Data My Love

One day, Dong-in contacted me. He said he wanted to collect data from Dafont. Upon reviewing the task, I realized it involved collecting all the data registered on Dafont. Dong-in often proposes interesting projects to me, and this was one of them.

The data included in this book was collected up to 4:19 p.m. on October 13, 2025. At the time of collection, the total number of data entries was 95,956, while the total number listed on the Dafont website was 96,157. There’s a discrepancy of about 200 entries—why did this happen?

Since Dafont’s website is structured to load data dynamically, it’s impossible to retrieve all the data at once. Dafont does not provide an API or any other method for retrieving data. Therefore, all this data was collected through crawling using an agent—a method that remotely controls a browser. Because new fonts were constantly being added and removed from Dafont even while the data was being collected, the sequence became out of sync, resulting in duplicates and omissions. Currently, there is a discrepancy with the actual data of 200 entries, but I wonder how much the discrepancy will grow as time goes on. New fonts will appear, and others will disappear. The data is constantly changing.

I would like to express my gratitude to the contributor who suggested this.