mempool

« first day (1704 days earlier) ← previous day next day → last day (2596 days later) »

SmokeDetector

01:28

[ SmokeDetector | MS ] Bad keyword with email in answer, email in answer, username similar to website in answer: Unable to remove bitcoins from watch-only address by orion on bitcoin.SE

3 hours later…

SmokeDetector

04:35

[ SmokeDetector | MS ] Username similar to website in answer: Is freebitco.in legit site? by JL BTC PTC on bitcoin.SE

2 hours later…

Nick ODell

06:16

Initial Coin Offerings Horrify a Former S.E.C. Regulator / NYTimes

Best quote:

>“I.C.O.s represent the most pervasive, open and notorious violation of federal securities laws since the Code of Hammurabi,” Mr. Grundfest said in an interview.

fredsbend

07:01

@NickODell Hyperbole much. Rofl.

Pieter Wuille

If you look at the amounts involved, I'm not sure it's that much of an exaggeration.

fredsbend

@PieterWuille I just laughed at Hammurabi.

Pieter Wuille

fredsbend

But if the coin does not represent ownership, then it's not really a security, right?

Pieter Wuille

In the US, any tradable asset pretty much is a security.

Security (finance)

A security is a tradable financial asset. The term commonly refers to any form of financial instrument, but its legal definition varies by jurisdiction. In some jurisdictions the term specifically excludes financial instruments other than equities and fixed income instruments. In some jurisdictions it includes some instruments that are close to equities and fixed income, e.g. equity warrants. In some countries and languages the term "security" is commonly used in day-to-day parlance to mean any form of financial instrument, even though the underlying legal and regulatory regime may not have such...

fredsbend

07:12

The SEC doesn't manage gold trade, four example, but they manage gold securities.

Pieter Wuille

And the SEC has already made a publishing to point out that they consider ICOs subject to security regulations.

fredsbend

That's news to me.

I wonder how they'd feel if I imitated kick starter with an ico? Buy these coins, then once we have a product like, you can buy them with the coins.

Pieter Wuille

fortune.com/2017/07/25/…

@fredsbend That's literally what nearly every ICO to date is. Except no product ever appears.

fredsbend

I don't actually have to let you use them, just like you can fail to fulfill the promise on kick starter.

Pieter Wuille

Yes, but kickstarter buys are (1) not tradable and (2) not marketed as something that will make a profi

fredsbend

07:15

@PieterWuille Then it's simple fraud, presuming they never intended to make a product. But if they intended, tried, and failed, I don't see how that's more than bad luck.

Pieter Wuille

The point is that nobody in their right mind should ever ever buy these ICOs except if they plan to use it as a speculative asset.

fredsbend

@PieterWuille Good point. Also not divisible. You buy a single promise on kick starter, not a division of unspecified promise.

Pieter Wuille

There have been a few "succesful" ICOs in that they actually resulted in interesting technology that was developed - Ethereum in particular.

But even for those, I feel like it's a doom hanging over the project, due to the vast amount of funds held by the foundation that created them.

fredsbend

@PieterWuille Well, if they're just supposed to be company coupons, it's not that big of a deal. Do companies make coupons non transferable for legal reasons?

@PieterWuille I did not realize ether was an ico.

What came out of it?

Pieter Wuille

08:11

They had a presale for ETH you could buy with BTC

SmokeDetector

08:48

[ SmokeDetector | MS ] Manually reported question: How to earn Free ARF Token by Samia on bitcoin.SE

4 hours later…

SmokeDetector

12:26

[ SmokeDetector | MS ] Email in answer: i have a blockchain wallet.aes backup can i get my password from that? by sagar amrutiya on bitcoin.SE

eponymous

13:23

think that's a new record for closeness.

@fredsbend Not a lot, lets be honest. Ethereum is just a platform for people doing sub-ICOs. The commonly posted defence is that people were buying 'crypto fuel' to write and use with contracts, but in reality there's no meaningful application of ETH.

It's a similar problem to what happened with MasterCoin. A bunch of people put money into something which had a high market cap and not a lot of liquidity, so everybody starting doing sub-ICOs with all of their bottled up but not realisable gains. The amount of money being shunted around in the ETH network is completely beyond reason.

1 hour later…

Max Vernon

14:43

Does bitcoin-core support multiple wallets yet? I can't seem to see any documented way to do it. @PieterWuille you impled v0.15 would support it - does that comment still apply?

2 hours later…

Nick ODell

16:33

@eponymous From a perspective of governance, having turing-complete contracts is strongly preferable. If someone comes up with a new use for a cryptocurrency network, they shouldn't need to convince every man, woman, and dog who uses that network to allow their new use.

@fredsbend Someone should make an ICO for the Coin of Hammurabi.

2 hours later…

fredsbend

18:31

@NickODell Yes! Awesome! It'll house laws.

5chdn

Sup guys, is this the "10k" edition of the weekly chat event? ;-)

Max Vernon

19:19

wow, for that price there better be free shipping

Pieter Wuille

@NickODell The goal is not computation, but verification. And you don't need a Turing-complete language to describe the verification of Turing complete statements.

I agree with the advantages of having a sufficiently flexible script language so that it supports the efficient validation of most or all conditions you want for spending, but I disagree that implies a Turing-complete language.

@MaxVernon Yes, it's in the 0.15 release notes: bitcoincore.org/en/2017/09/01/release-0.15.0

SmokeDetector

[ SmokeDetector | MS ] Link at end of answer: Ways to monitoring BTC trading profit and loss? by joshblum on bitcoin.SE

Max Vernon

@PieterWuille cool....

wallet=wallet_one.dat
wallet=wallet_two.dat

in the bitcoin.conf should work.

Nick ODell

19:36

@PieterWuille Why is verifying a computation easier than computing it? How would you verify that a computation is correct without running the same computation?

Pieter Wuille

@NickODell My favorite example is this: assume Bitcoin today had an OP_MUL operation (it doesn't, but it wouldn't be unreasonable to have it) but not an OP_SQRT.

And someone finds a use case for OP_SQRT, and starts arguing for a softfork to include such an opcode.

In that case, there is an obvious answer that this is unneeded.

Any program which uses an OP_SQRT somewhere in its execution can be rewritten to take as input an extra argument which is the expected result of that square root, and instead of recomputing it, verify it by multiplying it with itself and comparing it to the input of the square root.

And this will always result in lower execution cost: a multiplication is far cheaper than a square root.

Why would every full node in the world need to recompute it? The one signing the transaction already knows the outcome of that square root. He can just tell everyone, and let them verify it instead.

And this observation is more generic, actually. Many computationally expensive statements can be rewritten as the cheaper verification of something else.

In particular, there is one that we already rely upon, and nobody questions: signatures are a proof of knowledge of the private key. But nobody suggests that Script should be able to compute discrete logarithms!

This is what leads to signature aggregation: we don't actually care to see signatures for each public key. We just want a proof of knowledge, and the signature is one way to accomplish that. Aggregated signatures accomplish that in a more compact and computationally cheap way.

Another application of this idea is MAST: you don't need to show the entire program to a validator if only part of it is executed. Just commit to all possible branches ahead of time, and then reveal the one you choose.

And the most extreme idea is Turing-complete SNARKs - the idea that a constant-size expression could be a proof that a particular arbitrarily-complex statement was true, which can be validated in constant time!

2 hours later…

Murch

21:38

@fredsbend I'd suggest that you consider that BTC.top, Antpool, ViaBTC and BTC.com are all either owned or heavily influenced by Bitmain, and Roger Ver's Bitcoin.com is working closely together with Jihan Wu. :p

PHD LAB

22:13

hi guys

any help here

is much appreciated :D

PHD LAB

22:42

Q: issue in forking bitcoin 0.12 and internal miner

For education purposes(preparing online course) I am working on a bitcoin fork which enables the cpu mining. By looking in internet i've learned that internal miner was removed from bitcoin codebase since the version 0.13. So i have forked the version 0.12.1 and i have edited the chainparam.cpp f...