> they're holding off on going fully orbital until the Ship engines are relatively stable

They’re still re-flying Block 2 boosters. Hence intentionally leaving off heat tiles (and re-flying engines). Burning for orbit wouldn’t make sense on an, essentially, already-obsolete vehicle.

Block 3 launches on Flight 12. (It also validates a new pad.) Once that is debugged, SpaceX would be ready for an orbital attempt.

The crazy thing is right after orbital they go for propellant transfer. This is something our species has never done, and it’s ridiculously capability enabling if we can get it even within an order of magnitude of cost expectations.

I think they'll start launching starlink v3 satellites pretty soon, before perfecting reentry let alone rapid reuse. They've demonstrated a zero-gravity engine re-light several times and deployed dummy sats twice, that's all they need to put real satellites in orbit. We could see it on the second or third launch of the block 3 rockets.

I think the tile loss rate will still be important to them before that. Even in such low orbits, any tiles lost would take some time to come down (and might even survive all the way down).

If they can make it so they only lose tiles when in a suborbital trajectory, they may be safe to begin deploying real Starlinks as soon as V3 has proven engine relight.

Oh, do you think so? I thought they're looking very good outside the atmosphere at least, although it's difficult to really tell. I'd be surprised if that holds things up but you could be right.

As you say, it's just difficult to tell, the tile loss seems less dramatic than many used to expect back when the heat shield was relatively early in design, but ultimately only SpaceX knows how much they're passively losing during the coast phase.

All I'm saying is that that's one more factor besides relight that I think will need to be sorted (it might already be sorted, I wouldn't know) before orbit.