Ask HN: If you had 5 years of uninterrupted time, what would you build and why?

RPGs are hard because they require huge amounts of writing and art assets along with a farly complicated code base to allow for all the interactions.

I'd focus on creating a generic framework that would allow an author to write their game, define the rules in a simple DSL, choose from a set of standard art assets (or plug in their own).

The idea would be that no coding would be required, and weather you are creating a turn based game like fallout, or action game like Diablo it would all fit together.

So the author would provide their dialog, their quests, their item definitions and rules, and be able to generate a game playable on multiple platforms.

As far as why benefit to creating a browser in 2018 that targets 16 year old OSs, because there is no guarantee that we will be in a 64-bit (or greater) exclusive world by then. Without knowing the future, my goal setting off would to remain as cross-system ready as possible, which of course includes some 32-bit OSs

I may spend a lot of time in management-land, but I love coding, and still spend a lot of time in Ruby and CoffeeScript. In the past, I've coded in C, Perl, a smattering of Basic, and have a passing familiarity with Python, Scala, and Java.

Ruby is a nice place to be, mostly thanks to the community, but I get very frustrated with the Ruby core. There's a lot of bugs and inconsistencies in the standard library, and the VM should be a lot quicker. V8 is an existence proof that it's possible to build a fast dynamic language runtime.

Instead of a sane Net::HTTP, or some good GC instrumentation, or an actual grammar to try and do some static analysis, we get Refinements, which should never have made it into the core language.

I get a little jealous of the Python guys sometimes; not only for SciPy and NumPy, but for the fact that the Python core team does spend a lot of time fixing and improving the internals. They don't get it right all the time, but the level of engineering feels better somehow.

But I don't like the "one way to do it" attitude; one of the things I really like about Ruby is that the community feels more experimental, more tolerant of change, and less likely to criticize non-constructively.

Rather than just complaining about it, I'd like to try and scratch my own itch, and see if it's possible to build a language that can match V8 for speed, Ruby for creativity and expressiveness, and Javascript for portability.