Zatanna: the announcement title cut to eighty characters

There are two ways to miss information in an ad. The first is that it is absent: we see it, we look for it elsewhere. The second is that it is present but amputated, and that is much more dangerous, because a truncated string reads like a complete string. “Robert Jor” looks like a name. “Volume” looks like a word. There is no warning that the ending is missing.

The phenomenon described here is not an editorial mistake. It's a mechanical effect: three pieces of variable-length information are concatenated into a fixed-length field, and anything beyond that is removed. The result retains the appearance of a well-formed title, with its format statement neatly placed in parentheses at the end — precisely because this statement, added last, is not cut off.

The exact measurement

I surveyed the French market on August 30, 2026 on three queries, then duplicated: 300 distinct advertisements. Among them, 176 end with a format mention in parentheses — 169 with “(Pocket)”, 6 with “(Hardcover)”. It is this family that I examine here. The amounts quoted aresums claimedthat day and do not correspond to any sales; a single day reading shows no change.

The length of titles in this family ranges from 30 to 80 characters. Twenty-one of them measure 79 or 80 characters, which means they hit the ceiling. One hundred and twenty-six end, just before the format parenthesis, with a word of six letters or less—a short word at the end of the string being the most visible sign of a break.

The ceiling is not a hypothesis: no title in the sample exceeds it, and several stop there to the nearest character. This is what we expect from a technical limit, and not from an editorial choice: an editor who abbreviates stops on a whole word, a machine stops on a counter.

On the character that concerns us, the same mechanics produces short and therefore intact lines: “Jamal Campbell Zatanna (Poche)” at €18.42 then €20.49, and the same title in “(Hardcover)” at €27.24 then €29.47. These four advertisements are readable, and they show in passing that the mention of format clearly distinguishes two different objects - which gives the latter a value that the rest of the title no longer has.

Why a truncated string is worse than a missing string

An empty field triggers a check. A completed field does not trigger any. That's all the difference, and it pays for itself later. The Collector Who Seizes “Chuck Dixon, Robert Jor” in an author field will never see an alert: the value is present, it is plausible, it looks like a name. She just doesn't fit anyone.

The consequences then spread silently. An alphabetical sort puts this false name in its place. Grouping by author creates an additional entry, separate from the same author spelled correctly elsewhere. A search for the full name does not return this line. None of these operations report an anomaly: they do exactly what is asked of them on data that should never have been entered.

The case of the volume number is even clearer, because it is encrypted. “The Eye of the World: The Graphic Novel, Volume” stops before its number; the second announcement, cut one character further, stops on an isolated letter. In both cases the buyer does not know which volume is offered to him, and in both cases the title appears complete to anyone who does not count the characters.

It must be added that truncation does not occur randomly, and this regularity aggravates the problem. As the author prefix is ​​placed at the top, it is always the last elements of the chain that disappear: the end of the title of the work, the volume number, the subtitle. In other words, the most discriminating information — that which distinguishes one volume from another within the same series — is systematically the first sacrificed, while the names of the authors, common to the entire series, are systematically preserved.

This asymmetry explains why the process goes unnoticed for so long. The truncated titles remain perfectly identifiable at the level of the series: we recognize the work, we recognize the authors, we believe we have understood. It is only when putting the volume back in its exact place, or checking that you do not already have it, that the lack is revealed - that is to say well after the purchase.

Six effects on an inventory

What this statement does not demonstrate

I notice a ceiling of eighty characters and breaks that fall in the middle of words. I'm not sure where this limit applies: at the time the seller posts, at the time the marketplace saves, or in an intermediary tool. Nothing in an announcement provides information on the technical chain that produced it, and I am careful not to reconstruct it.

I also can't say that all lines ending with a short word are truncated. One hundred and twenty-six out of one hundred and seventy-six have this sign, but a title can legitimately end with a short word. The only certain clue is the length: the twenty-one announcements on the ceiling have lost something, and you can often see what.

Finally, the price differences cited – €18.42 and €20.49 for the same title in pocket format – cannot be interpreted from the title alone. Two advertisements for the same book at two different prices is the normal situation in a second-hand market, and I draw no conclusions about the condition or edition of the copies.

Translate this into a file

The instruction:never copy an ad title into a structured field without checking that it is complete.The test takes one second: does the string end with a complete word and a complete idea? A title that stops at “Volume,” an isolated letter, or an obviously abbreviated name should be treated as incomplete.

When you detect a truncation, the best practice is to capture what you know and leave blank what you don't know, rather than filling in at random. Reconstruct “Robert Jor” in a plausible name amounts to introducing a supposition into a field which should only contain facts. If the name seems obvious to you, write it down in a notes field indicating that it is a reconstruction—the nuance is slight to write, decisive to reread.

Also get in the habit of separating what the chain mixed up when entering: the authors on one side, the title of the work on the other, the format in its own field. This is the only way to take advantage of the most reliable information in the line — the format statement — without inheriting the fragility of the rest.

One last remark, for works in several volumes. When the volume number has disappeared in the clipping, it cannot be deduced from the price, nor from the place of the advertisement in a list, nor from the alphabetical order. Ask the seller, or read it on the item upon receipt. Any other method amounts to guessing, and a riddle recorded without mention becomes, a few months later, indistinguishable from a fact.