you.
My biggest reason is that I don’t have to finish. When I get bored with it, I move on.
I used to put pressure on myself to “finish” and would often miss and feel guilty. Not just computer projects, but electronics, learning musical instruments, etc.
Then I realized I was making my hobbies work and doing it badly. Real work (where I have to finish) got in the way of my hobbies. Hobbies are supposed to be fun, relaxing, and rewarding. What was I doing?
I gave up the pressure and it was liberating. I pick up the guitar for a couple months, then get lost in trying to make analog circuits. I tinker in the garden then teach myself enough CAD to do a small woodworking project. I started sewing masks and tote bags then set it aside. I fish a bunch in the spring and fall. Then, I’ll revisit them later when the muse hits and don’t feel like I wasted money.
Now my only goals are to have some fun, relax, and learn something new.
I really only program for fun in languages I don’t know or platforms I haven’t played with. I learn by picking up a project that scratches an itch.
Sometimes those feed back into my work (eg Go, Kotlin), but most don’t (various LISPs, low level C).
My issues are exacerbated by my ADHD, but I would recommend the approach to anyone with a steady job they enjoy. That and not using TV/movies as a hobby. There’s just so much cool stuff to explore.
Yeah learning things is so fun, and also so is just having a bunch of weird skills you're competent or at least knowledgeable at.
I won't post my list of accumulated learnings because it's about as good a personal identifier as my SSN. But it's similarly scattered as yours and I love them all even the ones I haven't used in years.
It's really become the way I orient my life. What bizarre or useful thing I'll get into next is always the big question now.
I haven't gotten any feedback on my github stuff (which isn't quite useless, just not useful to many), but it's pretty easy to ignore feedback in general, and it's not too hard to turn off issues; not sure if you can turn off pull requests though.
That is an excellent idea. I hope I'll never have to do that (haven't had to deal with this kind of thing so far), but I'll definitely try this should an entitled user show up in the future.
This is one of the downsides of programming-as-a-art: if you create paintings you may get feedback and requests for more (or less) - but you’re unlikely to get bug reports.
Try making actual art with programming, eg demoscene or quirky twitter bots or programs that automously paint paintings, and you'll get way fewer bug reports.
I use Github all the time to find code samples, especially for 3D stuff, shaders, audio, emulation, demoscene... It's more common to find solutions to my problems in zero-star projects than in big ones.
Lots of people star my code-as-art experiments too, and from the forks I can see they find educational value in it.
Not everything in the world has to fit perfectly into some preconceived mold. Heck, in that case, the most interesting/educational things are trying to avoid that.
There's nothing to be gained by anyone by using this approach, and a lot to be lost.
Gitlab/Bitbucket are not "dumps" for projects you deem useless, and "useless" projects are not useless at all from an educational point of view. Gitlab and Bitbucket are exactly the same as Github and should have both "serious" and "useless" projects the same.
Using a personal Gitlab instance or anything similar for that is a terrible idea because it costs substantial time and money of individual developers and severely hurts discoverability.
People should put their "useless" projects wherever they want and are permitted to.
Who says I want 'discoverability', after all we are talking about useless project born of a hobby no?
Not everyone wants to have to share everything that is useless with the world especially if it's only going to be you.
I also don't want people to have to bug me for feature requests as well on top of that.
There is a reason GitHub/Lab/Bitbucket has private repos for this purpose.
You can obviously choose to share it, but you shouldn't be whining if your 'useless' hobby repository suddenly becomes the focus of attention of developers requesting more features.
I never said you wanted anything. I'm the one who wants to have 'discoverability', both for my projects and other projects that might interest me. It's you, for some unknown reason, that don't want me or anyone to have it!
> Not everyone wants to have to share everything that is useless with the world especially if it's only going to be you.
Then don't. But don't demand that of others.
> I also don't want people to have to bug me for feature requests as well on top of that.
That's perfectly acceptable, and I never implied it wasn't.
> you shouldn't be whining if your 'useless' hobby repository suddenly becomes the focus of attention of developers requesting more features
That seems to be your problem, and you're the only one in this discussion complaining about it. Looks like you're projecting your issues onto me.
Anyone who thinks github code should be polished and supported or hidden from view is going to be sorely disappointed in my island of misfit experiments, but I really couldn't care less. I didn't put it on github for fame, exposure, or resume fodder, I put it there because they host my repos for free.
I would suggest you may not be the right person to use code in these repos.
For someone like me sometimes I'm looking for a small sample of code that fits a pattern. I could careless about the project. Sometimes you need an odd piece.
Saying his code doesn't belong on github tells me you might not belong on github.
I wonder why people are so annoyed or upset by this.
If you don't want to support an arch/language/platform, etc. just say so in the readme (all arch/languages/platforms besides ____ are unsupported.
People expecting things are pretty par for the course on the internet... Just be firm and curt. Close the issues quickly and refer to them if reopened.
If individual users are being persistent and causing problems, you can ban them. But if your problem is caused by a large number of different users, each coming across the project for the first time and creating annoying issues/PRs, banning doesn't help.
Or just ignore them. Don't engage with the message either physically as in typing a response, or even just mentally. Skim and move on.
That people send unwanted messages on the internet is the most common complaint I see regarding the internet that I just cant empathise with at all. I just really dont get it. Possibly has something to do with my first couple years online involving 4chan.
Wasn't really my experience. Besides, github have tons of useful smaller projects that either don't have any issues filed by strangers or have them while author happily ignores those.You have make something pretty big to have big enough community for it to cause issues.
And if that really bothers you, you can turn off issues in github.
I would image that satisfaction I'd get from having people actually use my code would greatly outweigh the inconvenience of being pestered about new features.
I'd imagine that it is satisfying, yes. But I think you may be underestimating how little other people will value your time in asking for help using that software. I have a colleague that maintains a popular open source project, and he spends a ridiculous amount of time answering questions about problems they're having, and very often those problems don't have anything to do with the software he's written, but the competence of the questioner. I had been considering developing a package, perhaps paid for with grant money, but the long term commitment that comes with it has scared me off.
There’s definitely something uniquely satisfying about knowing that code you wrote is saving other people time or helping them create things. The beautiful thing about programming is you can solve a problem once for yourself and also end up end up solving it forever for everybody.
That's precisely the point. They aren't free to make a pull request. A pull request puts a time demand on the project owner to review and merge the patch.
What they are free to do is fork the project, and do whatever the hell they want to in the fork.
