Hacker News Reader

Only 20k LOC in 3 years is impressive. And I agree that you should watch out for premature optimization an premature abstraction. What i found most interesting was using coffeescript that compiles to js that compiles to react that compiles to js. They might have been too focused on the technical side. You should probably pick the most boring and proven tech stack and focused more on the actual problem.

View on HN

Thank you! I'm one of the cofounders— keeping LoC low was one of our primary measures of code quality.

> we were too focused on the technical side

Yes. Oh boy we were.

> should probably pick the most boring stack

Agreed. I'd worked in the Coffeescript compiler so it was "boring" for me, but you should pick whatever's "boring" for you— likely not Coffeescript

Would you say the same thing if they were using Typescript instead of Coffeescript?

An actual problem might be how to generate a responsive design for "mobile" and different screen sizes. That's what you should focus on. The problem with picking new cool tools and languages is that you might end up solving dev/sys-ops issues, learning how to use them, and develop additional tooling, instead of solving actual problems. Unless you think it's a really good fit for the actual problem domain and thus worth the extra investments.

Reply on news.ycombinator.com