[HaseebQ]

This is a totally misleading calculation for the largest currencies. It's literally quoting the spot price for what's available on NiceHash and then dividing by the fraction you'd have of the total hash rate.

In reality, as you continued buying up hashing power, the price of the remaining hashing power would go up precipitously. This is basic supply and demand.

[fanzhang]

It's still a good proxy for how easy it is to do. The attacker could buy hardware, rent from people not on nicehash, etc.

It's like how companies' market caps are determined by the the last few trades, even though the last few trades probably represent .01% (or less) of the shares in the company.

Probably best to read the numbers a bit qualitatively. A coin that is 2% nicehashable would require substantial efforts (probably negotiating with a few private pools) to mount an attack; perhaps impossible for a not-connected miner. Wheres coins approaching the double digit percents probably is quite possible if you have even lukewarm connections. And those near 100% or above are likely super vulnerable.

[xur17]

For smaller coins it's especially scary because existing miners could easily switch to the coin for a few hours to mount an attack.

[acchow]

There's already a GPU and ASIC shortage with prices skyrocketing - even without someone attempting a 51% attack. Being a PC gamer is tought right now - you often can't even find a GPU in stock!

Can't imagine how difficult it would be to hoard a ton of hardware.

[xur17]

That is why I included the NiceHash % for each coin. NiceHash offers fixed price contracts, typically for ~30% of their supply that you can lock in for up to 24 hours.

It's definitely a lot less do-able for the larger currencies - I still think PoW is a good option for them honestly. This was more to show that 51% attack risk is problematic for smaller coins, and I'd love to hear a discussion on the best way to fix this.

[Iv]

> I'd love to hear a discussion on the best way to fix this.

Proof of stake (still problematic for very small cap) or proof of ID (my favorite) using something like the e-estonia crypto ID system. You can prove every miner is a unique person, award them coins on a deterministic pseudo-random order.

[swift532]

If you're going to use proof of ID, doesn't that kind of remove any need for a blockchain? How do you suppose you can have a censorship resistant currency if all a bad state actor has to do is punish anyone who mines a transaction they don't like.

I like crypto, but the greatest danger of it seems to be the potential for economic enslavement(we don't like you, therefore you can't buy bread anymore) through censorship and oppression.

[Iv]

Centralized government only provide means for a person to ID themselves, possibly through pseudonymous means.

If you are worried about targeted censorship, I seem to recall from the first days of BTC that there are ways to "shuffle" IDs: participate in a pool P, get a new ID that is untraceable to the original one but that is traceable to pool P and offers the guarantee that there are no duplicates.

But note that no cryptocurrency is "censorship-resistant" (I am of the opinion that financial transactions are not free speech, so calling it censorship is not the appropriate word and confuses issues). BTC can be (and has, in China) be forbidden. Forbid mining, jail people who do it. Easy in a country that bans also VPNs and Tor.

BTC is only as strong as Tor is and Tor is dependent on the governmental goodwill to let people use strong crypto. Keep in mind that until 2000 US companies were banned from exporting crypto tools that the NSA could not break [1] and even to this date, restrictions exist for some material and countries. USA is just one executive order away from making Tor, VPNs, and thus, anonymous BTC mining, illegal.

And I think that many countries will ban BTC. China is more energy-constrained than most of us, but the energy needs of the BTC netword is gargantuan. I think they have a year to solve that issue (I think they will use proof-of-stake) before states starts pulling the plug.

And always remember that as cool as crypto is as a tech, it does not exist in the vacuum. While it allows to navigate against an incompetent but permissive state, it does not fare well against a competent hostile one.

A part of the problem of maintaining cryptocurrencies (or anonymous networks) is political, not just technical. I too, as a geek, love the prospect of being able to solve political problems with technical solutions, but it only works up to a certain point.

[1] https://en.wikipedia.org/wiki/Export_of_cryptography_from_th...

[xur17]

Something like proof of ID would be amazing, but like you said it requires a secure digital id system to exist first, which few countries have. Is there anyone building something like this?

[Iv]

E-estonia will provide a certificate to anyone presenting themselves to an Estonian embassy with a valid passport (and, IIRC, $300). They thought many other nations would provide this service but to my knowledge they are still alone.

Estonian citizenship is not required. May crypto-geeks have an estonian ID certificate. If I were still into crypto I would have made one for myself I think.

[askmike]

That is about as far from a decentralised system as I can imagine, we now all have to trust the Estonian government (everyone part of the e-ID chain).

[Iv]

For ID systems, I have yet to find a way to do registration in a decentralized way. Until we get cheap unspoofable genome sequencing devices (which may never be a thing), we will have to trust some authorities.

Once registered, you don't need to trust the Estonian government at all: you get an IC card and a reader (all open source IIRC) and you can autonomously authenticate. If memory serves, you can even authenticate pseudonymously.

Note that you don't have to trust everyone, you have to trust the e-ID registration system as a whole. That is, a single flawed individual won't be enough to corrupt the whole thing.

[guy_c]

I wonder if we could see a spiral that effectively kills off smaller coins?

> Exchanges respond to these `weak` coins by increasing their confirmation requirements. Some of the really small coins would probably need huge numbers of confirmations. > Lots of confirmations, which would likely damage the value of the coin. > Lower price would reduce the miner hashrate > Lower hashrate would further lower cost of 51% attack. > Exchanges increase confirmations further

-REPEAT-

[contravariant]

Oh is that what it was. That wasn't entirely clear to me.

It would've made more sense to me to show the required capacity / available NiceHash capacity.

[Sir_Substance]

>This is a totally misleading calculation for the largest currencies.

That's not the only way it's misleading. Peercoin is listed as 8,559% available on nicehash, but peercoin is a proof of stake coin. It doesn't use hashpower to secure it's transactions, so it's not vulnerable to a 51% hash power attack (although it is vulnerable to other kinds of 51% attack).

[thisisit]

> Peercoin

As per wiki, it uses hybrid of PoW and PoS:

https://en.wikipedia.org/wiki/Peercoin

[Passthepeas]

yeah Im confused as to why a proof of stake currency is even in the list

[xur17]

I am using hashing rates from minethecoin [0] - if any are inaccurate, please let me know, and I will update them.

edit: I've removed Peercoin since it is a PoW + PoS coin as the GP points out, so these numbers are inaccurate.

[0] https://minethecoin.com/

[pjc50]

Is hash ownership transparent enough that you can say that? Given the possibility of shell company ownership..