Using speech recognition software adds to the likelihood. Dragon, for example, is really bad at its/it's, despite this almost always being eminently clear from the context. It's a weakness in the contextual match engine which remained in place last time I upgraded, though I've skipped the last two releases as they have become increasingly expensive for diminishing accuracy returns over the years. Maybe its [sic, for lols] fixed now.

Matthew Clapham
Matthew Clapham

Written by Matthew Clapham

Professional translator by day. Writer of silly and serious stuff by night. Also by day, when I get fed up of tedious translations. Founder of Iberospherical.

Responses (1)