That's not as true as it used to be, and there are still plenty of useful open journals/open science publications, though proper attribution would often be important.

[edit] you could pretty much say that on principle, any significant development should have a publication in the open.