For some time now I have been extremely dissatisfied with the standard format of scientific communications. Specificially, they contain unnecessary details and background information, and are too long. For this reason the majority of readers will read only the abstracts of papers. Looking back at my own papers I can see that as an author, I am completely guilty of this. Almost every paper I write begins with a paragraph "Quantum computing offers the potential to solve problems intractable on a classical computer... yadda yadda". Throw it out. My target audience is people who are specialists in the field of quantum computing. Anyone who doesn't already know that a quantum computer is likely to be capabable of solving problems intractable on a classical computer isn't going to be reading my papers. Next, mathematical proofs are irrelevant to the vast majority of readers. What 90% of my readers are interested in is what I do at a conceptual level, and the results that follow from that. They are not interested in all the gory mathematical intricacies.
All of this has motivated me to develop some guidelines for 'efficient scientific communication'. There are two overriding principles behind these guidlines. First, people should not have to read unneccesary padding that they are already well aware of. Second, the text should lend itself to being skim read. Here are some guidelines:
- It should be clear from the onset which audience the text is targeted at.
- The body of the text should include only details necessary to make the text understandable to the target audience. This means avoiding 'standard' introductions and other 'padding' that addds to the length but doesn't enhance the content.
- Extra background material, necessary to bridge the gap between the target audience and non-specialists is placed in footnotes. This way the target audience can avoid having to read unneccesary background material.
- The body of the text is in point-like form, devoid of lengthy elaborations, while presenting the techniques and results as concisely as possible.
- Elaborations that are not essential to understanding the content of the paper are placed in footnotes.
- The body of the text contains no mathemetical proofs, only results. All proofs are placed in the appendix.
- Language should be as compact as possible.
- Paragraphing and structure should be chosen so as to facilitate skim reading to the greatest possible extent.
I recently experimented with trying to write according to these guidelines. See my previous post for two papers where I've attempted to follow these guidelines as much as possible.