So, you want to be an astrophysicist? Part 2.5 - grad school

Gleðileg Jól!

Another blast from the past of Ye Olde Blogge

You're in grad school. Yay.
Now wtf do you do...

Well, you need to jump through the hurdles first.
Most places have some course and seminar requirements, you may in some cases test out of them or waive them, but think about whether you really want to. Odds are that your undergraduate curriculum was not complete or advanced enough in at least some areas. Yes, you want to get on with research, but you also want to be solid on the fundamentals.
Take the stoopid exams, whatever they call them, usually some variation on candidacy or comps or both.
Most places admit students they expect to pass, a few have a policy of admit everyone and then flunk whoever out, most don't.
So, trained experts have determined you have the innate capability of most likely passing those exams, with a bit of work. Or possibly shitloads of work.
So just do it. We all had to and it didn't do us any harm, bwahaha!

Take seminar courses. They are usually on the "hot topics", and usually very good.
As a grad student I liked to take classes, real classes (mostly math, I still want more math) and seminars, up through my final year. A lot of students want to be done with classes as soon as possible, a lot of advisors think classes are a waste of time and dilute research efforts. I disagree, but not strongly.

GO TO TALKS!

Seriously, and I don't just say that because I used to be in charge of colloquia.
Every place has some mix of formal colloquia, informal lunch talks and meetings.
Go to as many as you can stand. Especially at the beginning and at the end of your time. And in the middle... I know you're Real Busy, except, you know what, you're not. You will be... so now is the time for you to load up on actually hearing about nifty stuff.

The colloquiua are usually either "hot topic" talks, or major reviews of sub-fields. Lunch talks are more likely "in progress" talks. Either way you ought to want to hear about it. This is what the faculty think is hot, was hot recently or will be hot soon.
Go especially if it is outside your sub-field, you might learn something!
Introduce yourself to the speaker if possible. Go to lunch or dinner or coffee, or meet formally, if you can. Be nice.
These people will often remember you, they may be your future employers, or friends of your future employer, they are virtually certain to be reviewing a proposal or something of yours eventually. Astronomy is a small field and people talk.

While you're at it - if you have the oomph, go to stuff outside your department. University wide talks, seminars, things in other departments, even classes. A friend of mine in grad school claimed he could eat for free 5 days a week by attenting the right pattern of seminars and colloquia (not recommended, you can live on cookies and chips and coffee, but not well).

Ok, research. Yes, you should do some.
If you know what you want to work on, get going on it as soon as you can.
If you don't know, start exploring around as soon as feasible.
Go talk to faculty. Mostly they don't bite.

DO NOT ASK ABOUT MONEY! If you're doing it for money, then you're in the wrong business.
Ask to do research you are interested in doing.
Do your homework first, check the web page, glance through a paper or three. If you are not interested in doing research, you know, exploring the universe, adding new knowledge to mankinds inventory of nature and technology, then you should not be doing a PhD.
It is possible that what you think you're interested in is not suitable (in the stoopid profs opinion) for PhD research... it may be done, wrong, not interesting, undoable or just not interesting enough. The Prof may be wrong, but the word "advisor" has a meaning. You want to be advised, take it under consideration.
Try again, or ask for input or both. Some people have lists of projects to do, some people have some general ideas, and some people want to hear what you think. Radical thought.
If the answer from a potential advisor is "no" don't take it personally, and maybe try again, faculty have external lives that may preclude working with students or on particular projects at any given time.
If the project is good, a lot of the time money will be found...
Being a research assistant is good, but it won't kill you to be a teaching assistant for a while. You should be a TA for at least 6 months anyway. If you hate teaching, find out early in grad school, not when you become faculty!

If the department permits, or encourages it, then try doing a research project in a sub-field that is not what you think you want to do for a thesis.

Think very seriously about whether you want to do theory, observation, data analysis or instrumentation.
You may end up doing things you never imagined out of necessity (like theorists go take observations, cause if they don't no one else will; or observers running simulations, or building the instrument they need to do the observations etc etc).
Do you have the aptitude? Do you have the background? Can you learn?

Career goals should matter some, think about whether you want to stay in research? In academia? Industry? Observatories? NASA?
You want to pick up certain skills and experiences depending on which is your primary focus. Be ready to be disappointed, the competition is fierce. But most PhDs end up in some combination of interesting or well paid jobs. Opportunity cost is usually recovered even if you end up doing something other than you planned.

It is worth thinking about the medium term future of your chosen sub-field; is it expanding, new faciltiies coming online, is it peaking, or in decline? BUT, predicting the future is futile, particularly now. I don't know anyone who could confidently tell you which sub-field will see growth in the next decade. JWST is the biggest bet, if it launches and works.
Survey science and database mining look promising; so do far-IR and mm-wave obserbations; I also suspect computer simulations are a growth field again - BUT, if you had asked me 1-2 years ago my predictions would have been different and very, very wrong.

Pick and advisor you can get on with. Scientists are often "characters". Way high up on the list of "things you do not want to do" is being stuck with a PhD advisor you do not get along with.

BUG YOUR ADVISOR.
Er, but first make sure you pick an advisor who likes to be bugged...

There are two kinds of advisors - those who meet formally with students and expect progress; and those who bump into people in the corridor and blearily say "oh, you're working with me, right? Well, so, how is it going then...?"
KNOW YOURSELF: if you need the push of regular progress reports, then don't sign up with a hands-off advisor.
If you do want to bug your advisor, bring something to the meeting: progress, news, an idea, a question, a complaint. Something.

Self-motivate: particularly if you work with a hands-off type. Everyone procrastinates, but you're not going to have your hand held in most departments, you fizzle out, you're finished (eventually, when someone finally notices).

You're supposed to be working on new original stuff that you find interesting. So do it.

Finally: READ!!! Pro-actively.
Check arXiv regularly and thoroughly. Read the papers relevant to you and anything else that looks interesting.
Read the references! They are there for a reason.
Read the citations - if a paper is interesting, papers which cite it are also likely to be interesting. Use the ADS "C" option liberally and look through it quickly.
If in doubt ask you advisor, or just read it anyway.

Next, the slightly tricky issue of what we actually "do", research wise type of thingy. Might take a while...

PS: Rob's comment promoted
- this of course is how progress is made in astronomy...
I hope I am a type Ic in this new improved taxonomy, and I aspire to a possible Ib, but I fear I am a type II.

Rob says:
"There are two kinds of advisors - those who meet formally with students and expect progress; and those who bump into people in the corridor and blearily say "oh, you're with me, right? Well, so, how is it going then...?""

This is sort of like the two kinds of supernovae (I and II)... there is something behind it, but it is of course vastly oversimplified.

I would divide your first type into subclasses. Type Ia advisors are the ones who are very proactive bosses. They're watching you, keeping track of what you're doing all the time, nagging you to make progress, always prodding you.

Type Ib and Ic advisors are more like Type II advisors, but they are higher energy than your typical Type II (even, you might say, Type II plateau advisors). Where was a Type II advisors may see advising grad students as just one more thing that you have to do to stay in a University job, and generally regards you with benevolent neglect, the intentions of Type Ib and Ic advisors are similar to the intentions of Type Ia advisors, but the observational signature is different.

A Type Ib advisor will meet informally with students regularly, at his prodding. He expects progress, but doesn't really meet formally. He won't set any specific milestones, and he won't always tell the student where he should be, but he will give the student constant feedback on what he's doing.

A Type Ic advisor needs to be bugged, but (unlike the Type II advisors) wants to be bugged and (like Type Ia and Ib advisors) does sometimes lose sleep worrying about the current status and future of his students. He's too busy, like everybody else, so he isn't as proactive as a Type Ib advsisor. Plus, he has the philosophy that grad students have to learn how to be independent if they are going to succeed after grad school. He's more than willing to provide a lot of direction for grad students in their first couple of years, but expects the grad student to be increasingly self-motivated and self-directed as time goes by. He's always more than willing to provide advice, help, and support for the grad students, but expects the grad students to seek out help rather than sit around waiting for help to be thrust upon them.

Can you tell that I'm a Type Ic advisor? (That's the one I'm writing the most about.) I had a grad student who really needed a Type Ia advisor, who languished under me. That was really where I learned that I'm more of a Type Ic advisor, so I try to be very up front with students about that. I also remember being a grad student, where I was afraid to talk to my advisor-- if I didn't know something, or had anything undone, I was afraid to ask for help. I try to make clear to my students that I will not think less of them if they do that, that I expect and demand them to bug me when they need help, direction, and so forth. I'm willing to help give them direction, but they need to ask for it when they are adrift. (At least, once they're in their third or fourth years.) If I haven't heard from them for a while, I will prod them, but if they want to just coast along and not do much -- with me, they will be able to for a while, but it will be at the risk of their graduate career, for they will not be making the progress they need to be making.

(I've also known a couple of post-docs who evidently had extremely hands-on Type Ia advisors in grad school, and who as such did not seem to be able to operated as independently as I expected a post-doc to operate....)

All students should probably seek out some sort of Type I advisor, and find the sort that sounds most appealing to them. However, to be safe, students should assume that their advisor is a Type II advisor, and be proactive in bugging the advisor -- as you say. If the advisor is hostile to being bugged then, well, that's not a good advisor to have....