Proposals are an interesting beast. In astronomy, the proposals for funding and those for data are typically decoupled. In some cases (e.g., when apply for Hubble Space Telescope), support funding comes with the granted time for data acquisition. These applications are rare, and highly sought after.
I have a fellowship sponsored by the Chilean national government. This fellowship pays my salary and some other costs such as travel and publication costs associated with doing science. However, there is no guarantee that I will actually get time on telescopes in order to collect the data necessary for the project that the fellowship is intended to fund! Thus, in addition to the initial fellowship proposal, I have to write many observing proposals in order to get data. One would think that I am decently good at it by now.
In my experience, every proposal is a snarling jabberwocky that needs to be wrestled and tamed until it fits in a cage of predetermined size and material.
These times are often great for learning a lot about my science, as they require some level of reading. I need to know what I’m talking about right? They are also exciting times, because I think a lot about the projects that I am doing or want to do. But my main point for this post is this:
The amount of time you invest in a proposal needs to be carefully considered. I know people who work to the very last minute on proposals. The idea here is that every change you think of can only make the proposal better, and if that one change makes it good enough to get time, then it is worth making. But I think a little bit differently . . . in part, because that method leads to tremendous amounts of stress, late hours, and sweating bullets trying to get the proposal submitted in the last 10 minutes before the deadline and oh my god, did the website crash or is it just my internet - freak out! I have a typo!
No. Here’s the way I look at it:
The “goodness” of a proposal is a function of the amount of time/effort one spends on it. However, that function is not a power-law, not a quadratic function, not even linear. That function quickly closes in on an asymptote (see figure). And no matter how much you work on that proposal it’s never going above that asymptote. Now, somewhere on the “goodness” scale is a cut-off for how good your proposal has to be in order to be granted time, and many factors go into determining that critical value. As long as your proposal reaches that level, then it is “good enough” - you get the time, you get the data, you go on with your life. And it may in fact be that no matter how much time and effort you dump into your proposal, it will never meet that critical value.
Furthermore, it is nearly impossible to predict what that value is. You’ll go insane if you try to squeeze every last moment into proposaling. So, try not to go down the path of infinite time/effort. At some point there is nothing more you can do to make your proposal get time.
At least that’s what I try to do. Inevitably, I get sucked into the vortex of proposal deadlines.
Figure: Blue - your proposal "goodness". Green - the maximum "goodness" your proposal will ever have (determined by quality of proposed science). Red - critical value at which your proposal will make it past review and be granted time (determined by your writing ability and the mood of the review committee).
Kyle D Hiner
More blogs to come
Epilogue: Since the writing of that initial post I've found out the results of the applications I had submitted. One of my proposals was successful, while two others were denied time. I could probably do an analysis to see if I actually beat the odds, but that would require me doing some more research than I really care to do at this point. In any case, I (and my research group) was granted 3 nights of observing with the Magellan telescope at Las Campanas Observatory, Chile. That's a big deal. Each night costs about $30k to run the telescope. Effectively, I was given ~$90k to do some research. There's more to say on that topic, but this post is long enough already.
No comments:
Post a Comment