I work on Lighthouse (user of the Chrome DevTools protocol) and work with the DevTools team (obviously the main user of the protocol) - and when I first saw browserless I was blown away. So cool! Good job with your success.
What was the hardest part re: working with the protocol?
The protocol is pretty easy, I think coordinating the necessary “enable” calls is a bit cumbersome. Also the legacy JSON protocol is harder to support, but I understand why.
Hardest part is debugging crash issues and why they happened. You either just get a generic “Page crashed!” error (which I think is puppeteers handler message), or “browser disconnected!”. That and chromes logs are just crazy noisy, which I haven’t gotten a lot out of.
Those are probably the biggest, thanks for asking!
What was the hardest part re: working with the protocol?