(Disclosure: I'm getting a PhD in CS now)

> The hard part of startups is product: knowing what to build, and being able to build it

Funny enough, this is why I think a PhD is undervalued by many. You spend years focused on a single area, sometimes pivoting topics, and getting hands on experience with operating solo on the outskirts of a particular field.

The overall point of the degree is you learn how to learn. The nice part is you have the padded walls of university equipment, funding, and time; So failure isn't terminal, just an opportunity to figure out what didn't work.

There's a few reasons how and why this message got distorted and lost in modern academia (honestly it would be a great subject for a book). It's no mistake that better known programs and programs with ties to a stream of successful startups are also the ones where there's a culture of builders and makers.

In hindsight it was quite early in my career, but a bit over 20 years ago I felt burnt out in tech and wanted to try something new. I was share trading quite a bit so thought maybe I'd become a stock broker (online trading was only just starting to become a thing, so being a broker was still a real job at the time!). I finished the degree and got qualified, and while I loved the learning I discovered I hated the industry and that it wasn't for me.

But all of that discovery work, the financial planning, what their actual objectives were, having to dig into why someone wanted to buy a house, etc. proved to be some of the most invaluable and transferable skills I've learned (that, and SQL ;). A lot of that part of the course was learning how to ask questions, and learning how to follow-up to get past the superficial and often incorrect initial answers.

To your point I feel like there's these kind of meta skills that are transferable to many domains, but are sadly lacking these days from a lot of education that is hyper-focused on very role-specific qualifications. The ones that immediately come to mind:

- Learning how to learn (i.e., how to understand yourself, and improve)

- How to ask good questions (similar to above, but about others)

- And something I need a better label for because I just clumsily call it "sales", but in the Daniel Pink "To Sell is Human" style. Something like being able to structure a coherent and compelling narrative to other people. Clearly I fall short on this one.

> Funny enough, this is why I think a PhD is undervalued by many

no, the opportunity cost is too great for this field, for people that saw the light in just undergrad or wound up building software for other people without an undergrad CS degree