At 13:13 into the video, he mumbles:

"W-what I'm trying... I have a solve this... err... implement rigorously is this sort of 5 step process:

1. Your requirements are definitely dumb: everyone's wrong, no matter who they are [...] So try to make your requirements less dumb."

2. Then, try to delete the part or process.

The bias tends to be very strongly in favor of adding parts in case we need them... but you can basically make "in case" arguments for so many things! [...]

So, you gotta delete the parts or process steps that you don't need.

3. Whatever requirement or constraint you have, it must come with a name, not a department. Because you can't ask the departments, you have to ask a person

[...] otherwise you could have a requirement that basically an intern 2 years ago randomly came up with off the cuff... and they're not even at the company any more... but it came from the, let's say, Air Flow Department!

4 Only the third step is to simplify or optimize [he seems to be off-by-1].

Possibly the most common error of a smart engineer is to optimize the thing that should not exist. Everyone has been trained in school that you gotta answer the question -
convergent logic. So you can't tell a professor "your question is dumb" - you'll get a bad grade!

So everyone basically - without knowing it - they got a mental straight-jacket on: they'll work on optimizing the thing that should simply not exist!

4b. Accelerate cycle time: you're moving too slowly. Go faster!

But don't go faster until you worked on the other three things first!

If you're digging in your grave, don't dig it faster. Stop digging your grave!

So... but... it's... you can always make me ...$?#@... faster!

[he literally said that last sentence too fast for it to be discernible! 😂]

5. And then the final step is Automate.

Now, I have personally made the mistake of going backwards on all 5 steps [...] Yes, multiple times, on Model 3.

That is: I automated, accelerated, simplified, and then deleted. [but did he forget to question the requirements after deleting the part?]

This starts at 22:00 into the interview, and it's worth listening to even with the incessant beeping of the construction cranes in the background.

@codewiz His five steps were great to listen to.

What makes it so compelling is that he literally stands in front of a giant rocket that is being built according to those rules while we watch him talk.

@tsturm Yes! Talk is cheap, rockets are hard (or so we thought, until Elon made them so mundane).

@tsturm @codewiz We thought they were hard because the space agencies and the defense contractors they were providing welfare to had an interest in making it look hard so that people wouldn’t realize how slowly they were moving and how they were mostly flushing even their relatively meager budgets down the toilet.

@codewiz @tsturm I suspect the reason NASA killed DC-X is because that program had accomplished so much with so little money, making everyone else look bad. Al Gore put things right again by pushing for a complicated expensive design that would take decades to fail.


@freakazoid @tsturm I have to admit: I had never heard of DC-X before.

· · Web · 1 · 0 · 0

@codewiz @tsturm Even though what it accomplished was pretty amazing, and space nerds recognized it as such, it wasn't a big flashy expensive program, so it didn't get much media coverage. People want Star Trek, so STS and then Gore's pet National Aerospace Plane got most of the public's attention.

@codewiz @tsturm The problem is that you will never get Star Trek by trying to build the USS Enterprise directly. What you'll get instead is a failed program and people saying "space is too hard!"

Sign in to participate in the conversation

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!