Update:
Finals are over and I’ve taken the day to formulate my plan. After rereading PauloPalma’s post and further researching Dominic’s method, here’s what I’ve come up with: It seems like there are two fundamentally different approaches, alphabetical organization, which is helpful in reading but not so helpful in the productive aspect of language, and subject/grammar organization, which is precisely the reverse. Because I value the productive aspect more, I decided to go with the latter system. For vocabulary that can be easily grouped according to a particular class, I will let the definition dictate the palace used. For instance, food items can be stored in a restaurant. I think this sort of approach lends itself best to (concrete) nouns. For more abstract words, I’ll group them according to grammatical class (verb, adjective, etc.). I’ll be starting with French, and then I may extend this to Japanese.
For Japanese, I’m going to restart my Anki deck and try to come up with a mnemonic association for each word. If I’m successful then I’ll apply the same approach that I plan to use for French. If not, then I’ll just have to stick with the Anki cards and use spaced repetition.
edit I made a file to keep track of everything. Here are my notes so far:
Structure for image creation: composite image, image for semantic component and image for phonetic component
Groups considered:
Nouns (objects)
Verbs // starting palace: UNR DMS building, starting at the steps
Adjectives // starting palace: old Reno house, cul-de-sac
Adverbs // starting palace: Starbucks
Other connectors // starting palace: old Reno house, two-story
Expressions // starting palace: SM 64 level, Whomp’s Fortress
Subdivisions within nouns:
Food → BJ’s restaurant
Shopping items → Walmart
Household items → current house?
edit again It occurred to me that I can also pull from my repertoire of images from the Ben system. For instance, alors que ≈ lork = an acklay from Star Wars.