There's a difference between pointing at something and describing it. A catalogue number does the first. Everything you actually need to know comes from the second, and collections that skip it end up as long lists of numbers nobody can interpret.

Pointing versus describing

An identifier is a pointer. Its only promise is that it leads to one record inside one system. It doesn't claim the record is good, legal, correctly credited, or still there.

Metadata is the description attached to that record: who made it, when, in what language, in what format, about what. This is the part people search, filter, and sort by, and it's the part that keeps working when the pointer breaks.

The failure mode is a collection of hundreds of numbers with no descriptions. Two years later the catalogue has re-indexed, and the whole list is unusable — the decay process covered in our piece on why shared codes stop working.

The six fields that do the work

You don't need an elaborate schema. Six fields cover almost every practical need.

Tags deserve their own care, because most tag use is exclusion rather than search — the argument in the piece on how genre tags work.

Format is the field people underrate most. An animated short, a serialised release, and a printed comic ask for completely different attention, and a list that mixes them without saying so is a list you will misread every time you open it. One word per entry fixes that permanently.

Artist credit is not optional

Of the six, artist credit is the one most often dropped, and it's the one that matters most. A number with no artist attached makes the work anonymous, which suits sites that never licensed anything and harms everyone else.

Credit is also the most useful field for a reader. If you liked something, the artist is the thread that leads to more of it. A tag leads to a hundred unrelated works; a name leads to a body of work.

Stripped credit is usually a signal about the site you're on. Places that remove it tend to be the same places that skip age verification and host material nobody has reviewed, which is worth noticing before you go looking for a dead number.

What a usable personal collection looks like

Plain text is fine. One line per entry, six fields, and the catalogue number at the end as a shortcut rather than the anchor. Sorted by artist, it becomes something you'll actually reread.

Add one more column if you can: a short note on why you kept it. Collections without that column grow until they're meaningless. Collections with it stay small, which is the point.

Keep the file somewhere that does not depend on any single service. The whole reason numbers fail is that they live inside somebody else's database, and a record stored in the same place inherits exactly the same weakness.

The same discipline applies at scale. A catalogue that shows full metadata beside every identifier, retires numbers permanently, and gates explicit entries behind verified age is a library. One that shows a number and a thumbnail is a pile.

Why numbers alone never hold a collection together

Numbers are cheap to store and cheap to share, which is exactly why people over-rely on them. The cost arrives later, when the pointer stops pointing and there's nothing underneath it.

Write the description. Keep the credit. Treat identifiers as a convenience layer on top of a real record, and your list survives the next time a catalogue rebuilds itself. Our editorial standards for adult material sit in the control room, and connected creation, accounts, and moderation stay on the source platform through the production booth and models.