It's the classic chicken-and-egg problem. TV manufacturers don't want to push a feature (which increases the build cost of their product) that no one cares about, but the national networks that produce the high-value content carried by stations have little incentive to support ATSC 3.0 when hardly any consumers have equipment that can take advantage of it (and even if they did, only a sliver would use it anyhow since only about 1 in 6 households watch OTA TV). The only group that *really* wants ATSC 3.0 to succeed are the non-network-owned local station groups (e.g. Nexstar, Sinclair, etc.) but, without a government mandate to force a conversion from 1.0 to 3.0, there's really not all that much they can do to spur its adoption.
Except that "4K" has been a main buzzword that the National Association of Broadcasters has used in association with ATSC 3.0 since they began promoting the concept several years ago. If stations had the bandwidth and the networks were willing to offer them sports and primetime content in 4K, you can bet that they'd carry it. But, as I say, bandwidth is especially limited as long as its being divided between 1.0 and 3.0, forcing tower sharing. So 1080p HDR is the best one can hope for via 3.0 at any point in the foreseeable future.
The real value to local stations of having high-quality feeds (4K, HDR, etc.) is in being able to sell them to MVPDs for higher retrans rates. That's where the money is for local stations, not so much in free OTA broadcasting and the incremental ad revenue those viewers provide. That said, local stations could theoretically offer higher quality feeds to MVPDs than they broadcast OTA. This past summer we saw many NBC locals offer live feeds of the Olympics in 4K to certain MVPDs like Comcast and YouTube TV, even though they weren't even doing an ATSC 3.0 OTA broadcast. So those two things aren't necessarily linked.
As for targeted advertising via 3.0, yeah, I agree that that's of interest to stations but they'll likely need a minimum critical mass of 3.0 viewers before such an effort would be feasible/profitable. And of course, that feature has zero marketing appeal to consumers, so it does nothing to promote uptake.
Given that an antenna easily connects to a gateway/external tuner by just screwing in a coaxial cable, I doubt we'll see an antenna physically combined with a tuner. The tuner will need to connect to power, which limits where it can be placed. Better to have the antenna completely separate so that any length of coaxial cable (including pre-existing wiring already in the walls) can be used between the antenna and tuner, allowing for optimal antenna placement for signal reception.
