It is traditional that conference attendees will ask presenters questions about their work. Newcomers are often concerned as to the best way to respond. We suggest the following strategies:
- Present the slides that you did not have to present in the main talk because the chairman asked you to stop to allow time for questions.
- Answer the question for which you had a pre-prepared slide.
- Answer the question for which you had prepared the best answer.
- Attack some other competitive field in general.
- Ask a new question back.
- If there are multiple people standing waiting to ask questions, first pick the most friendly looking of them; answer their question extremely verbosely to soak up all available time.
- Say “That is a good question; let’s take this off-line”. [This is particularly good if you are scheduled to drive straight to the airport after your talk.]
- Refer the questioner to an obscure paper (preferably in Russian, unless you know that the questioner speaks Russian in which case choose French or Japanese etc) in which you assert there is an answer to a more general and better posed version of the question that was asked.
- Rhetorically ask the question back: “Well, what’s your opinion?”.
- Flick pass the question to an eminent authority figure (big-shot) in the audience.
For researchers who are really pushing the boundaries of knowledge (for example building an entirely new axiomatic basis of mathematics in order to properly address the fundamental questions in machine learning), presenting a talk is more difficult because the audience will be so far behind your thinking they will probably not be able to appreciate it all. If anyone is so bold as to actually ask a question, follow the advice below.
- Continually interrupt the questioner to prevent him from actually finishing his question – you do not want your pure ideas polluted by an external viewpoint implicit in a question.
- If the questioner does nevertheless manage to finish, you can just assert “I do not understand anything you have said – it does not relate at all to my talk” and refuse to answer. The very fact he asked question clearly indicates the questioner is not up to understanding the revolutionary new framework you have presented; you may ignore him.
Note that none of the above strategies require you to be able to hear the questioner. The following strategies do require you to hear the question.
- Say that the question refers to a part of the work performed by one of the coauthors (who should not be present at the talk).
- Say that the question is related to some other (unrelated question) and answer that question.
- If you are a non-native speaker of English, keep slipping into your native tongue randomly.
- Say that the question is dealt with in detail on your poster, and make sure you are not at the poster session.
- If the question admits a yes/no answer, toss a coin and give it: just say “yes” (or “no”).
- Answer the question asked, concisely and clearly.