5 ms·

DAOs are unable to perform any interesting calculation due to excessive cost of computation. It can't even solve a Linear Program to distribute resources effect

DAOs are unable to perform any interesting calculation due to excessive cost of computation. It can't even solve a Linear Program to distribute resources effectiently (let alone actual scheduling which requires mixed integer problems which often take many core hours on actual CPUs). Any function a DAO can perform a notary can perform cheaper, including running votes.

bsedlm 5y ago
> Any function a DAO can perform a notary can perform cheaper, including running votes. However the DAO can the notary's job more transparently and without corruption (real oversight of their activities is cheaper... heck, it's possible)
rmbyrro 5y ago
A notary can be bribed. You can theoretically corrupt a blockchain, but it's harder than one notary. Aside, notaries where I come from are extremely expensive and inefficient. I'd choose a blockchain in a split second if I could, for anything requiring human notaries now.
- bsedlm 5y ago
  I agree with this statement, people downvoting parent are almost certainly living in not-very corrupt countries.
- danvayn 5y ago
  Not to mention that the blockchain we're talking about is supposed to get a lot more performant very soon.
  - FridgeSeal 5y ago
    “It’ll be better soon! Just as soon as <x> gets implemented! Just you wait and see!” We’ve been hearing this for basically as long as cryptocurrencies have been around. I suspect we will see a currency that can do decent throughput, but I think it will require advances in the underlying theory for that to happen and that’s a few years away.
    - The_rationalist 5y ago
    - CyberDildonics 5y ago
      There aren't any advances necessary, blocks just need to be made bigger, happen more frequently or both. Bitcoin cash has 32MB blocks instead of the 700KB average that bitcoin has and monero has a dynamic system. Litecoin, dogecoin and whatever else could handle plenty more throughput by just upping their block size. There is no actual technical barrier to this. The only reason people think it's a problem is because bitcoin and ethereum have limited themselves so much.
      - bastawhiz 5y ago
        Just last week this site featured an article about Chesterton's Fence: https://fs.blog/chestertons-fence/ https://fs.blog/chestertons-fence/ Surely the people who design the protocols aren't intentionally sabotaging the usefulness of the blockchain with these limits? If the limits are raised, what problems are introduced, and what advances are needed to avoid them?
        CyberDildonics 5y ago
        This article is abstract, hand waving, vague prediction nonsense in this context. It already works to have large blocks, it always has. Why don't you tell me specifically and technically where you think the bottleneck is? The CPU is a fraction of one core, bitcoin's transaction fees have been at times more than the cost of hard drive space to store the entire chain, and only servers even need to sync chains in the first place. Bitcoin is crippled to sell a second layer. If you look at its throughput, it is literally less than a dial up modem. The average block size is 700KB every 10 minutes. You can print base64 characters on a laser printer and get more data throughput on paper. No one has ever been able to give me a credible answer here because it doesn't exist. The best anyone can do is gish gallop with unrelated nonsense hoping that other people who read their reply can't make sense of it.
- _vertigo 5y ago
  Notaries can be bribed, sure, but smart contracts can and will have bugs.
  - ipaddr 5y ago
    But will those bugs matter or not get patched or not exist? Open source is used daily and you could say the same about it.
    - sdenton4 5y ago
      Try rolling back a bug exploit on a blockchain... This is one of the fundamental problems with blockchains: Local problems often lead to global consequences, requiring forks of the blockchain and/or high level intervention to maintain consistency or to roll back exploits. Here's the classic example: https://www.coindesk.com/learn/2016/06/25/understanding-the-dao-attack/ https://www.coindesk.com/learn/2016/06/25/understanding-the-... 'Don't write bugs' is not a viable strategy for software development.
    - bastawhiz 5y ago
      If a notary lies or commits fraud, you still have legal means to undo the damage. You can't yank back funds or mutate data on a blockchain.
      - rmbyrro 5y ago
        Justice is never an option for me. I'm just an average middle class citizen. It's rigged against myself, telling from personal experience. If there's a risk for me to be defrauded, Justice is no recourse option. I'd rather avoid the risk in the first place.
- freemint 5y ago
  Notaries are liable with their property for negligence. Notaries are also ensured. You can actually recover damages when a notary did it.
Grimburger 5y ago
The only DAO I've participated in was Bisq and it works just fine going on years now, resources get allocated for development and ops, plus the work done on marketing, documentation, etc. This is all paid for by trading fees on the exchange. Some people turn up and do dev work for one round, others participate for years. Not sure what your threshold is for "interesting calculations" but running an entire tech business for years on end with no central governance or traditional corporate structure seems incredible to me. You can participate too if you wish, maybe try drive it into the ground for fun? Or maybe run a seed server with decent uptime and get paid for it instead? Rather than theoretical putdowns based on some strange misconceptions of how this stuff works under the hood, that's a real life counter example. https://bisq.network/dao/ https://bisq.network/dao/
- andrepd 5y ago
  I don't get it. What does that DAO do exactly? They talk about distributing revenue, but revenue for what?
  - zucker42 5y ago
    Looks like Bisq is a decentralized exchange. You run the software and it automatically connects with others trying to trade the same pair of cryptocurrencies in the other direction. I got this from the FAQ. Presumably the DAO manages the decentralized exchange.
    - Grimburger 5y ago
      Pretty much this. It's a P2P trading platform that runs on a native Java GUI with seed server backends, people pay fees to trade on there, this goes into a consolidated funding pool under no ones control, stakeholders vote on paying contributors for new development/running the servers/etc. Really anything can be applied for as a contribution, you can rock up right now and say you'll translate the docs into Esperanto for X amount and the community decides if that's worth paying out of the pool for this round, if rejected you can apply again next round, no hard feelings. No one is in control, it's all decided communally with votes recorded onchain for each proposal and paid out automatically if they pass, they have coloured coins as the proxy for the votes (sort of like stock voting but digitally and with cryptographic signatures), those same coins are also what people pay for trading fees, making a closed loop between funding and advancing the DAO. If people trade less, there's less money for dev work, creating incentives for a range of stakeholders to collectively make proper choices and produce good results. Never once have they raised money from speculators, no ico or anything. All funding has come from people using the product itself.

bsedlm 5y ago

> Any function a DAO can perform a notary can perform cheaper, including running votes. However the DAO can the notary's job more transparently and without corruption (real oversight of their activities is cheaper... heck, it's possible)

rmbyrro 5y ago

A notary can be bribed. You can theoretically corrupt a blockchain, but it's harder than one notary. Aside, notaries where I come from are extremely expensive and inefficient. I'd choose a blockchain in a split second if I could, for anything requiring human notaries now.

bsedlm 5y ago
I agree with this statement, people downvoting parent are almost certainly living in not-very corrupt countries.
danvayn 5y ago
Not to mention that the blockchain we're talking about is supposed to get a lot more performant very soon.
- FridgeSeal 5y ago
  “It’ll be better soon! Just as soon as <x> gets implemented! Just you wait and see!” We’ve been hearing this for basically as long as cryptocurrencies have been around. I suspect we will see a currency that can do decent throughput, but I think it will require advances in the underlying theory for that to happen and that’s a few years away.
  - The_rationalist 5y ago
  - CyberDildonics 5y ago
    There aren't any advances necessary, blocks just need to be made bigger, happen more frequently or both. Bitcoin cash has 32MB blocks instead of the 700KB average that bitcoin has and monero has a dynamic system. Litecoin, dogecoin and whatever else could handle plenty more throughput by just upping their block size. There is no actual technical barrier to this. The only reason people think it's a problem is because bitcoin and ethereum have limited themselves so much.
    - bastawhiz 5y ago
      Just last week this site featured an article about Chesterton's Fence: https://fs.blog/chestertons-fence/ https://fs.blog/chestertons-fence/ Surely the people who design the protocols aren't intentionally sabotaging the usefulness of the blockchain with these limits? If the limits are raised, what problems are introduced, and what advances are needed to avoid them?
      - CyberDildonics 5y ago
        This article is abstract, hand waving, vague prediction nonsense in this context. It already works to have large blocks, it always has. Why don't you tell me specifically and technically where you think the bottleneck is? The CPU is a fraction of one core, bitcoin's transaction fees have been at times more than the cost of hard drive space to store the entire chain, and only servers even need to sync chains in the first place. Bitcoin is crippled to sell a second layer. If you look at its throughput, it is literally less than a dial up modem. The average block size is 700KB every 10 minutes. You can print base64 characters on a laser printer and get more data throughput on paper. No one has ever been able to give me a credible answer here because it doesn't exist. The best anyone can do is gish gallop with unrelated nonsense hoping that other people who read their reply can't make sense of it.
_vertigo 5y ago
Notaries can be bribed, sure, but smart contracts can and will have bugs.
- ipaddr 5y ago
  But will those bugs matter or not get patched or not exist? Open source is used daily and you could say the same about it.
  - sdenton4 5y ago
    Try rolling back a bug exploit on a blockchain... This is one of the fundamental problems with blockchains: Local problems often lead to global consequences, requiring forks of the blockchain and/or high level intervention to maintain consistency or to roll back exploits. Here's the classic example: https://www.coindesk.com/learn/2016/06/25/understanding-the-dao-attack/ https://www.coindesk.com/learn/2016/06/25/understanding-the-... 'Don't write bugs' is not a viable strategy for software development.
  - bastawhiz 5y ago
    If a notary lies or commits fraud, you still have legal means to undo the damage. You can't yank back funds or mutate data on a blockchain.
    - rmbyrro 5y ago
      Justice is never an option for me. I'm just an average middle class citizen. It's rigged against myself, telling from personal experience. If there's a risk for me to be defrauded, Justice is no recourse option. I'd rather avoid the risk in the first place.
freemint 5y ago
Notaries are liable with their property for negligence. Notaries are also ensured. You can actually recover damages when a notary did it.

Grimburger 5y ago

The only DAO I've participated in was Bisq and it works just fine going on years now, resources get allocated for development and ops, plus the work done on marketing, documentation, etc. This is all paid for by trading fees on the exchange. Some people turn up and do dev work for one round, others participate for years. Not sure what your threshold is for "interesting calculations" but running an entire tech business for years on end with no central governance or traditional corporate structure seems incredible to me. You can participate too if you wish, maybe try drive it into the ground for fun? Or maybe run a seed server with decent uptime and get paid for it instead? Rather than theoretical putdowns based on some strange misconceptions of how this stuff works under the hood, that's a real life counter example. https://bisq.network/dao/ https://bisq.network/dao/

andrepd 5y ago
I don't get it. What does that DAO do exactly? They talk about distributing revenue, but revenue for what?
- zucker42 5y ago
  Looks like Bisq is a decentralized exchange. You run the software and it automatically connects with others trying to trade the same pair of cryptocurrencies in the other direction. I got this from the FAQ. Presumably the DAO manages the decentralized exchange.
  - Grimburger 5y ago
    Pretty much this. It's a P2P trading platform that runs on a native Java GUI with seed server backends, people pay fees to trade on there, this goes into a consolidated funding pool under no ones control, stakeholders vote on paying contributors for new development/running the servers/etc. Really anything can be applied for as a contribution, you can rock up right now and say you'll translate the docs into Esperanto for X amount and the community decides if that's worth paying out of the pool for this round, if rejected you can apply again next round, no hard feelings. No one is in control, it's all decided communally with votes recorded onchain for each proposal and paid out automatically if they pass, they have coloured coins as the proxy for the votes (sort of like stock voting but digitally and with cryptographic signatures), those same coins are also what people pay for trading fees, making a closed loop between funding and advancing the DAO. If people trade less, there's less money for dev work, creating incentives for a range of stakeholders to collectively make proper choices and produce good results. Never once have they raised money from speculators, no ico or anything. All funding has come from people using the product itself.