7 ms·

But what about blockchain apps?

36 points by xj9 9y ago

swatkat 9y ago
I am seeing this: Surge protection Warning:You triggered the wiki's surge protection by doing too many requests in a short time.Please make a short break reading the stuff you already got.When you restart doing requests AFTER that, slow down or you might get locked out for a longer time! Google cache: https://webcache.googleusercontent.com/search?q=cache:qGQvzKU5kBEJ:https://www.heropunch.io/xj9/2018/but-what-about-blockchain-apps+ https://webcache.googleusercontent.com/search?q=cache:qGQvzK...
- xj9 9y ago
 i'm being hugged to death! maybe i should add come caching to my nginx config
tw1010 9y ago
I had a real hard time reading this post without being disoriented by the lack of capitalization.
- freehunter 9y ago
  Yes, this is a stylistic choice some bloggers are using these days and it's pretty annoying. I'm all in favor of learning how to break the rules but in this case it just makes it harder to read without adding any benefit.
- xj9 9y ago
  i'll do the next one in all caps ;)
- ldiracdelta 9y ago
  I ran into a grizzled veteran on our company's IM platform who refused to use anything but properly capitalized and punctuated sentences inside of the IM chat. He converted me.
CoffeeDregs 9y ago
AFAICT, with current blockchain technology, in order to process PayPal level volumes, every single participant in the network would need to process PayPay level volumes... 1,000 transactions per second is not a problem for a single payments company; it seems like a giant problem for each of the 100,000 nodes on the network also to process 1,000 transactions per second. Certainly I am missing something obvious?
- wmf 9y ago
  There won't be 100,000 full nodes; more like 1,000 full nodes and 99,000 light nodes that only verify their own transactions.
- xwvvvvwx 9y ago
  This is one of the biggest problems with Blockchains right now. If it cannot be solved then their utility is extremely limited. There are two main approaches that seem likely to work: - "Layer 2" networks: Most transaction volume happens off-chain. POW blockchain remains as a settlement layer. Quite a few proposed protocols. Lightning network [1] is the most well known. First mainnet lightning transactions were sent in December [2] - Sharding: Split the blockchain into lots of different blockchains with a subset of miners validating transactions on each sub-chain. Miners are randomly assigned a new sub-chain on a regular basis to stop collusion. Each sub-chain has it's own address space. Sub-chains can communicate using asynchronus message passing. Ethereum has a draft specification [3][4] and is starting to implement some test clients [5]. [1] https://lightning.network/ https://lightning.network/ [2] https://twitter.com/alexbosworth/status/946175898029395968 https://twitter.com/alexbosworth/status/946175898029395968 [3] https://github.com/ethereum/sharding/blob/develop/docs/doc.md https://github.com/ethereum/sharding/blob/develop/docs/doc.m... [4] https://github.com/ethereum/wiki/wiki/Sharding-FAQ https://github.com/ethereum/wiki/wiki/Sharding-FAQ [5] https://blog.ethereum.org/2018/01/02/ethereum-scalability-research-development-subsidy-programs/ https://blog.ethereum.org/2018/01/02/ethereum-scalability-re...
eecks 9y ago
Surge protection warning
dvt 9y ago
Blockchain apps aren't going to happen, just how BitTorrent apps never happened. There's a bunch of reasons, but most of them come down to the fact that decentralization isn't really a boon. Consider a new social network I want to build. Making a simple prototype and hosting it on AWS is easy. Getting people to use it is hard. Getting people to use it and make them self-host some kind of "node" in this decentralized universe -- that's exponentially harder. Diaspora clearly showed that no one wants to set up and host their own nodes. And then you have to solve all kinds of really hard problems: double spending, consensus, 51%+ attacks, and so on. Some of these problems are simply unsolvable and we have to compromise. SSBT, for example, has no consensus, so everything is subjective[1] (which makes it almost impossible to do anything useful with). The people still fighting this war of decentralization remind me of the Japanese soldiers[2] who didn't realize WW2 was over. You lost. No one cares about decentralization. Google won. Amazon won. Facebook won. The great majority of people invest in BTC and ETH because they think it'll make them wealthy, not because of some noble desire to fight "the man." Besides, BTC isn't really decentralized anyway, the million-dollar-VC-backed miners basically control that ecosystem. [1] https://github.com/ssbc/secure-scuttlebutt/issues/86 https://github.com/ssbc/secure-scuttlebutt/issues/86 [2] http://www.cnn.com/2014/01/17/world/asia/japan-philippines-ww2-soldier-dies/index.html http://www.cnn.com/2014/01/17/world/asia/japan-philippines-w...
- Devagamster 9y ago
  Take a look at beaker. I'm not sure that decentralized apps wont happen. I agree that centralized servers have many benefits, but there are also benefits to giving control to users.
  - jstanley 9y ago
    I had a look at beaker earlier today. It doesn't appear to do anything IPFS doesn't do, it just has a pretty user interface. IMO it would be a much better project if it was IPFS-compatible so that it could benefit from all the existing work that is being done on IPFS.
- gsich 9y ago
  Email is decentralized and is probably the most used and most compatible system there is.
BinaryIdiot 9y ago
Not sure I entirely understand where this post is going. What does centralization versus decentralization have to do with a blockchain? Just because Bitcoin and others are using it as a centralized data store for a decentralized currency doesn't mean the blockchain has to be tied to any of that. I feel like people keep getting really far away from what a blockchain really is (a fancy linked list that can ensure integrity from start to end) and are confusing it for how Bitcoin uses it.
- fabianhjr 9y ago
  I read it as: Blockchains(PoW/PoS/etc) are not the solution for decentralized applications because blockchains are a global state and most decentralized application don't _need_ a global state. (Specially non-fintech ones that don't have a double-spend problem like social/websites/photos/etc) It briefly mentions Secure Scuttlebutt, a gossip protocol that works over meshnets without global internet access as an example of a decentralized technology without a global state.
fsiefken 9y ago
DAG (Directed acyclic graph) based currencies don't need a blockchain; Byteball/Dagcoin, IOTA, Rayblocks (Nano) and others.
jasondc 9y ago
I agree with some of the authors points. One item that is missing is the incentive structure. Let's take mesh networks for example (as mentioned in the article). Substratum pays users in coins to share their network, which creates an incentive to grow the network (as long as the coins are worth something). If the coins are worthless, the model doesn't work. It would take a lot of money to create the incentives without the coins.

36 points by xj9 9y ago

I am seeing this: Surge protection Warning:You triggered the wiki's surge protection by doing too many requests in a short time.Please make a short break reading the stuff you already got.When you restart doing requests AFTER that, slow down or you might get locked out for a longer time! Google cache: https://webcache.googleusercontent.com/search?q=cache:qGQvzKU5kBEJ:https://www.heropunch.io/xj9/2018/but-what-about-blockchain-apps+ https://webcache.googleusercontent.com/search?q=cache:qGQvzK...

xj9 9y ago
i'm being hugged to death! maybe i should add come caching to my nginx config

tw1010 9y ago

I had a real hard time reading this post without being disoriented by the lack of capitalization.

freehunter 9y ago
Yes, this is a stylistic choice some bloggers are using these days and it's pretty annoying. I'm all in favor of learning how to break the rules but in this case it just makes it harder to read without adding any benefit.
xj9 9y ago
i'll do the next one in all caps ;)
ldiracdelta 9y ago
I ran into a grizzled veteran on our company's IM platform who refused to use anything but properly capitalized and punctuated sentences inside of the IM chat. He converted me.

CoffeeDregs 9y ago

AFAICT, with current blockchain technology, in order to process PayPal level volumes, every single participant in the network would need to process PayPay level volumes... 1,000 transactions per second is not a problem for a single payments company; it seems like a giant problem for each of the 100,000 nodes on the network also to process 1,000 transactions per second. Certainly I am missing something obvious?

wmf 9y ago
There won't be 100,000 full nodes; more like 1,000 full nodes and 99,000 light nodes that only verify their own transactions.
xwvvvvwx 9y ago
This is one of the biggest problems with Blockchains right now. If it cannot be solved then their utility is extremely limited. There are two main approaches that seem likely to work: - "Layer 2" networks: Most transaction volume happens off-chain. POW blockchain remains as a settlement layer. Quite a few proposed protocols. Lightning network [1] is the most well known. First mainnet lightning transactions were sent in December [2] - Sharding: Split the blockchain into lots of different blockchains with a subset of miners validating transactions on each sub-chain. Miners are randomly assigned a new sub-chain on a regular basis to stop collusion. Each sub-chain has it's own address space. Sub-chains can communicate using asynchronus message passing. Ethereum has a draft specification [3][4] and is starting to implement some test clients [5]. [1] https://lightning.network/ https://lightning.network/ [2] https://twitter.com/alexbosworth/status/946175898029395968 https://twitter.com/alexbosworth/status/946175898029395968 [3] https://github.com/ethereum/sharding/blob/develop/docs/doc.md https://github.com/ethereum/sharding/blob/develop/docs/doc.m... [4] https://github.com/ethereum/wiki/wiki/Sharding-FAQ https://github.com/ethereum/wiki/wiki/Sharding-FAQ [5] https://blog.ethereum.org/2018/01/02/ethereum-scalability-research-development-subsidy-programs/ https://blog.ethereum.org/2018/01/02/ethereum-scalability-re...

eecks 9y ago

Surge protection warning

dvt 9y ago

Blockchain apps aren't going to happen, just how BitTorrent apps never happened. There's a bunch of reasons, but most of them come down to the fact that decentralization isn't really a boon. Consider a new social network I want to build. Making a simple prototype and hosting it on AWS is easy. Getting people to use it is hard. Getting people to use it and make them self-host some kind of "node" in this decentralized universe -- that's exponentially harder. Diaspora clearly showed that no one wants to set up and host their own nodes. And then you have to solve all kinds of really hard problems: double spending, consensus, 51%+ attacks, and so on. Some of these problems are simply unsolvable and we have to compromise. SSBT, for example, has no consensus, so everything is subjective[1] (which makes it almost impossible to do anything useful with). The people still fighting this war of decentralization remind me of the Japanese soldiers[2] who didn't realize WW2 was over. You lost. No one cares about decentralization. Google won. Amazon won. Facebook won. The great majority of people invest in BTC and ETH because they think it'll make them wealthy, not because of some noble desire to fight "the man." Besides, BTC isn't really decentralized anyway, the million-dollar-VC-backed miners basically control that ecosystem. [1] https://github.com/ssbc/secure-scuttlebutt/issues/86 https://github.com/ssbc/secure-scuttlebutt/issues/86 [2] http://www.cnn.com/2014/01/17/world/asia/japan-philippines-ww2-soldier-dies/index.html http://www.cnn.com/2014/01/17/world/asia/japan-philippines-w...

Devagamster 9y ago
Take a look at beaker. I'm not sure that decentralized apps wont happen. I agree that centralized servers have many benefits, but there are also benefits to giving control to users.
- jstanley 9y ago
  I had a look at beaker earlier today. It doesn't appear to do anything IPFS doesn't do, it just has a pretty user interface. IMO it would be a much better project if it was IPFS-compatible so that it could benefit from all the existing work that is being done on IPFS.
gsich 9y ago
Email is decentralized and is probably the most used and most compatible system there is.

BinaryIdiot 9y ago

Not sure I entirely understand where this post is going. What does centralization versus decentralization have to do with a blockchain? Just because Bitcoin and others are using it as a centralized data store for a decentralized currency doesn't mean the blockchain has to be tied to any of that. I feel like people keep getting really far away from what a blockchain really is (a fancy linked list that can ensure integrity from start to end) and are confusing it for how Bitcoin uses it.

fabianhjr 9y ago
I read it as: Blockchains(PoW/PoS/etc) are not the solution for decentralized applications because blockchains are a global state and most decentralized application don't _need_ a global state. (Specially non-fintech ones that don't have a double-spend problem like social/websites/photos/etc) It briefly mentions Secure Scuttlebutt, a gossip protocol that works over meshnets without global internet access as an example of a decentralized technology without a global state.

fsiefken 9y ago

DAG (Directed acyclic graph) based currencies don't need a blockchain; Byteball/Dagcoin, IOTA, Rayblocks (Nano) and others.

jasondc 9y ago

I agree with some of the authors points. One item that is missing is the incentive structure. Let's take mesh networks for example (as mentioned in the article). Substratum pays users in coins to share their network, which creates an incentive to grow the network (as long as the coins are worth something). If the coins are worthless, the model doesn't work. It would take a lot of money to create the incentives without the coins.