It was a moonless night, and the three of us were sitting on a desk facing a wall, all eating his leftover egg salad with bread and hot sauce. I mean, we were trying to eat that - but really, we were also trying not to get our hands too dirty, considering the mission we were on.
In the process, one of them had taught me the basics of xargs - or at least
they were trying. And so, using curl, we sequenced through the natural
numbers API we had once architected - hoping to rescue what remained.
I don't know if it was the beer or the egg salad - but something made us feel triumphant, although it would turn out that we had failed.
I mean, our efforts were not totally without effect, and so we were indeed able to back up some of the data. But in the grand scheme of things, we failed - because we didn't back up the artists' manifestations and because, years later, the metadata turned out to be useless.
This, my friends, is the story of a failed startup and how some colleagues, now friends, tried to back up digital art. It's the story of ascribe and "ascribed." And it's not fiction - What you just read is an actual account of a night we spent trying to back up data from an S3 bucket. Sadly though, we weren't diligent enough, and so when public access was revoked - almost all ascribe metadata but more importantly, all digital art files ended up being deleted.
But let's backtrack a second: In 2015, I started working as a front-end developer for a company called ascribe and what we did was registering digital art on the Bitcoin blockchain. Bitcoin because Ethereum had simply not been around yet. Digital art? Yes, really like a precursor to the modern-day NFT.
It were the early days, and browser plugins like Metamask didn't exist yet - but our idea was not uncensorable media. Instead, it was to decentralize digital art's ownership layer and make it ownable like you could own a Bitcoin. And we ended up doing an OK job on that as the SPOOL protocol is still accessible through the blockchain and so technically, ownership can still be tracked and viewed.
But then, in 2016, the early-stage startup ended up getting into funding issues as back then, having daily active web3 users wasn't a criteria VCs valued. And so this meant that we pivoted the company to build a scalable blockchain database called BigchainDB instead - ascribe as a product development focus was cast aside but remained online for quite some time post-pivot.
Still, and this is somewhat reasonable, when refocusing its attention, BigchainDB Limited eventually decided to shut down ascribe, and unfortunately, they took down most of the art too.
So that story at the beginning of this moonless night, that's from a moment shortly before ascribe's official shutdown date when, I think, I wasn't even working there anymore, but we still felt compelled to save what the artists had uploaded.
Today, I felt motivated to share this story publicly as I think it needs more recognition from those contributing to the NFT space. When NFTs popped in 2021, some of the OG's narration was on restoring ascribe and its pre-historical art. I and others were asked: What was ascribe? What were the artworks? How can I access them, and are there any archival or restoration projects?
The sad story is: We didn't have answers. In fact, we tried getting together on GitHub, and we did some reverse engineering on the open-sourced backend. With a restoration project we founded called "ascribed," we thought: Maybe it's possible to re-instantiate ascribe as an NFT-native platform by migrating the SPOOL protocol stored on Bitcoin over to Ethereum.
But my frank assessment is that it's going to be close to impossible, and potentially, that I'm now also OK with that. It's because I'm seeing some other good projects embodying the spirits and visions of ascribe - of making digital art permanent, to archive it, and to fundamentally alegalize it by using crypto-native protocols.
Today, seven years later, ascribe is an integral part of the space's story and a stark reminder that, contrary to what our parents kept telling us, namely to cautiously use the internet as it never forgets - the internet actually has forgotten a lot of important stuff, for example, the ascribe library.
So I want to take this moment and motivate two critical aspects of digital art and media platforms that we should strive to implement: It's (1) permanence and (2) alegality.
Permanence is the temporarily independent and content-integer accessability of content. IPFS implements permanence through content-addressability, hence creating integrity through file hashing and by making everyone capable of replicating the said file. The Internet Archive makes past websites available through the Wayback Machine, creating a permanent web of all versions throughout time. Bitcoin and Ethereum keep all ledger data available to all network peers so as to enable peers to recompute the entire accounting history and preserve the network's integrity.
Alegality is a subtractive measure to decrease reliance and dependency on institutional law. Alegality means putting a minor emphasis on legality - not in opposition to legality (as this would be "illegal") but rather as to conceptually solve legal problems without the law being a relevant aspect of the solution. Bitcoin solves the double spending problem by building an uncensorable permissionless consensus network that never requires the intervention of legal recourse. TheDAO's white hat hackers settled the recourse of the then-defunct smart contract. And NFTs solve the problem of attribution and monetization of art through on-chain registration financialization and not through copyright defense.
Permanence and alegality are some tools of Jaya Klara Brekke's hacker engineer towards creating autonomous digital policy. The story of ascribe shows us the importance of autonomy, alegality, and permanence - as had we treasured those aspects, then more of its art would be available today.
Tools of the hacker engineer are to be applied with care but are vitally important to understand, practice, and implement to preserve on-chain music, art, and culture.
originally published on neume.network on 2022-10-25 by timdaub