Yeah, if by "going over", you mean accelerating in the horizontal direction, then you're right.

Just to illustrate this: Consider we want to put 1 kg of mass into orbit.

First, we have to raise it by 100 km. That requires 1e6 J = 1 MJ of energy (formula is m*g*h).

Then, we have to accelerate it sideways, to a speed of 8 km/s. The energy to do that is 32 MJ (formula is ½*m*v²).

So, most of the energy (97%) is actually in the sideways movement.

I suspect it's because the hydrogen molecules are bigger than a single helium atom, which doesn't form molecules (since it's a noble gas).

So the hydrogen molecule only seeps through if it's oriented right (since the hydrogen molecule is a stick-shaped molecule).

isn't that Hitachi?

The Top Gear Reliant Robin launch reached 3000ft / 900m, although they were unable to stick the landing.

Also with starlink even one company's constellation is causing issues with astronomers and launches.

How bad will it be if there are 5-6 different companies with their own network floating around up there. And then other countries with their own network.

Reusable rockets, in particular.

Imagine having a reusable car in a world where they were all disposable.

My buddy’s 2 million mile ‘95 civic says this is a sure bet

Size is only a proof of logistics. Not tech. Physics don't change fundamentally between 6 meters and 120 meters. You learn a lot from scale modeling without the added costs. Starship's real challenge is actually the logistics necessary to fulfill the desired specifications and experimenting with engineering to reach the scale. The most innovative aspect of Starship would be orbital refueling, and they aren't there since the thing hasn't reached orbit yet. SpaceX problem right now is insisting on high turnover engineering, which doesn't work at scale without heavy costs, because it is a logistic problem, not a engineering problem.

Honda giving a whole new meaning to crotch rocket.
Oh wait, it's an actual rocket!

The satellite constellation is the natural consequence of cheaper rockets. It's a true paradigm shift, but the pioneer in this case has only the moat of being able to spend less money per launch. If someone else can deliver payloads to low earth orbit for less than $2,000/kg, then they'll easily be able to launch a Starlink competitor.

Physics don't change fundamentally between 6 meters and 120 meters

Yes it does. Mass to strength ratio of structural components changes with scale. So does the thrust to mass ratio of a rocket and its fuel. So does heat dissipation (affected by ratio of surface area to mass).

And I don't know shit about fluid dynamics, but I'm skeptical that things scale cleanly, either.

Scaling upward will encounter challenges not apparent at small sizes. That goes for everything from engineering bridges to buildings to cars to boats to aircraft to spacecraft.

You mean like Starship?

My personal opinion is that it fails because SpaceX, like a lot of space startups, embrace a silicon valley coding mindset of 'move fast and break things', which results on them spending much more of their time and effort on testing than on design. Make a change, test, make a subsequent change, test. It gets them to a working prototype more quickly than legacy space/ defense companies. However, there's no emphasis on modeling or design, which is problematic for solving complex problems that haven't been solved for 50 years already.

Yeah, it's a bad situation. I'm against monopolies, but I also see how filling the sky with redundant satellites is a terrible plan, so I don't like the idea of lots of competition either.

I think low orbit satellite communications is a pretty awesome concept. It has the potential to become like a second Internet backbone, but a backbone that can bring data directly to users without the additional router hops that local ISPs introduce. On paper, it's amazingly efficient and can distribute service to all of the world... But in practice the business and management side is deeply problematic. One company should absolutely not be in charge of global Internet service. And one country would not be any better.

The only solution I can see is to make it safe and feasible to have way more satellites operating in low earth orbit. I'm really not sure what that solution might look like...

Here's an off-the-cuff idea though: One solution could be an extremely robust low earth orbit maintenance and "pruning" system. All satellites would need to be monitored by third parties. And those third parties would need the authority and ability to quickly deorbit (prune) any satellite that deviates from its exact expected orbit. If satellites can ensure no deviation from their path and can safely maneuver to avoid collisions, it could be possible for many more satellites to safely share an orbital altitude.

Deorbiting is all well and good, but more and more we're finding that these satellites contain chemicals that are very disruptive to the ozone layer. It's going to be CFCs all over again, but with even more corporate capture of government.

Also a very capable downhill bike that was using a gearbox well before it got popular

You have it the wrong way round. SpaceX's accomplishments are impressive despite Elon.

Honda is cool they build robotics and the best motorbikes in the world per cost

That's a fair point. The alternative is taking things up to a "graveyard orbit" somewhere between LEO and GSO, to a particularly unpopular altitude, where nobody's fighting for real estate. Satellites can sit there indefinitely, you could even clump them up in a big ball, the tiny pull of gravity they have is actually enough to keep them bunched together.

The only problem with that plan is that it takes a lot of energy to raise an orbit that much, I'm not sure how to make that feasible.

Again, the only possible player that could do that any time soon is blue origin/Amazon.

Stoke Space is working on a fully reusable rocket though, I'm really impressed with their rocket concept, some very smart design choices were made. They do have working hardware and have demonstrated their core engine. But I have no idea how close they are to first launch tests, I expect it will be a while