3 ms·
I defended my dissertation in December 2010, and I started at IBM Research in January of 2011. Within weeks of being at IBM Research, I found myself having fant
by scott_s 15y ago
I defended my dissertation in December 2010, and I started at IBM Research in January of 2011. Within weeks of being at IBM Research, I found myself having fantastically productive, ad-hoc research discussions with my colleagues that lead to real work being done - code being written, experiments being run and papers written. Things moved much, much faster than they could have when I was working alone as a graduate student. Three highly motivated people working together on the same problem are more than just three times more productive than a single person working on the problem.
I heartily endorse the author's advice to collaborate. I didn't do it much as I should have as a grad student. But, in some ways, doing so is not encouraged. My last two years, I had one goal: to graduate. I knew what I had to do to get that done, so I plowed through and did it. Because you need to come up with a dissertation that is yours, you have many students who will rationally work on their own projects without much collaboration.
- evanj 15y agoI'm glad to hear I'm not the only one who feels this way. I agree that working alone seems like the easiest or best choice. However, I think it is a local maximum, and not a true optimal solution. At least for me, working with others makes me way more productive, so I think that my collaborations helped me graduate sooner. I were to start again, I would be more proactive about working with others. Its true that to graduate, you have to write your own thesis, but that doesn't mean you can't collaborate. I've found that parts of work naturally fall to people according to their interests, so it wasn't really a problem for the projects I worked on.