from Hacker News

Ask HN: Blockchain success stories that did need to be decentralised?

by tablatom on 12/7/21, 5:33 PM with 55 comments

• by JayStavis on 12/7/21, 7:51 PM

  I find the render network's ambition to be very compelling, and think the decentralized approach to rendering is necessary. Could something similar be accomplished with a centralized model? Theoretically yes definitely. Practically though, my stance is this can't happen soon with open standards.

  IMO the project aims to solve a couple fundamental problems:

  - Graphical computing power is prohibitively expensive

  - Graphical computing power is inherently scarce and inaccessible, while most graphics resources are in reality sitting idle.

  - Thermal/power constraints are incompatible with a mass-market form factor for VR/AR/XR.

  Looking past the shilling in this tweetstorm I think it's the best summary of the core concepts I've seen: https://twitter.com/arbvision/status/1460774216136024071

  [edit: appended "open standards" above]

• by endisneigh on 12/7/21, 5:58 PM

  There is no problem that cannot be solved with a trusted central body that could be solved by using a blockchain. People fundamentally misunderstand - a blockchain doesn't mean you don't have trust, it just means you're trusting something else.

  A hypothetical, un-compromisable entity would be more effective than any blockchain simply because the entity wouldn't be beholden by the inherent inefficiency of maintaining decentralized consensus.

• by MichaelRazum on 12/7/21, 7:28 PM

  For me definitely decentralized exchanges. Look at uniswap.

  Really look forward for the day where I can buy stocks there. In my perspective the user experience is so much better. You own your "tkns". You don't need a broker or whatever to get access to an exchange. It works 24/7. Especially would be useful for people without bank access. By the way trading volume is already huge there.

• by saurik on 12/7/21, 6:31 PM

  The core issue is that trusted, central bodies are generally already evil or will become evil, whether by their own volition or as authoritarian governments force them to use their centralized control to help them accomplish their goals.

  Here is a talk I gave to Mozilla Privacy Lab back in 2017 about centralization and all of the extremely-concrete evil that has resulted, with every single example from the (long) talk directly cited (as every slide is a news article). And the sad thing is that every few days there is some new high-profile abuse of centralized power that I always feel also deserves to be in an updated version of this talk... it is a never-ending issue as people suck.

  https://youtu.be/vsazo-Gs7ms

  So, I feel like your very question of trying to solve things that can't be solved with centralized systems is flawed; hell: I studied decentralized systems in grad school, and I have been saying for as long as I can remember that "anything you solve with a decentralized system I can solve cheaper and faster and better with a centralized one"... but you know what I can't do? I can't solve it in a way that doesn't lead to at least some moral landmines due to my now having chosen to take those shortcuts by having centralized control of the result.

  Inherently, then, I believe the reason to work on blockchain stuff is not to solve things that "can't" be solved using centralized systems... it is to figure out how to re-solve the things that people already solved using centralized systems without putting any individual or small cabal in charge of anything that would let them do evil things. And it is then to figure out how to do well enough with what results that people are willing to put up with it probably being a worse result on the surface because it is a better result for humanity.

  And like, that's a nearly impossible bar, and maybe to really get there will require regulation of centralized systems to make them illegal... which I realize will probably never happen :(. But that doesn't make working on these systems dumb, the way a lot of people here seem to think: it makes it all the more noble, as everyone who isn't doing this are actively making the world a worse place to live (and in some of the more egregious examples, such as everyone who chooses to work at Apple, have pretty direct blood on their hands from issues in countries like China).

  Hell... looking at your comment history, I see you recently being excited about the idea that Google could use their fleet of self-driving cars to create a mass road surveillance network capable of logging the license plates of "bad drivers" (and from thread context, "reporting them")... Google already is used as a stooge for governments (even as they push back, they still provide tons of data) but at least it is almost always about their own users: expanding their charge into watching other people is just evil and will lead to unprecedented ways to abuse this information stockpile :(.

• by infinitesunrise on 12/7/21, 5:44 PM

  DAOs, distributed autonomous organizations. Since you can script contract behavior on a blockchain, you can prefigure and automate the governance of an open organization in any way you desire: Cooperatives, worker ownership, covenants, voting by shares, etc. BBLLC (Blockchain-Based Limited Liability Corporation) laws are becoming a thing, for example Vermont recently legalized them, meaning that a DAO can now be legally binding and not just an organizational novelty.

  I think that over the next decade or so this will become the greatest impact of blockchains outside of finance.

• by smoldesu on 12/7/21, 5:37 PM

  Very few organizations will willingly apply decentralization to their system when they make insane amounts of money off of their centralized assets. It's a lose-lose scenario in their eyes, so the majority of successful companies eyeing the blockchain are really just pointing and laughing. "You mean they can buy our Netflix originals and not have to pay us a monthly fee to watch them? What a joke!"

• by raghavtoshniwal on 12/7/21, 7:32 PM

  Hard-pressed to come up with something concrete but blockchain is a way to build consensus. Mostly, it's more expensive than centralized consensus. In some cases though, it's cheaper.

  For ex - Cost of consensus by the govt in a stable democracy is much lower than a failed state. Blockchain could be used there.

• by scottiebarnes on 12/7/21, 5:54 PM

  Prediction markets, although this is becoming less necessary as governments come around to allow sports betting. Still, there are probably many markets that will never allow legal betting.

• by NicoJuicy on 12/7/21, 9:06 PM

  Energy transfer between consumers and producers cross border

  Similar to this but it's still too simplistic :

  https://www.frontiersin.org/articles/10.3389/fenrg.2021.6711...

  It also requires more data points than payments. Since the current solutions aren't better than VISA i don't think it can manage that much data.

• by high_byte on 12/7/21, 8:28 PM

  AMM - Automated Market Makers. Probably most (liquid) money is in those.

• by jallen_dot_dev on 12/7/21, 7:17 PM

  > problems that could not have been solved "the old way" (i.e. trusting a central body).

  If this is the criterion then the answer is going to be "nothing". Because anything a blockchain can do, somebody running 1 server could do. The whole point is to do things just without the central authority part.

• by matteroftime on 12/7/21, 7:01 PM

  I've been studying the space for a few years now and from what I gather is similar to what @saurik is saying. Blockchain-based applications are prime candidates for adversarial networks where a centralized entity will at some point be incentivized to take actions that benefit itself and not the network.

  This is why I believe that these applications have little use in spaces other than finance. For example, the NYSE is governed by a central entity that acts as a market maker, buying and selling securities to "make" the market. This is all done behind closed doors in a way that the vast majority of people do not understand. A Blockchain-based market maker provides much more transparency and stability than a centralized one governed by human decisions and thus emotions.

  However, very few, if any Blockchain projects are truly decentralized as they are still developed and changed by a central organization. We need more projects that are simple, efficient, financial tools that are immutable on the Blockchain that people can use and there is no incentive from a central authority to change the system in order to achieve their goals or boost their profits.