Most steps have LogMessageOnFailure as an option. But there are times, particularly on Extract steps, where I'd like to log a message or value even if that step doesn't fail. Unless I missed something, in order for me to know the value extracted to a variable on an Extract step I have to write a coded step to log the value. Troubleshooting without knowing the extracted value sucks and so does having to write a coded step to find out the extracted value.