An ancient guqin from the late Tang dynasty, called Feiquan. The Palace Museum’s craftsmen have been making a replica of it for six months, and they’re still scraping a ground layer mixed with deer-antler frost onto the body of the qin.
It’s not enough to make something that looks the same. The wood has to be old stock—hundreds of years old. New wood would have cracked long ago. The ground layer has to be scraped on, layer by layer: too thin, and it won’t protect the qin’s body; too thick, and the sound turns dull. They call this work “replication,” and it appears alongside “restoration” on their work badges. Six months, and they’ve only got this far.
I used to think the biggest fear with this kind of thing was making a copy that looked too much like the original. So what, exactly, are the rules afraid of? After reading the Measures for the Administration of Cultural Relic Replication and Rubbing, I realized they don’t care whether it looks the same at all. They care about something else.
Every replica must be marked as a replica and given a serial number. If it’s made for sale, it must come with an information sheet stating the relic’s name, its period, the institution that holds it, who made the replica, who supervised it, when it was made, and which number it is.
The rule on mass production is even stricter. It says the original relic must not be used. So if you want to make ten thousand, after the first one, you can’t touch the original again—you can only use replicas to make more replicas. For rubbings of steles and model calligraphy from before the Yuan dynasty, you first have to make a replica carved printing block.
You see? What they want isn’t “likeness.” It’s “identification.”
Every replica has to acknowledge that it is a replica. It also has to state its place in the sequence. That’s the costly part. It takes the question “Am I the original?” out of the mouths of appraisal experts and puts it on a piece of paper.
A lot of people think the Palace Museum avoids making near-perfect copies because it’s afraid of disrupting the market. But the museum won’t even “fill in” missing details. If a seal on a painting has worn away until it can no longer be made out, they can photograph it with hyperspectral and multispectral imaging and recover which strokes were dark and which were light before it wore away. And then? They save the image and tell you there used to be a seal there. They don’t add a single character to the original. They simply leave it alone.
They’d rather let it stay incomplete. I’ve thought about that for a long time. That’s real restraint.
Take that way of thinking and put it on the blockchain, and you run into a familiar face. The thing digital collectibles are proudest of is having serial numbers: ten thousand copies means ten thousand tokenIds, and no one can change them. They’ve managed one half—the numbering. What about the other half?
That’s the half that’s missing. The chain records “which copy this is”; it doesn’t record “whether this is a replica.” Mint one image ten thousand times, and the first and the ten-thousandth look exactly the same in the contract. Nowhere does it say the original is in a museum and these are all reproductions. And they still call that limited edition? Hardly.
Article 12 of those measures requires the information sheet to name the holding institution. There’s no such field on-chain. Open the item in your wallet and you’ll see a contract address, a number, and a holder. It won’t tell you who holds the original.
There’s an even finer distinction. Article 2 excludes “imitation” from the measures’ scope. Replication requires approval; imitation doesn’t. The line is drawn around the craft and the process. On paper, it’s clear-cut. When you’re looking at the actual object, the difference between the two sides of that line isn’t so easy to tell.
I’m not going to call it a loophole. Rules are for people. If people can understand them, that’s enough.
From what I can see, numbering answers “How many copies are there?” Attribution answers “What is this?” The first takes just one line of code. The second requires someone to be willing to write a sentence in the contract that works against their own interests.
As for the Palace Museum’s more than 100,000 high-resolution images, they are publicly available, but the various notices aren’t entirely consistent about what you can download or how you can use it. There’s no definitive answer. If you’re really going to do something with them, best read the fine print first.
So take another look at those numbered limited editions you have. Turn them over. Is there a line saying it’s a replica, which number it is, and who holds the original?
$BTC $ETH $BNB
#文物 #数字藏品 #故宫 #版权
It’s not enough to make something that looks the same. The wood has to be old stock—hundreds of years old. New wood would have cracked long ago. The ground layer has to be scraped on, layer by layer: too thin, and it won’t protect the qin’s body; too thick, and the sound turns dull. They call this work “replication,” and it appears alongside “restoration” on their work badges. Six months, and they’ve only got this far.
I used to think the biggest fear with this kind of thing was making a copy that looked too much like the original. So what, exactly, are the rules afraid of? After reading the Measures for the Administration of Cultural Relic Replication and Rubbing, I realized they don’t care whether it looks the same at all. They care about something else.
Every replica must be marked as a replica and given a serial number. If it’s made for sale, it must come with an information sheet stating the relic’s name, its period, the institution that holds it, who made the replica, who supervised it, when it was made, and which number it is.
The rule on mass production is even stricter. It says the original relic must not be used. So if you want to make ten thousand, after the first one, you can’t touch the original again—you can only use replicas to make more replicas. For rubbings of steles and model calligraphy from before the Yuan dynasty, you first have to make a replica carved printing block.
You see? What they want isn’t “likeness.” It’s “identification.”
Every replica has to acknowledge that it is a replica. It also has to state its place in the sequence. That’s the costly part. It takes the question “Am I the original?” out of the mouths of appraisal experts and puts it on a piece of paper.
A lot of people think the Palace Museum avoids making near-perfect copies because it’s afraid of disrupting the market. But the museum won’t even “fill in” missing details. If a seal on a painting has worn away until it can no longer be made out, they can photograph it with hyperspectral and multispectral imaging and recover which strokes were dark and which were light before it wore away. And then? They save the image and tell you there used to be a seal there. They don’t add a single character to the original. They simply leave it alone.
They’d rather let it stay incomplete. I’ve thought about that for a long time. That’s real restraint.
Take that way of thinking and put it on the blockchain, and you run into a familiar face. The thing digital collectibles are proudest of is having serial numbers: ten thousand copies means ten thousand tokenIds, and no one can change them. They’ve managed one half—the numbering. What about the other half?
That’s the half that’s missing. The chain records “which copy this is”; it doesn’t record “whether this is a replica.” Mint one image ten thousand times, and the first and the ten-thousandth look exactly the same in the contract. Nowhere does it say the original is in a museum and these are all reproductions. And they still call that limited edition? Hardly.
Article 12 of those measures requires the information sheet to name the holding institution. There’s no such field on-chain. Open the item in your wallet and you’ll see a contract address, a number, and a holder. It won’t tell you who holds the original.
There’s an even finer distinction. Article 2 excludes “imitation” from the measures’ scope. Replication requires approval; imitation doesn’t. The line is drawn around the craft and the process. On paper, it’s clear-cut. When you’re looking at the actual object, the difference between the two sides of that line isn’t so easy to tell.
I’m not going to call it a loophole. Rules are for people. If people can understand them, that’s enough.
From what I can see, numbering answers “How many copies are there?” Attribution answers “What is this?” The first takes just one line of code. The second requires someone to be willing to write a sentence in the contract that works against their own interests.
As for the Palace Museum’s more than 100,000 high-resolution images, they are publicly available, but the various notices aren’t entirely consistent about what you can download or how you can use it. There’s no definitive answer. If you’re really going to do something with them, best read the fine print first.
So take another look at those numbered limited editions you have. Turn them over. Is there a line saying it’s a replica, which number it is, and who holds the original?
$BTC $ETH $BNB
#文物 #数字藏品 #故宫 #版权