Days sheltering in place: 149.
Exercise: indoor bike, 8 miles’ish.
Keyboard practice: practice, practice, practice.
The joy of being a sole coder on a project is that you are your own boss, you are the architect, you set the standards and you decide whether they should be followed or what the exceptions are as you go along to met your own needs. As far as the code is concerned, you are god.
You have no architectural discussions, except with yourself. There are no code reviews because you already read the code when you wrote & debugged it.
It worked on your own machine and thus there is no QA.
Unfortunately you have to be DevOps as well as architect, designer, engineer and creator. DevOps isn’t the fun part, it’s the “somebody has to do it” and that somebody is me job.
And your code is flaky. Your architecture is weak. Your features are buggy. Your code is not extensible or often even maintainable. You can’t ask anybody else how something works, because you are the oracle and only one that knows anyway.
Having to discuss your architecture with somebody else, having your code reviewed for flaws and standardization, getting it rejected then is better than getting it rejected by QA, which is better than the bugs getting to the user.
Having to compromise on “your best way” to implement something, having to be a team player, creating code that lints, passes code review and QA feature and regression testing. These are things that are part and parcel of writing great, maintainable, extensible code that becomes a solid product.
Yes a single person can build a great product, if they have enough experience from not working alone, are a great architect, designer, coder, debugger, are hard on themselves, do their own QA & support. They are generally the exception.
Plus, it’s lonely when you are the sole programmer. Especially when your code is broken, and you don’t know why.