(1) You're on a program committee and one of the papers assigned to you presents an algorithmic technique that is a competitor to a technique you previously developed and published. You're upset to discover that their approach generally performs better than yours, and they emphasize that fact in the paper, boasting about how they've superceded your work. At the same time, they gloss over some issues with their approach that suggest yours might still be better in some cases. Do you - focus your review on criticism of the glossed-over issues, because that's the area where you have the most expertise? - champion the paper as an advance in the state of the art, while suggesting that the glossed-over issues merit more discussion in the final draft? - inform the program chair that you will have difficulty being an impartial reviewer, and request that the paper be assigned to someone else? - something else? (2) You've been working on a software system for six months, in hope of submitting a paper on it to the #1 conference in your field. A week before the main submission deadline, after having already submitted an abstract, you discover a bug in your code that would cause it to fail in certain corner cases -- none of which actually arose in the experiments run so far. The obvious fix imposes new costs in the common case, and seriously degrades performance. You think there's a way to fix it without the performance degredation, but it would require a redesign that can't be completed in time for the conference deadline. Do you - implement the obvious fix, re-run your experiments, submit what you have (perhaps with a sentence that says you have improvements in mind for future work), and hope that you can get better results before the deadline for final copy? - submit the results you've already collected, since the corner case never arose, and you're pretty sure the redesign that fixes the bug will have similar performance? - withdraw the abstract, start working on the redesign, and plan to submit to another conference? - something else? (3) A colleague on a program committee asks you to provide an external, expert review of a anonymous submission that's right in your area of expertise. You agree, and then while reviewing the paper you discover that it contains a solution to a problem that has been plaguing you for months. Obviously you can't steal the idea, but building on it will allow you to make much faster progress in your own work than would otherwise have been possible. Do you - build on the idea now, but wait to submit your derivative work until the idea it builds on is published in this or some other forum? - wait to even use the idea until it is published in this or some other forum? - go looking for the idea on arXiv, in the hope that the authors have posted it, and you'll be able to cite it (and use it) now? - something else? (4) While at a conference, you mention an idea you're working on to a famous senior colleage from another institution. He suggests a few avenues of research that might be worth pursuing. A week later, he sends you an email suggesting that the two of you "collaborate," and that he be listed as a co-author on the paper you're currently writing. You don't believe his suggestions over lunch at the conference were significant enough to merit being a co-author, and he doesn't seem to be volunteering to do any additional useful work. Do you - go ahead and let him sign on? After all, you'll end up on a paper with a famous co-author. - offer to mention him in an acknowledgment, but decline to make him a co-author? - report him to the ethics board of the professional association that hosted the conference? - something else?