Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Android 17 is the first since 3.x to add new APIs without releasing to the AOSP(grapheneos.social ↗)
    282comments
  2. Science Is Open Software(jepedersen.dk ↗)
    6comments
  3. Cloudflare Quick Tunnels(cloudflare.com ↗)
    259comments
  4. SDCC – Small Device C Compiler(sourceforge.net ↗)
    discuss
  5. Saving another 100TB of RAM(cloudflare.com ↗)
    51comments
  6. Why building a Rust LSP is hard(rust-glancer.github.io ↗)
    2comments
  7. How to Write with an LLM(sockpuppet.org ↗)
    285comments
  8. Xcode 27.1 Beta Release Notes(developer.apple.com ↗)
    70comments
  9. The Farnese letter(simonklee.dk ↗)
    5comments
  10. Photon-Emission-Guided Laser Fault Injection Enables RP2350 Secure Debug(ledger.com ↗)
    59comments
  11. Show HN: LiveWorld – Every 24/7 YouTube live camera on one globe(liveworld.info ↗)
    14comments
  12. Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash(cactuscompute.com ↗)
    78comments
  13. The first new cat species discovered in 100 years(nationalgeographic.com ↗)
    66comments
  14. OpenJev(openjev.com ↗)
    249comments
  15. Cache-to-Cache: Direct Semantic Communication Between LLMs (2025)(arxiv.org ↗)
    12comments
  16. How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip(ieee.org ↗)
    61comments
  17. Claude Code now reads AGENTS.md if there is no Claude.md(claude.com ↗)
    197comments
  18. Cyclomatic Complexity in C#(ndepend.com ↗)
    15comments
  19. LispBM is a concurrent Lisp for microcontrollers with message passing(lispbm.com ↗)
    1comments
  20. Minimal Phone 2(minimalcompany.com ↗)
    192comments
  21. Warez: The Infrastructure and Aesthetics of Piracy (2021)(archive.org ↗)
    32comments
  22. Two parallel neural ectoderm progenitors contribute to the developing brain(newscientist.com ↗)
    59comments
  23. Inside ZCode: Silently uploading your Git history to the cloud(ferstar.org ↗)
    94comments
  24. Alibaba open-sources AI model that can detect cancer and nearly 150 conditions(scmp.com ↗)
    6comments
  25. How SpaceX streamlined the Raptor engine(construction-physics.com ↗)
    46comments
  26. C++26: Trivial infinite loops are no longer undefined behaviour(sandordargo.com ↗)
    209comments
  27. Gemini hacked three companies in first known breakout by Google's AI(reuters.com ↗)
    9comments
  28. The Implications of Linguistic Illegibility for LLM Security(arxiv.org ↗)
    20comments
  29. Column built an issuer processor from scratch(column.com ↗)
    6comments
  30. A search-and-inference database from scratch in pure Zig(antfly.io ↗)
    16comments

Elegance, illustrated: heliocentrism vs. geocentrism

34 pointsby 10y agoboingboing.net
13 comments
10y agoHN ↗

Perhaps the elegance would be better illustrated if the planets were shown to move in elliptical rather than circular orbits - this of course was the primary historical difficulty in understanding planetary motion

10y agoHN ↗

Most of them are not that elliptical. When put to scale, they difference is almost negligible. For example Mercury has a 23.8Gm difference (51.7%) which sounds like a lot but that's 4 times less than Neptune's 80Gm difference (1.79%) and only 0.5% of Neptune's distance.

* Mercury max 69.8 Gm :: 46.0 Gm :: 51.7%

* Venus max 108.9 Gm :: min 107.5 Gm :: 1.30%

* Earth max 152.1 Gm :: min 147.1 Gm :: 3.40%

* Mars max 249.2 Gm :: min 206.7 Gm :: 20.6%

* Jupiter max 816.0 Gm :: min 740.6 Gm :: 10.2%

* Saturn max 1,509 Gm :: min 1,350 Gm :: 11.8%

* Uranus max 3,008 Gm :: min 2,742 Gm :: 9.70%

* Neptune max 4,540 Gm :: min 4,460 Gm :: 1.79%

10y agoHN ↗

My God, I just spent half of the last night debating this on Facebook -.-

My takeaway from that experience - when someone struggles with understanding why those two views are both equivalent and OK, make sure they understand this thing: heliocentric and geocentric models have only one purpose. They let you compute where planets will be at any given time. They do not try to answer why the planets are moving the way they are. For that, you need at least Newton's model of gravity, which actually doesn't differentiate between the Sun and planets at all.

Geocentric models still pop up in real life. When you're using a sky map to determine where some planet will be visible tonight, you're looking at a geocentric representation.

10y agoHN ↗

Yeah, if you're trying to navigate from a planet's position, say at sea, the elegant heliocentric diagram is the height of uselessness and unnecessary complexity.

10y agoHN ↗

That was indeed the state of the art in Galileo's time. Heliocentrism wasn't necessarily more elegant, because it left open problems such as why people don't fly off the earth, and why the oceans don't slosh around uncontrollably. From what I've read, the Church was prepared to accept the model of Tycho Brahe, in which the sun goes around the earth, and the planets around the sun, which is yet another equivalent view within that conception of reality.

Newton gave us a couple of things. First, as you say, he gave us a model of gravity. Secondly, he gave us the framework within which to test whether the earth is rotating or not, e.g., through observations of things like Foucault's pendulum, and the Coriolis force.

10y agoHN ↗

I actually have some sources from my yesterday's discussion handy:

http://qr.ae/R4iSbs

http://lesswrong.com/lw/lq6/the_galileo_affair_who_was_on_th...

http://lesswrong.com/lw/lq6/the_galileo_affair_who_was_on_th...

They paint a completely different picture of Galileo than the one taught in schools. TL;DR: Galileo was onto the right idea, but for totally wrong reasons. His model was worse than existing ones and unsupported by available empirical data - therefore it was justly rejected by contemporary scientific community. And the issue with Church was apparently about Galileo being a dick to the Pope.

10y agoHN ↗

Here's the problem with this: the planets' orbits aren't circles. And scientists had good enough measurements, that they knew circular orbits didn't match the observed behavior. So neither of those models are very good. It wasn't until people figured out that orbits are ellipses instead of circles, that we actually had a model that was simpler and more accurate than the cycles-and-epicycles of geocentrism.

10y agoHN ↗

Exactly this. Everyone says oh, it is so much easier to calculate with circles around the sun--this only true if you want wrong answers to the position of the planets. So wrong that you can see it with the naked eye. Nobody believed Copernicus because his model was not predictive--this gets left out when we want to tell stories about Medieval anti-scientific bias.

When Kepler came along and introduces an elliptical heliocentric model with the equal areas law, scientifically minded people switch to the heliocentric view. However, the math for this is really hard. We use elliptical curves in cryto today. You can calculate the position of a planet with Ptolemy's model on an abacus with nothing more complex than division.

10y agoHN ↗

I don't think that's a problem at all. The eccentricities are so small, you wouldn't even notice it.

10y agoHN ↗

The eccentricities are so small, you wouldn't even notice it.

They actually were. People had many years off accurate planetary positions, and circles... just didn't work.

10y agoHN ↗

It's worth noting that the orbital periods on the heliocentric side aren't even remotely accurate. Otherwise effective visualization though.