Sunday was Day 1 of the word-collection phase of the Lotud workshop. Not all of those who were scheduled to arrive on Saturday actually showed up, so we are now down to three word-collection groups for this week, since one of the no-shows was one of the team leaders.

This first day always consists of a great deal of training, and this workshop was no exception. The morning consisted of presentation of some housekeeping items, ground rules, and the theory of what we're here to do. Then in the afternoon, we assigned activities for the participants to work on in small groups, after which we compared the results from the different groups.

There are three primary "pitfalls" that we want the word-collection groups to avoid as they work.
1) misspellings
2) non-adherence to the standard form of the word (e.g., "running" instead of "run)
3) miscategorization

The first two of these pitfalls affect the dictionary directly, while the last affects the usability of the thesaurus component of the results. If a word is misspelled when it is entered into the database, it will create an incorrect entry and the computer will not recognize that it is the same word as the correct form, entered from another semantic category. The same is true if the word is entered into the database in a non-standard form. Either of these scenarios creates a lot of work for someone later on in order to make everything correct and consistent. If a word is entered in the incorrect semantic domain (or category), searching for synonyms of words will yield inconsistent or useless results.

After three days of training for the individuals in key positions and one day of training for the entire group, I can see that we have succeeded with regard to the first two of these three areas, but we still have a ways to go on the third one. So in Day 2, before we continue with the word collection, we'll need to look back at the results of the last exercise of Day 1 and help everyone better understand how to properly categorize the words they're collecting.