3 ms·

What about graduate school in computer science? Surely a Ph.D. in CS would have fallback options at places like Google or Microsoft where their graduate work wo

What about graduate school in computer science? Surely a Ph.D. in CS would have fallback options at places like Google or Microsoft where their graduate work won't have been wasted?

rms 17y ago
Most of the technical PhDs do just fine. There are too many biology PhDs though.
- ratsbane 17y ago
  I have the vague impression that PhD computational biologists, bioinformaticists, are in demand and probably will be for a long time.
  - rms 17y ago
    That's definitely true, I mean non-programming, regular biologists, molecular biologists, etc.
Nosferax 17y ago
I hope so ! I'm still pondering whether to do a Ph.D. or not myself. Almost done with M.Sc...
timwiseman 17y ago
There are generally plenty of positions for PHDs in technical fields, and I have known of many employers that will pay the tuition for at least the masters in a technical field if you don't already have one (mine is, for instance.) Humanities tend to have fewer direct applications in business though. History at least is seen as some value in certain types of analysis and historians are often used as consultants in some fields. On the other hand, perhaps I just run in the wrong circles, but I cannot think of any direct use for a PHD in say English Literature other than teaching.
anonjon 17y ago
The problem in humanities is that the market is saturated, also, the majority of undergraduate classes are taught by graduate students (they're cheap!), further reducing the number of potential hiring positions. So you have to actually be exceptional and have published something interesting to even have a good chance at making it as a professor. You could do other things as well, write books or work for various foundations (teach high school!), but you can do those things without a Phd in medieval English literature. You also are spending 8+ years not getting paid or putting yourself into debt, and that you still might end up working at Arby's (or what have you). He left out that the pay for a starting professor sucks. (Considering that you just spent 8 years of your life getting the degree). Computer science is a bit of a different ballgame in that it is applicable (well, vaguely) to software development. So you might be able to get a job that is orthogonally relevant to whatever it was that you passionately studied in graduate school. The sad truth, however, is that the majority of software development does not come anywhere close to requiring a Phd. And you still come out up to your eyeballs in debt or flat broke. The good thing about it is that it looks pretty good on a resume. (Although weighing 8 years writing software against 8 years in grad school, I might go with the guy who has work experience... is kind of a toss up). The google/ibm/microsoft research positions are few and far between, in any event.
- timwiseman 17y ago
  The sad truth, however, is that the majority of software development does not come anywhere close to requiring a Phd. The vast majority of current software development is guis for databases and can be done with relatively little training, not even requiring a BS in CS much less a PHD. But there is more advanced development where the more theoretical backing is at least useful. (Although weighing 8 years writing software against 8 years in grad school, I might go with the guy who has work experience... is kind of a toss up). I largely agree, but it depends on what kind of software the programmer was writing. Doing major kernel hacking for a Linux distro is highly impressive for instance, doing maintenance patches for yet another database front end is not. Also remember that the two is not an either or proposition. It requires a somewhat flexible day job and it may take longer, but you can work on a graduate degree while holding down a day job. I work a help desk part time for most of my undergrad degree and work as a DBA/Sr. Programmer while working on my masters right now for instance. On the flip side, you can go to class at day and contribute to open source or do contracting at night.
tom_b 17y ago
My anecdotal experience is that CS grad school is probably worth it. I have a MS myself and when I wrapped it (at the end of the Internet bubble in 2001), it was still worth 15-20% in starting salary at big corps. Some of my cohort continued onto the PhD - they <mostly> found employment in universities and research institutions over the last three or four years. However, the Taulbee survey (http://www.cra.org/statistics/ http://www.cra.org/statistics/) seems to indicate that we are seeing a large increase in the number of PhDs over the last decade (1800 a year? thats a bunch to get sucked into Microsoft and Google on a yearly basis). Several of my cohort had to take postdocs or lectureship jobs (at $3K per class or something else ridiculous) to stay in academia, so it might be a bit of a CS PhD bubble as well. Falling undergrad enrollments over the last few years might have bottomed out as last year there was an uptick in BS enrollments. The Taulbee data indicates that a large portion of CS PhD types (~60%) wind up in industry positions of one type of another. I would probably encourage talented hackers with the interest to try grad school, but to not feel compelled to go at all if they can pick up challenging work at a startup or even bigger company. Be damn careful about winding up an "IT professional" though. If you don't control that process, you'll wind up a jack of all trades in a job market that tends to reward specialization and scarcity of knowledge rather than general "get it done" skills. Of course, if you start your own company, you can ignore all the above advice.

What about graduate school in computer science? Surely a Ph.D. in CS would have fallback options at places like Google or Microsoft where their graduate work won't have been wasted?

rms 17y ago

Most of the technical PhDs do just fine. There are too many biology PhDs though.

ratsbane 17y ago
I have the vague impression that PhD computational biologists, bioinformaticists, are in demand and probably will be for a long time.
- rms 17y ago
  That's definitely true, I mean non-programming, regular biologists, molecular biologists, etc.

Nosferax 17y ago

I hope so ! I'm still pondering whether to do a Ph.D. or not myself. Almost done with M.Sc...

timwiseman 17y ago

There are generally plenty of positions for PHDs in technical fields, and I have known of many employers that will pay the tuition for at least the masters in a technical field if you don't already have one (mine is, for instance.) Humanities tend to have fewer direct applications in business though. History at least is seen as some value in certain types of analysis and historians are often used as consultants in some fields. On the other hand, perhaps I just run in the wrong circles, but I cannot think of any direct use for a PHD in say English Literature other than teaching.

anonjon 17y ago

The problem in humanities is that the market is saturated, also, the majority of undergraduate classes are taught by graduate students (they're cheap!), further reducing the number of potential hiring positions. So you have to actually be exceptional and have published something interesting to even have a good chance at making it as a professor. You could do other things as well, write books or work for various foundations (teach high school!), but you can do those things without a Phd in medieval English literature. You also are spending 8+ years not getting paid or putting yourself into debt, and that you still might end up working at Arby's (or what have you). He left out that the pay for a starting professor sucks. (Considering that you just spent 8 years of your life getting the degree). Computer science is a bit of a different ballgame in that it is applicable (well, vaguely) to software development. So you might be able to get a job that is orthogonally relevant to whatever it was that you passionately studied in graduate school. The sad truth, however, is that the majority of software development does not come anywhere close to requiring a Phd. And you still come out up to your eyeballs in debt or flat broke. The good thing about it is that it looks pretty good on a resume. (Although weighing 8 years writing software against 8 years in grad school, I might go with the guy who has work experience... is kind of a toss up). The google/ibm/microsoft research positions are few and far between, in any event.

timwiseman 17y ago
The sad truth, however, is that the majority of software development does not come anywhere close to requiring a Phd. The vast majority of current software development is guis for databases and can be done with relatively little training, not even requiring a BS in CS much less a PHD. But there is more advanced development where the more theoretical backing is at least useful. (Although weighing 8 years writing software against 8 years in grad school, I might go with the guy who has work experience... is kind of a toss up). I largely agree, but it depends on what kind of software the programmer was writing. Doing major kernel hacking for a Linux distro is highly impressive for instance, doing maintenance patches for yet another database front end is not. Also remember that the two is not an either or proposition. It requires a somewhat flexible day job and it may take longer, but you can work on a graduate degree while holding down a day job. I work a help desk part time for most of my undergrad degree and work as a DBA/Sr. Programmer while working on my masters right now for instance. On the flip side, you can go to class at day and contribute to open source or do contracting at night.

tom_b 17y ago

My anecdotal experience is that CS grad school is probably worth it. I have a MS myself and when I wrapped it (at the end of the Internet bubble in 2001), it was still worth 15-20% in starting salary at big corps. Some of my cohort continued onto the PhD - they <mostly> found employment in universities and research institutions over the last three or four years. However, the Taulbee survey (http://www.cra.org/statistics/ http://www.cra.org/statistics/) seems to indicate that we are seeing a large increase in the number of PhDs over the last decade (1800 a year? thats a bunch to get sucked into Microsoft and Google on a yearly basis). Several of my cohort had to take postdocs or lectureship jobs (at $3K per class or something else ridiculous) to stay in academia, so it might be a bit of a CS PhD bubble as well. Falling undergrad enrollments over the last few years might have bottomed out as last year there was an uptick in BS enrollments. The Taulbee data indicates that a large portion of CS PhD types (~60%) wind up in industry positions of one type of another. I would probably encourage talented hackers with the interest to try grad school, but to not feel compelled to go at all if they can pick up challenging work at a startup or even bigger company. Be damn careful about winding up an "IT professional" though. If you don't control that process, you'll wind up a jack of all trades in a job market that tends to reward specialization and scarcity of knowledge rather than general "get it done" skills. Of course, if you start your own company, you can ignore all the above advice.