Before submitting a paper… disclosure

Paul McEuen 1/19/05

Title: State Main Result Clearly (note: there is only one main result in any paper)

Abstract: Summarizes results leading to conclusion indicated by title

1. Introduction – gives historical perspective, background, leading to the central burning issue that this paper resolves.
2. System – describe devices, apparatus, etc used.
3. Data – describe with no interpretation the bulk of the measurements performed. (Just the facts….)
4. Massaging – describe any nontrivial analysis you did to make things clearer
5. Theory – present any theory needed to understand data
6. Interpretation – describe what you think the results mean.
7. New experiment - present one extra bit of data motivated by 4-6. (“If x is true, then y should happen. We measured, and indeed y happened….”)
8. Conclusion – repeat central result of paper and then say a few words about its implications for previous or future experiments/theory.

A few comments:

(4)-(6) can get mixed up sometimes, but keep them as segregated as humanly possible. What you think should not get mixed up with what you did. For this reason, ALWAYS keep (3) unsullied by interpretation. You are an experimentalist first—your job is to report the results of your experiments. After you’ve done that, you can comment on them. (and if you keep your comments separate, people are free to ignore them if they think you are full of hooey. But they can’t ignore the data!)

Be careful to properly reference other people’s work. Similarly, be careful to delineate what you did that was new, versus what other people did, versus you did that confirms what people did before. (note that this is commentary, and with few exceptions should not be allowed in (3)). Do not assume that the reader knows the field.

• Roman symbols (a, b, c…) are always italicized.
• There is always a space between a number and its units, for example 10 mV. (10 and mV are treated like two separate words).
• Abreviations and acronyms are writen out in full when they are needed at the start of a sentence. “Figure 3 and Fig. 1 both show CNTs. Carbon nanotubes are not seen in other experiments.”

A critical aspect of doing science is producing clear, concise writing. One of the most influential and best-known guides for writing clear, concise english is "Elements of Style" by Skunk & White.

One of the many tools that helps make writing clear and unambiguous is the comma. Bill Bryson has written an excellent short explaination of comma usage. This excerpt is taken from “Bryson's dictionary of troublesome words”.

When writing a scientific article, you must be familiar with the conventional framework of scientific articles. The only way to gain familiarity with this framework is to read many articles. You will start to notice techniques that good authors employ (especially as you start to write your own article). For example, in the article by Sazanova et. al, try reading only the first sentence of every paragraph. In this way, you can grasp the strucutre and logical progression of the whole paper in sixty seconds. The style ensures that you are never lost and you can quickly locate information you are seeking.

Preparing figures requires the same skills as preparing digital artwork. You have an 8.5 cm wide column to prepare a 300 dpi image that visually communicates a range of information. A single figure often includes multiple graphs, a cartoon of the experiment, an AFM image and many text labels. Powerpoint is often the best program to assemble all the components of a figure and create a balanced layout.

The weakness of powerpoint is the limited export options. The final format of the figure will be a high resolution jpg with the correct linear dimensions and pixel dimensions (unless you are submitting to Nature, which accepts powerpoint files). A useful way to export from powerpoint is the “print screen” button on your keyboard. This gives you 1200 pixels to work with when the screen capture is opened in photoshop. Linear dimensions and pixel dimensions are easily controlled using photoshop.

Ethan 2/14/2008

• Avoid writing detailed text. Most details are better explained in person. Let the figures and figure captions tell the story. Text which is absolutely necessary must be big.
• Graphs must be easy to read. Axis labels and tick numbers at least 16 pt and curves drawn in bold lines.
• Don't use more than 2 different font sizes. For example, 16 pt for labels and references, 20 pt for important captions.
• Imagine presenting the poster to someone, the poster must include all cartoon diagrams to support your explainations.