Do we need another open source webhosting panel?

classy

@cassa said:
Why not contribute to the existing panels?

1. Which one?
2. Where do I start?
3. If I don't like the UI, and I do change it, what then?

classy

@joepie91 said:
It is. There is one global event loop implementation with one set of primitives that everything uses.

What I mean is if you need to do heavy computing you're always better off with a native module. At which point using a native thread is trivial.

Btw, the node event loop is just an infinite while loop calling libuv and timer callbacks.

ricardo

It'd be worth taking all the popular panels out there and doing a feature comparison, what features or missing, what licensing it uses, whether they're still active, how they can be improved etc.

It'd perhaps take a few days to do but really gives you the groundwork needed for an informed decision/discussion.

Wiki does have a foundation to start with. https://en.wikipedia.org/wiki/Comparison_of_web_hosting_control_panels

What's maybe a good barometer is those panels who are IPv6 / HTTP2 and any emerging technology friendly. Fag packet calculation says there's probably only half a dozen worth considering into a 'final round' of comparison of features.

joepie91

classy said: What I mean is if you need to do heavy computing you're always better off with a native module. At which point using a native thread is trivial.

At that point, you're basically writing C++ rather than Node. Which is fine for the occasional bit of computation (eg. hashing a password), but if your application is all heavy computation, then you probably just shouldn't be using Node to begin with.

classy

@joepie91 said:
At that point, you're basically writing C++ rather than Node. Which is fine for the occasional bit of computation (eg. hashing a password), but if your application is all heavy computation, then you probably just shouldn't be using Node to begin with.

Agree. Now we're at the same frequency :-)