|
>I agree with you that trust is an open problem, but it's one that I think is getting solved automatically for free by ubiquitous surveillance, if we administer it sanely. The problem is, who watches the watchmen? And who watches the watchmen who watch the watchmen? And so on, because at some point you have to trust someone not to act maliciously. Not to mention that ubiquitous surveillance is a problem in itself, because again who can we trust to administer it sanely? It all goes back to the same root problem: trust, which is fundamentally incompatible with human nature.
The best course of action, IMO, is to do away with it entirely by using trustless systems. >The advantage of centralization is efficiency. Exactly! And because of that a certain degree of centralization is pretty much unavoidable, and it's why decentralization is not a silver bullet, and why the answer lies in the middle. A balance has to be found. A purely decentralized system is very slow to develop, maintain and update compared to a centralized solution; because of the glacial pace of development in the space, lots of detractors say that Bitcoin has been around for 10 years and nothing has changed, but I disagree with that! The ecosystem has matured a lot under the hood in these years, but it took a very long time to get to this point compared to what a centralized could do in the same span of time, with a lot of fragmentation and plenty of time spent coordinating actors towards a common goal. It's the complete opposite of "move fast and break things" :D >Speaking as an open-minded but skeptical outsider, you would have to convince me that the blockchain provides/solves decentralization better than existing alternatives. The only decentralized solution I see that works and has adoption is just torrents. Maybe git, if you stretch the definition a bit. But I don't see a way to build and deploy websites that doesn't involve blockchain, for example. What are the alternatives? The fediverse? Someone still has to maintain the instances, and once those go down the content is lost. Scuttlebutt seems to fit the bill, but barely anyone uses it. >Again, not to be a spoilsport, but it sounds like I could do a lot of that with git and some cryptography. Maybe I just don't get it? Well yes, but you'd just be reinventing web3 with none of the adoption. git can be thought of as a sorta blockchain, and wallets are simply public-key cryptography. The core issue is making it accessible to everyone and then actually getting people to use it, but luckily (and sadly) there is an economic incentive to drive adoption and usage, which is why crypto has experienced an explosive growth in the last years. Without that you're never going to put any kind of dent into the existing centralized systems, anything one would develop would just be a passion project at best. (again, scuttlebutt) >On that we both agree. This stuff you're describing sounds interesting, and nothing like the scam-fest "crypto" stuff. It is! I am one of the few who is unironically in it for the technology. Sure, the money would be a nice bonus, but these are shark infested waters so any money I throw in I just consider it a lost gamble. But it's interesting to see what people are building, there's plenty of people chasing stars and impossible pipe dreams, such as a fully decentralized society down to the government, and it's really hard to discern who is a snake oil salesman and who is actually interested in developing whatever they're peddling, but interesting projects do exist. The hard part is wading through the scams, of which crypto is 99% comprised of because of the lack of regulation.
Hopefully this crash will cleanse the ecosystem and finally let actually useful projects that aren't just overly complicated financial games to shine through. |
> The problem is, who watches the watchmen? And who watches the watchmen who watch the watchmen? And so on, because at some point you have to trust someone not to act maliciously.
I've thought about this and I think the likely answer is that the system has to be self-referential (a recursive function, if you will.) The infrastructure and personnel of the panopticon must be subject to the panopticon. E.g. members of the CCP must be subject to the Chinese "social credit" system.
It seems to me that if everyone is subject to the system then we still have a techno-tyranny but it's ruled by Mrs. Grundy. Otherwise, if there are privilege levels to the panopticon, you're going to wind up with masters and slaves.
> Not to mention that ubiquitous surveillance is a problem in itself, because again who can we trust to administer it sanely?
I don't think we can avoid solving that problem, I think ubiquitous surveillance is unavoidable due to economic factors (e.g. the fleet of self-driving cars is a de facto surveillance network. See also "sensing for wi-fi": your router can see you.) It's already happening, eh? Our cell phone track our locations, etc.
> It all goes back to the same root problem: trust, which is fundamentally incompatible with human nature. The best course of action, IMO, is to do away with it entirely by using trustless systems.
Here's where I strongly disagree. I don't accept the "it's human nature" argument against our higher values. I've lived in high-trust and low-trust enclaves and high-trust is just better. It's more efficient, but more importantly it's just fantastic! It feels great! It's just as natural if not more so to love and trust each other as it is to hate or fear each other. The best course of action, IMO, is to foster the higher human values and establish and grow trustworthy regimes within the chaotic larger systems and eventually establish a kind of techno-utopian Golden Age. :)
Well met!