New Video: "Can You Store Data on the Bitcoin Blockchain?"

Aug 2, 19

The new video can be found here:

Thanks for any support! Here's the script:

In a previous video, I argued that non-cash use-cases of the Bitcoin blockchain are both viable and legitimate. A blockchain is not, however, a miracle invention that makes every single electronic service or application better, or more secure, or even more transparent. Though the blockchain may be revolutionary, it’s use-cases are inherently limited by its nature.

As a distributed timestamp server, one of the main ways the blockchain provides utility outside of its electronic cash system is by allowing you to prove that you were in possession of certain data at the time a block was published. That means you could prove, for example, that you were the original source of an idea, be a song, a work of art, a book, a computer program, a product design, literally anything that can be represented in one way or another digitally. And you don’t actually need to add all of that prospective data to the blockchain. All you have to add is a cryptographic hash of that data. Once that hash is on the blockchain, signed with your private keys, others can independently confirm that it corresponds to the data in question.

There are those who suggest that adding data to the blockchain in its entirety can be a viable means of storing that data in a decentralized way. That approach to data storage is not particularly viable for several reasons.

The first problem is that adding large amounts of data to the blockchain is expensive. Even with the exceptionally low fees of one satoshi per byte on the Bitcoin Cash blockchain, with an exchange rate of 400 dollars per BCH it still costs four dollars to add a megabyte of data to the blockchain. That’s four thousand dollars per gigabyte.

1 sat/byte * 0.00000001 bch/sat * 400 usd/bch * 1000000 bytes

Then, once you’ve paid the miners that hefty fee to include your data in their blocks, there’s a second problem: The miners have no financial incentive to continue to store any of your data once the block that included it has been validated. All that a miner needs in order to continue mining is the hash of the previous block and an up to date set of all unspent transaction outputs, aka the UTXO set, which they update after every block. Miners can “prune” away the rest of the data, as described in section seven of the Bitcoin whitepaper titled “reclaiming disk space”. They might not all be doing this yet, but almost certainly will at mass adoption scales if storage costs become non-negligible compared to their other costs.

So not only is adding data to the blockchain expensive, but it holds no guarantee of its storage. If you want to be sure your data is stored as part of the blockchain, you have to run a non-mining node yourself, at which point you might as well have just added the hash of the data to the blockchain and stored the data yourself separately.

Now, it’s not unlikely that there will be non-mining “archival” nodes on the network that don’t prune at all and supply the blockchain in its entirety to those who want to validate or analyze its entire history… but they won’t be able to do so for free. Archival nodes will probably charge for access to the data they’ve hung onto.

So again, you might as well have just added the hash of the data to the blockchain, and then stored that data separately, either yourself, or using a paid service.

Adding data to the blockchain in its entirety, instead of just adding its hash, is, at best, potentially more convenient than storing it yourself or finding an independent service to do so, as long as there are nodes on the network that haven’t pruned the blockchain. Personally, I can’t imagine any use-cases where that convenience is worth the prohibitively high up-front mining fees, let alone the likelihood of having to pay for access to that data later on.

In a future video, I’ve cover some specific, viable non-cash use-cases of the blockchain.

Upvotes (1)

Comments (0)
sort by  /