Throwing away your code is the worst mistake you can make.
Comes down to a miriad of factors:
Code is easier to write than to read — because everyone writes code slightly differently;
Architectural problems — the layers of the program have not been properly defined or are poorly managed.
Mirage of inefficiency — probably overstated because of the advance in processing speeds, network speeds, etc. Even “messy” code that works, often works better than expected.
It’s ugly — using simple “Search and replace” can fix inconsistent naming, or it’s “too long/big/unwieldy” so where practical use things like syntactic sugar to cut down on the mental friction of ugly code.
Old code has been worked on over the years, bugs have been fixed, etc. Throwing away old code is throwing away all of that knowledge and progress.
Goes back to a comment in the Problem Solving document that says “It’s always easier to refactor a complete piece of code - then to write a perfect piece of code.”
Things I did:
Reorganised all of the markdown files that I have scattered across various apps (Wiki-JS, Zettlr, Notion, etc) into this git repo. Much simpler and centralised now.
Started documenting more shit, including what I did to fix things from start to finish.
Updated the problem solving document to include recent thoughts from Reddit.
Worked on the design documents for the User Interface and the Web3 Interface documents for my personal crypto project.