A little background before my question: I am an amateur in mid-development of a logographic writing system, not necessarily a whole language with phonetics. As I develop, I don't so much care about grammar, at least not yet. The language is heavily context-based, it requires you to have foreknowledge of the concepts it conveys, and ideas portrayed in the writing system are to be implied by the reader. It's memory-intensive, so you can't quite analyze it like, say, English, and determine its meaning through some roots or affixes. In a nutshell, virtually no grammar. No specific way to order your ideas, just implied context.
Ideas are distinguished by category: nouns, verbs, adjectives. (That should be used as a dominant identifier for the question yet to be asked)
Okay, that'll probably do it for background. My question is this: Is there an advantageous method to sort and organize characters of a logographic system, specifically for dictionary look-up? (Given the properties I mentioned above.)
My current method is just sorting by the kinds of ideas. Ex.: In the verb category, I may sort verbs whether they describe positional things, e.g., to be underneath, or motional things, e.g., to run, etc. For nouns, a similar system may be employed (haven't reached this point yet), as well as for adjectives.
However, I imagine that such an approach would/could create confusion. (Is an ice cream cake a cake or an ice cream? Why?)
Certainly, there must be other options I'm not seeing. This question, for example, seems to brush the idea just a smidgen, but it is centered largely on computer programs for organization. For this, I should specify that I am working entirely on-paper. Computerization is not an option quite yet. I am strictly searching for methods for organizing a kind of lexicon, entries similar to those arrayed in an English dictionary.