1) Make a UTF-8 database. a) Make a new database b) Edit the user options (uo from main menu) c) Add the line "codeset=UTF-8" (without the quotes) 2) Set your TTPATH in your config file (.linesrc on UNIX, lines.cfg on MS-Windows) to include the tt subdirectory from the distribution. UNIX sample#1 TTPATH=/home/joe/lifelines/tt UNIX sample#2 TTPATH=/usr/local/lifelines/tt MS-WIndows sample TTPATH:=C:\lifelines\tt [NB: The := syntax is to allow single backslashes to work] [If you don't have iconv, this gets you codeset translation tables for some popular codesets. Whether you have iconv or not, this gets you codeset translation tables for ANSEL and HTML entities.] 3) Run the self-test regression test st/st_all (or on MS-Windows, st\st_all). The third prompt is Do you want to output to a file (0/1), and answer 1 for that, to get a file copy of the results. Let me know how they turn out (or, certainly, feel most free to try debugging any failures!) 4) Create some users with some non-ASCII characters, in name, or anywhere else. Play around and see if things work. Try setting different codesets for report and gedcom output, and see if they come out correctly in the requested codeset. Eg, ReportCodesetOut=ASCII//html (should get html entities for non-ASCII characters) GedcomCodesetOut=ISO-8859-2 (should get only East European standard codeset characters)