I'd call that a "distinction without a difference". Those are generally fallacious, and more often pedantic. Sure, you can still technically use React if, say, you're based somewhere that patents aren't particularly enforced. But if that's the case, why do you have patents in the first place?
That is plain wrong. Software patents exist and are enforced in the EU and other member states of the European Patent Treaty. The European Patent Office even has a nice FAQ on helping you get your software patents granted at:
The situation is a bit more subtle than you make out. The EPT has clear language which excludes software from patentability, but that hasn't stopped lawyers from arguing that some software can be patented if it has a physical effect (even a minimal one). Wikipedia actually summarises it decently: https://en.wikipedia.org/wiki/Software_patents_under_the_Eur...
The situation isn't even that subtle. The "physical effect" AKA the German Teaching has eroded into a "further technical effect" in the past decade. And with the UPC looming, things could get even worse. But that's getting off topic.
You can't legally use React without a patents license. The difference between the patents grant and the copyright license is mostly in the timeframe they cover (20 years for patents, 100+ years or whatever for copyright).
That nobody seems to know whether there are in fact any patents licensed by the patent grant within React is one of the most frustrating things about it.
Agreed. It gives the entire debate a very metaphysical quality. How can I determine the value of a patent grant when I don't know what the patents are, or if they even exist?
Ceding your license to use any patents Facebook may or may not have on React. There's a difference.