Google Translate uses complex neural and numeric systems to find equivalencies in speech and text across languages. The tool is seemingly simple—at least for users. But on the backend, developers worked tirelessly to get the application to where it is now. One fascinating breakthrough came when Google employees applied Wittgenstein’s language theory: a posthumously published theory that words hold no supreme meaning and that meaning is derived from a word’s use. This theory helped developers give words more descriptive titles within the multi-dimensional system the translator pulls from. Read more at Quartz.