Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Don't Become a Scientist (1999)

wuphys.wustl.edu

Post reply on HN
Print view View original

31–40 of 167 posts

Re: Don't Become a Scientist (1999)

#31

Post by gcp »

UK-AL wrote: post #20
marcosdumay wrote: post #19

Earlier quoted context omitted.

It's a completely different situation. Programming leads to some serious concentrated rewards, while science leads only to dispersed rewards. As a result, there are plenty of people wanting to invest in programming, but mostly governments want to touch sciences. That of course does not mean that the job market for programmers will always be good, only that when it stops being, it will be for very different reasons.

I'm guessing massive oversupply will be the reason, as it is here.

Demand for faculty positions is pretty stable. Demand for programmers much less so.

Re: Don't Become a Scientist (1999)

#32

Post by eshvk »

marcosdumay wrote: post #19
msvan wrote: post #6

I wonder if/when this will happen to programmers. I keep thinking that the job market for programmers will be radically different in ten years, what with the enormous amount of people "learning to code" these days. He did however suggest people to become programmers in 1999, and if you followed his advice then, you would probably still be quite happy with that decision today.

It's a completely different situation. Programming leads to some serious concentrated rewards, while science leads only to dispersed rewards. As a result, there are plenty of people wanting to invest in programming, but mostly governments want to touch sciences. That of course does not mean that the job market for programmers will always be good, only that when it stops being, it will be for very different reasons.

> Programming leads to some serious concentrated rewards, while science leads only to dispersed rewards.

Can you explain what you mean by this?

Re: Don't Become a Scientist (1999)

#34

Post by UK-AL »

eshvk wrote: post #32
marcosdumay wrote: post #19

Earlier quoted context omitted.

It's a completely different situation. Programming leads to some serious concentrated rewards, while science leads only to dispersed rewards. As a result, there are plenty of people wanting to invest in programming, but mostly governments want to touch sciences. That of course does not mean that the job market for programmers will always be good, only that when it stops being, it will be for very different reasons.

> Programming leads to some serious concentrated rewards, while science leads only to dispersed rewards. Can you explain what you mean by this?

He means science is a public good, everybody wins when a scientist discovers some knowledge/cure/whatever.

Re: Don't Become a Scientist (1999)

#35

Post by dnautics »

ColinWright wrote: post #22

Extensive discussion from 6 years ago: https://news.ycombinator.com/item?id=122106 Scott Aaronson's reply: http://www.scottaaronson.com/blog/?p=312 That was submitted[0] six years ago, but got no discussion at all. [0] https://news.ycombinator.com/item?id=123584 There have been many, many more occasions when this was submitted,

Scott's quip "many PhD programs even see training students for industry as a fundamental part of their mission." Has long been true for chemistry and biology, and that has possibly staved off the phenomenon for a bit longer than it was for physics (original article, 1999), but it is certainly now true in these fields too.

Of course, these are all anecdotes, by which I mean data strongly filtered through confirmation bias. So it's worth thinking about the bias here; we have tons of grad students and postdocs comiserating, and a select few professors. One of which says it's miserable and is trying to warn people in the future, and the other who has had a fabulously successful career saying everything's more or less fine.

Re: Don't Become a Scientist (1999)

#36

Post by eshvk »

yaketysax wrote: post #8

"Leave graduate school to people from India and China, for whom the prospects at home are even worse." Even worse in terms of what?

I presume that he means that economic conditions are so bad in India and China that people would go through poverty in American Graduate schools instead of staying back home.

I think there is some truth and some hyperbole to that presumption. Especially in the past five years when both nations have seen strong economic growth. Not all people who come from India are the same. Not all American graduate schools are the same. I did a part of my high school and my undergrad in India and came to America as a part of a EECS PhD program at UT, Austin. All of the people who got accepted in my program from both India and China were either academically incredibly strong (top of their classes at strong undergraduate programs in India/China) or had done relevant research. This was true also for the American undergraduate students in my program. Economically speaking, none of the Indian/Chinese students were going to suffer if they didn't join a PhD program in America. They had very good jobs. (A job at Google India in 2009 was pretty fucking hard to get into what with dozens of interviews and so on.).

Now, this is only part of the story. There were at the same time enough lots of students in India who either found it hard to get a job or wanted to improve on their economic situation who came to (equally shitty) graduate schools in America in order to better their situation. Presumably the same is true for students from China.

Re: Don't Become a Scientist (1999)

#37

Post by dnautics »

jonathansizz wrote: post #25

To be successful in science, as in most other professions, requires a combination of direct skills (ability, knowledge, industriousness, perseverance, etc.), indirect skills (e.g. connections, political ability, salesmanship) and luck. As a field gets more overcrowded, the job itself becomes less attractive (lower pay, more hours, less interesting work) and the relative importance of the indirect skills and luck beco…

I don't think it works like that because public grants are not a free-market phenomenon.

Re: Don't Become a Scientist (1999)

#38

Post by 001sky »

gcp wrote: post #31
UK-AL wrote: post #20

Earlier quoted context omitted.

I'm guessing massive oversupply will be the reason, as it is here.

Demand for faculty positions is pretty stable. Demand for programmers much less so.

There is a huge imbalance in new phd/new tenure-job openings. also, the proportion of faculty jobs with tenure is shrinking. both of these are massive 'caveats' to the use of 'stable'.

Re: Don't Become a Scientist (1999)

#39

Post by namocat »

geekam wrote: post #2

Would like to know * how true this is, today? * Since I already have an alternative career path (computer science, programmer) can I now pursue my passion of science by going to grad school (physics)?

I have a PhD in Neuroscience and left the "postdoc treadmill" to be a software engineer. So far I am very happy with my decision and wish I would have read this article before getting my PhD. It's 100% accurate, and everyone in the biological sciences knows it.

Re: Don't Become a Scientist (1999)

#40

Post by gone35 »

IvyMike wrote: post #15
yaketysax wrote: post #8

"Leave graduate school to people from India and China, for whom the prospects at home are even worse." Even worse in terms of what?

It's been 15 years since this article was originally written, which doesn't seem like a long time. But China's GDP was $1T in 1999, and was $9T in 2013. I confess near-absolute ignorance but the job prospects for advanced degrees in China probably have gotten at least a little better in that time. https://en.wikipedia.org/wiki/Historical_GDP_of_the_People's...

And China's GNI per capita (PPP) in 2013 was still just about 20% that of the US: $10,900 in current intl dollars [1]; and India's is half of that, at $5,080 [op cit].

Notwithstanding all that moral panic about China and India, it's hard to fathom how desperately poor these countries were throughout the twentieth century in the first place --and how in many ways they still are.

[1] http://data.worldbank.org/indicator/NY.GDP.PCAP.CD

Post reply on HN