[cyrksoft]

I have a PhD from a top Econ department in the US (I finished it a couple of decades ago, so things might have changed).

Of course a PhD is not the best decision if you are only looking at optimising your path to industry. However, during my PhD I had a fantastic time. I studied what I wanted to, I had a lot of free time to develop any skill I wanted to.

It was also very useful to jump start my career. Instead of having to work for X years until o reach a certain position, I was able to go straight into a high paying job.

I came from a poor and rural part of South America. For me, the PhD was the best and easiest path to getting a top paying job in the US. I'm not saying this is for everyone, but it can be a very nice choice for a career in industry. I even met PhDs from low ranked universities during my career.

[timr]

> Of course a PhD is not the best decision if you are only looking at optimising your path to industry. However, during my PhD I had a fantastic time. I studied what I wanted to, I had a lot of free time to develop any skill I wanted to.

Same. And I would say that the PhD "rapidly" forced me to learn to take on large, ambiguous projects and see them from conception to completion. This was, by far, the most valuable result of the experience. Anecdotally, this seems like a rare skill, even in tech -- though it is not necessarily as recognized as it should be.

I put "rapidly" in quotes up there because if I have one piece of advice to PhD students, it's to get the hell out of grad school as soon as you can. Like the parent, I had fun exploring my interests, and it was a period of intellectual freedom that most people never get. I relish that time, but I also regret the costs. I was idealistic and unwilling to compromise my personal goals, which ended up delaying everything. If I could go back and do it over, I'd be more mercenary about finding an already successful collaborative project, contributing a little bit, and getting out. Grad school is a job -- a particularly abusive, low-reward job -- and the only goal is to finish. You get no trophies for knowing more stuff at the end, and spending time gratifying your curiosity or being a perfectionist might seem appealing ("why would I be putting myself through this if I weren't going to indulge my intellect?"), but it's ultimately a trap.

(Unfortunately, the nature of grad school is that nobody who is pre-disposed to wander aimlessly will listen to my advice, while those who understand what I'm saying probably just think I'm stating the obvious.)

[cyrksoft]

> I put "rapidly" in quotes up there because if I have one piece of advice to PhD students, it's to get the hell out of grad school as soon as you can. Like the parent, I had fun exploring my interests, and it was a period of intellectual freedom that most people never get. I relish that time, but I also regret the costs. I was idealistic and unwilling to compromise my personal goals, which ended up delaying everything. If I could go back and do it over, I'd be more mercenary about finding an already successful collaborative project, contributing a little bit, and getting out. Grad school is a job -- a particularly abusive, low-reward job -- and the only goal is to finish. You get no trophies for knowing more stuff at the end, and spending time gratifying your curiosity or being a perfectionist might seem appealing ("why would I be putting myself through this if I weren't going to indulge my intellect?"), but it's ultimately a trap.

I feel the same in many ways. I went into the PhD without a specific project and it was extremely complex. I was fortunate enough work in a theory heavy field, but had friends with more data-related interests. They had a really hard time finding data. Unless you are in top places in the US, data is almost impossible to get.

I highly recommend project PhDs (if you want to choose the PhD route), which are the norm in Europe/UK. You can choose very good universities in cheap cities and have some great 3/4 years. Travelling in Europe is quite cheap, food is cheap enough.

[timr]

Yeah, anecdotally, all of my European/UK friends with PhDs had a much shorter experience. The UK system, in particular, seems far more sane: 3 years is a pretty common duration, vs the 5-7 that's typical in the US.