Notes on the Electronic version

Motivation

In the course of my research, I needed to look up the definitions of many archaic terms, as I was using material dating from Linnaeus onwards. Many of these terms are not present on modern entomological dictionaries. Eventually I was directed to Jardine 1914, which was, of course, out of print. A photocopier solved my immediate problem, but did not permit the “cut and paste” required by modern document preparation, and it left much to be desired in the way of presentation.

Why a CD-Rom?

This is not a facsimile edition. It is an electronic reconstruction, allowing the modern researcher to “cut and paste” from a web page or word processor document. This was not a simple undertaking, as many processing steps were required to get from the paper original into the electronic form you are looking at now. The electronic form is readily search-able but not always convenient - once in electronic form, however, it is a simple matter to print paper copies as required.

By making the material available on a CD-Rom, it is accessible to researchers on the majority of computers. The material is presented in a format accessible not only to PC users, but also to Macintosh users and most other alternatives.

The Process

This work was created by scanning the individual pages into a computer. At this point there are only pictures of the pages in the computer.

Optical character recognition (OCR) software was then used to process the scanned page images and produce approximately equivalent text. Because of the complexities of reading, and the wide variation in typefaces over the last 87 years, this is not as simple as you may imagine.

The text was then checked for obvious recognition errors and converted to HTML format. This is the format commonly used on the World Wide Web. It provides a rich set of ligatures (things like æ) and it is accessible on a wide variety of platforms, not just the PC. (The only fly in the ointment is footnotes - HTML handles these badly because it has no concept of a page - compromises were made to keep them proximal to the reference and yet separate from the text.)

At this point, icons such as the one you see to the right were added. These icons mark the start of a page of the original. In this way you are able to tell exactly where in the original work the text comes from. By clicking on these icons, you are taken to the scanned image of the original page, for comparison.

The text was then processed by a program to insert “anchors” for each of the words or phrases defined by the dictionary. This is a method permitting the computer to find an exact location within the text. Such hints are necessary, because computers can't actually read, so they can't tell a definition from a reference. There was a degree of manual processing required after this step, due to typesetting inconsistencies in the original work.

The text was then processed by yet another program to locate all references within the bodies of the definitions to words and phrases defined elsewhere in the dictionary. These references were then “linked” to the definitions, in the same manner as World Wide Web pages. Thus, as you read the electronic version of the Dictionary, you are able to click on an unfamiliar term and be taken directly to its definition. This step required a great deal of manual work, due to things like (a) the computer couldn't tell that plural and singular forms are actually the same word, further complicated by the presence of several languages, (b) the computer couldn't recognise abbreviations, and (c) there are a number of spelling errors in the original (hardly surprising when you consider the book was typeset manually and the typesetter was working with technical jargon and several languages).

The final step was to take the web pages and assemble them into a Microsoft Word document. This was not essential, but it gives a simple way for users to print the whole book. This also allowed the footnotes to be treated properly.

The labor involved was about 4 hours per page, approximately 1000 hours for the whole book. This was spread over evenings and weekends for the best part of a year.

At the end of this tale, I must add that I have nothing but admiration for all of the work Jardine did to prepare the original using only pen and paper. Even with modern computerised marvels, simply reformatting and representing his work was a big job. Type setting it manually must have been a formidable task. However, this is easily dwarfed by the labor of assembling definitions and then cross checking by hand and eye and wit. What an astonishing accomplishment! I can only hope that my efforts have done Jardine's work justice.

Variations from the Original

As mentioned above, this is not a facsimile edition. Once the manuscript was in electronic form, a number of decisions had to be made concerning the text.

Jardine was very inconsistent with plurals. The majority of the time they were given as “singular, pl. plural”, however at times they were given as “singular, plural”, or “singular, (plural)”, or “plural, sing. singular)” e.g. epimera. They have all been regularised to use the first form.

Jardine used a number of terms in the definitions which are not common English, and which he failed to define. Where entries have been added, they are annotated with [PM] to make this clear. Entries have only been added for words or phrases used in definitions present in Jardine's original text.

Jardine didn't always sort the definitions into strict alphabetical order. This problem has not been fixed. The order in which words and terms appear is the original order used by Jardine.

There are numerous instances where the body of a definition used an apparently non-existent word, which was very similar to a word defined in this Dictionary and, by context, appeared to have meant to use the defined word. Such instances are assumed to have been an error, and have been replaced by the defined word.

In all cases where an alteration has been made, only the reproduction (HTML and PDF) has been changed, the facsimile images are unaltered.

Peter Miller,
9 December, 2003