Sites that hold large collections of adult comics and animation get lumped together, but they fall into two quite different categories with different goals. Telling them apart is useful, and it takes about a minute.

What an aggregator optimises for

An aggregator exists to collect traffic. It gathers material from wherever it can, presents it in a browsable grid, and monetises the visits. Its incentives are volume, page views, and search visibility.

That's not automatically bad — plenty of useful services work this way — but it means preservation is not a goal. Entries get added because they attract visits and removed when they cause problems, and nothing about the database is designed to last.

The tell is metadata. Aggregators generally show a title, a thumbnail, and a number, because that's all the scrape captured and nothing more is needed for the business model.

What an archive optimises for

An archive is trying to keep things. Its identifiers are stable, retired numbers stay retired, old links redirect rather than break, and the metadata is complete enough to identify a work independently of the site.

Archives also care about provenance: who made this, when, in what edition, from which source. That's the difference between a record and a copy, and it's the difference that matters in three years.

The two categories can look identical on a phone screen. Both show a grid of thumbnails, both have a search box, both hold a lot of files. The difference only becomes visible when you open one entry and see how much the site actually knows about it.

Those six metadata fields are the ones we set out in the piece on the metadata a number can't carry, and their presence is the fastest way to sort one category from the other.

Why the distinction matters to creators

Credit is the mechanism that turns attention into income in this field, because most work is funded by patronage and commissions rather than by sites hosting copies — the picture in our piece on who gets paid.

Aggregators break that mechanism at scale. Not usually out of malice; the metadata simply gets discarded during copying and nobody restores it, since restoring it would cost effort that produces no traffic.

Archives that maintain credit keep the link intact, which is why artists tend to be far more relaxed about them even when the material is the same.

Removal handling is the other difference creators notice. An archive with a stated takedown route and a working contact address is something a creator can deal with. A site with neither is a wall, and the material stays up regardless of who wanted it down.

Why the distinction matters to readers

References built on aggregators decay fast. Numbers get reused, entries vanish, and the site itself may be replaced by an imitator using the same name — the failure chain described in our piece on why shared codes stop working.

References built on proper metadata survive all of that, because they identify the work rather than its position in somebody's database. That's the whole argument for writing down artist, title, year, and format.

The moderation difference

Preservation and moderation tend to travel together. A site that maintains records also tends to have someone reviewing what's in them, and a site that scrapes at volume generally doesn't.

That matters because the floor is not optional: characters must be fictional and unmistakably adult, with no ambiguity about age, no real-person sexual depiction, and no coercion presented as romance. Our standards are written out in the control room, and connected creation with real account and moderation controls stays on models.

Sorting a collection site in one minute

Look at one entry. Is the artist named, and does the name link anywhere? Are there six fields or two? Does an old identifier still resolve? Does the age gate load before content?

Four questions, all answerable from the outside, none requiring you to trust anything the site says about itself. Size tells you nothing here — some very large sites are pure aggregators and some small ones keep excellent records.