Hobbes: Do you have an idea for your project yet?

Calvin: No, I'm waiting for inspiration. You can't turn on creativity like a faucet. You have to be in the right mood.

Hobbes: What mood is that?

Calvin: Last-minute panic.

The “point” of gradual school should be taking long walks with friends from completely different departments, talking about abstract ideas that aren’t related to your work at all. Maybe later in the shower you realize some subtle connection to your work that gives you an idea for an innovation, but maybe not. The point of graduate school is play. The switch from “play” to “product” is difficult but necessary. Both phases are important to the budding PhD. You need to demonstrate the ability to think creatively AND the ability to process that creativity into a deliverable. That’s the value of a PhD. If you’re interested in a focused grind to a specified finish line, skip graduate school.

I agree with you and I'm surprised I haven't seen anyone else make this point. Graduate school, and to a lesser extent post doc, is when you're supposed to dilly dally. Of course at the time it doesn't quite feel like that, but you are paid a meagre wage for a reason--to survive, but to not be comfortable. No extra money to go party, travel, etc. The devotion to the craft is supposed to be an almost pious experience. Even though it's simple, it is an incredible luxury most have never and will never get to experience (only a small percentage of the world obtains or even attends grad school for a PhD). With an eye on "getting out", you miss out on part of the beauty. I had many contemporaries through my studies like that. They were always stressed out and pushing to get home at 5pm, unlike us slackers who hung around and discussed tangential things but always with an eye toward connecting the dots. The ones who really excelled post grad school were the ones with vision and drive, not just drive alone, and the former always hung around shooting the shit because they appreciated the value derived from informal, idle chat.

One of my colleagues was in grind mode and switched to a 6 day life week by living 28 hour days so that he could be on certain machines all night when normal humans were sleeping.

Yeah he burned out quickly and left with a masters.

I wish I knew the answer to this. When everything is totally screwed up and on fire and I need to get something done by 5:00 today that would normally take a week and a half, by God, it becomes apparent every millimeter of fat that can be trimmed and I can usually scrape together what's needed by then. Then the deadline is over and I get a similar project the next day and it's back to taking 5 days to finish.

There are real quality differences between the two scenarios, but they aren't nearly as bad as what you'd think they are. The reality is that I can just work a lot more efficiently when stressed. If I could do that all the time I'd be able to work 3 hours a day and remain as productive.

It's not typically sustainable to operate at such high levels of stress, but the stress brings with it a myriad of hormones and behavioural adaptions that, for some people, aid in sustaining focus and effort.

Because in normal time we spend most of our time reading HN rather than working on our urgent stuff.

Whelp that's my signal to get back to work.

It’s selection bias. There are probably far more instances where rushing through causes problems than there are where rushing through doesn’t.

My supervisor would often say: Don't rush, we don't have time.

Funny because he was often the reason we had to do the last minute rush :)