Encoded from the sources with the greatest care for quality

In order to make sure that our database corresponded exactly to the original sources we proceeded as follows.
One team encoded only the symptoms into an Excel file. Once a book was fully encoded, we converted that file into an empty Repertory, which could be used with Radar.
Another team then added the remedies to these empty rubrics. In the process of doing so they checked the symptoms once more against the original book.
When this new repertory was finished we printed it in a repertory format. A third team then checked this printout against the original book. Symptoms and remedies had to correspond 100 %.

One difficulty we encountered is that different books used different remedy abbreviations. In addition, these abbreviations were often different from the ones used in Synthesis.
To avoid this apparent source of mistakes, we created special remedy catalogues for each book. As a result, the person who encoded remedies could encode exactly the remedy abbreviations as in the printed book. In bg2, for example, one would enter "aco., alu., ap., bro., cocl., etc." instead of "acon., alum., apis, brom., cocc., etc."
The person who checked the final printout, had to check for the exact correspondence of the remedy abbreviations between the printout and the original book.

After this procedure the remedy abbreviations were converted to the Synthesis format so that each repertory now uses the same, known Synthesis remedy abbreviations.
As a result we have a separate repertory for each book that exactly corresponds to the original source.