I left to co-found a start-up - never regretted it as I was "hyper cynical" (a direct quote from a colleague) about the whole academic world and how easy it was to play the "publish or perish" game.

I like building stuff that people use, not writing papers that mostly went unread.

Circa 2010, as I was finishing up my BS in CS, one of my professors asked me to have a meeting in his office. He was trying to recruit me into a graduate program under him, since he liked my performance in his class. I said I wasn't very interested, since I wanted to make things people will actually use. His response was there was a team in Utah that helped make part of the USB spec the decade previous. I was attending the University of Minnesota.

Yeah, no thanks.

Source: I’ve worked on both.

Those aren't the only two options and even they were, it isn't always true.

My guess is that if you want to be paid full-time to work on USB, you'll probably find more (and better-paid) positions in industry.

I had the same feeling!

And as it turns out, many of the skills are transferable: https://twitter.com/mizzao/status/1505529213612609536

From a personal perspective I don't regret my PhD, though it was charity funded so it was topped up to be around £20 000 instead of the usual £14000ish per year. It was an opportunity to engage with a small group of experts in research that was engaging and of use to the wider world, and present it internationally. I also tried to start a start-up during this time, but the underlying idea failed to pass preliminary trials (I work in healthcare). So the two are not mutually exclusive.

People's experience tends to vary. I think a lot depends on how much you research the group and supervisors you will be working with beforehand, and how much you engage them in the research you do.

I was lucky enough to get involved in the web very early (1992) and present papers at the first few web conferences - though that wasn’t my day job.

Edit: I worked on a couple of EU Esprit 2 projects.

I imagine that must have been quite interesting!

What I find curious: I get paid to do lots of things which are clearly complete nonsense, but the funding comes with the requirement that these specific nonsensical things must be done. I could do lots of useful things, but I do not get paid for those.

I'm far enough into the process that I've decided to just collect the stupid credential before I move on. Which is funny because I did not get into this for the credentials, but dropping out at this point would just look weird on my CV.

Developing all sorts of platforms that will theoretically benefit the public / researchers / whoever, but in actuality will never be used by anyone and in fact will be shut down as soon as the corresponding project ends.

Those are the two generic examples I can think of, for everything else I'd have to give too much context.

But yeah, you gotta do what pays the bills. I'm not making triple figures in dollars, but in my country it's six times what I would get during a PhD program.

In the US, PhDs can take 6-8+ years. There are some shitty advisors who delay their students' graduations to get additional low-wage labor. Also, the academic job market is terrible and has been for 30+ years, so it takes a full year just to do the search.

Because the duration is so dependent on the program, and the only end is when the advisor agrees that sufficient work has been done to merit a PhD, it can stretch on without much hope in sight.

The former step is intended to clearly define what is sufficient work. In effect, it is the contract that eliminates some of the risk of the advisor's whim.

That seems unfair. By the time many people get started on their dissertation they have often done years of research and have gotten several papers published. Starting your dissertation basically means that you're 'done' with your PhD and ready to summerize what you've discovered.

So with those advisors, it was normal for people to graduate six months after the prelim (because the department required a minimum of 6 months between the prelim and the defense).

At least one world famous research university has a fix for this problem of "... when the advisor agrees":

The official statement of the university is that the Ph.D. student must submit "an original contribution to knowledge worthy of publication".

So, if a student believes that they have done such work, then they are free to, and maybe should, submit it for publication. When the work is accepted for publication, the student can then so announce to the professor, department school, and university.

> So, if a student believes that they have done such work, then they are free to, and maybe should, submit it for publication. When the work is accepted for publication, the student can then so announce to the professor, department school, and university.

For many fields, this would not work. My advisor required a journal publication to get the Master's degree, and several to get the PhD. He (and many others) would have failed any candidate's defense if they had just one paper if they were in their committee.

(Yes, it's crappy).