User:Jadrian Miles/Thesis proposal feedback/dhl meeting notes
Don't come off as glib. dhl suggests:
- pause after every question so it's clear you're thinking and/or say their question back to them to make sure you've got it right, and never cut people off;
- if a question isn't clear, say you don't understand and ask them to re-ask;
- end every answer with "did that answer your question?" or something similar, even (especially?) if your answer is just, "that's a good point; I'll think about it";
- answer the question asked; don't spiral off into a discussion of something else (like the clustering bit with Chad) unless it serves a point you're about to make;
- remember that people just mention stuff you didn't think of, not to suggest that you screwed up, but to bring it up. Respond positively and don't try to justify why you didn't do it. Trying to show engagement with the idea by talking about how it relates to what you already did can rob your commenter of a sense of ownership over their comment. Reinforce their ownership.
Clear up misunderstandings early.
Don't apologize in a talk. If you forgot something, don't mention it.
Answering a single question in the middle of the talk isn't the time for a conversation. Defer to the questioner; just say they're right. Leave discussion for the closing Q&A. Spike said "unambiguous" seemed weird, so just agree with him and show you mean it: edit in place or something. A further discussion can wait until after the talk or a private meeting later.
Balance the desire to answer questions completely with the desire to make sure the talk is over on time. Ending your talk late pisses people off.
---
Spike said to number the slides. Helps some people, doesn't bother most others.
---
Try video-recording your talk. You'll be surprised and might get a better feel for how you come across when answering questions.
---
Chad sez: most of the questions derived from misunderstandings. Make input and output of every step clear. Make your goal clear. Make your terminology clear. Context doesn't matter as much to CSers as clear "problem statements" for each contribution.