Ambiguous, or subjective, organization schemes are tailored to a specific information space's content and audience rather than relying on universally understood categories. They include topical, categorical, audience-specific, task-oriented, and metaphorical schemes. Despite being harder to design and more demanding to maintain than exact schemes, ambiguous schemes are generally more useful to people, because they support exploratory browsing, the user doesn't need to already know exactly what they're looking for.
Topical schemes are the most common and broadly useful, though devising clear topic boundaries for a large, evolving content collection is genuinely difficult. Task-oriented schemes work well as a primary scheme for a single-product company's support site, but multi-product companies typically need a topical primary scheme with task-oriented as secondary. Metaphorical schemes are rare and generally not recommended: they're difficult to sustain as content grows, since not everything fits neatly into a chosen metaphor.
An ambiguous scheme costs more to build and more to maintain than an exact one, but the payoff is a space that works for people who don't yet know the right words for what they want.
Topical is correct. The content itself, codes, materials, case studies, continuing education, is exactly the kind of broad subject-matter variety a topical scheme is designed to organize, and it supports exploratory browsing for members who don't have one single fixed target in mind. Metaphorical is a weak fit here specifically: the source material is explicit that metaphorical schemes struggle to scale and sustain themselves as content grows, and a "building structure" metaphor would strain badly trying to accommodate something like continuing education requirements, which has no natural physical-structure analog.
Choosing topical comes with a real, ongoing cost: someone, an information architect, or whoever is filling that role, has to continually decide where each new code or case study fits, refine topic boundaries as the field evolves, and watch for topics becoming either overloaded or too fragmented as the collection grows. This is precisely the trade-off the source material describes: ambiguous schemes are harder to design and maintain than exact ones, even though they serve users better.A reasonable way to manage this over time is periodic review, scheduling regular audits of the topic structure as new content accumulates, consolidating topics that have grown too crowded, and splitting or renaming ones that have become too broad, rather than letting new topics proliferate indefinitely without anyone revisiting the overall structure.