eppo

the sky is high, the emperor is far away

[eppo @EppoRyan]

@anduriltech It’s only ‘hypersonic’ if it comes from the air-breathing region of NASA/AFRL. Otherwise it’s just ‘sparkling rocket’

[eppo @EppoRyan]

@Dylan_Morri Phones got way smaller until settling on roughly iPhone-sized.

The argument for modular/factory produced (eg SMRs) vs bespoke is that learning rate from mass quantity >> lost economies of scale (eg bigger turbines are cheaper/watt).

Moore’s law is also bc of learning rate

[eppo @EppoRyan]

@DenverRayburn @reindsummit Just ordered ten! Can we get some kids and infant sizes next?

[eppo @EppoRyan]

@1llegalEngineer @MehdiHacks Including lens?

[eppo @EppoRyan]

@mattparlmer Turns out saying “let’s take money from the top x% (where x <50%) and give it to ourselves” is a pretty bulletproof way to form a majority.

As x goes down (eg at 1%), your odds of winning go up

[eppo @EppoRyan]

@NikoMcCarty @mkoeris Why don’t they just strap a small neodymium magnet to the test pigeons and an equivalent size/weight nonferrous chunk to the control pigeons?

Seems like they’re overcomplicating it and introducing confounding variables by messing w/pigeons’ immune systems

[eppo @EppoRyan]

This isn’t just an SF thing. Less than half of the US now are net taxpayers—we’ve crossed a critical point where the majority now has incentive to increase rather than reduce govt spending.

“A democracy can only exist until the voters discover that they can vote themselves largesse from the public treasury.”

https://t.co/8vtkXb9qO8

[eppo @EppoRyan]

@jared_western Hell yeah, congrats!

[eppo @EppoRyan]

@CollinCornwell EPGY or bust

[eppo @EppoRyan]

@Longodelian @scottastevenson This sounds like Accelerando

[eppo @EppoRyan]

@atelicinvest @larpcapitalwc In 1960, ~85% of US homes had indoor plumbing and >90% had TVs. Also, OWNING land and a house that doesn’t have the same level of tech as 2026 is subjectively preferable to renting an apartment forever

Another interesting fact—avg airliner speeds are actually slower now vs 1955

[eppo @EppoRyan]

@atelicinvest @larpcapitalwc Buick made ~5x more of those cars than Chevy made Bolts.

In fact, the entire point of midcentury American design was that it was mass produced and available to everyone

Some things were simply more beautiful in the past—when people cared and were optimistic

[eppo @EppoRyan]

@JesseTayRiver https://t.co/ph0kjMHqHm

[eppo @EppoRyan]

@83dollaroring they don’t want you to know this but once you have a wife you’re allowed to talk about JDAM and synthetic aperture radar as much as you want

[eppo @EppoRyan]

@connorkapoor

[eppo @EppoRyan]

@AI_in_LEO @HalenMattison @PhilipJohnston To be clear, even if they own <5% of the eventual ODC share, Starcloud is still a multi billion dollar co. The pie is going to be so massive that even the scraps SpaceX doesn’t eat will be worth 10s of billions

[eppo @EppoRyan]

@AI_in_LEO @HalenMattison @PhilipJohnston Okay, so now the critical path includes not only “be better at building satellites than SpaceX” but also “be better at building launch vehicles than SpaceX”

I see a world where Starcloud is like Rocketlab bc people want redundancy, but my bet is SpaceX owns >>50% of orbital DCs

[eppo @EppoRyan]

@AI_in_LEO @HalenMattison @PhilipJohnston Nobody has asked the hardest question which is how you compete with SpaceX, who is not only by far the best in the world at building satellites but also has access to launch at cost vs at a 3x markup

[eppo @EppoRyan]

@PhilipJohnston @AI_in_LEO @HalenMattison When Spacex builds compute sats and launches them at cost, wouldn’t yours have to be ~3x cheaper than theirs to match their compute costs?

(66% margin for F9 seems reasonable based on public data; Starship could likely be much higher in the medium term bc fewer competitors)

[eppo @EppoRyan]

@bryancsk @tautologer Surely anything that doesn’t require breaking physical laws is in the cards though?

Eg there’s no apparent reason why, with sufficiently advanced technology, we couldn’t simply repair DNA molecule by molecule

Solving FTL seems meaningfully different than solving cancer