In Plain Sight
This person collects what is known and then puts it in order. They are not hoarding. They cross-check names and dates, and keep going until a pile becomes something anyone can look things up in. You'd find this in the neighbor whose notebooks end up being the only record the village has of who lived where.
The Pattern at Work
A local bird group has been going since 1978 and has always kept records, which is to say that forty-one people have kept forty-one kinds of notes in forty-one notebooks. One member spends four winters turning that into one thing. The work is not transcription. It is deciding that two observers writing reed warbler in 1983 and 1991 meant the same bird in the same reed bed, working out which of three place names refers to one flooded field, and marking the thin years so nobody later mistakes an absence of observers for an absence of birds. What comes out gets used by a county survey and then a planning inquiry, and settles an argument about a drainage scheme nobody in 1978 could have imagined.
A factory runs four machines that are older than most of the staff. One maintenance fitter has spent a decade writing down what is actually true about them: that number three runs hot above twenty-six degrees, that the manual's torque figure is wrong for the bolts everybody now uses, which supplier's belts last. None of this is in any document the company owns. He indexes it by symptom rather than part, because a person at seven in the morning with a machine down does not know the part, they know the noise.
A woman inherits three boxes from an aunt and spends two years on them. The output is not a family tree, which anybody can buy. It is a cross-referenced file where every claim has a source and the uncertain ones are flagged, including the two that contradict the story the family has told itself for sixty years.
What the Examples Show
The obvious reading is that these are people who like collecting, and the word used is completist, usually with a small laugh.
Collecting is the easy half, and not the half that took four winters. The work is reconciliation: deciding when two records refer to the same thing, which is never mechanical. Two names for one field, one name for two birds, a torque figure right for a bolt nobody uses. Each is a judgment, and a pile with none of them made is not a resource but material.
The second half is designing for the person who arrives later and is in a hurry. Indexed by symptom, not by part. Thin years marked as thin. None of that improves the collection; all of it makes the collection usable by somebody who does not already know what is in it, which is the only test that matters and the one most collections fail.
Going Deeper
The Library of Alexandria's real achievement was not the scrolls, which were merely expensive. It was that Callimachus produced the Pinakes — tables listing what was there, by author, so a scroll could be found rather than merely owned. That is the first catalogue anybody records, the moment a heap became a library. Every version since has faced the same problem in new clothes.
Pliny wrote down everything he could gather and was wrong about much of it, and the Natural History was still read for fifteen hundred years because it was organised. Domesday put a kingdom's assets into a form a clerk could query. Diderot's Encyclopédie added the cross reference, which sounds minor and is the whole thing: an entry pointing to another turns a list into a structure. The Oxford English Dictionary was built from paper slips posted in by thousands of volunteer readers, one quotation per slip, and the editors' work was almost entirely deciding which slips were about the same sense of the same word. Dewey put a number on a shelf. Vannevar Bush described a machine for following trails between documents in 1945, and gave everybody afterwards the idea.
The people who do this share a particular fate. The index is never finished, because the material keeps arriving and the standards keep improving, so no version can be declared done. Almost nobody funds it, since it produces no finding and settles no question. And it is credited to its users: the planning inquiry remembers the ecologist who cited the data, the county survey remembers the survey, and the four winters are remembered by one person. That is not injustice exactly. It is what a good index does, and doing it well means disappearing into it.
The Image
The index nobody asked for.
Somebody made the thing you are looking something up in. Not the facts — those were lying around — but the decision that this entry and that entry are about the same thing.
You see the work most clearly when it is absent. Any pile you have given up searching was missing one person who cared enough to cross-reference it.
Where It Stops
Having a good memory is not this. Neither is accumulating, which is the opposite in one respect: a hoard grows by addition and an index by judgment, and the second is much rarer.
It goes wrong as an order that only its author can use. A scheme fine enough to hold every distinction becomes one nobody else can enter, and the four winters are then a private pleasure.
It goes wrong again as collection replacing conclusion. There is a point at which the material is enough and somebody should say what it shows, and this pattern postpones that because one more box is always arriving.
It also fails where the knowledge is in hands rather than words. A great deal of what a fitter knows cannot survive being written down, and a file that claims otherwise is worse than an honest gap.
Take the plainer explanation first. Any archivist does all of this because it is the profession. The test is whether it appears with no obligation at all, on a subject nobody is paying for.
Where It Pays
Inside a job. Archives and libraries plainly, but the pattern earns most where no such function exists: technical documentation, maintenance and plant records, quality and standards, clinical coding, legal knowledge management, and the enormous amount of an organisation's working knowledge that currently lives in four people's heads. The contribution is that a company stops being able to lose what it knows when somebody resigns, and the four people who hold it are always about to retire.
It pays badly anywhere the measure is output this quarter, since the value of an index is realised entirely by people who have not arrived yet, and often not for years.
Outside one. Local and family history, natural history records, club and society archives, collections of every kind, and the practical business of being the person who knows where the documents are when somebody dies. The cost worth naming is that the work is invisible until needed and then treated as though it had always existed. The second is more personal: a mind that wants to reconcile every record finds unfinished piles genuinely uncomfortable, which turns other people's shoeboxes into an obligation and makes it very hard to walk past a mess that is nobody's responsibility.
Try This
Take a pile you have — photographs, receipts, a folder of articles — and do not sort it. Sorting is the trap.
Instead, write the finding aid. One page saying what is in there, roughly how much, what the gaps are, and how somebody else would locate one specific thing.
Then hand the page and the pile to somebody else and ask them to find something. Watch where they fail. That failure is the only reliable sign of whether your order is an order or a habit, and it beats another weekend of filing.
The index is a tool, not a self. Pick it up where the material will outlive the person holding it. Put it down for things nobody will ever need to look up, which is most things.
If This Isn't You
Most people keep what matters, lose the rest, and carry what they need in their heads, which is how nearly everybody has always managed. Feeling no urge to cross-reference a shoebox is an answer, not a gap.
Where To Go Next
Its near-twin — Story Keeper. Both hold something so it survives the people who made it. The Story Keeper carries a few things whole, in the telling, and the integrity of each is the duty. This one carries a great many in pieces, and the duty is that any can be found.
Its shadow — Truth-Seeker. The Truth-Seeker tests a claim and discards it when it will not stand. This one keeps the doubtful claim, flags it, and files it, on the grounds that the reason it was believed is itself a fact worth having.
Most often confused with — Evidence Mapper. Both work patiently with material nobody else has connected. The Evidence Mapper builds one event from fragments and stakes a claim. This one builds an order over everything and stakes none.