The whole point of being a scientist is, indeed, writing papers and giving talks.

… and of course writing grants rounds out the top three.

Not only will this get them used to taking the initiative when they get to grad school, but hopefully it’ll help with the advisor search.

Mind you, I’m in astro, so it’s slightly different but I started with research between my freshman and sophomore years of undergrad. That in turn got me publications (yay, always good, including one first author pub my sophomore year) and experience which got me jobs which then got me into grad school with a topic and an advisor. My working during undergrad for various different people (whilst sticking with my official research advisor whom I loved as a mentor) also let me know the type of advisor I can do well with- the kind that is generally hands off, but is there when I finally say “Enough! I’ve been trying to figure out this problem and I’ve google-whacked, emailed people, talked with people and I’ll be buggered if I can find anything that might help.”

And I’d like to say that in the “be like a postdoc you’d want to hire
” and “be a finisher” sections should perhaps mention proposals. Much as we may not like to admit it, the holy NSF/NASA/DOE/DOD grant money, is what funds much of the sciences. Writing proposals can always be a trying experience, however doing them before your last year of your grad career get you so much more experience in writing and have the chance to improve before your funding depends on you being able to secure it.

I can clarify a bit about the hot topic issue. What I always told students was that they had to define their problem narrowly enough that by the time they were done with their thesis, they knew more than anyone else in the world about their topic. The thesis topic also has to be couched in the form of a problem that admits of a solution (that is, a hypothesis), not in the form of a research area.

Generally, finding something about which to be the expert means choosing a topic that’s much narrower than the student would like. The trick then is to situate the narrow project as part of a bigger, longer-term project. Some professors are geniuses at pulling together small projects into empires. That management role of seeing the big picture is much more important than students usually think, particularly in raising funding.

The missing piece of advice here is collaboration. The best scientists I know all collaborate heavily. I used to let students collaborate on any course work they did as long as they cited who did what. That’s what the real world of science is like. Collaboration is particularly useful if there’s a mismatch in skills, such as writing, solving particular kinds of equations, etc.

-Mark

(1) Writing and giving talks are skills that you learn just like doing physics problem sets is a skill, only they’re open-ended and require initiative. Before you’ve learned them well, you’ll be uncomfortable with them; that’s normal. Grad students don’t like giving talks because they have to get up in front of profs and might embarrass themselves. Okay, but they have to learn to get over that. Talking in front of grad-student-only audiences is a good way to do that.

(2) It is sort of possible to do science without writing papers and giving talks. You can work in a research group on a project that somebody else directs, in a more technical capacity: writing software, building instruments, doing lab work, and so on. You won’t be independent or self-guided. But even then, you’ll find that you may have to write internal reports, give talks within the group, and so on. You can’t be an effective research professor or PI if you don’t write papers, proposals, and grants, and give talks.

Go skiing mid-week when there’s fresh powder and no lift-lines. Use profanity when you TA a lab section. Sign up for a conference just because it’s in an exotic location. Play foosball–everyday. Drink.

I think you have just described 90% of the UW Astro grads.

(Note that I think this is a good thing.)