[ben_w]

> Won't a photon climbing out of a huge gravity well have a huge redshift,

Yes

> thus confounding estimates of distance from us and estimated age?

Not if you know, or can get a good estimate of, the potential well it's climbing out of.

That said, Brian Cox does sometimes joke that astronomers round π to 1; while I wouldn't know about the reality, it's probably safe to infer at least some frustration on his part about the precision of things in this field.

[marcosdumay]

The joke was way more true at the 20th century. There were many really important measurements where we got unprecedentedly precision, enough to say it's X, 100X, or something in between.

Nowadays astronomy got a lot more precise. But there are disagreements on how much confidence to put on that extra precision.

[dylan604]

sometimes, i wonder if astronomy/physics were to only use unsigned numbers, if things would just make more sense. you'd get much more precision, and then you wouldn't have to worry about "but the math says it's possible" issues by taking everything by * -1.

[labster]

Jesse, what are you talking about?

[dylan604]

Well, Mr White, if you look at all of the "weirdness" in physics with theories coming out because "the math says it's possible", then you'd see that it would be much more simple if we were not allowed to have negative numbers because we're only using u64 integers. we'd have more digits for precision as well.

[nickpeterson]

I find that weird outliers are usually evidence that the underlying model is flawed in some fundamental way, but sometimes I think people believe the exceptions are fundamental, and a lot of effort goes to the wrong place.

[gpgn]

These are somewhated dated but well worth watching; the professor does all kinds of approximations.

Frontiers/Controversies in Astrophysics with Charles Bailyn: https://youtu.be/ZiK4lg3Tzfs?t=1628

[hutzlibu]

I also don't know how accurate that description is, but this comic comes to mind:

https://xkcd.com/2205/

But I mean, space is really, really big and we are observing from one single spot with (on cosmological scale probably) primitive technology. So of course most of it is guessing and when you "guess" a lot of things, it maybe does not matter a lot, if you have 3.41 Pi or 1, when the data you have are rough estimates anyway. But sure, when you do sloppy math, when you could have precision - that would be just wrong and unscientific.

[dylan604]

What if you have 3.14Pi?

[dotancohen]

Then you're so close to 10 that you can just use 10.

I actually did that on a physics exam once. Somehow I had a value with pi squared. I just replaced it with 10. I don't remember if I got that question right or wrong - I'm hard pressed to think of a situation where pi squared actually legitimately shows up.

[rmu09]

Surface area of a Torus?

[dotancohen]

Not that I remember, or really relevant to undergrad physics - I'm pretty sure it was physics. Fusion and accretion disks are above my pay grade!

[olddustytrail]

That's pretty much pi squared. Which, for weirdly good reason, is close to the value of G in meters per second squared.

[dylan604]

only on this particular rock

[olddustytrail]

Nope, on all rocks.

[quickthrower2]

I was confused... the symbol π is actually Pi, but it renders as something that looks like a different alphabet entirely on HN.