Forgotten Industries // L'Œuvre // Entry 009
Why a vinyl collector can't trust the label, the barcode, or the database, and what a record that carries its own history would change.
Pick up a record and you are holding three questions, whether you mean to or not.
What is this, exactly? Where has it been? How do we know?
Everything a collector pays for lives within those three. The music is the same on every copy. What is not the same is which copy this is, whether it really is what the seller says it is, and whether anyone in between can actually prove it. Vinyl tends to answer those questions badly. The tools collectors trust most don't actually identify a record, and the one database that tries to has been filled in by the same people buying and selling.
The number on the spine
Start with the catalog number, the code printed on the spine and the label. It may look like an identity. It is closer to a family name.
mewithoutYou's Brother, Sister came out on vinyl through Burnt Toast Vinyl in 2007 as BTV081. It was pressed again in 2009, 2011, 2012, 2013, 2015, 2016 and 2017. Discogs, the reference collectors use for this, lists sixteen separate entries under that one catalog number, test pressings included. Nearly all of them carry the same barcode: 634457189010.
So the two things printed on the jacket that look the most like an ID, the catalog number and the barcode, tell you which album you have. They don't tell you which of sixteen records you're holding. Those sixteen are not priced alike. A first pressing and a late repress of the same album can sell for several times different amounts, and the jacket alone won't tell you which one you've got.
The dead wax
Experienced collectors know to look past the jacket, into the dead wax: the smooth band between the last groove and the label, where the cutting engineer etches a matrix code by hand. It's the closest thing a record has to a fingerprint. It's also the first place to check whether the vinyl and the jacket do, in fact, belong together.
But a fingerprint only identifies something if it is unique to the thing being identified. On Brother, Sister, the side-A runout reads btv081A (Re-1) mewithoutYou : Brother, followed by a small drawn sun, and that same etching is recorded on entries from 2007 through 2013. The etching travels with the metal parts that stamp the record, and when a label presses again from the same parts, the etching comes along unchanged. The dead wax tells you whose parts pressed the record, but tends to come apart when you try to pinpoint when.
What finally separates Brother, Sister pressings is the colour of the vinyl itself, because the runs came in different colours. That's the real rule of identifying a pressing, and most of the market doesn't apply it: the catalog number and the barcode never identify a pressing on their own. Colour and runout together usually can.
Alas, even that has gaps. A sealed record can't be checked without opening it, which destroys exactly what the buyer was paying extra for. Colours fade, photograph badly, and get described loosely: one seller's "oxblood" is another's "maroon." Two runs can share a colour, and then the evidence runs out.
The database that holds the answers
When the object can't settle the question, collectors go to Discogs. It's the closest thing vinyl has to a public registry, and it's remarkable: millions of releases, documented by volunteers, with runout transcriptions and pressing notes that exist nowhere else.
It is also maintained by its own users, and sometimes you can see the damage. Some limited-edition entries show more users claiming to own a copy than the edition ever had copies. That isn't fraud. It's people filing their copy under the entry with the right name, but the wrong pressing, one honest mistake at a time, until the entry's numbers stop meaning anything. An edition-size figure is supposed to be a fact. Once enough copies are misfiled, it isn't one anymore.
A database records releases, meaning the types. Nobody records the individual copies. There is no record anywhere that says this copy was pressed in that colour, was sealed until 2019, passed through these three owners, and was graded at these three conditions along the way. Each time a record changes hands, the buyer starts the investigation over from nothing.
A record that remembers
This is the gap that object-level authentication is built to fill. Put a secure chip in the jacket, scan it with a phone, and the scan returns a history attached to that one copy rather than to the release it belongs to.
For vinyl, that fixes several things the tools above can't.
Identity gets settled once. If the pressing is identified when the copy is registered, by the label, the plant, or a collector who checks the colour and runout at intake, then every later owner inherits that answer. Nobody has to break a seal to find out what's inside.
Ownership has a chain. Every sale adds a link. A copy with a documented line of owners is a different object from one that showed up out of nowhere, and the market already prices that difference. It just has no way to show it.
Condition gets a history. Grading is subjective, and records get worse with every play. A dated series of condition notes tied to the copy says more than any single grade.
The registry gets cleaner. If each copy is registered against its pressing, the count of copies in the world comes from registered objects rather than from users choosing an entry from a dropdown. An edition of 500 would show 500.
What a chip can't know
A chip proves the history recorded on it, and nothing more. Vinyl has a few specific ways to fool a careless system, and anyone building one for records would likely benefit from planning for them.
The jacket is not the record. A chip in the sleeve authenticates the sleeve. Discs get moved between jackets all the time, sometimes innocently and sometimes not. A vinyl registration has to tie the chip to the disc as well: a photo of each side's runout and the vinyl colour, taken at registration, so a later owner can check that the record in their hands is the one on file.
The record starts when the chip goes in. A copy pressed in 2007 and chipped in 2026 has eighteen years of history the chip never saw. For older records, how carefully the copy is identified at registration matters more than anything that happens afterward. If the chip is attached to a wrong identification, every later scan repeats the mistake with confidence.
Not every tag is secure. Basic NFC tags can be copied. Authentication depends on tags that produce a cryptographic response that changes with each scan and can't be cloned, and collectors will eventually ask which kind is in the jacket.
None of these limits weakens the idea. They're the design spec. Collectors already trust the dead wax because it's hard to fake. A chip earns the same trust by recording the dead wax rather than replacing it.
Pang
The copy on my shelf is Brother, Sister on teal and lavender split vinyl, with the sun and moon labels. Side B's dead wax reads btv081B (Re-1), next to a hand-drawn crescent moon. Discogs files it as the seventh pressing, the 2013 "Easter" Moon variant. The feeling was new to me: if my research was right, I owned a very rare pressing of a record I love. I checked the pressing date against my own memory, to see whether it matched when I would actually have bought it. It did. Then I searched for any other copy for sale, anywhere, and the search came back with nothing. That was the high point.
It was also the moment I realized how much of it rested on "if." The sixth pressing came out the same year and also had a Moon variant, listed as blue and purple split. Same catalog number. Same barcode. Same etching. Same year. Same moon. What separates a sixth pressing from a seventh is whether a stranger looks at a photo and reads the colour as "teal and lavender" or "blue and purple." I know which one I have because I'm holding it. I couldn't prove it to a buyer looking at a screen, and nothing on the record itself can either.
Against forgetting
Records already carry evidence of where they came from: the colour in the vinyl, the etching in the dead wax, the shrink-wrap, the hype sticker, the price tag from a shop that closed in 2011. The trouble is that the evidence is scattered, easy to misread, and it disappears each time the record changes hands.
A record that carries its own history doesn't replace any of that. It keeps it together. A thing documented is a thing not yet lost, and for a record that's been through several owners, the documentation is part of what you're buying.
Sources: Discogs database, catalog number BTV081, queried 2026-10-30. Runout transcriptions as recorded by Discogs contributors.
Pang is defined in Entry 001: the moment an abandoned object becomes evidence again.
-- Forgotten Industries // L'Œuvre // Entry 009