IOTA - page 746. | Bitcointalksearch.org

TPTB_need_war

sr. member

Activity: 420

Merit: 262

Quote from: Come-from-Beyond on October 28, 2015, 06:16:11 PM

Quote from: TPTB_need_war on October 28, 2015, 06:10:33 PM

Can someone link me to that and/or the relevant page of the white paper? Seems to be mining could be used to generate check points. That was one of tweaks I had in mind.

There is no about that in the whitepaper. There is no a formal proof that coin generation is impossible, intensive search for a coin generation technique was done to have 2% annual inflation (because it's a near-optimal number) and none of the ideas let to keep the security of the system at an acceptable level. This is a well-known problem of proof asymmetry, it's hard to prove that unicorns don't exist while it's trivial to prove the opposite if you have such a unicorn. If you have an idea how to add 2% inflation, share it, please.

You can have an orthogonal block chain which records a consensus on the state of the tree (hash). Voila checkpoints and debasement. Since the trees are the objective reality, then the block chain can't lie with a 51% attack. The block chain could be PoW or PoS.

I shouldn't be giving away ideas for free, but any way you all have shared a lot for free and you'd eventually figure this out.

Come-from-Beyond

legendary

Activity: 2142

Merit: 1010

Newbie

Quote from: tromp on October 28, 2015, 06:37:30 PM

Simple. Just allow special coinbase txs that additionally need to reference the previous coinbase tx and need so much PoW that they can only happen on average every 10 mins:)

This magically moves Nash equilibrium towards superwide DAG.

Come-from-Beyond

legendary

Activity: 2142

Merit: 1010

Newbie

Quote from: TPTB_need_war on October 28, 2015, 06:34:47 PM

Now you know why ... They were ... And ...

Insults are not welcome in this thread, without an evidence I treat your words as insulting.

tromp

legendary

Activity: 990

Merit: 1108

Quote from: Come-from-Beyond on October 28, 2015, 06:16:11 PM

If you have an idea how to add 2% inflation, share it, please.

Simple. Just allow special coinbase txs that additionally need to reference the previous coinbase tx and need so much PoW that they can only happen on average every 10 mins:)

Come-from-Beyond

legendary

Activity: 2142

Merit: 1010

Newbie

Quote from: mthcl on October 28, 2015, 06:29:40 PM

Remember that the "longest path algo" is not what's written in the whitepaper

Anyhow, let's finish our private discussion in slack and only then make the results public.

It's still worth discussing my algo publicly even if we don't adopt it.

mthcl

sr. member

Activity: 376

Merit: 300

Quote from: stdset on October 28, 2015, 06:30:46 PM

Quote from: Come-from-Beyond on October 28, 2015, 06:26:07 PM

Possible? I don't see how it's possible to draw a picture to have longest-path-as-the-score rule to be broken by an adversary.

"Longest-path-as-the-score" differs from what is proposed in the whitepaper. We were talking about another algo.

Exactly

Just wait a bit until we finish our internal discussion...

TPTB_need_war

sr. member

Activity: 420

Merit: 262

Quote from: myagui on October 28, 2015, 06:24:39 PM

Re: Premine vs PoW vs ICO vs User ID vs 'a life of crime':

Why not simply make a fixed number of tokens available, at a fixed price per token?
(If not sold out, any unsold tokens would then be provably burned)

This way, developers can still buy their own tokens, but in doing so, they are competing with the other users/buyers. Any tokens bought up by the developers, are tokens that become unavailable for someone else to buy. This is much better than the usual premine, in the sense that the developers are trading a portion of potential outside funding, in exchange for whatever tokens they buy for themselves (aka, putting their money where their mouths are, because they then become a truly interested party, after funding).

Tying up to an existing coin (or coins) seems interesting as well. That would likely attract the widest user foundation, though possibly at the expense of most (all?) of the funding potential...

Sorry there is no difference from a premine. They can buy up most of the coins thus limiting the supply and thus they can set an artificially higher price per share for the ICO (some fewer investors are willing to pay a higher price than other investors, i.e. not all investors are equally astute). Review the math of my post again. Remember all ICO from other investors money ends up in their pocket, no matter how many coins they buy.

Now you know why Ethereum's sale was so large yet they ran out of money so fast. They were buying their own coins recycling the same money over and over. And suckering investors into thinking they needed to rush before the priced moved higher on the next pre-timed price increment.

stdset

hero member

Activity: 572

Merit: 506

Quote from: Come-from-Beyond on October 28, 2015, 06:26:07 PM

Possible? I don't see how it's possible to draw a picture to have longest-path-as-the-score rule to be broken by an adversary.

"Longest-path-as-the-score" differs from what is proposed in the whitepaper. We were talking about another algo.

mthcl

sr. member

Activity: 376

Merit: 300

Quote from: Come-from-Beyond on October 28, 2015, 06:26:07 PM

Quote from: mthcl on October 28, 2015, 06:19:06 PM

He would have to attach it "below" the merchant's tx, but yes, you're right, it's a possible attack vector. Anyhow, the referencing algorithm is not yet finished, so we are discussing it with CfB right now.

Possible? I don't see how it's possible to draw a picture to have longest-path-as-the-score rule to be broken by an adversary.

Remember that the "longest path algo" is not what's written in the whitepaper

Anyhow, let's finish our private discussion in slack and only then make the results public.

TPTB_need_war

sr. member

Activity: 420

Merit: 262

Quote from: smooth on October 28, 2015, 06:23:36 PM

Quote from: TPTB_need_war on October 28, 2015, 06:18:44 PM

Quote from: smooth on October 28, 2015, 06:17:20 PM

The track record in trying to do that is pretty bad. Limit per buy-in just raises the stakes to create/recruit/impersonate more straw buyers. See Stellar and probably some others.

That is more difficult if you can't pay with crypto and have to pay with a fiat account. And if the maximum amount per buyer is say $500. Not saying they should do that. We all hate the fiat world right? (I paid my rent today with BTC)

For paying that may be true (the earlier attempts at "per person" distribution were mostly free). But then that raises other legal, institutional, and logistical costs and obstacles, as you have previously reported with respect to your social media experiments.

Agreed it could. Paypal often withholds funds, especially if there is a sudden surge of payment volume. Selling via BTC is much easier.

But it is possible to verify an account with Paypal (or Amazon payments) without taking payment via Paypal (or only take a small payment), then take the rest from that person via BTC. So then you don't care what Paypal does.

mthcl

sr. member

Activity: 376

Merit: 300

Quote from: mthcl on October 28, 2015, 06:19:06 PM

Quote from: stdset on October 28, 2015, 06:14:19 PM

Quote from: mthcl on October 28, 2015, 06:07:21 PM

Quote from: stdset on October 28, 2015, 06:06:25 PM

Quote from: mthcl on October 28, 2015, 06:03:52 PM

But the cumulative weight of that tx is not so big, so why the merchant should accept it?

NP, the merchant waits of course for normal amount of confirmations.

On your picture it has 1 confirmation only?..

I can draw more pictures. But I don't think it's necessary. Imagine that the attacker started preparing for the attack a month ago. He spent the whole month to accumulate PoW on top of the second doublespend. He published no transactions during the month. Then he publishes the first doublespending transaction, provides the first confirmation, thus attaching it to recent part of the tangle, waits for the merchant to send him his puchase. Then publishes his secret subtangle and attaches in to the legit subtangle. The first doublespending transaction now is rejected by the network, the second doublespend has more weight.

He would have to attach it "below" the merchant's tx, but yes, you're right, it's a possible attack vector. Anyhow, the referencing algorithm is not yet finished, so we are discussing it with CfB right now.

... there is also a complication that the current height calculation algorithm is not very computationally efficient. We'll have to modify it in any case, and there are already several good candidates.

Come-from-Beyond

legendary

Activity: 2142

Merit: 1010

Newbie

Quote from: mthcl on October 28, 2015, 06:19:06 PM

He would have to attach it "below" the merchant's tx, but yes, you're right, it's a possible attack vector. Anyhow, the referencing algorithm is not yet finished, so we are discussing it with CfB right now.

Possible? I don't see how it's possible to draw a picture to have longest-path-as-the-score rule to be broken by an adversary.

Sebastien256

hero member

Activity: 715

Merit: 500

Quote from: TPTB_need_war on October 28, 2015, 06:22:41 PM

Quote from: smooth on October 28, 2015, 06:15:25 PM

Quote from: TPTB_need_war on October 28, 2015, 06:10:33 PM

As for complying with laws, my current thinking is to sell the software with some tokens for use at set prices to end users (not investors)

It is not at all clear there are "end users" for something like this for years to come. It will take a significant period of time to build a market with supporting devices, services, and a large enough population of users, all interacting with each other using the token as a tool to accomplish something else. So I doubt that anyone buying it now could be viewed as anything but a speculator (including long term investor).

That might be different for some other kind of token such as one backing assets in a game or even maybe decentralized storage. Something with actual ready-to-go uses. But that seems a bit off topic on this thread.

Developers are an example of early adopter users who need access to a real world system in order to build out the ecosystem. Yes the number of users and people who want to play with such a system is small at the start. That is why if you sell to end users and not speculative investors, don't expect to raise $10 million at the ICO. You might raise $50,000 perhaps ($500 x 100).

If only raise 50k, it would be reasonable to allow certain amount premine for the devs, but everything would be clear from the start.

smooth

legendary

Activity: 2968

Merit: 1198

Quote from: TPTB_need_war on October 28, 2015, 06:22:41 PM

don't expect to raise $10 million at the ICO. You might raise $50,000 perhaps ($500 x 100).

I would guess that $50 000 would not be sufficient to develop hardware to support IOTA, but I could be wrong.

myagui

legendary

Activity: 1154

Merit: 1001

Re: Premine vs PoW vs ICO vs User ID vs 'a life of crime':

Why not simply make a fixed number of tokens available, at a fixed price per token?
(If not sold out, any unsold tokens would then be provably burned)

This way, developers can still buy their own tokens, but in doing so, they are competing with the other users/buyers. Any tokens bought up by the developers, are tokens that become unavailable for someone else to buy. This is much better than the usual premine, in the sense that the developers are trading a portion of potential outside funding, in exchange for whatever tokens they buy for themselves (aka, putting their money where their mouths are, because they then become a truly interested party, after funding).

Tying up to an existing coin (or coins) seems interesting as well. That would likely attract the widest user foundation, though possibly at the expense of most (all?) of the funding potential...

smooth

legendary

Activity: 2968

Merit: 1198

Quote from: TPTB_need_war on October 28, 2015, 06:18:44 PM

Quote from: smooth on October 28, 2015, 06:17:20 PM

The track record in trying to do that is pretty bad. Limit per buy-in just raises the stakes to create/recruit/impersonate more straw buyers. See Stellar and probably some others.

That is more difficult if you can't pay with crypto and have to pay with a fiat account. And if the maximum amount per buyer is say $500. Not saying they should do that. We all hate the fiat world right? (I paid my rent today with BTC)

For paying that may be true (the earlier attempts at "per person" distribution were mostly free). But then that raises other legal, institutional, and logistical costs and obstacles, as you have previously reported with respect to your social media experiments.

TPTB_need_war

sr. member

Activity: 420

Merit: 262

Quote from: smooth on October 28, 2015, 06:15:25 PM

Quote from: TPTB_need_war on October 28, 2015, 06:10:33 PM

As for complying with laws, my current thinking is to sell the software with some tokens for use at set prices to end users (not investors)

It is not at all clear there are "end users" for something like this for years to come. It will take a significant period of time to build a market with supporting devices, services, and a large enough population of users, all interacting with each other using the token as a tool to accomplish something else. So I doubt that anyone buying it now could be viewed as anything but a speculator (including long term investor).

That might be different for some other kind of token such as one backing assets in a game or even maybe decentralized storage. Something with actual ready-to-go uses. But that seems a bit off topic on this thread.

Developers are an example of early adopter users who need access to a real world system in order to build out the ecosystem. Yes the number of users and people who want to play/experiment with such a system is small at the start. That is why if you sell to end users and not speculative investors, don't expect to raise $10 million at the ICO. You might raise $50,000 perhaps ($500 x 100).

mthcl

sr. member

Activity: 376

Merit: 300

Quote from: stdset on October 28, 2015, 06:14:19 PM

Quote from: mthcl on October 28, 2015, 06:07:21 PM

Quote from: stdset on October 28, 2015, 06:06:25 PM

Quote from: mthcl on October 28, 2015, 06:03:52 PM

But the cumulative weight of that tx is not so big, so why the merchant should accept it?

NP, the merchant waits of course for normal amount of confirmations.

On your picture it has 1 confirmation only?..

I can draw more pictures. But I don't think it's necessary. Imagine that the attacker started preparing for the attack a month ago. He spent the whole month to accumulate PoW on top of the second doublespend. He published no transactions during the month. Then he publishes the first doublespending transaction, provides the first confirmation, thus attaching it to recent part of the tangle, waits for the merchant to send him his puchase. Then publishes his secret subtangle and attaches in to the legit subtangle. The first doublespending transaction now is rejected by the network, the second doublespend has more weight.

He would have to attach it "below" the merchant's tx, but yes, you're right, it's a possible attack vector. Anyhow, the referencing algorithm is not yet finished, so we are discussing it with CfB right now.

TPTB_need_war

sr. member

Activity: 420

Merit: 262

Quote from: smooth on October 28, 2015, 06:17:20 PM

The track record in trying to do that is pretty bad. Limit per buy-in just raises the stakes to create/recruit/impersonate more straw buyers. See Stellar and probably some others.

That is more difficult if you can't pay with crypto and have to pay with a fiat account. And if the maximum amount per buyer is say $500. Not saying they should do that. We all hate the fiat world right? (I paid my rent today with BTC)

smooth

legendary

Activity: 2968

Merit: 1198

Quote from: Sebastien256 on October 28, 2015, 06:14:02 PM

Quote from: iotatoken on October 28, 2015, 06:04:34 PM

In short: yes we could take up loans by people to buy up IOTA token. There are plenty of these opportunities if one wants to and this would not change if we had premine either, it would just give us even more IOTA.

As for verification process: there is no way to beat sybil, there would be numerous ways to fool this third party too. We could just pay some random people to buy coins via their IDs and then send them to us.

We wont cheat the system, it would be a detriment to IOTA and thus a larger detriment to ourselves than the profit we could get out of manipulating the ICO. This boils down to how inclined you are to believe in conspiracy theories. Fortunately participating in ICO is 100% voluntary.

I understand you can fool to some extent the verification process, but requiring a limit buy-in per user complicated the fooling process a lot.

The track record in trying to do that is pretty bad. Limit per buy-in just raises the stakes to create/recruit/impersonate more straw buyers. See Stellar and probably some others.

Topic: IOTA - page 746. (Read 1473405 times)