Comment by theptip | Hacker News Reader

Comment by theptip | Hacker News Reader

Parent Full thread

Poking around in that project, I don't see any specific tests of the Wallet contract... would be interested to see what UTs/STs were included if anyone knows where they are -- or was it just manually tested on the testnet? It seems that a basic set of UTs for this contract should have caught this issue.

It's not even in the category of weird timing attacks or complicated internal states allowing unexpected state transitions -- this is a trivial bug that could be spotted by viewing the contract in Remix (the Solidity IDE).

There are certainly some weird edge cases in the Ethereum VM programming model that are hard to UT, and I think that dev efforts in this area are under-resourced (e.g. it's currently impossible to test time-dependent contracts as there's no way of mocking out the clock in the simulated blockchain).

If you are considering writing a smart contract, read and re-read https://github.com/ConsenSys/smart-contract-best-practices, and then write UTs for each of these cases so that you can convince yourself that you've covered them.

Relevant to this bug:

> Understand that your public functions are public, and may be called maliciously. Your private data is also viewable by anyone.

As I said here last week "Also the lead developer is some kid who believes very much in moving fast and breaking things. This philosophy may work for a social network site but not for other people's money."

This philosophy permeates the community of Etherum development.

Bitcoin disabled most of the smart contracts op codes for a very good reason.

I wouldn't recommend putting anything of value into an Etherum smart contract and find a project with more responsible developers who take a more "wait and see" approach.

Do you have examples of this, or just hearsay?

Beyond the one we're already discussing?

the story you are commenting on is a good example

it should not be this easy to lose sixteen million dollars

I feel like we shouldn't be expressed in dollars, but in percentage of total funds. In this case 16 million is about 1% of the total. This would be a $16 billion case if the US banking system was the target and used this technology.

That... doesn't make a difference in my mind. Losing 1% of people's money is _very_ unacceptable and worth being laughed at for.

Financial institutions are accurate to tiny tiny fractions of percentage points. Not "eh we got 99% of it right".

Especially when the odds are that it will happen again soon.

Yes. If a similar scale thing were to happen in the U.S. banking system, 1% would be 160 billion dollars, not 16.

I'm amazed that otherwise intelligence people really believe this is a better system than fiat currencies + banks.

*intelligent

Microtransactions.

You should ask yourself what the Bitcoin developers are doing while Ether takes their market share. The answer is the exact same thing, at https://elementsproject.org

Now why hasn't there been as much publicity about this? Is it because Vitalik Buterin is a wunderkind genius who simply beat them to market? Nope, the Bitcoiners are simply being more responsible and making sure they get the formula right before rushing to market.

You sound very biased on this.

I am very biased because I know the people at Blockstream have been thinking about this problem longer than the developers of Ethereum.

Please see this article to get an idea of the personalities involved:

http://www.newsbtc.com/2016/08/17/gregory-maxwell-vitalik-bu...

People write this stuff off so fast. Here we have a parity developer, probably one of the most competent in the Ethereum ecosystem, and he screwed up to the tune of losing $30m of other people's money.

If HE can't get it right, what business do you have running around saying that anyone who can make a webpage can make a decentralized application?

We are playing with money. Dealing with attackers is not as simple as rebooting your server. Mistakes mean irreversible loss.

The devs in this industry are way over their heads and nobody is willing to admit it, because doing so destroys the entire core value prop of Ethereum, the poster child of the current bubble.

Dear god, I've been thinking crypto is a bubble since 2010. At some point, we have to admit this is not just a bubble. It's a new unproven evolving technology

I actually think eth has a lot of potential, but I don't think he's saying crypto is a bubble. He's probably talking about the ICO "bubble", which isn't a controversial thing to claim. Eth has already halved in value in the last month which is mostly put down to the crazy value being pumped into ICO's slowing down.

I would argue ICO phenomenon is not a bubble Rather, some ICOs are scams But many are legitimate business ventures. Buying into an ICO is similar to buying deep out of the money options

Crypto has had at least 3 distinct bubbles. There's the current one, marked by Eth's absurd rise to $300. The previous one was when Bitcoin made an absurd rise to $1200 in 2014. There was a bubble before that where Bitcoin rose to $266 in 2011. And one before that I believe where Bitcoin rose to $32 in 2010.

Though crypto has risen past the high point of each of these bubbles, each price point can be definitively classified as a bubble in it's own time, and was followed by a crash that wiped out >70% of the market cap over the coming months/years.

Though it did recover, that doesn't mean it wasn't a bubble.

I believe cryptocurrency is going to be bigger than e-commerce is today. But it's not there yet, and it shouldn't be valued as though it has changed the world. It hasn't yet, and it's still a few years away at least from doing so.

It's only a bubble if you can identify it ahead of time. Stating that something became much more valuable over a short period of time doesn't mean you know how to identify bubbles. Bitcoin is 70-80 times more valuable now than when it was at 32, and had you bought it then you would be up a lot. So, to say crypto is in a bubble today is the same thing as to predict that in a few years it will be worth less than today. Do you know that definitively ? I don't

Downvoters, since you know how to indentify bubbles, I'm assuming you shorted ethereum on margin on GDAX when it hit 400 and made a ton? No? If not, why not, if you knew definitely it was a bubble ?

Because it's very hard to figure out where the top of a bubble is, and if you short a bubble asset too long before the peak, you go broke, even if you are correct.

Don't short hopium.

Well, now we are talking semantics. You don't get to claim you are correct if your forecast turns out to be wrong. So, if you identify a bubble, part of the requirement to say that you succeeded is that you can identify the top. Otherwise, your "knowledge" that it's a bubble is useless

I think people agree that a price rise was a bubble after it pops, and I don't believe that prediction is a necessary condition.

It is, otherwise how do you distinguish a bubble from a rise justified by fundamentals ?

He may be brilliant, but he played fast and loose without real testing or concern, so I wouldn't call him the "most competent". For a wallet contract they knew would secure many millions of dollars, there should have been dozens of people auditing it before rolling out to users.

You're "no true Scotsman"ing so hard Mary, Queen of Scots doesn't even qualify.

I wish I could see how the op is employing the no true Scotsman fallacy, but I can't.

https://twitter.com/SatoshiLite/status/887781929726038016

"If the creator of Solidity, Gavin Wood, cannot write a secure multisig wallet in Solidity, pretty much confirms Ethereum is hacker paradise."

Charlie is trolling. The architect isn't going to be the best craftsman, that requires very different set of skills.

Parity devs may be academically brilliant, and they put out a fantastic client, but the wallet issue shows how lax their development lifecycle has been.

Gavin Wood didn't not make the code change that caused this...

I can't edit my post above to add this now, but looks like Charlie retracted this as Gavin was the original author, but not the one who made the mistake.

Someone's loss is anothers gain. That money didn't disappear, just got reallocated. And if spent ,will generate economic activity. It's certainly a hit in the trust of the core dev though.

This gets awfully close to the broken window fallacy https://en.wikipedia.org/wiki/Parable_of_the_broken_window

I agree, if they did not address the concerns it would become unworkable.

> Bitcoin disabled most of the smart contracts op codes for a very good reason.

Bitcoin is also stuck in the past. Yes, Ethereum breaks things, but for me it's not about the money but about the technology stack. We're too early in the process to have the luxury of being able to stop innovating.

If I want to play with cool state-of-the-art technology I use Ethereum. For financial transactions I use my bank and/or Transferwise. Haven't quite yet figured out why I'd use Bitcoin for something.

I've been writing some smart contracts over the last couple weeks and have learned a lot from OpenZeppelin's solidity repo–you can either use them directly (importing and extending off of them) or use them as a good resource for building tokens, crowdsales, etc–can't recommend them enough.

https://github.com/OpenZeppelin/zeppelin-solidity

Agreed, it's definitely the way to go. Hope the project will continue to evolve. Both Aragon [0] and Storj [1] are making active use of them, if you look for example usage.

[0] https://github.com/aragon/aragon-network-token

[1] https://github.com/Storj/storj-contracts

The 20 cases for underflow and overflow [0] show how difficult it is to develop error-free contracts. Would have been great if the VM would detect such cases and thus make the contract code less polluted with boundary and overflow checks.

Some projects have started sharing audit results publicly directly in their repos [1]. It's a great read to see common errors and yet unsolved issues that originate in the ERC20 token specification.

[0] https://github.com/ConsenSys/smart-contract-best-practices#i...

[1] https://github.com/status-im/status-network-token#reviewers-...

Reply on news.ycombinator.com