Details of hand-crafted and categorised database of 10,000 tulips created for the Mosaic Virus project. 

The politics and process of creating database is something that I am very interested in, particularly in relation to machine learning. Care needs to be taken. If it is too big the results will be too good and the the quirks and oddities that make it an interesting medium to explore will disappear. If it is too small it will not have enough information and become flummoxed, either producing nothing or one or two variations from the training set again and again. The process becomes almost like craft - repetitive, time-consuming but necessary in order to produce something beautiful.