You are here: Articles --> 2001 --> Effective
onscreen editing. Part three of a four-part
Vous êtes ici : Essais --> 2001 --> Effective onscreen editing. Part three of a four-part series.
by Geoff Hart
Previously published in a different form as: Hart, G.J. 2001. Effective onscreen editing. Part three of a four-part series. Corrigo, Newsletter of the STC Technical Editing SIG. 1(3):1, 4–6.
Click here to read Part 1 of this series.
Click here to read Part 2 of this series.
One nice thing about word processors is that they make it much easier to find and fix the inevitable errors that arise from authors who can’t spell and editors like me who get fumblefingered under deadline pressure.
It often helps to start your edit with a quick use of the spellchecker, since you can fix global problems (e.g., replacing British spellings with American equivalents for an American audience) that would otherwise interrupt the flow of your editing if you had to correct each instance manually; you also avoid annoying authors by requiring them to review dozens of simple corrections that don’t really require their approval. But use this tool with considerable skepticism, since spellcheckers remain fairly primitive in their capabilities. For example, they can’t detect the following problems:
Before you do any spellchecking, confirm whether the author has excluded blocks of text from being checked; authors often do this if (for example) they’ve included foreign-language text or a large expanse of jargon that wouldn’t appear in the regular dictionary, and haven’t installed the appropriate dictionary to check the spelling of these sections. (Using the standard dictionary would result in every word being flagged, and that’s sufficiently time-consuming that some authors can’t be bothered to check the spelling.) Fortunately, the word processor’s search function can often find protected text for you so you can manually check the spelling. The best word processors even let you define what dictionary they should use for each block of text, which is a godsend if you’re producing a bilingual publication with both languages in the same file.
Spellchecking a document before you begin editing substantively can save time, but it doesn’t eliminate the need to repeat the spellcheck once you’ve completed your editing; it’s all too easy to add your own typographical errors when you’re in a hurry.
If you work in a field with specific jargon, you can improve the effectiveness of your spellchecker by adding your own words to the software’s personal (or “custom”) dictionary or by purchasing specialized vocabularies to supplement the software’s original set of dictionaries. If you choose to add words to your personal dictionary, double-check your spelling of the new words using a printed dictionary. Many authors (and the occasional editor) inadvertently add misspelled words to their personal dictionaries, thereby ensuring that the spellchecker always accepts that particular typo in the future. The specialized subject-matter dictionaries (e.g., foreign languages; medical or legal terminology) available for many word processors greatly improve spellchecking for editors who work in those fields. Your software’s developer can generally provide a list of the dictionaries they supply, as well as a list of third-party developers who produce specialized dictionaries.
In contrast with spellcheckers, modern grammar checkers provide surprisingly little benefit even for unskilled writers, and often suggest incorrect changes. Occasional discussions of this subject on the copyediting–l discussion group for editors and the techwr–l discussion group for technical writers have thus far failed to identify any grammar software that can take the place of a good editor, and this will likely remain the case for many years yet. Even so, don’t write off grammar checkers completely: most let you customize which rules they use when they perform their checks and—more importantly—which rules they ignore. This flexibility lets you select only rules for types of problems that you’re weak at detecting; for example, if your particular blind spot involves sentences written in the passive voice, leaving only that rule selected will help you find problem phrases you might otherwise miss so you can decide yourself whether to fix them.
Using the built-in search and replace function is one of the primary advantages of doing onscreen editing; it’s very productive when you can find a single error or inconsistency and then search the entire document to find and correct all other instances of the problem, confident that you won’t miss any. With a little creativity, you can find lots of other ways to use the search and replace function to improve consistency. For example, in scientific or academic editing, editors must confirm that all the references cited in the text appear in the bibliography, with the correct author names and dates. (They may also have to do the reverse, and ensure that every reference in the bibliography has been cited somewhere in the text.) To do this, use the search tool to specify the pattern you’re searching for, such as “19” if all references resemble “Hart (1999)”; in other reference systems, you may have to search for different patterns, such as “[“ if the references in the text resemble “”. To do these checks, move to the top of the document, find the first reference, then immediately check that it’s present in the bibliography; put a checkmark beside it so that you can scroll through the bibliography and thereby identify which references haven’t been cited. Use the “find next” function to find the next reference, and continue in this manner until you reach the end of the file. Keeping the references open in one window and the main text open in another greatly facilitates this technique, but works best if you have a large monitor.
When you do find a problem that you decide to fix throughout the document, how can you return to where you paused to fix the problem? Insert a “bookmark”! Most software has a bookmark function, but it’s just as easy to insert your own bookmarks by typing [-] or some other combination of characters that won’t appear in the text. When you’ve finished your consistency check, simply find that bookmark again with the search function, delete it, and pick up where you left off.
Search and replace is particularly helpful if you want to use a certain word in one type of situation, but not another, and need to be consistent in this usage. Once you find a questionable word usage, you can stop what you’re doing and search the entire file to find every occurrence of that word. If it’s a simple problem, you can automatically replace the word everywhere in the file; if not, you can instead search through the file one instance at a time and decide whether to make the change each time.
If you choose to replace a word everywhere, do so carefully. An inauspicious choice for the “find” part of the “find and replace” equation can result in the replacement of far more words than you’d anticipated, particularly if the document includes a bibliography; although you can sometimes exclude the bibliography from the replacement operation, this step is easy to forget. The problem arises from “stemming”, in which you specify a search word that forms the root of one or more other words; in that case, both the search word and all other words that contain that stem will be changed. For example, a colleague once replaced all instances of “day” with “night” to correct a problem involving time; unfortunately, she forgot to type a space in front of “day” when she entered it in the Find field, and thus changed all days of the week too (e.g., Friday became Frinight). If you opt for “replace all”, pay close attention to the number of changes the software reports once the search is complete; if the number is larger than you expected, you may have chosen the wrong word, and most word processors will let you undo the change if you do so before you begin any other editing. To avoid this kind of problem, you can search for the text to be replaced before you actually commit to doing the global search and replace, or search for and replace one word at a time so you can confirm each case individually.
As you become more expert in using the software, you can learn to constrain the search to find only exact matches (e.g., only the whole word, only the identical pattern of capitalization, only italicized forms of the word) and thereby further reduce the risk of error. Fortunately, search and replace errors are usually obvious during subsequent editing or spellchecking, so whenever you use a global search and replace, be sure to leave time for a final pass through the manuscript. Never do a large-scale search and replace as your final editing task!
The search function offers far more power than simply finding words, however; you can often find specific named styles, hidden characters such as page breaks, and even formatting of individual words (e.g., italics). So not only could you find the name of your product, but you could also find all instances where you’ve forgotten to italicize the name. Combined with features such as “wildcards” and “regular expressions”, which let you describe patterns of characters rather than just specific words, you can even do sophisticated “find anything that resembles this word” searches. For example, a search for “edit*” would typically find “editor”, “edited”, and “editing”, whereas a search for “edi?” would find “edit” and “Edie”. (Details on how this works vary among programs; consult your software's user manual for details.)
Click here to read Part 4 of this article.
To learn the details of onscreen editing that are only summarized in this article, see my book Effective Onscreen Editing.
©2004–2017 Geoffrey Hart. All rights reserved