4 ms·

Look at Namecoin or Sia for example - Namecoin makes the ability to buy NS records scarce. Sia makes it possible to buy and sell digital storage. This only app

Look at Namecoin or Sia for example - Namecoin makes the ability to buy NS records scarce. Sia makes it possible to buy and sell digital storage. This only applies to a few things for which ownership of the content is not the content itself and for which one party has more interest than another. If I pirate a TV show, I have all the value of the content. On the other hand, if I read someone's DNS name, I do not have the value of controlling what it's set to. Similarly, if I want to buy storage, someone needs to be guaranteeing I can hold that or putting scarce resources that have value on the line if they fail.

acdha 9y ago
Both of those are well-established service categories with tons of options at various levels. What exact advantages come from adding a blockchain?
- problems 9y ago
  It differs a on a per-service basis. In the case of Sia, existing services are unable to take advantage of unused disk on other people's machines that is of little value to them, but more value to someone else - making storage extremely cheap compared to other existing services. The virtual scarcity provided by the blockchain provides a completely decentralized way to sell this storage with no 3rd party involvement. In the case of Namecoin, it removes the need for centralization of naming services and in doing so prevents external parties from interfering with naming services, the owner decides where it goes, not a registrar, not ICANN, not the US government. Generally speaking, decentralization is in one way or another the major argument for services done this way. Blockchains enable decentralization of scarce resources - but that's it. We need to stop proposing them for everything when they're really not appropriate for everything.
- Taek 9y ago
  Speaking as a co-founder of Sia, the primary advantage of a blockchain is the inability of any third party to interfere with your infrastructure. This also means you aren't depending on anybody to have uptime. Today, when you put your data on a cloud, you select and trust a service provider. They essentially own all of your infrastructure at that point, and can easily disrupt your business or life if they decide that they don't like you. They can also do things like change price based on income level. Etc. Sia removes that uncertainty from the equation. It's based on algorithms instead of people, and there's no controlling authority. It's a free market for data infrastructure with low barrier to entry and fewer places to introduce unfair competition. But most of all, it's going to be stable in the event of political chaos. You don't have to be an aws based logistics company worrying that Amazon is moving into the logistics space and now has a conflict of interest. You don't have to be a company in Turkey worried that the US is going to apply sanctions that revoke your access to the cloud. Blockchains in general are a response to the systemic risk that we've built into the internet. I don't think that most people realize that they are useful for that (mostly because blockchains like ethereum really aren't useful for that)
  - acdha 9y ago
    > They essentially own all of your infrastructure at that point, and can easily disrupt your business or life if they decide that they don't like you. They can also do things like change price based on income level. Etc. These are theoretically possible but never happens in practice because those are competitive markets in countries with established legal systems. Amazon directly competes with Netflix but if they were so inclined they'd be gifting business to Google, Microsoft, etc. and almost certainly facing significant legal repercussions. The flip side of that is that you have contracts and SLAs. Most of the questions I'd have about something like Sia come down to the same issue: the homepage doesn't seem to make precise statements about durability, response times, etc. What guarantees does a potential user have for any of that, especially if, say, someone has infrastructure issues or decides to stop participating and removes their capacity? Since you mentioned political instability, how do you avoid the Tor problem as soon as someone uploads legally risky material? Does that mean that the entire service is blocked because there's no more granular mechanism, or that participants need to factor possibly significant legal costs into their business?
    - problems 9y ago
      All files are encrypted on the hosts and only can be decrypted with the key, usually present only on the storage buyer's machine. All files are uploaded to 3 different hosts to avoid issues if someone flakes out, and could be to more if needed and every hoster puts some money on the line if they fail to host the files, providing a sort of financial insurance policy. This may not be good enough for every scenario, but I think for most it offers the possibility of covering things quite well. There are only a few major cloud storage providers around, the market just isn't that big and they're all quite expensive comparatively - for long term storage of backups and low-value data I find it quite a good concept. Publicly accessed data like S3 is often used for, not so much, that's more of a target for a traditional cloud provider or something like IPFS in the distributed world. Sia's more creating a new market for extremely cheap, reasonably reliabile storage rather than replacing one for expensive very high reliability storage in my view. But by nature of the way the network works, it should be possible to tell the # of hosts needed to match Amazon or Google in reliability - it's not in a state where all of that's possible yet though. > facing significant legal repercussions. There's nothing illegal about price discrimination, it's a common buisiness practice and something companies strive for - they want to make sure you pay the maximum price you're willing to. To do otherwise is essentially a failure to their investors.

acdha 9y ago

Both of those are well-established service categories with tons of options at various levels. What exact advantages come from adding a blockchain?

problems 9y ago
It differs a on a per-service basis. In the case of Sia, existing services are unable to take advantage of unused disk on other people's machines that is of little value to them, but more value to someone else - making storage extremely cheap compared to other existing services. The virtual scarcity provided by the blockchain provides a completely decentralized way to sell this storage with no 3rd party involvement. In the case of Namecoin, it removes the need for centralization of naming services and in doing so prevents external parties from interfering with naming services, the owner decides where it goes, not a registrar, not ICANN, not the US government. Generally speaking, decentralization is in one way or another the major argument for services done this way. Blockchains enable decentralization of scarce resources - but that's it. We need to stop proposing them for everything when they're really not appropriate for everything.
Taek 9y ago
Speaking as a co-founder of Sia, the primary advantage of a blockchain is the inability of any third party to interfere with your infrastructure. This also means you aren't depending on anybody to have uptime. Today, when you put your data on a cloud, you select and trust a service provider. They essentially own all of your infrastructure at that point, and can easily disrupt your business or life if they decide that they don't like you. They can also do things like change price based on income level. Etc. Sia removes that uncertainty from the equation. It's based on algorithms instead of people, and there's no controlling authority. It's a free market for data infrastructure with low barrier to entry and fewer places to introduce unfair competition. But most of all, it's going to be stable in the event of political chaos. You don't have to be an aws based logistics company worrying that Amazon is moving into the logistics space and now has a conflict of interest. You don't have to be a company in Turkey worried that the US is going to apply sanctions that revoke your access to the cloud. Blockchains in general are a response to the systemic risk that we've built into the internet. I don't think that most people realize that they are useful for that (mostly because blockchains like ethereum really aren't useful for that)
- acdha 9y ago
  > They essentially own all of your infrastructure at that point, and can easily disrupt your business or life if they decide that they don't like you. They can also do things like change price based on income level. Etc. These are theoretically possible but never happens in practice because those are competitive markets in countries with established legal systems. Amazon directly competes with Netflix but if they were so inclined they'd be gifting business to Google, Microsoft, etc. and almost certainly facing significant legal repercussions. The flip side of that is that you have contracts and SLAs. Most of the questions I'd have about something like Sia come down to the same issue: the homepage doesn't seem to make precise statements about durability, response times, etc. What guarantees does a potential user have for any of that, especially if, say, someone has infrastructure issues or decides to stop participating and removes their capacity? Since you mentioned political instability, how do you avoid the Tor problem as soon as someone uploads legally risky material? Does that mean that the entire service is blocked because there's no more granular mechanism, or that participants need to factor possibly significant legal costs into their business?
  - problems 9y ago
    All files are encrypted on the hosts and only can be decrypted with the key, usually present only on the storage buyer's machine. All files are uploaded to 3 different hosts to avoid issues if someone flakes out, and could be to more if needed and every hoster puts some money on the line if they fail to host the files, providing a sort of financial insurance policy. This may not be good enough for every scenario, but I think for most it offers the possibility of covering things quite well. There are only a few major cloud storage providers around, the market just isn't that big and they're all quite expensive comparatively - for long term storage of backups and low-value data I find it quite a good concept. Publicly accessed data like S3 is often used for, not so much, that's more of a target for a traditional cloud provider or something like IPFS in the distributed world. Sia's more creating a new market for extremely cheap, reasonably reliabile storage rather than replacing one for expensive very high reliability storage in my view. But by nature of the way the network works, it should be possible to tell the # of hosts needed to match Amazon or Google in reliability - it's not in a state where all of that's possible yet though. > facing significant legal repercussions. There's nothing illegal about price discrimination, it's a common buisiness practice and something companies strive for - they want to make sure you pay the maximum price you're willing to. To do otherwise is essentially a failure to their investors.