There are three common ways to refer to a book in this field, and people argue about which is best as though one has to win. They do different jobs. A numeric identifier addresses an item, a tag filters a group, and a title describes a thing. Comparing them properly makes it obvious which to reach for in which situation, and the answer is usually more than one.
Three tools, three jobs
| Code | Tag | Title | |
|---|---|---|---|
| Points at one item | Yes | No | Usually |
| Describes content | No | Partly | Partly |
| Works across sites | No | Loosely | Yes |
| Survives a closure | No | No | Yes |
| Short to type | Yes | Yes | No |
Read down the columns and the trade is clear. Codes buy precision at the cost of portability. Titles buy durability at the cost of length and occasional ambiguity. Tags buy grouping at the cost of accuracy, for the reasons on how tag taxonomies really work.
Where codes win
Exactness. When several translations of one book exist with slightly different titles, an identifier points at precisely the one you mean. No amount of describing beats that when precision is what you need, which is mostly in short-term conversation.
Precision matters for another reason too. Reposts and re-encodes circulate widely, often with pages missing or in a different order, and an identifier distinguishes the version you actually read from the six copies that resemble it. A title cannot make that distinction at all.
They also win on typing effort, which matters more than it sounds on a phone at one in the morning. Six digits versus a forty-character romanised title is not a close contest.
Where titles win
Everywhere that involves time. A title can be searched next year on a site that did not exist when you wrote it down. It can be recognised by someone who read a different translation. It survives a library closing, which in this field is not hypothetical, as set out on the page about link rot.
Titles also carry a little meaning. Not much, but a title tells you something about tone in a way a number cannot, and occasionally that is enough to decide.
The weakness is ambiguity. Short generic titles repeat, and translations diverge, which is exactly the gap an identifier fills.
Where tags sit
Tags are the only one of the three that helps you find something you have not already heard of. That is a genuinely different job and the other two cannot do it at all. It is also the least reliable of the three, because tags are applied inconsistently and inflate over time.
They also have a social function nobody plans. A tag with a strong community around it becomes a place rather than a filter, where people who like the same thing accumulate and start recommending to each other. That is worth more than the tag’s accuracy and it is entirely accidental.
Used as a first pass and then abandoned, they work well. Used as a description of what a piece is like, they mislead, which is the argument on what a tag really promises.
Using all three
Share a code with a title. File a title with an artist. Browse with tags and then judge by description. Each tool covers another's weakness, and the combined effort is roughly one extra line of typing per item.
Doing all three feels redundant right up until one of them fails, which happens on a schedule. The site moves and the code dies. The translation changes and the title stops matching. The tag inflates and stops filtering. Any two surviving is usually enough to recover the third.
For a personal list, title and artist are the two that must be there; the identifier is a convenience that will eventually stop working. That is the record-keeping argument on building a gallery of your own.
Picking the right reference for the job
Precision now: use a code. Durability later: use a title. Discovery of the unknown: use tags. Nobody has to choose one permanently, and treating it as a debate has kept a lot of conversations going long past the point of usefulness.
Whatever you point at, the standard holds: fictional adults, unmistakably grown, credited makers, nothing scraped. That is on the understanding of this house.