Code doesn't exist in a vacuum. It's only by context that code is provided with value and meaning. If you're the only person working on the code and you know exactly what you want to achieve then just wanting to code might work. In almost any other case you have to communicate about what the code is actually supposed to do because otherwise you're likely to create wasteful instead of useful code.
Being lower in the hierarchy but making decisions whether a meeting is worth your time pushes a ton of risk onto you if there is a mishap.
How do you respond to "You would have known if you stayed at the meeting"? Would the CEO have your back? How many times before you start getting a "problem employee" reputation.
I think you'd quickly realize to just stay at meetings but now you have no grounds to bring up criticism since it is assumed you could just walk out.
If a meeting is structured such that an attendee has enough time to realize it's pointless and leave before the meeting gets around to the important part, then it's a bad meeting. That meeting sounds like it should have been an email, or a message in an IM channel, or even more than one smaller meetings.
A manager who structures meetings so that people have to sit through pointless drivel in order to receive one or two important nuggets of information, is a bad manager who's scheduling exactly the kind of bad meetings that Musk is talking about.
Any guesses what the corresponding point I was trying to make was?
Productivity is the greatest source of waste. Coding the wrong thing is worse than being in a long meeting which might have led you to code the right thing. And by write/wrong I mean for the user.
Coding is the easy part of software creation. You can teach anyone any technical skill in a fairly short amount of time. Teaching someone systems thinking, and UI design, and just plain good taste is much harder.
The most efficient path to doing the right thing for the user is just to do something, get feedback, and iterate quickly.
Yes, sometimes the most efficient way to feedback is to not write code at all. Maybe it is a design document, maybe it's a prototype or mockups.
Sometimes it's creating software that the user can actually control and understand and solve their own problems vs waiting for a developer to create that one feature they need.
But that would require a little bit of thinking outside of code... Those boring meetings are a chance for your mind to wander and play around. The outcome of the actual meeting is less important.
But yes, most meetings are terrible, but you can't always know that ahead of time. Elon Musk is a perfect example of someone jumping head first without knowing to fill the pool first. Some of their production problems have been solved by other manufacturers a hundred years ago.