I know I'm late to the party, but a few observations about Netchoice v. Paxson (TX social media decision). It is a very frustrating opinion since it gets a bunch of stuff right (IMO), but goes way off the rails in other ways. /1

So what does the majority get right? At bottom, the principle that speech to enjoy First Amendment protection must be expressive. The rise of 1A as a general weapon against regulation has been facilitated by eliding this point. /2

For example, the push against laws dealing with collection of personal information should not raise 1A implications because they aren't expressive. But increasingly privacy is seen as raising 1A concerns anyway on a theory of researching an audience. (See US West v FCC) /3

IMO where the 5th Cir runs off the rails in failing to understand how platforms work. There actually is real editorial discretion/expression/curation going on here. So we should at least be dealing with intermediate scrutiny. /4

The majority invents an activity of "censorship," which it treats as not speech. But the decision on what to make available and how to promote are editorial decisions. The fact that they may occur after initial publication should not matter (and not all do). /5

Likewise, the decision of which individuals to carry is clearly an editorial function. So while 5th Cir is correct (IMO) on principle that 1A needs expressive conduct, it fails because the platform is engaged in expressive conduct. /6

5th Cir has now created a new type of expressive activity, "censorship," that is not protected by 1A. Hard to see how to distinguish "censorship" from protected editorial discretion. /7