It intentionally flew a just-barely-suborbital trajectory, because the in-flight relight of the Raptor engine (required for deorbit burn) is still in the testing phase. The relight succeeded, so they may be cleared for a full orbital trajectory on upcoming flights. Note that Starship achieved the near-orbital trajectory with a significant amount of fuel still onboard, which was intentionally vented before reentry, rather than burning it for a few extra seconds to achieve full orbit. The difference is negligible from a difficulty perspective.

$9 million for what? With Starship, there's no downrange retrieval of booster/fairing, no remanufacturing of an entirely new second stage for each flight, and minimal refurbishment (implied by the first sentence of my above post). Perhaps $2-$3 million for operating costs around Starbase, sure, but not $9 million. Not once the launch cadence approaches and then quickly dwarfs Falcon 9 levels, at any rate.

If Starship achieves full and rapid reusability, then it will be highly cost-effective even with a 16-ton payload. The only marginal material cost is the fuel itself, which is about $1M per Super Heavy launch. Contrast this with $10M simply to manufacture each new Falcon 9 upper stage, which is expended after each and every flight, and Falcon 9's payload to LEO is only about 18 tons. Not to mention, Starship supports a much more flexible payload form factor, to accommodate e.g. Starlink v3 satellites; no other existing launcher can do that.

Agreed that Pica probably makes the most sense for Starships returning to Earth from Mars; I expect there will be very few of these, due to the exorbitant cost and complexity (for the next few decades at least) of ISPP on Mars, thus there would be minimal economic benefit for being able to reuse them without refurbishment. And it's not clear at all whether the ceramic tile design will be able to withstand a MUCH higher-energy reentry, returning from Mars (or even from the Moon, for that matter).

Another question for the short-to-medium term is whether it may make the most sense to simply replace the aft flaps after each flight (pop out the old burned-through ones, pop in new ones), rather than overbuilding them for reuse? If the burn-through damage is limited to the flap itself, the tower could probably be outfitted to perform such a replacement in a matter of 10-20 minutes, while the ship is being refueled. Of course, IFT 10 flew an intentionally demanding reentry trajectory to stress the flaps; we'll have to see what the damage looks like for a nominal reentry profile, and also after a few more design iterations.

Thermal expansion. Long strips would warp and buckle as they expand. Small tiles have slight gaps in between to accommodate the expansion.

The gravity gradient (degree of warping of space) near a black holeâ(TM)s event horizon is vastly greater than the gravity gradient near (or within) an equivalent-mass star. I believe thatâ(TM)s what theyâ(TM)re talking about.

Not quite, because two events outside each otherâ(TM)s light cones cannot be relatively ordered this way. For such events A and B, there are frames of reference in which A was first, frames of reference in which B was first, and frames in which they were simultaneous, but none of these frames can be said to be a privileged âoecorrectâ frame.

That's because they did not come from the patients themselves. See my post above.

The stem cells were from a separate donor. The very first sentence of the study's abstract is: "Zimislecel is an allogeneic stem cell–derived islet-cell therapy."

allogeneic: denoting, relating to, or involving tissues or cells that are genetically dissimilar and hence immunologically incompatible, although from individuals of the same species.

Ironically this is not the case.Type 1 Diabetes by definition is an autoimmune attack by the body on its own cells. Using more of those same cells won't stop the immune attack. It's possible that a more selectively targeted immune suppressant may be sufficient to use if it's the patient's own cells; e.g. the calcium-channel blocker Verapamil has been shown to slow down the attack, but periodic re-infusions of beta cells would probably still be necessary.

It is emphatically not a once-a-day thing.Type 1 diabetics (who aren't on an insulin pump) use both a base long acting insulin AND a rapid mealtime/correction insulin. This can be 10+ shots per day, and requires constant vigilance to keep blood glucose in range. The insulin pumps use only rapid insulin, but have other downsides; scarring, internal 24/7 exposure to a teflon cannula, and 100% dependency on the technology working properly at all times to keep the patient safe. If the continuous glucose monitor fails (e.g. by sending incorrectly high readings), the pump can silently overdose insulin, leading to very bad outcomes. To reduce the chance of this happening, most pumps are set to maintain an unhealthily high blood sugar level, which can have its own long term health consequences. (E.g. with manual injection one can target an optimal range of 80-90 mg/dL, but pumps generally won't allow a target below 110 mg/dL.)

This is true, for now. Targeted immunosuppression therapies may eventually make patient-derived cells a more viable option; e.g. Anokion is currently working on "inverse vaccines" for T1D, which in conjunction with patient-derived stem-cell-based beta cells may provide an effective cure. Verapamil (a calcium-channel blocker) has also shown effectiveness in slowing down beta cell destruction, without broadly compromising the immune system. That sort of drug, combined with periodic stem-cell reinfusions, could also in principle provide a practical cure. My 7yo daughter was diagnosed with T1D at age 4 (no family history), and we are hopeful that one of these therapies may become available for her in the next decade or two.

Dang formatting. 2GRT269, 8DYX247, 3XIV159, respectively.

A family friend was once assigned â2GRT269â(TM), and had to replace it with a custom plate to avoid embarrassment. Then thereâ(TM)s â8DYX247â(TM), which some unfortunate Tesla owner is currently driving around with. But I actually wouldnâ(TM)t mind having â3XIV159â(TM) for my car. (Get it? A Pi plate! â3 14 159â(TM).)

This is eugenics. (With all the baggage that entails.) There are very few cases where such an approach could even theoretically do actual good, and they all depend on a degree of scientific knowledge and technology (and global availability) that we don't have yet.

Imagine we had a magical free CRISPR shot that could repair a "broken" genetic base pair. Only a tiny fraction of diseases, those caused by a single well-defined gene (e.g. Tay-Sachs, cystic fibrosis) would be fixable this way. Most "genetic" diseases (such as T1D) are caused by constellations of tens or hundreds of interacting genes, with all sorts of unrelated side-efffects, and with environmental and probabilistic components as well. There is no straightforward genetic fix for them; that's why they haven't been bred out of the population despite millions of years of natural selection.

In any case, pretty much everyone on the planet (including you) has a handful of not-so-great genetic mutations, and probably two or three badly screwed-up recessive ones. The only current practical (to the wealthy) approach, if pre-screening is available, is to choose not to have kids with someone who shares the same problematic genes that you have, or if you are a match and want kids anyway, to use IVF and pre-screen embryos. (Ethical takes on this vary.) It's also a slippery slope; how bad does a gene have to be in order to take this approach?

Tl;dr: the genome is far more complicated than you think, and for 99% of human ailments, genetic manipulation would not even theoretically work to solve them.