ICP has a flawed endgame

Watercooler
22

You raise many valid points and to avoid a “1984” type of dystopian future happening in some countries we are beginning to think in the right direction. Obviously Bitcoin has been successful in being “unstoppable” and uncontrollable by any one government. It cannot be inflated unnecesarily by the FED. However, it is fully traceable and attempts to anonymize any crypto are met with extreme displeasure by governments. Right now I believe the EU is “banning” cryptos that are anonymous or untraceable. So much for Monero and Bitcoin Private died a quick death and Tornado Cash was under massive attack. The excuse is always the same. We need to prevent money laundering by cartels and criminals. The real reason however is to control everyone esp. in regards to taxation and in other countries it would be more to control people’s freedom of speech very carefully. What people don’t realize is that the banking system was designed to collect data about you. You have 0 privacy and they can even hold your funds without due process. Money in the bank is not really yours. The bankers are in cahoots with the authorities and we are all considered guilty criminals in need of being controlled and watched by big brother.

The fact that some people would need true anonymity when expressing their true thoughts or financially due to oppressive regimes is a sad statement about the state of our world here in what should be “the future” 2025. Basically these people are being treated exactly like the true criminals who these authority figures are saying they do not want to enable but at a very steep cost which is to put us all into a prison state that is neverending. Slavery has not been abolished in our world. The majority of people are slaves and many people are not even allowed to move about freely being cruelly denied visas to places like the United States despite being model citizens in their own countries.

If your goal is to use crypto as a form of rebellion like in “Mr. Robot”, then yeah the IC does not currently fulfill this use case because if it did, it would inevitably be in the crosshairs of regulators who would seek to destroy its progress.

You have to start somewhere. The ideas are good and the implementation is unique. What we need is adoption by business to get the ball rolling and then later on, you can hopefully get to a point where it can be used in the way you seek and hopefully to get people to truly rise up against these oppressive regimes and to make the real changes in the real world that are needed to eliminate all of these bad actors and oppressive governments once and for all.

23

I guess our disconnect is that I’m not a business person, and not interested in conducting a business on the internet. I have to admit I got into it because of investing, but got stayed for the technology, back in 2017 or something. I’m interested in creating dissident infrastructure. I’ve written off the money part already since it flopped via regulations, so even if you use monero, you can’t really do anything with it because you can’t cash out properly without KYC, and can’t directly buy with it, either. That leaves only completely money-less, or at least unmonetised infrastructure that might have money as an add-on, as candidates for building such technology.

I once wanted to build my own blockchain for decentralised, encrypted file storage, along with some colleagues from university, but it ended up being only me who did all the work, so it never went anywhere. I’m on and off researching better alternatives to PoW style consensus in my spare time for years now, and I came to the conclusion that the best way to remove the entry barrier for participation is to completely separate the money and transaction verification from the consensus, and turn the blockchain itself into a pure message ordering layer. You can then add arbitrary processing logic for the embedded messages as a service that builds on top of that. And for good latency and lower node requirements, you would want something like an L2 anyway. And like Nervos, I think the design should be so that the intended use of the chain is solely to facilitate the coordination of L2 protocols. And if many separate decentralised services run on the same L1 message ordering layer as a fallback for when their L2 breaks, or something, then they all contribute to strengthening the same L1 security / trust anchor, while not competing over storage capacity or computation bandwidth.

My vision is an ecosystem of many decentralised services, that runs on raspberry Pis, with maybe an attached HDD or SSD, where the L1 is “dumb” but can fundamentally be run by any internet-connected device (let’s say 100MiB of memory and 100KiB/s bandwidth). And then if it’s a lightweight L2 protocol, such as a DNS-like service, or similar, it also doesn’t need additional hardware. But if you have something stateful which also allows freely programmable comutations or something, or maybe AI or whatever, you’d also hook up a separate device to that Raspi, which then needs to fulfill the L2 system specs. But every node that maintains any L2 connected to the general-purpose L1, would also strengthen the L1. My L1 consensus design also does not require mining power or stake to decide the truth; instead, it relies on the ability to identify anyone who does not agree with the obvious state of the system, and the ability to outright ban those nodes from participating in the gossip network. I won’t get into the details because it is a complicated and novel approach has little in common with existing approaches, and it would take too long to explain.

Such a setup would not fragment security across chains and not force monetisation on money-less systems, while still achieving global consensus for things such as DNS or other things that need to maintain some kind of public, modifiable/appendable state. I’m more interested in decentralised intermediary-free email, anonymous distributed hosting, and decentralised media networks that rely on volunteer-based replication of content. I don’t really find the time to build it all myself, though.

24

However, in the future, if fiat starts dying out and crypto is widely accepted directly, then anonymous cryptos may become more useful except there’s a saying that no one can escape “death and taxes” with the latter being the worse outcome. :wink: (LOL)

Are you using a consensus algorithm similar to Hedera since you speak about gossip?
What do you think about Arweave? Isn’t that more aligned with your particular censorship-resistant needs and desire to have anyone be able to participate? I believe Odysee (a decentralized and now ad-free version of YouTube was using LBRY and has now been purchased by Arweave)…It seems to work pretty well and allows for anyone to participate?

25

Yeah, taxes are definitely the worst, lol.

Hedera uses UTXOs and stake, if I’m not mistaken, and forms a DAG for finality. I’m relying on pre-timestamped messages for ordering (so, the sender timestamps it), and honest nodes with correctly calibrated clocks can then reject messages with timestamps that are too far outside the allowed deviation. The original sender of the message signs it with a key pair that is bound to a proof of space, which makes identities hard to generate. If a message is too far from the current time, then a dispute is started. Either the disputer or the original sender will be banned. The thresholds are chosen so that all honest nodes with reasonable clock accuracy will make the same choice. It gets a bit more complicated than that because it also has to take into account propagation latencies etc. Basically in the happy case, the timestamps are enough for ordering all messages (and messages with the same timestamp will get ordered by hash), and after the window for timestamp deviation elapses, the history up to a certain timestamp is final. In the malicious case (someone broadcasts an invalid timestamp), a dispute is started and either the accuser or the accused gets banned (because all messages are signed, and identities are hard to generate, this is a semi-permanent exclusion from the P2P network). In regular intervals, all honest nodes will additionally certify the valid finalised history with a proof of space (everyone targets the same hash, so the mining power of all honest nodes adds up). However, that is just so that outsiders that join the network can gain more confidence. Honest nodes that are already part of the network will not be influenced by the issued checkpoint.

LBRY was pretty good, but its incentive structure was bad. You pay once to make a public metadata record, and then “someone” is supposed to store the actual file forever, for free. And the Odysee frontend is censored, and the LBRY client app is buggy. I haven’t yet looked into Arweave. I basically just want something like torrent, but smarter: it should have a native search function, it should have a concept of authors and related works, as well as the concept of folders or websites. And content should be changeable by the author. For example if you hosted a blog on such a network, you want the blog to be mutable. And I think the concept of “forever” storage is not possible to achieve long-term. Rather have timed/rented storage that you pay for, or voluntarily replicated storage with no hard guarantees or expectations regarding availability. So that if you want your 1000 vlogs to be online, you either pay someone to host it, or host it yourself, or get enough followers so that they volunteer to host them for you, or that they pay someone to host it, or something along those lines. And ideally all that without having to register everything on the blockchain. You can still have a globally searchable index of content without a blockchain, if you can traverse the P2P network graph and do discovery queries with limited reach. That scales way better, and doesn’t need cryptocurrencies or blockchains, but it is much more technically complicated to pull off well.