"Good news all around. How do you start learning these things?"

"Ooh, that's – hm, so, for the most part…" She pauses. "It varies a lot between who you get instruction from. Some ways are utterly ridiculous."

"I'll take your word for it. There's only so many ways to start on engineering, they all involve numbers."

"Numbers how? Or – I guess it's probably complicated, I've never actually looked into it."

"Almost every engineering task involves doing a fair bit of math. How big do the gas cells on my ship need to be? Buoyancy equation. Math. How thick does this piece of wood need to be? Stress calculations. Math. How do I set up this chemical process to make soap or whatever? Molar balance, math. Even if that one might just look like multiplying a recipe."

"… I mean the last one doesn't sound that bad. The others sound reasonably straightforward if you already have the equations derived."

"Oh, they're not, really. A truss calculation can take three, four pages of paper. Because everything is balanced against everything else, sort of? Let me think of an example... The wood beam needs to be thicker? Okay, make it so. Except now it weighs more and that other wood beam over there needs to be thicker. And now the gas cells need to be bigger because of the extra weight so all the wood beams need to be longer, and more heavier, which means more wood again. That's just tedium, though. I was being a bit sarcastic. What you really need to think about is design philosophy - what does the thing you're trying to design need to do? What do you have at your disposal to accomplish that? Materials, tools, space and lift. Maybe the best way to start it is to actually design a thing from scratch and then figure out all the ways in which your design is terrible."

"… Okay, I see your point, now. Equations leading into equations meaning you have to adjust a ton of things to get it to meet up neatly."

"Yep, that's about the size of it. There's ways to make it simpler, but... Yeah."

"I suppose it's probably worse if you plan on using the finished result in a way that makes it a life-or-death situation." Pause. "I mean, that sounds kinda obvious, but still."

"We call that 'fail-safe thinking'. If it fails, make it fail in a safe way."

"Sounds hard to achieve with some components… Backups, too, I assume?"

"Yes. And you have to worry about material properties, and repeated stress causing failures, and this and that and the other thing. You know, if my engines caught fire, they would burn through the things keeping them attached to the ship first, and hopefully fall off without igniting the rest of it."

"… And then the ship would still stay aloft because of the weird atmosphere?"

"We'd go higher, actually. The majority of our lift comes from the gas bags, not aerodynamic effects."