If 1 dose, followed by a 2nd some time, can cut that number down a lot (some people won't get Covid, and few will get a bad one), hospitals can be freed and start returning to their usual work - cancer treatments etc. That's a big win. I think it's still the case that more people die of cancer than of Covid.
But that said, I also worry whether the vaccines will work with the longer spacing - I hope results like this show us they would.
Preventing the spread of COVID19 has an exponential effect on all stats.
If 1-dose only reduces the symptoms (but continues the exponential spread), then 20% vaccination rate means that you have 20% fewer hospitalizations after a month.
-----------
In contrast, if 2-dose prevents the spread, then 10% vaccination means 10% fewer spread *PER GENERATION* of COVID19.
That's 90% of the hospitalizations in one generation (~1-week period). After 2-weeks, you have ~81% hospitalizations. After 3-weeks, down to 73%. After 4 weeks, only 65%.
Controlling the spread is the #1 health priority that has exponential effects. I'd take 1/2 the vaccination rate if we have a better chance at tackling the exponential-spread issue.
Does 1 dose prevent the spread of COVID19?
If it requires 2-doses to get a true "efficacy" (lower spread), then it might be more worthwhile to push 2-doses, rather than 1-dose (where everyone gets "less sick", but the virus continues to spread exponentially).
Real problem: 1-dose wasn't really tested on a large "phase 3" style 10,000+ person test. Its impossible to answer this question (yet). It will take further testing to know.
----------
I think there's something to be respected about sticking with what was tested, even if the scientific theory argues for a different approach.
Just because the theory says so, doesn't mean that the theory is true (at least, not until the theory is well tested).
Doing something different and hoping for the best just because its kinda correct theoretically is... overly optimistic. I wouldn't necessarily be against it, but it makes sense to stick with what was formally tested.