Here and there I have been asked whether or not one should apply to a PhD position. After thinking for a while in these questions, I perceived that they can be translated to a single question: “should I apply for this position at University X”. This question is obviously important, because the decision to go to University X instead of University Y can pretty much drive your PhD. For instance, if University X does not provide basic funding to cover your expenses, it would be much more difficult to finish a PhD. However, if funding is guaranteed, the decision should not be taken in terms of the place you are going, but instead with who you will be working.
For sure, it will be very impressive to state in your CV that you have a degree from that top notch university. These universities may also have beautiful labs, with a lot of funding for your research. You may also have family or close relatives that live in the city that you are planning to move, and this would greatly reduce the transaction cost. However, in my rather little experience, the most important factor is your advisor. What if your advisor is a very busy person, that needs to coordinate several projects, and several people? What if your advisor is not a very friendly person? What if your advisor is planning to do some startup and now does very little research?
In other words: Is it worth to work in the top CS school in the world if you advisor is not available to support you accordingly?
Moreover, for many kind of research works (at least in software engineering) you may not need a multi million dollar cluster neither thousands of dollars to travel for field study. That’s awesome if you have. But if you don’t, it is not the end of the work. You can also find some very interesting research topics that you can work on your 2013 notebook.
IMO, More important than this is actually an advisor that could properly guide your work. That can devote some week hours to talk to you. That trust on you. That help you.
Unfortunately, it may be very hard to note these qualities in a person that you may not even know in person. Next are some tips that may give a hint along these lines.
Find his/her personal webpage: This is the very first rule of thumb. If your prospective advisor does not have a webpage, he may be very, very busy. Important: webpages such as Google Scholar or DBLP are not personal webpages. The bare minimum personal webpage should clearly make available his/her email address. If one does not provide a way to others contact him/her, s/he may not want to receive your contact.
Talk to his/her former students: Former students are an excellent source of information. Perhaps no one knows better than them how their former advisor works and behaves. Important: many advisors made available a list of his/her former students. This list can be available on the advisor personal webpage, or on their CVs. Cold email a bunch of former students and ask their perceptions regarding work with your prospective advisor. Many former students will be glad to help (in particular if they enjoyed the time that they worked with your prospective advisor. Moreover, there is nothing wrong to ask for an insider opinion (your prospective advisor does the same when he asks for recommendation letters).
Follow him/her in a social network: As of today, Twitter is probably the de facto on line way that researchers broadly communicate with their peers. Obviously email plays a role. However, email is private and you can hardly interact with someone else who is already having a discussion. But you can do this on Twitter. If your prospective advisor has a Twitter account, follow him/her to get a glimpse about what he thinks about many things. Whenever possible, interact with him/her.
Ask for an Skype meeting: A short online meeting can help a lot. If you have a negative answer, this may indicate a carelessness. Even very busy/famous professors can find the time for a quick online meeting. However, before the meeting, think about 3–5 question to ask the professor. Asking questions may indicate an interest on you side. If you have no interest, why should s/he have?