Gilles Fauconnier

Grabbed this book mostly due to an old Norvig article mentioning the potential promise of its contents before publication. I think there's plenty in here to satisfy anyone who's unsatisfied (or just bored) with modern statistical language models; at least in their attempts to imbue machines with any sort of linguistic understanding.

The structure's easy enough to follow, definitely has some parallels and connections with [a:Lakoff George|14868465|Lakoff George|https://s.gr-assets.com/assets/nophoto/user/u_50x66-632230dc9882b4352d753eedf9396530.png]'s [b:Metaphors We Live By|34459|Metaphors We Live By|George Lakoff|https://i.gr-assets.com/images/S/compressed.photo.goodreads.com/books/1388194058l/34459._SX50_.jpg|34433] that make a few of the earlier concepts a bit easier to digest. There were a few places where I felt notation was a bit awkward or confusing in that it already has meaning in the technical languages of other tangential fields of study (eg. mathematics, computer science) but this by no means reduced the overall presentation.

I hesitate to say much on the _theory_ mostly because I have yet to follow its development up to the present and as such, don't want to critique concepts or ideas that may be addressed or expounded upon in more recent developments.… (mehr)