The physics behind rockets requires the structure of the rocket to be extremely light, so structures can’t have more than an absolutely minimal safety factor. And the will experience extremely high vibration loads and g forces. And almost all the mass of the thing is extremely volatile fuel, which needs to be consumed at fantastically high rates and burned directly next to where it’s stored.

Effectively, the margins have to be so small and thus to leave a much higher risk of failure than most other engineering problems, even when done completely correctly.

They are incredibly complicated machines, and you don't have much chance to test them in realistic conditions without just firing them. They tend to fail much more early in their life, so you'll see a lot more explosions with newer rocket designs. (this is this design's first flight) Most launches to orbit are successful, I think by a wide margin.

Worth noting too, that even the rocket going off course usually means it has to be intentionally exploded by an automated system or manually by a range safety officer since that's the safest option.

I guess that's the equivalent of writing code perfectly and expecting no compilation errors or logical errors, with the requirement that it should run perfectly.

They're not.

The vast majority of rockets that simply work don't make the news.

There's a really good reason why caricature of a highly intelligent individual is a rocket scientist. Engineering rockets that go where you want them to go safely is still a really hard problem!

A lot of these companies are either trying different things or doing the same thing and no one is sharing their secret sauce.

They are not insurmountable challenges but each team building rockets from scratch have to go through them. I think it's vital to the long-term success anyway. Easier to learn from your own mistakes than someone elses.

They're fresh designs, just like how if 10 people reimplement the same 100 page specification, they'll have a stage of fixing crashes unique to their implementation, fresh rocket designs need a few rounds of iteration to sort out all the kinks, and after a point you're just better off doing a test flight, especially as a private company. You just don't really hear about similar issues for things like cars because you don't get large fireballs and air closures indicative of testing.

Once you get the issues sorted out though, they can be almost airplane like in terms of operations. See SpaceX's crazy launch cadence and reliability streak with Falcon 9 for an example.

Rockets essentially mostly consist of things that are intended to go boom with very little prodding. On top of that, the force on a rocket engine makes it prefer to make the rocket engine go through the rocket rather than pushing the entire rocket itself, and it requires careful engineering to make it not do that. Combined, this means that if anything goes slightly wrong, it tends to cascade into going horribly wrong and cause the rocket to explode.

In this case, though, it seems that the guidance system or something similar was going wrong, and the rocket was intentionally blown up in a safer location than an uncontrolled trajectory might have taken it.

It's the flight termination system. i.e. it's an intentional detonation of the rocket to limit risk.

The FTS triggers for unmanned crafts either manually from mission control or automatically based on some set of onboard checks and safeguards.

Given it detonated so quickly after launch, some piece of equipment stopped responding or a sensor started reading a dangerous value which then triggered the FTS.

This really isn't an unusual thing since this is the first launch of this rocket design (and if I remember correctly the first launch by this org).

But yeah TLDR it's better that a rocket explodes in a predictable manner that gets rid of all the fuel and breaks everything up into reasonably small parts while they are still in control of the rocket than for things to go wrong and it crash or otherwise cause more damage.

I think it's mostly new rockets. KAIROS was a new rocket and this was its first launch.

Old rockets that haven't been changed in decades like the Soyuz stopped exploding long ago.

As another example, Toyota Tacoma trucks are known to be reliable... because they have barely changed for years but now are kind of dated. Ford, on the other hand, keeps changing the F-150 so you get a lot of modern comforts, but there "bad model years" and "good model years."

New stuff just doesn't have the kinks worked out, especially when you have budgets and deadlines. Just how engineering goes.

Which ones? The ones that do are almost all prototypes (for instance SpaceX Starship) that have new technology innovations. Unless they are manned the safest thing to do if they are operating outside limits is to use the FTS (flight termination system) to make a large fast object into much smaller pieces. Unlikely most other vehicles, rockets can't just coast or glide to a stop if they go wrong.

Because they are essentially bombs

are rockets always blowing or is the occurrence of launches/new platforms increasing? I think it's the latter and we should expect more launches/new platforms and explosions over the next decade in comparison to the last decade.

Because we're not making rockets with inches thick steel hulls and fission cold gas engines.

I think cars will stop crashing sooner than rockets stop blowing up.

Design of those rockets is based on design of missiles, where time to hit is the most important thing. Fully electric space launcher or space lift will be much slower, but safer.

Yeah, it's not like it is rocket science. Oh wait, it is. The art of making big containers full of explosives as fast as possible. It rather is a technical miracle that most launches do not produce a big fireball.

Also it is not open source, but rather mostly secret technology. One company or country solving it, does not mean everyone will benefit. Everybody new in the game, has to invent quite some wheels again.

Hardware != Software