If you want security, use a secure OS. But it's not so easy, often decisions are made not to have the best technical solution, but to fulfill business agreements.
You probably had to compile libsodium and build a shared object. That can take a long time. But the scripts that run when npm installing modules can contain malicious payload, yes.
I always try to find something that runs in web-assembly, but it's better to avoid nodejs altogether if you want high security.
Go is much better for these kind of things overall.
"weak stdlib": You can use the Java ecosystem, most of the things you need are there
"it can be very hard to follow": I agree, if you are not a lisp developer it's really hard, but if you develop clojure everyday, it's as easy as anything else.
The answer is usually clean code when it comes to clojure. Keep your functions small.
"more memory intensive, and generally slower": yep, the JVM is more memory intensive than Go. No surprise there and Clojure adds up on top of that. Startup times are pretty slow too. But compared to python it's still fast. Apples to oranges.
I would say Clojure and Go are both great languages that tackle different problems so it's not a fair comparison.
Yeah, also the complete ruin of the internet. I don't want to read generated content. Feed that back to AI and you got a ruined internet and AI can't even learn anymore.
Yes because people are stupid and vulnerable and will believe things a computer tells them.
They can make life decisions on chatGPT advice, instead of asking friends or family. Heck maybe they don't have family, vulnerable people like that can be exploited with convincing chatbots easily.
For startups, subscriptions can stack up fast and it's the same as B2C, they can't afford it. Subscriptions are liabilities.
But for medium sized businesses or larger, it depends on how much value the tool provides but should be fine, especially if they can fire somebody now that a fancy AI does his job.
Enterprise pricing is a thing for a reason.
A lot of people have tons of subscriptions already, I would go with usage based pricing. It's a deal breaker for me, if I need to add a monthly subscription for something I don't use a lot.
But the most important thing is to get the product right.
It happens sometimes. Just the other day a local TinyLlama instance started asking me questions.
The chat memory was full of mostly nonsense and it asked me a completely random and simple question out of the blue. Did chatbots evolve a lot since he was created.
I think you can get models to "think" if you give them a goal in the system prompt, a memory of previous thoughts, and keep invoking them with cron
I often build project specific tooling, to generate code or just transform some data, but maybe someday I create a FOSS version I can share.
Stuff like parse some XML or JSON and output Go structs and functions and Typescript types and functions, HTML,React Components, SQL Tables, Stored functions, Pl/PGSQL.
Mostly to avoid writing boilerplate when I use the same data structures in the database, middleware, client.
For simple CRUD apps, it works well. I use local KV to track changes so I don't have to rerun things I don't need.
But Deno is great for CLI reporting tools or Scheduled tasks, fetch and aggregate data.
I think of Deno as a little swiss army knife. It's a tool that got everything built in.
I use the Repl a lot, just for a specific task, get it done then move on.