Audit your programming language. If you are writing in Python, PHP, Ruby or server-side JavaScript, you are optimizing for quick prototypes, at the expense of performance, which means that you will need to throw in more cloud resources to scale. Consider faster languages (e.g. Rust is typically 10x-30x faster than Python, but also Go, Java/Scala or C#/F#).
I get where this sentiment comes from, but I use Ruby to build web-things because it turns our Ruby and its ecosystem of tooling, are really good for building stateful (i.e., DB-backed) web-things! This is increasingly true for small teams, or even solo devs, as Rails continues to push forward.
Yes, other languages might be faster, but that’s also not always the bottleneck. And, as we know, it’s all tradeoffs, all the way down. For me, I’d look at the other optimizations well before I worried about this one. But you do you. 😄
I like this advice, though I would add a little more on theme of resilience. Many of the items considered in the risk analysis ultimately boil down to a loss of connectivity.
If an application permits it, make it work offline
Otherwise make it do as much as possible while disconnected (local-first, peer-to-peer)
I’m uncomfortable with the definition of a Zerg: is it anything that consumes finite resources (e.g., energy, disk space, food, attention, democracy, etc.) without providing something at least as valuable in return, thereby pushing us closer to exhausting our resources or societal collapse? I hope I’m not a Zerg!
If you don’t take any money and can’t expect contributions, where are the resources coming from to invest in creating really high-polish software? Audit audit audit invest invest. It sounds great, but it costs something. If you have 10 users the way to maximize the chances that your project survives at all is to invest in getting more users first so that it becomes worth investing in all this other fit and finish (and a Rust rewrite 🙄)
I get where this sentiment comes from, but I use Ruby to build web-things because it turns our Ruby and its ecosystem of tooling, are really good for building stateful (i.e., DB-backed) web-things! This is increasingly true for small teams, or even solo devs, as Rails continues to push forward.
Yes, other languages might be faster, but that’s also not always the bottleneck. And, as we know, it’s all tradeoffs, all the way down. For me, I’d look at the other optimizations well before I worried about this one. But you do you. 😄
I like this advice, though I would add a little more on theme of resilience. Many of the items considered in the risk analysis ultimately boil down to a loss of connectivity.
I’m uncomfortable with the definition of a Zerg: is it anything that consumes finite resources (e.g., energy, disk space, food, attention, democracy, etc.) without providing something at least as valuable in return, thereby pushing us closer to exhausting our resources or societal collapse? I hope I’m not a Zerg!
If you don’t take any money and can’t expect contributions, where are the resources coming from to invest in creating really high-polish software? Audit audit audit invest invest. It sounds great, but it costs something. If you have 10 users the way to maximize the chances that your project survives at all is to invest in getting more users first so that it becomes worth investing in all this other fit and finish (and a Rust rewrite 🙄)