As a professional editor, I annually edit about 1.5 million words of research on computer science and engineering topics. The tools I use must work well, and believe me, I’ve tried a few! In this article, I share the platforms and programs I frequently use for editing research, which I suspect would be useful for writing research as well.
Microsoft Word and LaTeX
In the good old days when research workstations were real Sun Sparcstations and monitors were black and white and shaped like sugar cubes, if you wanted a computer-science research paper to look good, LaTeX was your only option. Although I have a lot of affection for LaTeX, decades have passed since then, and Microsoft Word is now my primary editing tool. I often edit LaTeX documents directly, but every company I work for actually converts the TEX files to Word documents first.
This is primarily because of Word’s tracked-changes feature, a tool that marks deleted text by changing it to red strikethough font and inserted text to red underlined font. As far as I know, there is nothing comparable elsewhere (OpenOffice’s Word clone excepted).
Also gone are the days when precise control over documents and the editing interface was not possible in Word. Although many quirks remain, templates and macros allow you to customize your document to a high degree. In fact, many journals now accept Word documents and even provide Word templates.
LaTeX is still often used for papers with heavy mathematical content. It is also used by researchers who do not wish to use Microsoft products for various reasons. It outputs beautifully typeset documents, but has a steep learning curve. Currently, I use TeXworks with MiKTeX. There are many options, but this one is adequate.
If you do use LaTeX, TeXonWeb is good for checking short snippets of TeX code without having to start a whole file or project in TeXworks.
Tools for References, Journals, Conferences and Referee Suggestions
For checking references (or finding them), there are also many options. PubMed has a lovely search interface, but it is primarily for biomedical research and does not include some computer science sources. JANE is also excellent for finding potential reviewers, journals, and related work using the abstract of your paper. Unfortunately, it only searches PubMed.
In contrast, Google Scholar is very inclusive (and accessible without an institutional license). Search in Scholar can be frustrating and fiddly, and the trick is to do a general search using the title text only. Scholar’s references also often contain typos and incorrect/missing information, so if there is any question of an error, it is best to cross-check the information against the original document.
To find journal abbreviations, Journal Title Abbreviations is the most comprehensive interface for the ISO 4 general standard, but very slow and fiddly! I download the data to my computer, delete the non-English content, and search it using Excel. IEEE has their own list of abbreviations. It is important to be careful with abbreviations because ISO 4 is not the only standard, and many incorrect abbreviations on the Internet.
Finally, I use two additional tools. Neither are free, but they aren’t too expensive. PerfectIt is a Word plug-in for checking Word documents. It checks the consistency of terms (e.g., “back-propagation” versus “backpropagation”) and abbreviations. It can be customized. I also use FlashPaste, which will insert customized text into a document. It can insert some limited control characters, so performing tasks such as inserting a Word comment with specific text is possible. This is much easier than writing the equivalent Word macro from scratch!
The quest to increase productivity, especially in the gig economy, is a never-ending process. If there are any tools you think I’ve missed, please let me know!