One of the greatest productivity boosts you can get from using Instant Text comes from its ability to generate custom-made glossaries based on what you have typed in the past. This process is called Glossary compilation and is entirely automatic. It allows you to obtain glossaries that reflect the subjects on which you are writing, with their frequent terms and phrases. With compilation, it is no longer necessary to guess what words and phrases you might need in the future. You only need to let Instant Text analyze what words and phrases you’ve used in the past.
Glossary compilation is a very fast process: Creating a glossary with thousands of entries can be done in seconds on current computers, whereas it would take weeks to do it manually. So this is something that you can do very easily every time you need to cover a new subject or a new specialty. In addition, you can easily control how the compilation works. For example, you can set thresholds to retain only words that appear with a certain frequency.
There are two main ways to compile a glossary. If you have a large file covering a given subject, you can use the process called Single File Compilation. For example, you may ask Instant Text to compile a 250-page document on Copyright Law to obtain a glossary on Copyright terminology, or a lengthy request for proposals to produce a glossary with the client’s terminology for preparing your response.
The second form is called Multiple Compilation, and it consists in compiling several documents that have a common subject and that are in the same folder. For example, you may ask Instant Text to analyze the files of a folder of medical reports by a given doctor or covering a given specialty, to produce a glossary of frequent words and phrases used by this doctor or in that specialty.
Both forms use similar techniques. Multiple Compilation actually produces a temporary combined document and the analysis then proceeds as for Single File Compilation. In both cases, documents may be in a wide variety of formats and the first step is a conversion to Ansi text format before processing the contents.