I need to get better at failing

I have about 50 projects that I started and never got around to finishing. I've been working on these on and off for about 2 years now, so that makes on average 1 project every 2 weeks.

I've had a few major projects that I intended to launch, projects that I had worked on for a "whole month(!)" but eventually lost interest in.

My latest project has achieved liftoff.

0. I quit my job.

1. I started with a new, interesting, and simple open source web framework. I made sure that the code was understandable -- I should be able to bend the framework as if I had written it.

2. I chose a new, interesting, and simple database. I chose a NoSQL database which forced me to design the architecture differently.

3. I wrote many of the components that already exist in web frameworks and libraries -- a simple relational ORM with caching, object models for forms and server-side validation and the client code to tie everything together.

4. I started soliciting feedback from people. I made a couple of friends along the way and they helped me stay focused as well.

Since I chose new technologies I was forced to reinvent the wheel quite a bit, but the advantages are that (1) it's really fun and (2) since I understand everything from the client down to the database and framework, I can write websites incredibly fast.

Keep at it -- I wouldn't worry too much about abandoning your projects as long as you're doing what interests you. Don't worry about what your project will ultimately become. Rather, have a general sense of where you're going and just keep hammering at your codebase. Make it squeaky clean and think hard about the design. On the other hand, keep it simple and be prepared to restart especially when you've made the mistake of over-designing. Most importantly, have fun.

Once you have a complete understanding of all the layers you will be able to execute a million miles an hour when you do have a great business idea. Then you're flying.

- J