Jonas Windey it’s a con for two reasons:
- It adds mass to what you have to deliver over HTTP/HTTP2. An author that is using async/await a lot, is liable to create a much larger package right now. At this point, to the tune of dozens of lines of code per async function.
- It’s more to execute. Those dozens of lines of code need to be called. As I stated, abstraction comes at a cost. More to execute generally means less performance. This isn’t going to matter for 95% of use cases, but when you’re looking at scaling Node or a high performance real time app, it’s extra overhead for no real gain other than feeling good about how the code looks.