Hacker News Reader

Comment by naturalgradient | Hacker News Reader

As a current PhD student in a highly competitive program, I wish I could say this is reasonable advice, but it's wishful thinking in the hyper-competitive environment at top tier universities (where the goal of a PhD student would be to graduate as faculty material).

There is just no way around grinding out the experiments. You literally have to will something to work in multi-month long pushes where you iterate again and again until you figure it out.

Deep work is good and nice for the theoretical parts, but all the academic superstars/upcomers I know work super long AND super hard to make conference deadlines all the time.

Of course, if you just want to graduate without stressing all that much or much ambition, this might work.

View on HN

>You literally have to will something to work in multi-month long pushes where you iterate again and again until you figure it out.

Or instead of iterating 5 times, you take your time and do it right on the 2nd attempt, figuring it out in 1/3 of the time. Hurrying only leads to mistakes. It makes no more sense than attempting to continue working when you really need to go to the bathroom. You are are a human being and have biological limitations. Ignoring them is of no benefit and usually significant detriment.

You can say that but sometimes there is just no way to get it right without trying stuff that doesn't work, improving on that, and throwing the old stuff away. You can't plan ahead and take your time to do things the right way because you don't know what the right way is. Iterating quickly is a way to work through those approaches quickly, and throwing away bad approaches is better than refining the wrong one.

John Carmack:

Focused, hard work is the real key to success. Keep your eyes on the goal, and just keep taking the next step towards completing it. If you aren't sure which way to do something, do it both ways and see which works better.

> Or instead of iterating 5 times, you take your time and do it right on the 2nd attempt

You're assuming the majority of failures are human-error induced.

Academia is infamous for only rewarding positive results, even when the underlying is exploratory and planning couldn't have saved you. "X does not cause Y" is a legitimate outcome of asking "does X cause Y?", but it's (usually) not an acceptable result for a thesis. It's all too possible to end up needing all of those iterations even when you don't make mistakes.

My answer from the trenches is that it just does not work that way. You have to try out if something works. Something can make sense forever in your head until you try it in the real world.

I'd say not testing your ideas and instead spending forever overthinking/planning them is actually a leading cause of procrastination for PhDs.

> Or instead of iterating 5 times, you take your time and do it right on the 2nd attempt, figuring it out in 1/3 of the time. Hurrying only leads to mistakes.

In an abstract sense this is true, but there are some fields were you do just have to do it again, and again, and again. I know an exceptionally smart biochemist who routinely has 16 hour days. There's no substitute.

In my own field, someone who would later go on to win a Nobel for the work, was working on something that simply required a huge amount of time and iteration to get right. This is often the case.

This is not true in all fields, of course. I just want to caution against extrapolating from experience in one field to all PhDs, which is the sense I get from a lot of comments in this thread.

I don't know a single biochemist that puts in 16 hours of work a day. I know of plenty of biochemists that routinely start and end their days 16 hours apart because they have a growth protocol that takes 12 hours to complete prior to harvesting, so they get to the lab, prep, set up the experiment, then leave for a large chunk of the day while their flasks are hard at work, then get in at 8pm to finish the protocol, record results, tidy up and head home for sleep.

The above timeline is common and routine as are claims of having just done a '16 hour day' by those who are on the above schedule.

I'm not saying biochemists couldn't run 16 hours of experimental protocols at their bench per day, but I've never seen it happen.

In undergraduate a classmate of mine spent 3 months doing electrical engineering research on the properties of various chemical doping regimes.

At the end of the summer it came to light that the chemical supplies they received were not the claimed purity invalidating all of their research.

In my experience, if you want your PhD to be enjoyable you probably shouldn't do it at a top-tier university. Top-tier universities are pressure-cookers. On the other hand, the job market favors folks with Ivy League (or Ivy equivalent) degrees...

One of the moments that caused me to nope out of science was when I (as a research assistant) received a work e-mail from a postdoc at a high profile northeastern uni at 11:00 PM on a Saturday night. There's no way in hell that's an appropriate time for work in my book.

Reply on news.ycombinator.com