4A. Populate the Category List

There are several reasons for pre-populating the Category list in FLEx before importing:

• By looking at the list carefully, you can see which items need to be normalized (correct spelling, consistent choice of abbreviation and punctuation).

• Some categories have been localized into other analysis languages.  If you create the category by selecting it from FLEx's Category Catalog, then it will be pre-populated with all the translations that are available for the analysis languages in this project.

• If you create the category from FLEx's Category Catalog, there will be a GUID (identifier) attached to the category.  Although this is not visible to the user, it could be used to facilitate cross-linguistic research in some applications in the future.  (Obviously, this doesn't apply to categories that are not in the catalog, that you have to create as custom categories.)

• Asking the linguist to review the list gives them a chance to see if they are happy with their choices of which categories to include, and what abbreviations to assign to them, thinking both in terms of their analysis, and in terms of what will appear in a "list of abbreviations" in the front matter for the dictionary.

When setting up the categories in the empty project that you will be importing into, it is good to fill in both the Abbreviation and the Full Name.  Again, this gives the linguist a chance to look at their list and confirm it.  If there is more than one analysis language, this is a good time to ask the linguist to provide translations of the categories (both full name and abbreviation) in the other analysis language(s), or to confirm the ones that FLEx supplied, for items that are already translated in FLEx.

When you create a category by choosing it from the Category Catalog, keep in mind that you are not required to keep any of the information as supplied. There is still value in getting initial values for Name, Abbreviation, and Description, but you should feel free to customize it for this project.  There are several ways you might customize it:

1. The most important step is that you need to change any abbreviation that does not match what is used in the data for that category.  For instance, in FLEx, the default abbreviation for "Pronoun" is "pro".  However, many linguists have used "prn" instead.  For the purposes of the import, it is much easier (and more reliable) to change the abbreviation once in the Category List, rather than changing it in all of the data.  After the import is done, if the linguist wants to consider using "pro" instead of "prn", they can (for instance, if they want to follow Leipzig Glossing Conventions).  But it is better to let the linguist make that decision after the import is done.  For now, the priority is to get the data into FLEx reliably.

2. It can also be helpful to adjust the Name to match how they think of it.  For instance, they may prefer "Verb Intransitive" instead of "Intransitive Verb".  Although the linguist can easily change this after the import is done, it is nice if you can make it match their preferences before you do the import.  There will be enough things that look strange to them as they begin to use FLEx; it is helpful if this won't be one of them.

3. Some of the categories are "close approximations", or use different terminology.  For instance, the term called "Connective" in FLEx is more often called "Conjunction" by many linguists.  If the linguist has a category abbreviation "conj", you could add the category "Connective", and then change the name from "Connective" to "Conjunction".  You might also tweak the Description related to this:  Instead of "Also known as a conjunction, a connective is a class of..." you could change it to read "Also known as a connective, a conjunction is a class of..."

The Category Catalog does not claim to be exhaustive.  There are many common category labels that are missing (for instance, Copula or Negative Particle).  Thus, you will inevitably need to create some custom categories.  To do this, use the blue link at the bottom of the Category Catalog that reads "The category I need is not shown in the Catalog.  Create a custom category."  Then fill in the Name, Abbreviation, and possibly Description, and place it at the correct place in the hierarchy (for instance, a Negative Particle might be a child category of Particle).

When you do the import and review the log file, if it tells you that it had to create some categories, then you will know that you didn't populate the list completely.  Look at the ones it created, and determine if you just forgot to normalize some of the data, or if you need to create new categories in the empty project before importing again.