Will this still be readable in thirty years?

A journal fails to survive in one of four ways, and they are not equally likely. In rough order of how often they actually destroy volumes: storage, then ink, then light, then paper. People buy for the fourth and lose journals to the first.

Most common cause of loss

Damp, or a flood, or a house move

Most common cause of illegibility

Ink that was never meant to last

Slow, invisible damage

Sunlight on a shelf

The thing people research

Paper, which is usually fine

Paper can be rehoused, damp can be dried, a torn page can be repaired. An ink that has faded is gone, and no conservation will bring it back. This is the decision to get right on day one.

• Pigment beats dye, for permanence — Broadly, inks fall into dye-based — colourant dissolved in liquid, brilliant, huge colour range — and pigment-based, where solid particles sit on and in the fibre. Pigment inks are far more resistant to light and to water. Almost every vivid, unusual fountain pen ink is dye-based, and that is exactly what makes it vivid.
• Look for the words, not the price — "Pigmented", "archival", "permanent", "document ink", "waterproof when dry". A ballpoint or a pigment liner labelled archival is a better bet for permanence than an expensive bottled ink chosen for colour, and costs a fraction as much.
• The classic answer is iron gall, and it has a catch — Iron gall ink is why medieval manuscripts are still legible, and modern versions of it exist. The catch is that it is mildly acidic and, historically, has eaten through paper where applied heavily. Modern formulations are much better behaved than historic ones, but this is a case where the romantic answer needs the caveat attached.
• Test it yourself — the whole test takes ten minutes — Write a line on a scrap of your usual paper. Cut it in half. Run water over one half; tape the other in a sunny window for a month. You will learn more about your actual ink on your actual paper than any specification will tell you, and you will do it once.
• Pencil, unexpectedly, is excellent — Graphite does not fade, does not run and does not react with paper. It smudges and it can be erased, which are real objections — but a pencilled journal from 1950 is generally in better condition than a biro one from 1980.

The paper horror story everybody has heard is the yellowed, brittle newsprint that crumbles when handled. That is acidic wood-pulp paper degrading, and it is real — but it is characteristic of cheap bulk paper, not of the paper in a typical bound notebook.

• Acid-free is worth having and is often already there — Many decent notebooks use acid-free or buffered paper and say so somewhere. If a notebook says nothing, it may still be fine; if it feels like newsprint and the pages are already slightly cream at the edges when new, it is not.
• Heavier paper survives handling better — Not because it is more chemically stable but because thirty years of page-turning is mechanical wear. Thin paper tears at the gutter first.
• Binding matters as much as paper — Glued paperback bindings fail — the block separates from the cover, then pages come loose in ones and twos. Sewn bindings survive. This is the single most useful thing to look for and it is visible: open the notebook flat in the middle and look for thread.
• Loose sheets and ring binders are the least durable format — Not because the paper is worse but because pages get lost, reordered and separated from their context. If a journal matters, bind it.

Ask anybody who has lost old volumes and it will not be fading. It will be a leak in a loft, a garage that flooded, a box that went to a charity shop during a house move, or mould in an unheated spare room.

• Not the loft, not the garage, not the cellar — The three places people put boxes of journals and the three worst environments in a house. Lofts swing thirty degrees between seasons and condense; garages and cellars are damp. Paper wants what people want: steady temperature, moderate humidity, no direct sun.
• Upright on a shelf, in an ordinary room — A bookshelf in a heated room is genuinely close to ideal, and it is free. Standing rather than stacked — a stack presses the lower volumes out of shape and makes any individual one hard to get at, which means they get handled roughly.
• Out of direct sunlight — This is where light does its damage: not to the ink you are writing today but to the spines and edges of the volumes on the shelf for the next twenty years. A shelf that gets afternoon sun will bleach a row of covers noticeably within a couple of years.
• Label the spine and date it — A box of unlabelled identical black notebooks is how a run of journals gets split up or thrown out. Dates on spines take a minute each and are the cheapest insurance here.
• Tell somebody they exist, and what you want done with them — The one entirely non-material step, and the one that decides more outcomes than any of the above. Journals are thrown away by well-meaning relatives who did not know they were meant to be kept. A note in the front of the first volume is enough.

A photograph of every page, stored somewhere else, defends against the one failure mode nothing on this page can prevent: the whole shelf being destroyed at once. It takes a few minutes per volume with a phone.

What it does not do is replace the original. A digital copy is a hedge against catastrophe, and it comes with its own thirty-year problem — files need to be readable, findable and actually still there in three decades, which means an ordinary image format, a plain folder structure, more than one location, and something that gets checked occasionally rather than a hard drive in a drawer.

The honest summary is that paper stored sensibly is the more reliable of the two over thirty years, and that the copy is worth having anyway because the failure modes are completely different.