Intel's IWD Linux Wireless Daemon Continues Picking Up New Features

[jwh7]

How does it compare to the current implementation? I have no issues at all with the current implementation, so what does IWD do better?

By "current implementation" do you mean versus wpa_supplicant? It is much simpler to use, faster, and for those who care*...smaller (iwd is 1.5 MB installed on arch, vs 4.1 plus at least 1.7+5.8 for deps). This article is a bit dated now (Nov 2018) with all the work done already this year, but a good summary of the development. I've had no issues using iwd with systemd-networkd on three machines (*including an old netbook running arch32). :-)

[starshipeleven]

Good to hear! How does it compare to the current implementation? I have no issues at all with the current implementation, so what does IWD do better?

It's easier to work with if you are using it directly, and requires less babysitting as it is not stateless by design like wpa_supplicant.
(example: it will auto-reconnect to the wifi if for some reason the connection drops, with wpa_supplicant it's all on you to detect that there is no more connection, rescan and try to connect again)

But if you are using already a software frontend to it (NetworkManager or Connman, or whatever), you won't really notice much as this is just moving the leg work from one program to another.

[Vistaus]

It's easier to work with if you are using it directly, and requires less babysitting as it is not stateless by design like wpa_supplicant.
(example: it will auto-reconnect to the wifi if for some reason the connection drops, with wpa_supplicant it's all on you to detect that there is no more connection, rescan and try to connect again)

But if you are using already a software frontend to it (NetworkManager or Connman, or whatever), you won't really notice much as this is just moving the leg work from one program to another.

Ah, thank you so much for the explanation! In that case there's no need to switch as I use NetworkManager.