Or am I misunderstanding something (or everything) ?

Maybe somebody can help me understand: If websockets can be used as an attack vector as described in the article above, then the log4shell-problem seems like a small issue in comparison. A malicious server which returns javascript, where the javascript can both talk back to the server (ajax/XHR), and at the same time can open arbitrary websockets to the victims localhost (or any RFC1918 addresses for that matter), can potentially probe and exfiltrate information and change things on the victims host or intranet (not everyone has implemented zero trust..). Do the latest browsers like Chrome/Firefox/Safari/Edge really allow untrusted javascript to initiate webstocket connections to ANY ip, including localhost and RFC1918, out of the box? Seems like a major security issue to me... I was under the impression the this kind of access was disabled years ago, but I have to admit I have never actually tested it..

Got any links to that explains this SSL issue? I would like to know more