The precision of the timing and dosing depends a lot on what kind of diabetic you are. A type 1 with no endogenous production is going to need precision, and I can't imagine somebody wanting tight control giving up their pump and going with a pill!

type-2's on the other hand, have insulin resistance often accompanied by astonishingly high endogenous production. They can sometimes achieve nice control by just having a base of exogenous insulin that's just high enough for their pancreas to supply the rest of what they need -- the precision is supplied by the still largely functioning pancreas. I know at least one type-2 that has been put on insulin with instructions to inject X units with dinner, with no consideration of the actual content of the meal! Presumably a pill would work fine for that group.

Doesn't Type 2 already have pill-based treatments?

Yes, there are pills that will both increase insulin sensitivity (metformin), and to stimulate insulin secretion (the sulfonylureas). These are basically the first line of treatment. But the disease tends to progress to the point where they are no longer enough. The insulin resistance keeps increasing, and at some point insulin secretion also starts to fall off -- though it's usually still very high compared to a non-diabetic. At that point, exogenous insulin is all that's left. Unfortunately, it also seems to be used as a bit of a threat by MDs ("you'd better loose weight and start exercising, or you're going to need injections!")

BTW, I'm not an MD, just a somewhat informed layman. I just happen to be married to a type-1, and have type-2 running through my family, so I've been motivated to learn a bit about it all (mainly in an attempt at self-preservation!)

We put people on insulin glargine all the time (lasts 24h, peakless). This medication sounds potentially very useful. It wouldn't mean that all insulin could be oral, but if it works then perhaps we could save people from having to do their glargine injections.

I'm a type 1 diabetic. If this wouldn't work with rapid-acting insulin, might it not be a good thing for long-acting insulin, e.g. Detemir/Levemir (which I currently take)?

I'm type 1 too, there's potential, but the same problem with degradation exist. If you eat something that slows down or speeds up digestion or alters it in any way, it could give you a lot of highs and lows. It's like how certain foods take longer for the carbs to be absorbed and cause lows first then highs later. Alcohol could be effected even more too.

For basal-bolus therapy, you do want to have a long-acting, "peakless" dose which is then augmented by "spikes" of insulin when needed. So this pill is more likely to replace the long-acting insulins like Lantus/Levemir.

I think any forward progress in oral insulin is worth a close look.

Estrogen and progestin are given in slow-release IUDs or subdermal methods. Is science not able to apply the same?

Estrogen and progestin need to be at a certain, constant level. Insulin needs to be at varying levels based on blood sugar. To much insulin and you go into hypoglycemic shock.

Insulin is reactive. It regulates the glycemia (amount of sugar in the blood), which would vary highly in its absense depending on the time elapsed since the last meal.