Imagine that you had endless resources to devote to a single project, what would it be - and why?
For further actions, you may consider blocking this person and/or reporting abuse
Read next
The Fastest Way to Start Your AI Project–Quickstart ModelKits
Jesse Williams -
The whole “shameless plug” conundrum and stigma associated with sharing one’s own blog posts
Prahlad Yeri -
Open-Source will make you richhhh! 🤑💸
Rohan Sharma -
AWS Lambda: OCR and Text Translation in the AWS Cloud
Wojciech Lepczyński -
Top comments (4)
It would be a video game -- an amalgamation of some of my favorite sandbox space games, with various kinds of automation (production, logistics, trade, defense), fun ships, epic story, economy simulation and so on. Inspired by games such as X3, Factorio, Empyrion, Subnautica, etc.
As part of it, I also want to develop a functional-programming-based game engine -- an open suggestion by John Carmack (but I would not do it in Haskell). I love building framework-less FP systems, and so far I have not liked the needless framework overhead in existing popular engines that I researched. So if I ever build a game, I was planning on doing my own engine anyway.
Then on top of the engine there should eventually be some design tools to visualize assets and other concerns that improve productivity.
It would be a cherry on top if I could financially get away with releasing the engine and tools libre. I also think it is an interesting model to get the game source code when you purchase it.
You know, little things like that.
That sounds really cool! Out of curiosity, what language would you implement the game engine in?
My favorite language is F#, and I feel really comfortable with what I could do there. Rust is also something I have my eye on for going lower level (no GC / runtime). From what I can tell so far, I would program in Rust similar to the simplified way I already use in F#. I wrote about my F# tactics here.
I would build a distributed relational in-memory HTAP database.
In my past few roles, the most difficult problems have been surrounding timely analytics on operational data. The most recent solution that I built involved a system that could shard analytical queries between an OLAP and OLTP database and a complex ETL to constantly migrate data between the two. While it works well, it is quite a bespoke solution, and I would love to create something that could solve the problem more generally.