FajerElements of paper writing

Elements of paper writing

  • “you write a paper when you have no more than 1/3 of data”
    Roger Cooke

General

  • Paper writing is a long process not necessarily pleasant.

Length

  • Till the manuscript is perfect.

Writer’s ego

  • Egos play no role. Do not take any criticism personally. I rewrite my own sentences many times.

Final polish

  • Go through every sentence and see whether you can say the same but more succinctly.

Style

  • Succinct and precise. Scientific manuscript is not the place for literary prose. Each statement can be interpreted in one and only one way.

Flow of argument

  • Each paragraph is supposed to lead into next one or initiate a new argument.
  • Each paragraph is supposed to make a point. If it does not, remove it.
  • Keep the sequence straight: A,B,C …. avoid comebacks to points previously made, avoid arguments of type A,B,A type

Relevance

  • Each statement has be connected with the paper. A good check is to ask whether the paper would be poorer, conclusion less clear if you threw the whole statement out. If not then throw it out.

Reference book

  • Book of all relevant references. Make sure you have tabs for easy finding of any reference

Manuscript Structure

Methods

  • Anybody must be able to repeat the experiment by following the Methods section.
  • Enough detail to repeat if the method is new
  • If the method is copied from another paper w/o modification  reference to that paper is enough
  • If the method is modified either describe the whole method from scratch (still with reference to original paper) or describe the modifications in detail.

Results

  • Poster is by far the best outline, if you don’t have one, the first step in any paper writing is to summarize the data in a poster form.
  • Need headings, problems/objectives of each experiment, results in figure or tabular form, conclusions
  • Narrative takes through objectives, experiments and controls in that order.
  • Avoid interpretation of the data if the conclusion depends on it. Data are saint – your interpretation is not. However, ….
  • End each subsection with the one sentence conclusion
  • Be explicit in description of Tables and Figures this is what this section is about – taking the reader through the Figures and Tables. Don’t assume that if it is shown in the table you don’t have to talk about it.
  • If you do not talk about piece of data in the narrative then remove it from the Figures or Tables.

Introduction

  • This is not a section to show off what know about science and everything else. Every statement, quoted data must have a direct relevance to your paper.
  • usual flow: general statement of the objective  global review of state of the field wrt the objective  statement of a specific object  illegal but useful – what we did and what we found

Discussion

  • Most fun and the most difficult part:
Summary
  • Executive summary of the data with significance: why we bothered at all ?
Relation to other data
  • Science is not done in vacuum, a lot of people did something similar, we have to give them credit
  • Emphasize what new your paper brought in, how did you extend the previous body of knowledge
Modeling/Significance
  • Discussion is about the data interpretation: you are allowed to speculate as long as it is stated as such. While the Results had to be very refrained, here let your imagination fly.
Concluding remarks
  • The only thing people often read

Abstract

  • Leave to the end writing the paper
  • The most important part of the paper shows up on the Medline. Other people on the basis of abstract will decide whether or not to read whole paper, whether or not quote your paper and when they hunt for the piece of data they read sometime ago they will look mostly at the Abstracts

Word processing

Outline

  • Use collapsible outline
  • Can use auto numbering for easy hierarchy and bullets for salient points
  • Initial version ( “0” ) of the manuscript consists of remarks: sloppy, non-grammatical statements, often not in English which conveys what you want to say. They are “for your eyes only”
  • do not write paper narrative till everything you want to say is said in the Remark style.
  • Paper writing is about translation of your Remarks into narrative - not the other way round

Style sheets

  • Format/Styles menu allows to change the attributes like: font, spacing, color, shadowing
  • Automatic update: all paragraphs with given style are changed
  • Attach specific Template file or redirect the source directory of default Normal. dot template via Tools/Options/File locations/User templates
  • Do not leave empty lines to separate the paragraphs – use the space before/after attribute of a given style
  • Remark style
  • Figure style

Mailmerge variables

  • Use mailmerge variables to denote anything what can change in the manuscript. This includes: (a) figures and tables; (b) quantities which are still being “firmed up”; (c) references unless you use Endnote.
  • Define mailmerge variable: CTRL-F9; type within the brackets “set variable_x “value_of_variable_x”
  • Refer to mailmerge variable: CTRL-F9; type within brackets “variable_x”

Figures and Tables

  • Embed them in text for easy reference. Do not use Corel Draw objects: the resolution is lost after few transfers. The PowerPoint object linking is more reliable.
  • Anchoring of objects is difficult  use Format Object/Layout/In line with text to keep figures in the same place at all times
  • For safety (lost object linking) keep the copy of figures in the separate PowerPoint file.

References

  • Endnote is easy to connect to Medline and format in journal style (or any other user defined style). However, the linking to Endnote is iffy ! Modifications (like removal of duplicates) of the reference Endnote file screws up the text embedded references

Spelling

  • Use own custom dictionaries but redirect via Tools/Options/Spelling/Dictionaries otherwise each time you change computer you loose your definitions
  • Don’t ever show me draft w/o running through spelling checker

Elements of paper writing B9/8/20001