Sunday, February 27, 2005


About writing presumptions in papers

While I was reviewing GECCO papers, it reminded me of different styles of writing presumptions. Some researchers write presumptions implicitly, and others write them explicitly. However, I found that some people misunderstood the meaning of those presumptions. For example, one of presumptions of Dirk Thierens' convergence-time model (1994) is infinite population size, and I've heard people think the result is not useful because infinite population is unrealistic in real problems. However, it is already an good approximation for even not too large population size (central limit theory). Of course, there exist many papers about finite population size correction, and they have different strength and weakness (usually more accurate predictions and less insights, see Dave's post here.). But my point is, the results in a paper might still be very helpful for your applications even if not every presumptions is fulfilled, and I like it when people write clearly about their presumptions because I feel that they really know what they are talking about.

I was driven to write something similar, if slightly stronger last year when I was reviewing AAMAS and GECCO papers. I actually think people do know what they're talking about -- at least in their specialty.

It's just that some days I despair that many are incapable of expressing what they know in a way that makes them seem other than an ill-read nincompoop or a lazy plagiarist....

I don't think there will ever be much improvement, frankly. Part of the point of submitting papers to meetings is still pedagogic: MSS and the reviews they receive are de facto part of the training process for young academics.

Unfortunately, there doesn't seem to be a course in remedial technical writing.

Keep reminding them!
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?