Disclaimer: This post was originally written on April 28th, 2015; any links, references, or content within may no longer exist.
There are a ton of things I want to learn - languages, architectures, designs, systems...the list goes on and on. There are also new ideas I would love to build. A recent post on Hacker News, titled A web-reading bot made millions on the options market caught my eye. For a while, as I perused the comments, I was thinking through the idea of building a bot that would be capable (in a non-production, "oh-wow-isn't-that-cute" kind of way) of parsing news stories, tracking posts from Twitter for specific companies, tracking their stock price movements, and determining when would be a good time to buy based on the criteria I input.
There was one comment that got me to start thinking about the full scope of the project - the generic "anyone can do this over a weekend" statement. So I started digging in...could this really be something I could get done over a single weekend? I tore into...
- Information on options trading (just the basics),
- Web parsing and scraping
- Language selection (Elixir or Clojure would probably be the final two to choose from)
- API and UI design (I need a way to interact with my bot)
- Market simulations with API's I could interface with
And I came up with the solid answer of this is more than just a project for a single weekend. Even gathering and consuming enough knowledge of a moderately-sized domain would take more than a weekend, let alone one the size and depth of investing and the options market in particular. However there are many developers out there that will blindly throw the "shouldn't take more than a weekend" statement out there and let it wreak havoc.
With that said, I still think its somewhat of a fair statement - with one modification:
With a solid grasp of the required knowledge, this project could be completed in a weekend.