Any non-ASCII characters in the output of a build script (batch script, calling a PowerShell script calling MSBuild - which are fully unicode) are not displayed in the web console output or in generated e-mails. Instead, that little box with the inverted question mark appears, which usually marks an encoding error. It seems that Jenkins converts the script output from Unicode to Windows1252 (or something similar) and then passes it on to something expecting UTF-8. In any way, since PowerShell and MSBuild generate Unicode output, there should not be any encoding issues or character set imcompatibilities at all. I'm not sure how good Java performs in this, though.