Every large catalogue advertises the same thing: scale. Thousands of entries, everything in one place. Past a certain point that stops being a benefit, and most readers have felt the moment it flipped.

Why more stops helping

A catalogue with a hundred entries can be browsed. One with a hundred thousand cannot, so the interface becomes the product and the collection becomes a backdrop.

At that scale, discovery depends entirely on sorting, filtering, and recommendation. If those are weak — and they usually are — the reader sees the same few hundred popular entries repeatedly while the rest of the catalogue is functionally invisible.

So the marginal entry adds nothing for the reader. It adds a page for the site, which is a different interest entirely — the browsing problems set out in our piece on browsing versus searching.

What curation actually means

Not filtering. A filter removes things mechanically; curation is a person choosing a small set and saying why they chose it.

That last part is what makes it useful. A recommendation with a reason attached lets you decide whether the reason applies to you, which a ranked list never does. Ten entries with a sentence each beat a thousand sorted by popularity.

Reasons also survive disagreement. If a curator's taste turns out to differ from yours, knowing why they chose something still tells you whether to look — which is more than a five-star average has ever managed for anybody.

Why so few sites do it

Money, mostly. An advertising-funded catalogue earns per page view, so more entries is straightforwardly better regardless of whether anyone can find anything — the funding pattern described in our piece on what free actually costs.

Curation also costs labour that produces no additional inventory. Someone has to read the material, form a view, and write it down, and that person has to be paid whether or not the recommendation performs.

And it creates accountability. A curated list is an argument someone can disagree with, whereas a popularity ranking is nobody's opinion and therefore nobody's problem.

What readers do instead

They curate for themselves, usually by following artists rather than sites. An artist is a coherent body of work with a recognisable sensibility, which is exactly what a catalogue isn't.

That's why artist credit does so much work here — it's the reader's main curation tool as well as the creator's income link, as we set out in our piece on who actually gets paid.

Personal collections are the other substitute, and they work best when kept small and annotated rather than exhaustive.

Recommendation between readers does the rest. Most people find things through someone they trust mentioning them, which is curation performed informally and for free, and it works better than any ranking algorithm this field has produced.

What curation can't replace

Choosing well doesn't substitute for moderation. The floor holds regardless of how carefully a list was assembled: characters are fictional and unmistakably adult, with no ambiguity about age, no real-person sexual depiction, and no coercion presented as romance.

Nor does it replace metadata. A recommendation without title, artist, year, and format is a dead end the moment the link breaks — the point in our piece on metadata. Our standards are in the control room, and connected creation with real account and moderation controls stays on models.

What a smaller catalogue could offer

A site that held five hundred entries, described each one properly, credited every artist, and explained why anything was included would be more useful than almost any large catalogue in this field.

It would also be harder to fund, which is the whole reason it doesn't exist. Until the economics change, the practical answer for readers is to stop treating catalogues as a source of judgement and use them only as a lookup — then get the judgement from artists directly.