Sure, and I'm not complaining too much as a guy who studied Math/CS in my undergrad days (and I'm even considering graduate work!). The author also gives a nod to the industrial paths.

On the other hand:

* I think if you consider "tech, finance, data scientist", some reflection reveals a discouraging fact: we're usually not talking about science or even engineering at all, but essentially support functions allowing new kinds of productivity or scale for mundane business functions. So there's careers out there for us educated in these quantitative/technological arts... but it's generally not a career in science.

* More concretely, my math/CS education is almost entirely superfluous to my six-figure job doing CSS and wrangling squirrelly JavaScript frameworks/apps (mostly as a support function to sell automobiles). Meanwhile there are tenured faculty making half what I do (and postdocs/grad students on poverty incomes) actually working to figure out how life works, cure cancer, and build space elevators.

I don't know much about research related to tenure achievement and salaries, but I found these:

http://chronicle.com/article/Average-Faculty-Salaries-by/126... http://www.opia.psu.edu/sites/default/files/AIR_Tenure_Flow_...

My takeaway from skimming is that (a) faculty salaries vary, those in areas with the industrial options we've talked about seem to be higher, but an average of $65-70k seems credible (b) data on tenure is hard to come by but everyone agrees it's very difficult to get at high status institutions.

As someone who did a PhD and 6 years of post-doc in nuclear physics at a high status institution, I would disagree that tenure is difficult to get at a high status institution. I would argue that tenure at a high status institution depends on skills that are wholly orthogonal to scientific research. Yes, I'm bitter. And so are all my (ex)-colleagues.

Tenure is a social position for promoting an establishment.

The concept entirely predates modern ideas of science and research, so it is no real surprise that it does not serve that purpose.

You can get several interesting jobs that are not just doing CSS.

As a quant, you can be doing ML to predict the stock market or hard core network coding to make placing orders faster.

In a data scientist job you'd be doing data mining to gather insights about whatever area your company works in.

At a company like Google or MS, you could be developing/implementing interesting algos for search, ad targeting, etc.

This is not even to mention Google/FB research, and even more so, MSR where you could be doing even more researchy stuff.

Now it's possible that the above jobs are much scarcer than the support function jobs but I mean most of the world does mundane jobs and earns way less than you. A PhD gives a much higher chance of getting a less mundane job.

>As a quant, you can be doing ML to predict the stock market or hard core network coding to make placing orders faster.

I think he was pointing out that what is doing is not that relevant to advancing human knowledge. Proprietary trading strategies and optimizations that will most likely never be made public are not any better.

All of your examples are the same until you hit Google/FB research where you might actually be working on something that isn't directly promoting the bottom line of your employer.

About the fact that the majority of professors get tenure. Lets put some numbers to this statement.

64% get tenure according to http://www.sciencemag.org.libproxy.mit.edu/content/335/6070/... at least. Median time to exit the tenure track? 7 to 10 years depending on the area.

So now imagine this. You work as a PhD student for 6 years, as a postdoc for 2-3, as a faculty member for 7 years. And at the end (you're just about 40) someone flips a coin. Either you get tenure, or you beat it and get nothing. Along the way your industry peers have outearned you by hundreds of thousands of dollars by this point.

So yes. The situation is that bleak and the article is not overstating its point.

Along the way you have to keep in mind that things are only getting worse. Grant funding is much harder to get today than it was 10 years ago. Tenure rates are much lower than they were 10 years ago. So if you're planning on a career in science you have to factor in the fact that the situation you will face when you come up for tenure will likely be far worse than the already-poor statistics above indicate.

Well put.

The author here is differentiating between Science oriented careers (industry, government, consultation, practical research) and hard science jobs (research at the edge of science with no immediate practical benefit.)

The point is that with almost any Science or Engineering degree you can get a great job in industry but that in the world of hard science research, a place people go because of their ideas, not because of the profits, you will struggle for little chance of serious reward.

Perhaps this is good for an economy. Most people will be flushed into practical things and those who are driven to be artists, teachers, scientists and performers will be carried there by their passions despite the hardship.

The biggest problem with this is that we are probably herding our most talented young people straight into the open maw of industry and away from things that lay the foundation for human advancement.

"The point is that with almost any Science or Engineering degree you can get a great job in industry" this is far from guaranteed in the life science. It can be true, but very frequently in studies of job satisfaction among STEM graduate students (where the Bio folks tend to fall toward the bottom) one of the primary concerns is that there's not clear paths to industry on a large scale.

'"almost any"'

I reserve the right to later redefine this threshold but for now, call it %85.

I wish people would break this out by discipline for exactly the reason you mention.

Sadly the articles describing a lack of STEM graduates and the ones describing how shit being a scientist is both paint all science with one brush.

I agree. Reminds me of "X% of published results can't be replicated", "Y% of researchers don't understand statistics", etc., most of which turn out to be problems in particular fields like sociology, psychology or medicine.

> a CS perspective

S is different from TEM in STEM...and CS leans more towards TEM than S.