[conformist]

I like quaternions as much as the next guy (I’ve used them in numerical computations etc), but what is it about them that makes them show up on the front page every few weeks?

[ajkjk]

Lots of engineers are lapsed physicists who have hobbyist interest in this stuff.

Also lots of HN readers are actual physicists or mathematicians. It's not all techies.

Also, lots of engineers have at some point learned some computer graphics and so been exposed to quaternions in that setting. Since they're mysterious and hard to wrap your head around most people don't really 'get it', leaving a sort of standing curiosity that articles like this tap into.

[bee_rider]

I wonder if it is also some engineers who are nostalgic for the feeling of hard things becoming easier that complex numbers provided.

[jesuslop]

Baez wrote some ideas in [1], one I'm liking connects Lorentz group in dimensions 3,4,6 and 10 with the modular group SL(2,Z) that is at a crossroads of several hardcore math themes. For Lie algebras:

sl(2, R) ≅ so(2,1)

sl(2, C) ≅ so(3,1)

sl(2, H) ≅ so(5,1)

sl(2, O) ≅ so(9,1)

Dirac equation is the C case, the other cases have their uses.

[1] https://arxiv.org/abs/math/0105155

[AlecBG]

You can describe spinning particles in flat space and AdS space really nicely with twistors using this

[oceansky]

The word "quaternion" just rolls of the tongue. I always upvote it.

[ajross]

Because in 1985 Ken Shoemake dropped the idea like a bomb on the computer graphics industry and it changed the way hackers thought about rotations forever. https://www.ljll.fr/~frey/papers/scientific%20visualisation/...

I mean, there are practical reasons too (which are mostly just isomorphic to the stuff in the paper). But really that's why. It's part of our cultural history in ways that more esoteric math isn't.

[phkahler]

Because people like me use quaternions but have never attained a full understanding like 3x3 rotation matricies. I will be reading the above link since its only 12 pages and someone indicated its an easier read.

[aardvarkdriver]

Thank you for the reference, this is one of the more approachable sources for learning the nuts and bolts of quaternions that I've seen!

[westurner]

That quaternions also solve for what we normally have 3D+time for.

And Lewis Carroll (Oxford (Math)), preferred Euclidean geometry over quaternions, for "Alice's Adventures in Wonderland" (1865).

Quaternions:

  q = a + bi + cj + dk
  -1 = i^2 = j^2 = k^2

Summarized by a model:

> In a quaternion, if you lose the scalar (a) — the "real" or "time" component — you are left with only the three imaginary components (i, j, k) rotating endlessly in a circle.

(An exercise for learning about Lorentzian mechanics, then undefined: Rotate a cube about a point other than its origin. Then, rotate the camera about the origin.)

4D Quaternions (a + bi + cj + dk) are more efficient for computers than 3D+t Euclideans. Quaternions do not have the Gimbal Lock problem that Euclidean vectors have. Quaternions interpolate more smoothly and efficiently, which is valuable for interpolating between keyframes in a physical simulation.

Why are rotations and a scalar a better fit?

Quaternions were published by William Rowan Hamilton (Trinity,) in 1843, in application to classical mechanics and Lagrangian mechanics.

Maxwell's (1861,1862) original ~20 equations are also quaternionic; things are related with complex rotations in EM field theory too. Oliver Heaviside then "simplified" those quaternionic expressions into accessible vectors.

Is there Gimbal Lock in the Heaviside-Hertz vector field reinterpretation of Maxwell's quaternionic EM field theory? Maxwell's has U(1) gauge symmetry.

And then quantum has complex vectors and some unitarity, too

History of quaternions: https://en.wikipedia.org/wiki/History_of_quaternions

[westurner]

Shouldn't there be symmetry and unitarity given energy conservation? And quaternions express this with rotations in SO(3), but is there a better model than quaternions for EM field theory since 1861?

  -1 = i^2 = j^2 = k^2
  q = a + bi + cj + dk
  q = a + xi + yj + zk

(Edit) Quaternions and spatial rotation: https://en.wikipedia.org/wiki/Quaternions_and_spatial_rotati...

QED: Quantum electrodynamics: https://en.wikipedia.org/wiki/Quantum_electrodynamics :

> Mathematically, QED is an abelian gauge theory with the symmetry group U(1), defined on Minkowski space (flat spacetime). The gauge field, which mediates the interaction between the charged spin-1/2 fields, is the electromagnetic field. The QED Lagrangian for a spin-1/2 field interacting with the electromagnetic field in natural units gives rise to the action [...]

And QED is the basis for the Standard Model of particle physics and for some theories of n-body quantum gravity.

[koolala]

Hamilton came up with the term vector for the 3 component part of the quaternion.

[K0balt]

They sound shiny and mysterious?

[chpatrick]

I know you know, just practical intuition for 3D graphics in case someone finds it useful:

There's a 1-1 mapping between complex numbers and 2D rotation matrices that only do rotation and scaling. The benefit is that the complex number only has two coefficients, not four like the matrix. Multiplying these complex numbers is the same as multiplying the equivalent matrices. Quaternions are the same idea just in 3 dimensions (so with 3 imaginary units i j k, not just i, one per plane).

[Scene_Cast2]

Then why not geometric / Clifford algebra?

[stared]

In short:

scaling -> real numbers

1d rotations and scaling -> complex numbers

2d rotations and scaling -> quaternions

In the case of quaternions, there is called double-covering, which turns out (rather than being an artefact), play fundamental role in particle physics.

[conformist]

Yeah sure SU(2) up to sign is isomorphic to SO(3) and whatnot… I think it’s probably mostly the computer graphics history and the cool name that gets people excited about quaternions?

[the_fall]

Honestly, with all my love for the HN community, I think we have a couple of topics that just get upvoted without reading because they signal that you're in the ingroup. Few years back, another reliably upvoted thing was anything with "Bayesian" in the name. In the past couple of years, "busy beavers" would also get upvotes even though they have no practical use, their mathematical significance is dubious, and few people understand them in the first place.

[socalgal2]

That people haven't gotten used to geometry algebra yet.