7 ms·

Everything you want to do can be done on a simple database. You can mint tokens / share it / explore other tokens / message people / view feed. It isn't clear w

Everything you want to do can be done on a simple database. You can mint tokens / share it / explore other tokens / message people / view feed. It isn't clear why you want to forcibly put this on Ethereum blockchain and pay a fee to do every transfer (The transaction charge currently is ~ USD 0.5-1 per transaction)

dharness 9y ago
Can't win a free ETH if you use a simple database.
- streulpita 9y ago
  Haha, well actually we could do that regardless. The purpose of the giveaway is to source more ideas for what people actually want to do with cryptocurrencies, and to make sure our product can serve those use cases.
  - chris_wot 9y ago
    All of the examples you gave don’t need a decentralised cryptocurrency and I reckon could be done more easily with a regular database. Unless you can give a reason why those examples require decentralisation then, as much as I respect your attempt at finding an actual use for decentralised Merkel trees, I ultimately think it’s doomed to failure.
    - streulpita 9y ago
      I can offer a few reasons, but they might not convince you. Tokens are provably scarce, meaning the supply is limited and can't be changed by the creator of the currency. They're easily exchangeable, which means they can move between platforms and communities. Finally, they can interact with smart contracts, meaning people can enforce the way the token behaves. I'm actually interested if you can think of a reason too. This is an incredibly difficult brainstorming exercise. So many ICO's have failed to justify their use-cases, but tokens are so interesting that I have to believe there are good reasons to use them. Is there any case that you do find particularly helpful?
      - kristianov 9y ago
        If you think your users are smart enough to figure out all you've mentioned, then they are smart enough to use databases for their usecases.
        bjelkeman-again 9y ago
        I think there are lots of people out there that can come up with interesting use cases, without being software developers and not wanting to learn about databases. If something like this can make it easier for more people to be inventive, then I think that is a good thing.
streulpita 9y ago
That’s a question I have too. I don’t want to forcibly put anything on the blockchain. I’m specifically curious about use cases where the decentralization, scarcity, and smart contract pieces of this can become useful. I can’t predict exactly what, but I do think those things will be used for interesting projects and our tools will keep making that easier for people to play with. We’ve listed a few examples of things people are doing above. But right now, I agree that in many ways this is less functional than a database. Right now people use it for fun, and to try out the technology for new things.
cslarson 9y ago
> USD 0.5-1 per transaction This is absolutely not true. Gas prices have been floating between 1 and 4 gwei for a while now. An erc20 token tx may cost 150k gas so between .00015 and .0006 ETH per tx. So that's $0.09 to $0.37. A simple ETH tx is 25k gas so about a 1/6 of this. You can check ethgasstation.info for the latest standard and "safe low" gas prices.
- fourstar 9y ago
  So do you pay to post a comment on HN? Normal people will never hold an ethereum wallet and pay for transactions.
  - kirillseva 9y ago
    HN comments are free but the platform is owned by YC and YC enforces certain rules as they see fit. In Ethereum there is no owner, but looks like you have to pay for that
    - pavlov 9y ago
      The entire value of HN is that it’s owned, guided and actively moderated by YC. An Ethereum-based alternative would lose that aspect while also costing money and being very slow. I can’t imagine paying for what is effectively negative value. The test for blockchain “dApps” is this: is someone so desperate to have thing X decentralized that they’ll pay for it in both real money and usability inconvenience? If so, it’s a dApp candidate. In the real world, the only thing that seems to pass this test is some types of inter-company ledgers.
      - kirillseva 9y ago
        I'm no saying that HN is bad and Ethereum is good. For one thing, here I am posting on HN, so it's pretty telling :) I just wanted to highlight the difference and that that HN is not free. It's free as in beer. As far as inconvenience goes, it's temporary. Give it some time (2-5 years, I'd say) and it's only going to be about paying real money for it. And if you think that EOS is decentralized, and that it will ever see the light of day, then even paying money won't be a thing for the users as they have a different business model for dApps, where the dApp owner has to stake EOS coins to get a share of EOS network's resources. Which is more similar to existing business model of renting EC2 servers
CryptoPunk 9y ago
An intermediary solution would be to use a DappChain, which acts as a free-to-use sidechain of Ethereum. Loom is going to be releasing an SDK for their own sidechain, DelegateCall, soon: https://docs.google.com/forms/d/e/1FAIpQLSdlnBsw1iRAks1F2pD64K7JybsSIck38GoXl-iXrEzf-MF4Aw/viewform https://docs.google.com/forms/d/e/1FAIpQLSdlnBsw1iRAks1F2pD6... A similar initiative is POA Network, which has US public notaries acting as trusted validators for an Ethereum sidechain: https://poa.network https://poa.network
woah 9y ago
Simple answer, your centralized database can go away tomorrow, or be arbitrarily modified. That can’t happen with the blockchain.
- fwdpropaganda 9y ago
  Does that happen to your databases a lot? That's just bad engineering.
  - edanm 9y ago
    That's not the troubling case. The troubling case is the company hosting the DB shutting down, and therefore the entire system becoming worthless. This is what prevents you as the customer from trusting such a system.
    - arcticfox 9y ago
      but this is what exports are for. I would be fine if gmail etc. shut down tomorrow. there's no reason to add a massive overhead to everything just to solve an already-solved problem.
      - crispyporkbites 9y ago
        What if it’s not a massive overhead? What if the tooling is as simple as setting up a database nowadays, and the performance is a known/acceptable trade off? Surely you can see there are some advantages of blockchain based tech then?
    - beefield 9y ago
      That is preventing you to trust a system where "a gaming streamer is rewarding his most loyal viewers with tokens they store in their own database and you can redeem for shoutouts or merchandise."? I would somehow think that the database technology is quite secondary to the creditworthiness of the gaming streamer when people decide whether to trust or not a token system of a gaming streamer.
berberous 9y ago
I think it's interesting to standardize these things as ERC20 tokens. It's still very early days, but that allows you to leverage the entire ETH econsystem which is still growing. For example, I could store those ERC20 tokens on my hardware wallet, or phone wallet, send them to others without using a platform's website, etc. Also, your comment reminds me somewhat of the infamous Hacker News comment when Dropbox posted ("everything you want to do can be done simply with rsync"). Sure, perhaps it's simple for tech folks that know how to build a database and a website that lets you mint and share tokens. But what if you are a popular Youtube vlogger? You either don't know how to do that, don't have time to, or don't want to spend the money. But perhaps you'd use this service, quickly mint tokens, give them away to users, and invent interesting use cases for them. Personally, I love this idea and think it has great potential. Will anything useful come of it? Who knows -- it's early days. I think the most interesting use cases, if they exist, haven't been thought of yet. I look forward to seeing what people do with it.
- anilshanbhag 9y ago
  There are two parts: - Is the idea interesting ? Yes possibly. The idea of making it easy to create tokens is interesting. I can see twitch streamers giving away tokens to their subscribers / viewers. - Does it need to be on blockchain ? For all the use cases described, the answer according to me is no. It is easier to simply have a frontend through which users can interact vs maintaining a wallet, installing metamask, etc. This is going in the direction opposite of the Dropbox example - making things harder for users.
  - streulpita 9y ago
    It’s harder now but it’s going to get there! You gotta believe! We’re chasing something interesting and hoping we can keep making it easier and more useful. Right now, we do know it’s harder and people are still choosing to try it for fun! I agree with all your feedback above though and it’s actually really helpful to keep those realities in mind.
    - banachtarski 9y ago

dharness 9y ago

Can't win a free ETH if you use a simple database.

streulpita 9y ago
Haha, well actually we could do that regardless. The purpose of the giveaway is to source more ideas for what people actually want to do with cryptocurrencies, and to make sure our product can serve those use cases.
- chris_wot 9y ago
  All of the examples you gave don’t need a decentralised cryptocurrency and I reckon could be done more easily with a regular database. Unless you can give a reason why those examples require decentralisation then, as much as I respect your attempt at finding an actual use for decentralised Merkel trees, I ultimately think it’s doomed to failure.
  - streulpita 9y ago
    I can offer a few reasons, but they might not convince you. Tokens are provably scarce, meaning the supply is limited and can't be changed by the creator of the currency. They're easily exchangeable, which means they can move between platforms and communities. Finally, they can interact with smart contracts, meaning people can enforce the way the token behaves. I'm actually interested if you can think of a reason too. This is an incredibly difficult brainstorming exercise. So many ICO's have failed to justify their use-cases, but tokens are so interesting that I have to believe there are good reasons to use them. Is there any case that you do find particularly helpful?
    - kristianov 9y ago
      If you think your users are smart enough to figure out all you've mentioned, then they are smart enough to use databases for their usecases.
      - bjelkeman-again 9y ago
        I think there are lots of people out there that can come up with interesting use cases, without being software developers and not wanting to learn about databases. If something like this can make it easier for more people to be inventive, then I think that is a good thing.

streulpita 9y ago

That’s a question I have too. I don’t want to forcibly put anything on the blockchain. I’m specifically curious about use cases where the decentralization, scarcity, and smart contract pieces of this can become useful. I can’t predict exactly what, but I do think those things will be used for interesting projects and our tools will keep making that easier for people to play with. We’ve listed a few examples of things people are doing above. But right now, I agree that in many ways this is less functional than a database. Right now people use it for fun, and to try out the technology for new things.

cslarson 9y ago

> USD 0.5-1 per transaction This is absolutely not true. Gas prices have been floating between 1 and 4 gwei for a while now. An erc20 token tx may cost 150k gas so between .00015 and .0006 ETH per tx. So that's $0.09 to $0.37. A simple ETH tx is 25k gas so about a 1/6 of this. You can check ethgasstation.info for the latest standard and "safe low" gas prices.

fourstar 9y ago
So do you pay to post a comment on HN? Normal people will never hold an ethereum wallet and pay for transactions.
- kirillseva 9y ago
  HN comments are free but the platform is owned by YC and YC enforces certain rules as they see fit. In Ethereum there is no owner, but looks like you have to pay for that
  - pavlov 9y ago
    The entire value of HN is that it’s owned, guided and actively moderated by YC. An Ethereum-based alternative would lose that aspect while also costing money and being very slow. I can’t imagine paying for what is effectively negative value. The test for blockchain “dApps” is this: is someone so desperate to have thing X decentralized that they’ll pay for it in both real money and usability inconvenience? If so, it’s a dApp candidate. In the real world, the only thing that seems to pass this test is some types of inter-company ledgers.
    - kirillseva 9y ago
      I'm no saying that HN is bad and Ethereum is good. For one thing, here I am posting on HN, so it's pretty telling :) I just wanted to highlight the difference and that that HN is not free. It's free as in beer. As far as inconvenience goes, it's temporary. Give it some time (2-5 years, I'd say) and it's only going to be about paying real money for it. And if you think that EOS is decentralized, and that it will ever see the light of day, then even paying money won't be a thing for the users as they have a different business model for dApps, where the dApp owner has to stake EOS coins to get a share of EOS network's resources. Which is more similar to existing business model of renting EC2 servers

CryptoPunk 9y ago

An intermediary solution would be to use a DappChain, which acts as a free-to-use sidechain of Ethereum. Loom is going to be releasing an SDK for their own sidechain, DelegateCall, soon: https://docs.google.com/forms/d/e/1FAIpQLSdlnBsw1iRAks1F2pD64K7JybsSIck38GoXl-iXrEzf-MF4Aw/viewform https://docs.google.com/forms/d/e/1FAIpQLSdlnBsw1iRAks1F2pD6... A similar initiative is POA Network, which has US public notaries acting as trusted validators for an Ethereum sidechain: https://poa.network https://poa.network

woah 9y ago

Simple answer, your centralized database can go away tomorrow, or be arbitrarily modified. That can’t happen with the blockchain.

fwdpropaganda 9y ago
Does that happen to your databases a lot? That's just bad engineering.
- edanm 9y ago
  That's not the troubling case. The troubling case is the company hosting the DB shutting down, and therefore the entire system becoming worthless. This is what prevents you as the customer from trusting such a system.
  - arcticfox 9y ago
    but this is what exports are for. I would be fine if gmail etc. shut down tomorrow. there's no reason to add a massive overhead to everything just to solve an already-solved problem.
    - crispyporkbites 9y ago
      What if it’s not a massive overhead? What if the tooling is as simple as setting up a database nowadays, and the performance is a known/acceptable trade off? Surely you can see there are some advantages of blockchain based tech then?
  - beefield 9y ago
    That is preventing you to trust a system where "a gaming streamer is rewarding his most loyal viewers with tokens they store in their own database and you can redeem for shoutouts or merchandise."? I would somehow think that the database technology is quite secondary to the creditworthiness of the gaming streamer when people decide whether to trust or not a token system of a gaming streamer.

berberous 9y ago

I think it's interesting to standardize these things as ERC20 tokens. It's still very early days, but that allows you to leverage the entire ETH econsystem which is still growing. For example, I could store those ERC20 tokens on my hardware wallet, or phone wallet, send them to others without using a platform's website, etc. Also, your comment reminds me somewhat of the infamous Hacker News comment when Dropbox posted ("everything you want to do can be done simply with rsync"). Sure, perhaps it's simple for tech folks that know how to build a database and a website that lets you mint and share tokens. But what if you are a popular Youtube vlogger? You either don't know how to do that, don't have time to, or don't want to spend the money. But perhaps you'd use this service, quickly mint tokens, give them away to users, and invent interesting use cases for them. Personally, I love this idea and think it has great potential. Will anything useful come of it? Who knows -- it's early days. I think the most interesting use cases, if they exist, haven't been thought of yet. I look forward to seeing what people do with it.

anilshanbhag 9y ago
There are two parts: - Is the idea interesting ? Yes possibly. The idea of making it easy to create tokens is interesting. I can see twitch streamers giving away tokens to their subscribers / viewers. - Does it need to be on blockchain ? For all the use cases described, the answer according to me is no. It is easier to simply have a frontend through which users can interact vs maintaining a wallet, installing metamask, etc. This is going in the direction opposite of the Dropbox example - making things harder for users.
- streulpita 9y ago
  It’s harder now but it’s going to get there! You gotta believe! We’re chasing something interesting and hoping we can keep making it easier and more useful. Right now, we do know it’s harder and people are still choosing to try it for fun! I agree with all your feedback above though and it’s actually really helpful to keep those realities in mind.
  - banachtarski 9y ago