Why you shouldn't parse the output of ls(1)

The article doesn't mention "ls -1" at all, which has served me quite well in various more or less complex scripts over the years.

Or how about making a file name out of one complete line from ‘ls -l’?