I have been thinking about this a lot lately. It's been an immense problem for me and it's destroyed almost all my side projects.
The problem is that you're interested in the wrong aspects of your side project. You're interested in learning and solving the code problems that come with your project. You have to shift your perspective to the human side of your project. You have to learn to see the user's issues as the problem to be solved.
Let's use an "reviews" site as an example. Your users cant find reviews, that annoy them. What you're solving is not a programming problem; your problem is "make it easy for users to find reviews". That is your problem. Your problems is not what the best algorithm for ratings. None of the code stuff matters. These users do not care about code.
When you've shifted your perspective to the human side you'll start to see hacky solutions differently. A crappy algorithm for ratings isn't a poor solution, it's a perfect solution. You've solved the reviews problem. You have a perfect solution.
The problem is that most of aren't actually excited about the human side of our side projects. This is because we get excited about things that interest us, and what interests a programmer is usually programming. It's not surprising we choose projects with interesting code problems.
Start reflecting on the human/sociological side of things before you choose a project. You might be surprised at what you find. Projects that seemed interesting may suddenly become dull and the one's you thought were dull might suddenly become interesting. A human perspective doesn't exclude building stuff with a code focus, just make sure coders are your audience. You and your users just need to be excited about the same things.
There is this classic saying about "building something for yourself" resulting in the best products. As with most sayings, they left out some important information. They forgot to tell to you to make sure that what you're building for yourself is the same thing you're building for your users.
Make your goals align with your users. Otherwise, you'll find yourself trapped down the rabbit hole.
Some extra thoughts
Don't forget that all your tricks to solve programming problems work on human problems. For example, for a reviews site it's easy to phrase the problem as "make it easy for users to find the best reviews in the most efficient and enjoyable way possible while allowing them to simultaneously book and view and and and....". Break it down into it's simplest elements first, just like you would a programming problem. The simplest element of the problem is "make it easy for users to find reviews"; best reviews are a separate problem.
You're probably good at solving programming problems, breaking them down and being productive. You know all the 37signals posts, all the design patterns, and can quote re-work by heart. Use those principles you have learned and apply them to the human side. Everything your learned about productive programming applies to productive life.
The problem is that you're interested in the wrong aspects of your side project. You're interested in learning and solving the code problems that come with your project. You have to shift your perspective to the human side of your project. You have to learn to see the user's issues as the problem to be solved.
Let's use an "reviews" site as an example. Your users cant find reviews, that annoy them. What you're solving is not a programming problem; your problem is "make it easy for users to find reviews". That is your problem. Your problems is not what the best algorithm for ratings. None of the code stuff matters. These users do not care about code.
When you've shifted your perspective to the human side you'll start to see hacky solutions differently. A crappy algorithm for ratings isn't a poor solution, it's a perfect solution. You've solved the reviews problem. You have a perfect solution.
The problem is that most of aren't actually excited about the human side of our side projects. This is because we get excited about things that interest us, and what interests a programmer is usually programming. It's not surprising we choose projects with interesting code problems.
Start reflecting on the human/sociological side of things before you choose a project. You might be surprised at what you find. Projects that seemed interesting may suddenly become dull and the one's you thought were dull might suddenly become interesting. A human perspective doesn't exclude building stuff with a code focus, just make sure coders are your audience. You and your users just need to be excited about the same things.
There is this classic saying about "building something for yourself" resulting in the best products. As with most sayings, they left out some important information. They forgot to tell to you to make sure that what you're building for yourself is the same thing you're building for your users.
Make your goals align with your users. Otherwise, you'll find yourself trapped down the rabbit hole.
Some extra thoughts
Don't forget that all your tricks to solve programming problems work on human problems. For example, for a reviews site it's easy to phrase the problem as "make it easy for users to find the best reviews in the most efficient and enjoyable way possible while allowing them to simultaneously book and view and and and....". Break it down into it's simplest elements first, just like you would a programming problem. The simplest element of the problem is "make it easy for users to find reviews"; best reviews are a separate problem.
You're probably good at solving programming problems, breaking them down and being productive. You know all the 37signals posts, all the design patterns, and can quote re-work by heart. Use those principles you have learned and apply them to the human side. Everything your learned about productive programming applies to productive life.