« Why Oh Why Are We Ruled by These Idiots? (Yet Another Bushies-in-Iraq Edition) | Main | Signs that the Fed Believes Interest Rates Still Need to Rise »

July 12, 2005

Schroedinger's Orzel

This is a very good example of something that is not seen often enough: an intelligent outsider's guide to how a particular subdiscipline's research seminars work. Somebody should encourage the writing of these, and compile them.

From Chad Orzell's "Uncertain Principles", as he provides such a guide for Synthetic Chemistry (that doesn't mean "not-natural chemistry" or "not-real chemistry" or "ersatz chemistry," but "chemistry where the point is to make (synthesize) things":

Notes Toward a User's Guide to Synthetic Chemistry Talks: In principle, I think this is a very good idea.... In practice... I wind up sitting through a lot of nearly incomprehensible talks, most of them dealing with the synthesis of some molecule or another.... [M]y conclusions about synthetic chemistry talks....

There's no foolproof way to know for sure what you're in for.... Various "-tion" words ("methylation," "intercalation," "purification") are pretty good markers.... Active verbs are likewise a hint.... These talks always follow the same basic form, and can be broken down into four stages:... "Here's this thing we're trying to make." This is usually accompanied by a picture consisting of a bunch of hexagons.... "Here's the stuff we start with." This will include a couple of diagrams showing different arrangements of hexagons... almost all the strange words will be names of different parts and sub-parts of molecules.... "Here are the steps in the process." This will include at least one slide showing multiple diagrams of hexagons, with arrows between them... all the strange words will refer to methods of sticking pieces of molecules together.... "Here are some graphs to prove we ended up what we wanted."... pictures of chart recorder traces, blobby photographs of electrophoresis gels, or pictures of pencil marks made on chromatography films.... The key to interpreting the data plots is that they always come in pairs (at least). There will be one picture showing the signal from the initial reactants, which will consist of a set of peaks, or little photographic blobs, or pencil marks. Then there will be a second set, showing the signal from the same method applied to the products of the reaction. This will be a different set of peaks, blobs, or pencil marks.

The entire point of this section of the talk is to note that the peaks, blobs, or pencil marks in the second picture are in different places than the peaks, blobs, or pencil marks in the first picture. Success is defined as the disappearance of the peaks, blobs, or pencil marks corresponding to the reactants, and the appearance of the peaks, blobs, or pencil marks corresponding to the products.

Peaks, blobs, or pencil marks that are in the same places in both pictures are invariably due to solvents. The speaker will often pretend that these don't exist. Humor them....

If you absolutely need to ask a question.... If the speaker hasn't mentioned the yield specifically, you can't go wrong asking "What's the yield like?" If they have stated the yield, ask "How does the yield stack up against other methods of producing this stuff?" If they have stated the yield, and compared it to existing methods, and you still feel a need to ask a question, ask about the solvent peaks/ blobs/ pencil marks. Questions of the form "Why are you trying to make this stuff in the first place?" are usually considered unsporting....

[S]imilar guides could easily be prepared for various categories of physics talks... the Generic Quantum Information Question is either "What about scalability?" or "What about the decoherence rates?"

I take it that the answer to "What about scalability?" is: "It doesn't scale. If it did, we would have turned the Moon into Smart Matter by now, it would have taken over, and we would be its enthusiastic willing slaves--if it deigned to notice us at all." I take it that the answer to "What about the decoherence rates?" is: "If the decoherence rate was slow enough to be really interesting, you would now know what it's like to feel quantum interference from different versions of yourself listening to different versions of this talk."

Posted by DeLong at July 12, 2005 03:48 PM