5 ms·

Really sad that crypto has such a bad reputation in HN. As hackers we’re supposed to be challenging the entrenched Big Tech monopoly with its silos, but every t

by EGreg 4y ago

Really sad that crypto has such a bad reputation in HN. As hackers we’re supposed to be challenging the entrenched Big Tech monopoly with its silos, but every time someone like IPFS or BitTorrent or MaidSAFE releases a token it’s somehow considered worse than the extremely centralized “cloud offerings” or privately owned networks. It’s crazy. You say “a16z content is excellent” but a16z is the largest VC firm, with the VC industry funding closed source proprietary companies that extract rents and capture the market. People like Peter Thiel who say “competition is for losers, build a monopoly” and make sure Mark Z never builds wirehog, a peer to peer file sharing network, because that wouldn’t extract enough rents. People like Adam Neumann who build money-losing companies and dump the shares on the public, how are they any better than say the folks behind MaidSAFE who have been toiling since 2006 to bring everyone a FREE and open source secure file sharing network? Crypto is cryptography powering open, byzantine fault tolerant protocols. It is the future. People governing things together (currently called DAOs). Code you can trust. And so on. What sucks is blockchain, not crypto.

logisticseh 4y ago
The characterization of crypto stuff as an alternative to Big Tech is, frankly, one of the most bewildering "Emperor Has No Clothes" situations I've seen in my life. And I've seen a few tech hype cycles.
- EGreg 4y ago
  What do you think byzantine-fault-tolerant networks are? With redundancies built in, so your data is guaranteed to be online regardless of any third party, and so on? For example, everything I've built and given away over the last 10 years has been building towards an alternative to Big Tech: https://intercoin.org/overview.pdf https://intercoin.org/overview.pdf
  - colinsane 4y ago
    > What do you think byzantine-fault-tolerant networks are? With redundancies built in, so your data is guaranteed to be online regardless of any third party, and so on? as for data availability, bittorrent works stupidly well today without any monetization. it turns out that upload/download ratio works as a pretty decent metric to combat tragedy of the commons for larger torrent communities that really want to protect against it. for websites, i've been surprised in the past when i let a domain expire, and then it pops up a month later because somebody who enjoyed it had archived it and brought it back online themselves. that's more satisfying to me than if it had been kept available because i paid some distant/dispassionate person $1 to do so. point is, the future i want is largely one of reciprocity. decrease hierarchy and encourage participants to act as both consumer and producer. financialization doesn't do this: it encourages large, efficient producers to compete with each other to serve consumers. i don't want my 20,000-person retro-arcade community to be served by 5 people who've provisioned 4x10Gbps uplinks charging us for BW. then it's just some extractive thing (these producers are likely directing their proceeds to things unrelated to the community) and the magic's gone. financialization helps when the capital cost of entry is otherwise too high. say, using loans that are repaid by usage-based micropayments as a way to get mesh power/data networks off the ground since these require costly physical equipment. such networks would be beneficial both from the redundancy/fault-tolerance perspective you care about and also from the social angle i care about of being able to participate in more of the things around me. but after deployment, continued monetization works against that latter aspect. the one crypto project which financialized data networks (Helium) completely failed at the participatory aspect. its users (gateway operators) are merely providers for a different set of consumers (LoRa users). you may have good intentions, and you may build something better than FB/AWS, but if it's built around monetization it will only ever be a stopgap. stopgaps are fine and often good, but the places we need them are already shrinking. federated chat/social networks and peer-based media/streaming applications are developed and advancing rapidly enough for me to take them seriously today. another "big tech" alternative -- open SW/HW stacks for mobile phones as an alternative to Apple -- is in a similar state and crypto hasn't even started any parallel project that i know of. search alternatives are the furthest behind, but is anyone using crypto to build an alternative to Google? my understanding of search is that it's a tough problem to solve because of the money incentives (if a click results in profit via commerce or ads, then there's an incentive to pay for more clicks, either via backchannel payments or SEO), and that social ranking of search results is one of the only ways to overcome this.
    - EGreg 4y ago
      Thank you! You gave me a lot to think about. Is there any chance we can connect (by email or zoom) to discuss this further? My work has brought me in contact with people ranging from economics (Thomas Greco) to politics (Andrew Yang, Forward Party, Noam Chomsky) to distributed systems (Leslie Lamport, Petary Maymounkov) to regulations (Sara Hanks) and so on. I'm constantly looking for what systems work best for certain things, how they work, etc. I'm often seen advocating for UBI and single payer health insurance, and collaborative open source movements instead of closed-source silos. My work is at the intersection of Web1 (publishing and speech) Web2 (collaboration and communication) and Web3 (transactions and governance) so naturally a lot of this comes up. Example: https://community.qbix.com/t/freedom-of-speech-and-capitalism-in-2021-interview-with-noam-chomsky-political-commentator https://community.qbix.com/t/freedom-of-speech-and-capitalis... So I'd be very interested to hear and discuss your ideas about fostering a participatory economy that helps protect against the emergence of hierarchies and extraction of rent. How can we prevent people banding together and forming organizations that dominate others? You can contact me through the email in my profile. Or is there a way I can contact you?
      - colinsane 4y ago
        i've added to my HN profile a link to my about page, which has email/matrix/etc. my answers to your economic questions may be less concrete or justified than you'd hope for. they're generally aligned by some broad goal of "decrease the cost of baseline human existence to $0, and leave the culture to figure out the rest." in a previous era that would be phrased in terms of "post scarcity" of the base needs like food and shelter. from the scarcity angle, automation is the way out and if it's widely replicable then many of the incentives that drive hostile behavior are reduced. in a previous life i was active in personal manufacturing, and creator/maintainer of some 3d-printing firmware that later expanded into open source pick-and-place machines. today i'm more active in the base layer of computation: how can we fabricate logic gates/integrated circuits without costly, centralized fabs. i'm wrapping up some experiments around ferromagnetic logic gates. such logic gates would use materials that you can assemble safely on a SLS printer; for now i have only the repo -- i hope to write something more accessible in a month or so. https://git.uninsane.org/colin/fdtd-coremem https://git.uninsane.org/colin/fdtd-coremem
  - logisticseh 4y ago
    > What do you think byzantine-fault-tolerant networks are? LOL. I've been working on distributed systems for... a while. I've even had a few conversations with Lamport, pre-Turing award. Apparently today BFT is a buzzword used by alt-suits to hawk unnecessary solutions to fake problems. > With redundancies built in, so your data is guaranteed to be online regardless of any third party, and so on? 1. That's... not what byzantine fault tolerance gives you. 2. Fault tolerance was a huge problem in the 90s and 00s but today is in most practical cases a solved problem (and without any use of "crypto"). > For example, everything I've built and given away over the last 10 years has been building towards an alternative to Big Tech: https://intercoin.org/overview.pdf https://intercoin.org/overview.pdf Lots of nudity AFAICT.
    - EGreg 4y ago
      I know that BFT doesnt automatically imply the data has redundancies built in, which is why I mentioned it separately. As for fault tolerance being “solved”, how’s Google Reader treating you? Are you able to control your Facebook or Instagram experience? I hear they know their new changes will irk people but will do them anyway. You think trusting others and giving them all the control to change something under you at any time is good because they trust themselves? That’s cute. Maybe they’ll even call their offering “the cloud” and you’ll believe it is like the Internet. Put all your photos there and don’t make any backups. Heck, put all your legal agreements there too! Who needs multiple distrusting parties to secure a ledger? https://www.washingtonpost.com/technology/2022/07/27/instagram-video-shift-kardashian/ https://www.washingtonpost.com/technology/2022/07/27/instagr... Don’t worry, though. You’ll own nothing and be happy. Software as a service. DRM books that can be removed. Centralizing all the data in one place can attract all kinds of government agencies and advertisers hungry for cheap data. Heck, if government agents want, they can even impersonate you, and all they have to do is send one letter to people you never met! How nice: https://tutanota.com/blog/posts/australia-surveillance-bill https://tutanota.com/blog/posts/australia-surveillance-bill How can Hackers support these “anti Hacking bills” that let the governments and police hack your accounts and impersonate you… I guess I should come out with several flavors of ice cream for many in the HN crowd — it’s bootlickin good!
      - 4y ago
spaceman_2020 4y ago
Even as someone who believes in the blockchain, its a little rich to pose it as a counter to Big Tech. Its already being centralized and co-opted by the big guys.
- EGreg 4y ago
  Yes, as I said, blockchain sucks. It requires large nodes to run the gateways and people just use those gateways. And so the Big Boys (TM) can come and run those. For example OpenSea may as well be a centralized site with an API for everyone, since all the wallets hit that API anyway. Is that what you want? Are you just throwing your hands up in the air and saying we can never get away from extreme centralization, or are you saying we SHOULD have the extreme centralization? But the idea of crypto isn't just to work on a blockchain. It's about having public-private keys that you use to issue transactions on a decentralized, permissionless network. For example the Dat protocol / hypercore. What about all the other architectures? Is BitTorrent being centralized and co-opted, for instance? Is IOTA tangle or MaidSAFE or any of the other architectures?
  - rglullis 4y ago
    That "large node to run the gateways" cost me a total of $250, hard drives included. If you want to spend even less, people can get an old laptop and install dappnode on it. They can even share with their friends. On my dappnode I bought a single 3TB drive to set up Storj. It paid for itself in the first year, now it generates enough revenue to pay for its power consumption and turn a very modest profit. Based on many different nodes just like mine, Storj offers a S3-compatible service at a fraction of the cost from AWS.
  - UncleMeat 4y ago
    > It's about having public-private keys that you use to issue transactions on a decentralized, permissionless network. PGP has been around for many decades. Not exactly lighting the world on fire.
jrmg 4y ago
‘Byzantine’ is good now?
- EGreg 4y ago
  Yes, Byzantine Fault Tolerance is very good.
  - jrmg 4y ago
    Ah. Quite. I not only misunderstood that, I fell into the trap you’re complaining about, read your comment too fast, and associated ‘crypto’ with ‘cryptocurrencies’. Related, and perhaps to salvage something productive from my confusion: I think we’re well past being able to rescue the word ‘crypto’ from being synonymous with cryptocurrencies in general, and blockchains in particular, in the minds of most. For general ‘crypto’, I think it’s time to start using ‘cryptography’.
- yjftsjthsd-h 4y ago
  Probably not being used the way you think it is: https://en.wikipedia.org/wiki/Byzantine_fault https://en.wikipedia.org/wiki/Byzantine_fault
b1ue64 4y ago
Crypto depends on blockchain. You clearly don't know much about the technology you're peddling.
- EGreg 4y ago
  Really? Generating crytographic public keys, encryption, decryption, signing, ring signatures, Merkle Trees, cryptographic hashes, all this depends on blockchain? WOW!
  - decode 4y ago
    I too found it confusing when people started using "crypto" to be short for "cryptocurrency" (and later, all blockchain-related things) rather than "cryptography". However, that ship has sailed and as Moxie Marlinspike points out in his NFTs post [0], that's now a Pepperidge Farm Remembers meme. I've come to accept it as a shift in usage and using the older meaning of crypto will just result in miscommunication, like in this thread. So, to answer your question: no, those things don't depend on blockchain, but they're also not "crypto", according to the current usage of the word. [0] https://moxie.org/2022/01/07/web3-first-impressions.html https://moxie.org/2022/01/07/web3-first-impressions.html
    - EGreg 4y ago
      Well, these things are all used in crypto. And lots of other Web3 applications. None of them require a blockchain ! https://community.intercoin.org/t/web3-moxie-signal-telegram-and-why-decentralization-matters/2230 https://community.intercoin.org/t/web3-moxie-signal-telegram...
nonrandomstring 4y ago
People have a limited number of boxes into which to fit a very rich set of concepts emerging in the world of applied computing today. Unfortunately "tech people" are no better than average in dealing with breadth and nuance - and may actually be worse since we specialise. We're also equally vulnerable to the mass-media distortions of emerging concepts ("AI", "Crypto", "Social Networks" etc.), often at the behest of incumbent powers and players. All you can do as an aware hacker is patiently, politely, persistently explain in clear language. Eventually some of it sticks.
paulryanrogers 4y ago
> People governing things together (currently called DAOs). Code you can trust. And so on. Didn't the first DAO cause a fork of Ethereum? And isn't the most important part of investing understanding what you're investing in? Apparently these contacts as code are so complex in practice (DSLs require understanding it and all its dependencies.) they tend to be foot guns, waiting for someone more dedicated to find and exploit them.
- rglullis 4y ago
  The DAO didn't "cause a fork". The fork happened because there was a community consensus that the best solution to mitigate the hack was by migrating to a chain where the bug could be fixed. IOW, it was an example that governance is more sociological than technological thing.
  - paulryanrogers 4y ago
    So if the only recourse for bugs in DAOs is for the community to fork the underlying tech then it doesn't sound like a particularly robust system.
    - rglullis 4y ago
      Your mistake is thinking that the solution that was adopted for one instance of an issue is the only possible solution that can be used forever. None of this is static. The DAO hack showed a flaw in the then-current practices of smart contracts. Developers learned from it, changes were made to the system. The system became more robust after it. Much like any organism living in an ambient with evolutionary pressure, different mutations come up. After the DAO hack, some prefer to write their contracts in a way that allow some level of control by trusted admins, some of them still don't like the idea of privileged access to the contracts but will favor adoption of contracts only after formal verification. Some saw that as opportunity to specialize in contract security and become auditors. None of these approaches are perfect by themselves. Other bugs will come, other "hacks" will exist, and hopefully each of these will act as evolutionary filter on the existing system. Without a central authority, this is the only I see for systems to evolve. And if the whole point is to create an alternative that does not depend on a central authority, I'm fine with the current approaches to risk management.

by EGreg 4y ago

logisticseh 4y ago

The characterization of crypto stuff as an alternative to Big Tech is, frankly, one of the most bewildering "Emperor Has No Clothes" situations I've seen in my life. And I've seen a few tech hype cycles.

EGreg 4y ago
What do you think byzantine-fault-tolerant networks are? With redundancies built in, so your data is guaranteed to be online regardless of any third party, and so on? For example, everything I've built and given away over the last 10 years has been building towards an alternative to Big Tech: https://intercoin.org/overview.pdf https://intercoin.org/overview.pdf
- colinsane 4y ago
  > What do you think byzantine-fault-tolerant networks are? With redundancies built in, so your data is guaranteed to be online regardless of any third party, and so on? as for data availability, bittorrent works stupidly well today without any monetization. it turns out that upload/download ratio works as a pretty decent metric to combat tragedy of the commons for larger torrent communities that really want to protect against it. for websites, i've been surprised in the past when i let a domain expire, and then it pops up a month later because somebody who enjoyed it had archived it and brought it back online themselves. that's more satisfying to me than if it had been kept available because i paid some distant/dispassionate person $1 to do so. point is, the future i want is largely one of reciprocity. decrease hierarchy and encourage participants to act as both consumer and producer. financialization doesn't do this: it encourages large, efficient producers to compete with each other to serve consumers. i don't want my 20,000-person retro-arcade community to be served by 5 people who've provisioned 4x10Gbps uplinks charging us for BW. then it's just some extractive thing (these producers are likely directing their proceeds to things unrelated to the community) and the magic's gone. financialization helps when the capital cost of entry is otherwise too high. say, using loans that are repaid by usage-based micropayments as a way to get mesh power/data networks off the ground since these require costly physical equipment. such networks would be beneficial both from the redundancy/fault-tolerance perspective you care about and also from the social angle i care about of being able to participate in more of the things around me. but after deployment, continued monetization works against that latter aspect. the one crypto project which financialized data networks (Helium) completely failed at the participatory aspect. its users (gateway operators) are merely providers for a different set of consumers (LoRa users). you may have good intentions, and you may build something better than FB/AWS, but if it's built around monetization it will only ever be a stopgap. stopgaps are fine and often good, but the places we need them are already shrinking. federated chat/social networks and peer-based media/streaming applications are developed and advancing rapidly enough for me to take them seriously today. another "big tech" alternative -- open SW/HW stacks for mobile phones as an alternative to Apple -- is in a similar state and crypto hasn't even started any parallel project that i know of. search alternatives are the furthest behind, but is anyone using crypto to build an alternative to Google? my understanding of search is that it's a tough problem to solve because of the money incentives (if a click results in profit via commerce or ads, then there's an incentive to pay for more clicks, either via backchannel payments or SEO), and that social ranking of search results is one of the only ways to overcome this.
  - EGreg 4y ago
    Thank you! You gave me a lot to think about. Is there any chance we can connect (by email or zoom) to discuss this further? My work has brought me in contact with people ranging from economics (Thomas Greco) to politics (Andrew Yang, Forward Party, Noam Chomsky) to distributed systems (Leslie Lamport, Petary Maymounkov) to regulations (Sara Hanks) and so on. I'm constantly looking for what systems work best for certain things, how they work, etc. I'm often seen advocating for UBI and single payer health insurance, and collaborative open source movements instead of closed-source silos. My work is at the intersection of Web1 (publishing and speech) Web2 (collaboration and communication) and Web3 (transactions and governance) so naturally a lot of this comes up. Example: https://community.qbix.com/t/freedom-of-speech-and-capitalism-in-2021-interview-with-noam-chomsky-political-commentator https://community.qbix.com/t/freedom-of-speech-and-capitalis... So I'd be very interested to hear and discuss your ideas about fostering a participatory economy that helps protect against the emergence of hierarchies and extraction of rent. How can we prevent people banding together and forming organizations that dominate others? You can contact me through the email in my profile. Or is there a way I can contact you?
    - colinsane 4y ago
      i've added to my HN profile a link to my about page, which has email/matrix/etc. my answers to your economic questions may be less concrete or justified than you'd hope for. they're generally aligned by some broad goal of "decrease the cost of baseline human existence to $0, and leave the culture to figure out the rest." in a previous era that would be phrased in terms of "post scarcity" of the base needs like food and shelter. from the scarcity angle, automation is the way out and if it's widely replicable then many of the incentives that drive hostile behavior are reduced. in a previous life i was active in personal manufacturing, and creator/maintainer of some 3d-printing firmware that later expanded into open source pick-and-place machines. today i'm more active in the base layer of computation: how can we fabricate logic gates/integrated circuits without costly, centralized fabs. i'm wrapping up some experiments around ferromagnetic logic gates. such logic gates would use materials that you can assemble safely on a SLS printer; for now i have only the repo -- i hope to write something more accessible in a month or so. https://git.uninsane.org/colin/fdtd-coremem https://git.uninsane.org/colin/fdtd-coremem
- logisticseh 4y ago
  > What do you think byzantine-fault-tolerant networks are? LOL. I've been working on distributed systems for... a while. I've even had a few conversations with Lamport, pre-Turing award. Apparently today BFT is a buzzword used by alt-suits to hawk unnecessary solutions to fake problems. > With redundancies built in, so your data is guaranteed to be online regardless of any third party, and so on? 1. That's... not what byzantine fault tolerance gives you. 2. Fault tolerance was a huge problem in the 90s and 00s but today is in most practical cases a solved problem (and without any use of "crypto"). > For example, everything I've built and given away over the last 10 years has been building towards an alternative to Big Tech: https://intercoin.org/overview.pdf https://intercoin.org/overview.pdf Lots of nudity AFAICT.
  - EGreg 4y ago
    I know that BFT doesnt automatically imply the data has redundancies built in, which is why I mentioned it separately. As for fault tolerance being “solved”, how’s Google Reader treating you? Are you able to control your Facebook or Instagram experience? I hear they know their new changes will irk people but will do them anyway. You think trusting others and giving them all the control to change something under you at any time is good because they trust themselves? That’s cute. Maybe they’ll even call their offering “the cloud” and you’ll believe it is like the Internet. Put all your photos there and don’t make any backups. Heck, put all your legal agreements there too! Who needs multiple distrusting parties to secure a ledger? https://www.washingtonpost.com/technology/2022/07/27/instagram-video-shift-kardashian/ https://www.washingtonpost.com/technology/2022/07/27/instagr... Don’t worry, though. You’ll own nothing and be happy. Software as a service. DRM books that can be removed. Centralizing all the data in one place can attract all kinds of government agencies and advertisers hungry for cheap data. Heck, if government agents want, they can even impersonate you, and all they have to do is send one letter to people you never met! How nice: https://tutanota.com/blog/posts/australia-surveillance-bill https://tutanota.com/blog/posts/australia-surveillance-bill How can Hackers support these “anti Hacking bills” that let the governments and police hack your accounts and impersonate you… I guess I should come out with several flavors of ice cream for many in the HN crowd — it’s bootlickin good!
    - 4y ago

spaceman_2020 4y ago

Even as someone who believes in the blockchain, its a little rich to pose it as a counter to Big Tech. Its already being centralized and co-opted by the big guys.

EGreg 4y ago
Yes, as I said, blockchain sucks. It requires large nodes to run the gateways and people just use those gateways. And so the Big Boys (TM) can come and run those. For example OpenSea may as well be a centralized site with an API for everyone, since all the wallets hit that API anyway. Is that what you want? Are you just throwing your hands up in the air and saying we can never get away from extreme centralization, or are you saying we SHOULD have the extreme centralization? But the idea of crypto isn't just to work on a blockchain. It's about having public-private keys that you use to issue transactions on a decentralized, permissionless network. For example the Dat protocol / hypercore. What about all the other architectures? Is BitTorrent being centralized and co-opted, for instance? Is IOTA tangle or MaidSAFE or any of the other architectures?
- rglullis 4y ago
  That "large node to run the gateways" cost me a total of $250, hard drives included. If you want to spend even less, people can get an old laptop and install dappnode on it. They can even share with their friends. On my dappnode I bought a single 3TB drive to set up Storj. It paid for itself in the first year, now it generates enough revenue to pay for its power consumption and turn a very modest profit. Based on many different nodes just like mine, Storj offers a S3-compatible service at a fraction of the cost from AWS.
- UncleMeat 4y ago
  > It's about having public-private keys that you use to issue transactions on a decentralized, permissionless network. PGP has been around for many decades. Not exactly lighting the world on fire.

jrmg 4y ago

‘Byzantine’ is good now?

EGreg 4y ago
Yes, Byzantine Fault Tolerance is very good.
- jrmg 4y ago
  Ah. Quite. I not only misunderstood that, I fell into the trap you’re complaining about, read your comment too fast, and associated ‘crypto’ with ‘cryptocurrencies’. Related, and perhaps to salvage something productive from my confusion: I think we’re well past being able to rescue the word ‘crypto’ from being synonymous with cryptocurrencies in general, and blockchains in particular, in the minds of most. For general ‘crypto’, I think it’s time to start using ‘cryptography’.
yjftsjthsd-h 4y ago
Probably not being used the way you think it is: https://en.wikipedia.org/wiki/Byzantine_fault https://en.wikipedia.org/wiki/Byzantine_fault

b1ue64 4y ago

Crypto depends on blockchain. You clearly don't know much about the technology you're peddling.

EGreg 4y ago
Really? Generating crytographic public keys, encryption, decryption, signing, ring signatures, Merkle Trees, cryptographic hashes, all this depends on blockchain? WOW!
- decode 4y ago
  I too found it confusing when people started using "crypto" to be short for "cryptocurrency" (and later, all blockchain-related things) rather than "cryptography". However, that ship has sailed and as Moxie Marlinspike points out in his NFTs post [0], that's now a Pepperidge Farm Remembers meme. I've come to accept it as a shift in usage and using the older meaning of crypto will just result in miscommunication, like in this thread. So, to answer your question: no, those things don't depend on blockchain, but they're also not "crypto", according to the current usage of the word. [0] https://moxie.org/2022/01/07/web3-first-impressions.html https://moxie.org/2022/01/07/web3-first-impressions.html
  - EGreg 4y ago
    Well, these things are all used in crypto. And lots of other Web3 applications. None of them require a blockchain ! https://community.intercoin.org/t/web3-moxie-signal-telegram-and-why-decentralization-matters/2230 https://community.intercoin.org/t/web3-moxie-signal-telegram...

nonrandomstring 4y ago

People have a limited number of boxes into which to fit a very rich set of concepts emerging in the world of applied computing today. Unfortunately "tech people" are no better than average in dealing with breadth and nuance - and may actually be worse since we specialise. We're also equally vulnerable to the mass-media distortions of emerging concepts ("AI", "Crypto", "Social Networks" etc.), often at the behest of incumbent powers and players. All you can do as an aware hacker is patiently, politely, persistently explain in clear language. Eventually some of it sticks.

paulryanrogers 4y ago

> People governing things together (currently called DAOs). Code you can trust. And so on. Didn't the first DAO cause a fork of Ethereum? And isn't the most important part of investing understanding what you're investing in? Apparently these contacts as code are so complex in practice (DSLs require understanding it and all its dependencies.) they tend to be foot guns, waiting for someone more dedicated to find and exploit them.

rglullis 4y ago
The DAO didn't "cause a fork". The fork happened because there was a community consensus that the best solution to mitigate the hack was by migrating to a chain where the bug could be fixed. IOW, it was an example that governance is more sociological than technological thing.
- paulryanrogers 4y ago
  So if the only recourse for bugs in DAOs is for the community to fork the underlying tech then it doesn't sound like a particularly robust system.
  - rglullis 4y ago
    Your mistake is thinking that the solution that was adopted for one instance of an issue is the only possible solution that can be used forever. None of this is static. The DAO hack showed a flaw in the then-current practices of smart contracts. Developers learned from it, changes were made to the system. The system became more robust after it. Much like any organism living in an ambient with evolutionary pressure, different mutations come up. After the DAO hack, some prefer to write their contracts in a way that allow some level of control by trusted admins, some of them still don't like the idea of privileged access to the contracts but will favor adoption of contracts only after formal verification. Some saw that as opportunity to specialize in contract security and become auditors. None of these approaches are perfect by themselves. Other bugs will come, other "hacks" will exist, and hopefully each of these will act as evolutionary filter on the existing system. Without a central authority, this is the only I see for systems to evolve. And if the whole point is to create an alternative that does not depend on a central authority, I'm fine with the current approaches to risk management.