I recently read an article titled Deno’s Decline. I shouldn’t have. I’ve never really understood the negative atmosphere around the Deno project.
In my opinion the only claim in the article that holds any weight is that their marketing on “edge” doesn’t match their current region offering. The article spends 10 paragraphs on this trivial concept, concluding with:
This downward trajectory is obviously not a good look
Can we expect a “Vercel’s Decline” article as they move from “Edge” to “Regional” with Fluid Compute. Look, Deno Deploy may be struggling, it might not? But come on, do we really need to pile on? How is this productive in any way?
It’s comical the author mentions alternatives. Viral articles like this run completely counter to alternatives existing. Seriously, if you don’t like their product, just use something else. It’s a good thing that we have choices and new ideas.
Deno KV looks like nothing short of abandonware.
It’s a minimal data API. Does it work or not? How does it perform in production? No, none of that. Instead, let’s talk about the most important thing in software - is there a recently tagged version number?
Maybe instead we could talk about how they’re offering a unique API to FoundationDB with at-least-once semantic queues where message handlers automatically scale using v8 isolates. Or how you can atomically enqueue messages and interact with KV state in the same transaction.
The article then goes on a tirade against JSR. Let’s have a look:
- Is JSR open source? YES
- JSR is designed to be a public good for the JavaScript community, and will thus always be free to use.
🤦♂️ It’s time for the OSINAY link.