Hacker News Reader

Comment by jedberg | Hacker News Reader

So a while ago there was an article that someone was getting close to controlling 50% of the network, which means that they could make arbitrary changes to the blockchain. Whether true or not, if NASDAQ was using the blockchain to track stock ownership, it seems like there would be a HUGE incentive to try and control 50% so you could manipulate it. This seems bad.

EDIT: As I type this my comment sits at 0 points, as does every comment I make even remotely critical of bitcoin and the blockchain. Why does HN hate people that are critical of bitcoin?

View on HN

I downvoted you because - while you raise a generally valid concern - your statement is exaggerated to a degree that makes it grossly misleading. You say:

> which means that they could make arbitrary changes to the blockchain

This makes it sound like the 51% attacker could make arbitrary changes to the ledger, thereby stealing arbitrary shares. That is not the case. For example, such an attacker cannot forge any signatures and thus cannot manipulate transactions. All the attacker can do is censor new transactions or slowly unroll past transactions. In case of 51% of the computing power, unrolling could happen at a speed of (51% - 49%) = 2% of time, i.e. if the attack lasts for 50 days, the attacker could unroll one day worth of transactions. What makes the unrolling problematic, is that it enables the attacker to double-spend. One potential measure for NASDAQ to prevent this is to demand known signatures to be used for transactions (i.e. whenever you transfer shares, you must use a signature you previously registered with NASDAQ). With colored coins, the issuer can enforce such restrictions by refusing to redeem shares that were not properly obtained. So even in case of such an attack, NASDAQ could identify the attacker (or the person collaborating with the attacker in order to attempt a double-spend) and initiate legal measures (i.e. sue the attacker for damages).

> I downvoted you because - while you raise a generally valid concern - your statement is exaggerated to a degree that makes it grossly misleading

Ok, that's fair, however, I specifically said "whether that is true or not" because I didn't know, I was just repeating the article, which is clearly wrong.

In reference to the voting I'd like to suggest you abandon this strategy of down-voting because it's bad information and instead reply in an informative way like you just did because in this case I and other people learned something through discussion which I otherwise would not have.

Doesn't a technique like selfish mining (http://arxiv.org/abs/1311.0243) mean that you can undo past transactions faster than 2% of the time, though?

Your point is valid but not a reason to downvote - on HN downvotes are supposed to weed out aggression, trolling and other unwanted behaviours, not to express disagreement.

Sadly, that's not the case, according to Paul Graham: https://news.ycombinator.com/item?id=117171

Fortunately that was ~8 years ago. I think the current interpretation is upvote="thank you", downvote="less than uninteresting".

That's what the interpretation should be. But because of a comment Paul Graham wrote ~8 years ago, people think they can downvote just if they disagree.

Anyway, we're not allowed to talk about downvoting because it's against the rules. It's the great HN Catch 22.

The pretext of the discussion of what sorts of downvotes are okay is that someone is watching and is going to do something about it.

That seems like it might happen in egregious cases, but I'm not sure why people think someone is going to spend a couple of minutes analyzing whether the 4 down clicks on their tossaway comment were fair minded (whatever that means) or not.

So the problem isn't just that it's a boring discussion, it's also just not a reasonable expectation that every single click be fair (whatever that means).

How do you know who the attacked is to then be able to sue them?

Bitcoin isnt anonymous, so you could potentially figure it out.

I work in a co-working space and I see random people coming in to use the Bitcoin ATM here - there's no requirement for identification before putting their cash in and getting Bitcoin in return. How isn't that anonymous?

> they could make arbitrary changes to the blockchain

It's not that bad. See https://en.bitcoin.it/wiki/Weaknesses

"An attacker that controls more than 50% of the network's computing power can, for the time that he is in control, exclude and modify the ordering of transactions. This allows him to:

Reverse transactions that he sends while he's in control. This has the potential to double-spend transactions that previously had already been seen in the block chain. Prevent some or all transactions from gaining any confirmations Prevent some or all other miners from mining any valid blocks

The attacker can't:

Reverse other people's transactions Prevent transactions from being sent at all (they'll show as 0/unconfirmed) Change the number of coins generated per block Create coins out of thin air Send coins that never belonged to him"

Regarding your "edit", perhaps voters don't consider your comment that insightful, especially when you spread negative information about bitcoin, such as 50% attacks allowing attackers to "make arbitrary changes to the blockchain".

>As I type this my comment sits at 0 points, as does every comment I make even remotely critical of bitcoin and the blockchain. Why does HN hate people that are critical of bitcoin?

No offense, but you don't appear to understand these subjects very well and you seem to have limited knowledge of them, yet you persist in commenting on them and thereby spreading misinformation.

Possibly you believe you understand the subject better that you actually do? You're not the only person falling into this trap, and I'm sure you intend no malice, but this may be why you get a few downvotes.

I happened to notice this a few days ago, when you made a statement about bitcoin that a few seconds on Google could have shown you was poorly-informed. That's OK, everyone makes mistakes like this sometimes, but when I pointed out the error, you unfortunately tried to divert attention with an extremely weak strawman argument, and then just stopped responding. It's tiresome.

https://news.ycombinator.com/item?id=9837379

There are many, many criticisms that you could make of blockchain technology and bitcoin, not least of which is that nobody even knows for sure if they will work as intended as they scale up. There's no need to leap into the debate with such poorly-informed statements

They wouldn't use the actual Bitcoin blockchain, I think, as they have nothing to gain from the pseudo-anonymity and have better ways to prevent bad actors than proof of work. Instead they'd use what they call a "distributed consensus ledger" which would require one to get permissions (i.e. be known and sign legally binding agreements) to be part of.

NASDAQ is indeed using Bitcoin's blockchain, via a colored-coins implementation called the Open Assets Protocol [0].

If you have a good reason to use a public, decentralized, open-access, append-only ledger, then you very probably want to use Bitcoin's blockchain. Roll-your-own blockchains make very little sense, as yours will be much less secure than Bitcoin's without massive investments—investments which would give you still better security if they were directed at Bitcoin's blockchain instead.

On the other hand, if you don't want a public, decentralized, open-access, append-only ledger, then you should probably just use Postgres. Like DTC already does.

0. https://github.com/OpenAssets/open-assets-protocol

If NASDAQ started operating in bitcoins blockchain, the influx capital into it would be so insanely astronomical that the scale of the operation would be impossible to 50%.

When people talk about 50% control of bitcoins blockchain today, they are talking about mining pools, where the ones promising the best dividends get all the registrations and thus get a huge chunk of the mining community. Which just means they always sit at 40 - 49% share in the market, and never 50%, because miners know better.

I would have to doubt the feasibility of even an entity as insanely huge as NASDAQ being able to justify enough computational power to get basically one falsified transaction in bitcoin, before the fact NASDAQ was 51%ing bitcoin caused that huge investment in bitcoin mining I mentioned earlier.

So 51%ing bitcoin is a chicken and egg problem. Any source of capital large enough to monolithically 51% the network is also large enough to bring so much attention to bitcoin that new miners would drive the network size up immediately afterwards. You would probably never be able to maintain continuous 51% control over that kind of runaway interest.

If you wanted to attack bitcoin, wouldn't it make more sense to attack or conspire with the owners of the largest mining pools instead of trying to amass all that computational power?

Not really - because the pools are collections of individual miners that each can switch which pool they participate in on a whim. There aren't some much "owners" as facilitators. In practice we've already seen this, when several pools have gotten close to the 50% mark, individual minors have switched pools, or even split their participation across multiple pools.

The entire idea of the Bitcoin blockchain is to provide anonymous consensus. This is a tremendous achievement, but it comes at a great cost, in terms of energy consumption, confirmation time, and risk of reversibility.

Participants in regulated financial markets are all known, which makes the entire design of the Bitcoin blockchain moot. There are much better fits in the design space for financial institutions.

The only reason you see NASDAQ playing with such "colored coin" solutions is that there is currently very little expertise in this space in financial institutions, and there is a monumental hype machine behind Bitcoin.

I love Bitcoin for its counter-economics potential, but "colored coin" solutions for established institutions are a horrendously bad idea.

> The entire idea of the Bitcoin blockchain is to provide anonymous consensus...

>Participants in regulated financial markets are all known, which makes the entire design of the Bitcoin blockchain moot.

My understanding is that one of the major ideas of the Bitcoin blockchain is to solve the Byzantine generals problem, where it's irrelevant whether the participating generals are known. What is relevant is that the generals cannot/do not necessarily trust each other, yet they must still reach consensus. It's the trust issue that's important here, not the pseudonymity that one particular blockchain (Bitcoin's, in this instance) can provide.

> The only reason you see NASDAQ playing with such "colored coin" solutions is that there is currently very little expertise in this space in financial institutions, and there is a monumental hype machine behind Bitcoin.

I can't read the minds of any NASDAQ VIPs, but I suspect they have more than one reason for experimenting with emerging technologies. Maybe they don't want exposure to counterparty risk such as we find in TFA with regards to DTC/Cede & Co. Or perhaps they're attempting to quantify the tradeoffs in risks among different possession and clearing mechanisms. Or, sure, maybe they're being distracted by the newest hotness.

I welcome being proven wrong, but I'd be very surprised that the "monumental hype machine" behind an emerging technology -- whose market cap is less than half of NDAQ's -- is what convinced a group within NASDAQ to consider some use of Bitcoin's blockchain.

That's a common misconception. Solutions to the Byzantine generals problem have been known for a while. Researchers are still looking for more efficient solutions, but PBFT introduced in 1999 is already, well, practical.

The main problem with anonymous consensus are Sybil attacks. You need a way to limit the size of the consensus pool. Bitcoin does that using proof-of-work (quite elegantly, Bitcoin's PoW does actually a lot more than that, but this is really it's raison d'etre).

I recommend reading Ben Laurie on the topic. http://www.links.org/files/decentralised-currencies.pdf

---

If you look at the current problem with DTCC, you'll notice is has little to do with centralization, and more to do with the fact that there is no technological transparency. If all of these systems ran with well defined API, most of the issues would go away.

What financial are mostly after is a shared ledger with programmatic access. Distributed is just a nice add-on.

Not sure why you were downvoted. But in response to your query, there is always a risk of someone obtaining more than 50% of the netwrok, however with that comes cost, to attack the blockchain by misusing your 50% of the network causes as much damage to yourself as it does to anyone else. Right now if you contolled 50%+ you could double spend your coins, or perhaps you could swap a load of shares into your addresses if NASDAQ were using it to track share ownership, however if you did how long do you think it would take for them to shut down using the blockchain and use e.g. a centralised vrsion they run themselves, you then lose a lot of value in bitcoin, of which you ar emining over 50% of the daily allowance.

Yes it would be possible but it would be expensive and a very short lived way of making money. of course a nation state that wanted to get rid fo BTC would like this but for private individuals or firms there would be no incentive to under take an attack like this.

I've never understood why people whine about upvotes. If there were no upvotes, would you not even post? Do you only post so you can be validated by mouse clicks of people you don't know?

It's a discussion forum not a standalone blog and in discussions people don't like to be ignored or silenced. Just because it's typed out on the internet doesn't change that. I do understand why people whine about people whining about votes but it doesn't make it any less grating.

Bitcoin is crazy polarizing because it fucks with trust. Basically.

Reply on news.ycombinator.com