What's good about it? Half the post is about Bitcoin. A large chunk is about the ENS token distribution.
There's no link to any of the contracts or code. There's no comparison with existing providers and the trade-offs. There's no discussion on how DAOs are still very immature and evolving. There's no acknowledgement of how ENS has behaved since inception.
Yes, because Bitcoin is an actually working, properly designed system used by people for real-world ends.
> A large chunk is about the ENS token distribution
That would be two lines just below "Which one is ENS?", plus the image text?
> There's no link to any of the contracts or code.
What is there you want me to link to, specifically? I mean, I suppose I could give a source for "they could crank up the fee for renewing only your name to $1,000,000,000, and then allocate it to whomever they please once it expires" if you insist. But this is not contentious, surely? Is there any specific contract or code you think the article would be improved by linking to?
> There's no comparison with existing providers
Unstoppable Domains is arguably even less serious, Handshake doesn't allow you to own domains at the second level but is otherwise solid, [project] is good but there is a conflict of interest associated with me promoting it, MoneroDNS and Emercoin seem all right but I haven't looked into them deeply.
> the trade-offs
The trade-offs only exist because ENS exists as a "crypto-asset," the governance of which has to be freely traded on the open market for the economic structure to work. If ENS wouldn't have the purpose of making a profit for the insiders, there wouldn't be any reason to make all of these strange trade-offs in the first place.
When you sacrifice censorship-resistance and decentralization, what is it that you get in exchange? What are you trading it for?
> There's no discussion on how DAOs are still very immature and evolving.
How do you expect them to "evolve" in the future? Is there a specific issue that will be solved in a certain way, or is it like Proof of Stake, where "new developments" at an unspecified point in the future are supposed to fix all the problems, so you shouldn't ask too much tricky questions today?
> There's no acknowledgement of how ENS has behaved since inception.
Well, I note that ENS made direct lies about their decentralization status prior to November 10, 2021, I note that ENS publicly announced their intention to censor if required, and I note that a few hundred keyholders is better than seven.
Is there something else you think I should acknowledge?
> Flash loan voting? Clueless.
Does ENS have some special precaution against this? Or is it simply outside of the threat model?
> What is there you want me to link to, specifically? I mean, I suppose I could give a source for "they could crank up the fee for renewing only your name to $1,000,000,000, and then allocate it to whomever they please once it expires" if you insist. But this is not contentious, surely? Is there any specific contract or code you think the article would be improved by linking to?
I can imagine raising the price of renewal on a domain. But how would "they" control allocating the domain when it expires?
For example, by raising the new registration price to $1,000,000,000. Then, the DAO can register it at zero cost (it will get 100% of the fee refunded instantly), but anybody else would have to pay sticker price.
I believe there's also more complex schemes, where you directly restrict who can register it, but I'm not sure about those.
Oooh good one. Yes you need to spell out just how devious you are getting with these exploits. Love it. Skip all the broad sweeping stuff and get specific.
> Launch censorship-resistant decentralised websites with ENS. Upload your website to IPFS and access it with your ENS name.
And from the evidence given here [1] and [2]:
> It takes 4 out of 7 geeks to bypass the blockchain in case of emergency, court orders or kneecaps.
Not even 7. What is the point of the multisig then? At least ICANN requires all keyholders, but this is less than the maximum holders to bypass the blockchain.
> “Moving forward, we want to be as responsible as we can. This includes possibly seeking to register .ETH through the normal ICANN process” - note that the “normal ICANN process” for gTLD regitration requires compliance with trademark law.
So ENS is not 'censorship resistant' and it is basically a blockchain version of GoDaddy selling .eth domains which you do not own and 'can' take your domain away (or redirect it) whenever they want.
Sounds like decentra-lies all over again and foreshadowing the web3 delusion, unless they take away the 'censorship-resistant' and 'decentralised websites' claims in [0].
I mean, these days you have the DAO. So on the positive side, it's not 4 of 7 people anymore. But on the negative side, it's theoretically anyone, depending on how badly they messed up the code, concentration of ownership, and other economic/social factors.
You also have the problems that it's all resolved through a centralized service, in practice. So if Infura gets a court order to NXDOMAIN thepiratebay.eth, that's pretty much good night.
That's because ENS was set up as a delegated voting system. If one of those representatives starts losing their favor with the community, people can switch their delegates to someone else.
> This includes possibly seeking to register .ETH through the normal ICANN process”
This is a non-starter anyway. The new gTLD program has not been open for applications since 2012. Even if it were to reopen tomorrow, ICANN is highly unlikely to approve a gTLD that does not operate under a standard registry/registrar model. I'm not certain that they'd even be willing to consider it.
They do, actually. Charleston Road Registry (an entity owned by Google which also operates other TLDs like .APP, .DOG, and .FOO) is the registry; Google is the sole registrar. As far as I'm aware, all of the private TLDs operate on some variant of this model.
But it goes beyond just having those two distinct entities -- any ICANN TLD is expected to follow a set of operating policies (e.g. domain registration, renewal, transfer, expiry, and redemption; dispute resolution; WHOIS verification... etc) which is fundamentally incompatible with ENS.
For enough money, they might. If I'm not mistaken, the "community treasury" has 50% of all ENS. This comes to 0.5 * 100MM = 50MM ENS. At current exchange rates (assuming no slippage), that comes to $17.92 * 50,000,000 = $896,000,000.
I'm assuming ICANN approved all this gTLD nonsense because they like the money. This naturally raises the question: how much money would they need to approve this? And is it more or less than $0.9 billion dollars?
Money isn't the issue. If it were, other parties (like the various claimants to .app) would have b̶r̶i̶b̶e̶d̶ convinced ICANN to favor their applications by now. The real issue is simply that ICANN is a highly bureaucratic organization; they don't do anything quickly. They only published a report on the outcome of the 2012 gTLD offering last year.
They're in the process of adding layer 2 support so that you'll be able to register on rollups without having to pay high gas. Should be 1/10th of the cost in gas or less after its implemented.
In the meantime, I've generally spent about $60 or so on gas fees, and since it's $5 a year for the actual registration, I tend to offset the gas by registering for more years. If you spend $100 you get a domain for 8 years. That's as good as cheaper TLDs.
> [In Bitcoin] the power is moved from a monarch, a single point of failure, but it’s moved to nowhere. For example, who owns Bitcoin? Nobody!
What? Ownership/power is distributed among all the miners, albeit a number much bigger than 7. It still exists and can (and will eventually) be consolidated by a majority cartel.
The article is correct about ENS, but implying that N=1000000 power shares (approx # of Bitcoin mining individuals, likely trending down as mining margins decrease) is different than N=7 power shares (ENS sigs) in the realm of security guarantees is just smoke and mirrors. Neither system abolishes power nor is immune to manipulation.
Even if you control 70% of hashrate, that doesn't mean you can just do anything. You still have to abide by the rules. You can't approve invalid transactions or directly steal people's money, only re-order past transactions.
This is in contrast to ENS, where the equivalent of a 51% attack allows you to do just that - steal names, change registration parameters, basically anything.
Furthermore, in PoW coins, 51% attacks are fleeting by nature - they usually only run for a short time. Compare this to ENS, where if you get 51% control once, you can give yourself 100% control for perpetuity, because you can execute arbitrary code on the smart contract.
> Compare this to ENS, where if you get 51% control once, you can give yourself 100% control for perpetuity, because you can execute arbitrary code on the smart contract.
Maybe I don't understand crypto well enough, but since everything is in a public ledger, can't 'everyone' just decide to go on a hard fork right before the moment the bad actor took control, thus de facto invalidating all transactions after that point?
If that is true, in a way the power over a crypto resides in wherever (rather, whatever fork) the majority of users believes it resides.
Yes (actually, it would only have to be a soft fork), but that would require serious coordination. Ethereum has done it before, so it's not literally impossible, but it would require the whole of the Ethereum network to do it, not just the ENS people.
If it did happen, it would raise serious questions about the nature of the decentralization of Ethereum, and how suitable it is as a deterministic platform to build stuff on. Then again, it already happened and nobody seemed to really care.
If Ethereum is willing to fork over X, what's to say they're not willing to fork over Y? And the answer to this question is probably to do with "the big guys with the money at the top like X, but not Y". So much for decentralization.
Yes, in fact this is exactly what happened when a hacker found a bug in a DAO that allowed them to steal tens of millions of dollars. The outcome flies completely in the face of the idea of decentralization, censorship resistance and "law as code". But don't let that get in the way of the ponzi scheme.
Additionally to the points raised by other replies, Ownership/power currently is in Binance and Coinbase, because very few people transact directly in Bitcoin, and those that do realize gas fees are so high so Binance and Coinbase provide convenience ledgers that they control. This is exactly the same process as if you were trading stocks on the stock exchanges. Everything eventually becomes centralized, to solve the problems like transacting and punishing bad actors that develop in the bazaar.
I am sorry my man, but you sound like an old boss who was ranting about not being to print the electronic mail. Bitcoin doesn't have gas fees. That's Ether.
> What? Ownership, and power, is distributed among all the miners, albeit a number much bigger than 7. It still exists and can (and will eventually) be consolidated by a majority cartel.
That's not quite true. To the extent that the Bitcoin network is 'owned', it is owned by the full node operators, of which there are many more than there are miners.
See the SegWit2x vote[0], for example, wherein many large exchanges and miners representing < 83% of the hashing power of the network voted to increase the block size, but it failed for lack of adoption among node operators.
Notably, anyone can operate a full node. There is no specialized hardware required.
>Notably, anyone can operate a full node. There is no specialized hardware required.
Great! The 51% cartel summons their crony Beff Jezos who runs a billion minimal & globally-distributed (compromised) BTC full nodes with his enormous compute resources. How does this affect the integrity of the network?
How does my new full node connect to the network? Are there hardcoded relay/coordinator IPs that it tries to connect to? Who controls those?
>How does this affect the integrity of the network?
it doesn't. operating a node isn't a vote.
>How does my new full node connect to the network
there are boostrap nodes that the client initially connects to, but afterwards it connects to nodes based on previously seen nodes and nodes it has heard from other nodes.
The prior comment implied that full nodes had some power in ensuring the integrity of the network even in the event of a 51% attack. You are suggesting the opposite. Which is true?
It's related, yes. The main difference is that rather than your power being determined by how much bitcoins you hold, your power is determined by how much influence/power/contribution you have in the overall bitcoin ecososytem. A major merchant might not hold that many coins compared to the average whale, but probably holds more influence.
>so they require even more trust.
I wouldn't say it requires more trust. It's not like bitcoin holds an election every few years for determining which party holds the economic majority.
> I wouldn't say it requires more trust. It's not like bitcoin holds an election every few years for determining which party holds the economic majority.
But it's at least possible, from the protocol, to determine the balance of each address, and for each address to vote (with a message on the blockchain) for or against a given proposal.
Compare that to trying to determine which address "probably holds more influence", which seems like something you'd have to trust someone else's (off-chain) judgement on.
Yes, but that doesn't work if what you're voting on is consensus-critical (e.g. the ordering of blocks). If you're talking about other rule changes, sure. In practice, I think most would consider Bitcoin's sluggishness of rule changes to be a feature, not a bug. Because the process is not entirely deterministic, it is much harder for a bad actor to corrupt.
> Because the process is not entirely deterministic, it is much harder for a bad actor to corrupt.
As in software, a lack of determinism means a lack of reproducibility, and without reproducibility it's harder to have consensus.
In particular, if bitcoin makes a rule change (or refuses to implement a rule change), such a decision lacks the transparency and accountability needed for good governance, because the bad actor can just say "Well, that's just how things worked out that time. Blame the RNG.".
I don't think this is a theoretical concern when it comes to a bad actor corrupting the bitcoin protocol development process.
Proof of stake has a very specific definition. In Bitcoin, it's possible for a client to bootstrap with only knowledge of protocol rules. This is not possible in a proof of stake cryptocurrency, which also requires a trusted third party.
The author seems to be under the impression that changing from a 4-of-7 multisig to DAO control is not an improvement?
> Indeed, as long as a proposal gains more than 50% of the votes with a 1% quorum, it can execute abitrary code on behalf of the DAO - even such code that will give anyone dictatorial control over it in perpetuity.
Of course this is how it works. The DAO can do whatever it wants, and is controlled by the thousands of users who vote using their tokens. This is a huge improvement over a small multisig!
It may very well be better in relative terms. After all, arsenic is clearly better for your health than strychnine. But neither is especially good, now is it?
They said the DAO is an improvement, but it looks like the 11 core stakeholders effectively control the DAO. A contentious proposal could probably pass with only 6 votes. Over time those tokens may get spread among more holders.
I think that's ultimately what everyone wants. However, I think we need more research because once you throw away the keys they're gone forever. This is something that needs time to get right.
You don't ever want one person, one vote. The amount of power someone has in an organization should be proportional to the amount of skin they have in the game.
If I borrow 51% of the outstanding supply, how much skin do I have in the game? What if I buy $100 million worth of ENS, while simultaneously shorting it on DeFi?
What if I tell people that if they vote for my proposal, I'll raid the Community Treasury and bribe them with the money I get from that? Do they have skin in the game? And if so, which game?
If you try to borrow 51% of the supply, you'll find the cost to borrow goes up the more you get (until you're paying 10,000%+ apr.)
If you successfully pass a proposal to raid the treasury, all the $ENS tokens in there would likely crash in value as the market reacts to the news. Those who already have skin-in-the-game (by holding ENS and voting) would likely be worse off overall, so they wouldn't agree.
These are great questions! DAOs are essentially a big experiment in game theory and new forms of organization. I think you'd genuinely enjoy Vitalik's discussion about DAOs: https://vitalik.ca/general/2021/03/23/legitimacy.html
That certainly is spooky. Perhaps the quorum should be increased. (I imagine it would, if you tried to do something like this.)
Either way, there are some barriers. Each vote is subject to a 7-day period, so other delegates have time to react; votes must first reach some level of interest through Snapshot off-chain voting before moving on to the on-chain vote.
Are you referring to vote delegation? ENS holders 'delegate' their votes towards a person or entity, which then votes on their behalf, and this delegation can be changed at any time if the delegates are malicious.
This might not feel as democratic as direct one vote-per-token, but that scheme tends to have much lower turnout (as seen with compound.finance and curve.fi voting).
It looks like the top 11 delegates control around 40% of the vote[1], so nearly effective control. That does seem like a small group, but if a delegate was acting maliciously, I'd expect a number of their holders to change delegation.
> You “own” your domain, but you do not own your ownership of that domain.
I was having trouble forming examples of why I felt it important to be able to have full participation (e.g. running a full node for a given thing, or whatever else) before I involved myself in a "decentralized" thing
This is a well noted example of investigating how far someone can partake, and finding it woefully short, so thanks for this
>ENS is not decentralized in the sense that there is binding, non-human, trustless enforcement of what you may consider to be desiderata, such as:
> not seizing people’s names
This is false. Before control over the ENS protocol was handed over to the DAO, they made it impossible for the controller of the protocol to revoke any ".eth" name.
Of course, but what happens when they selectively raise the renewal fee for your domain to $500 quadrillion? (Also, what prevents them from changing the controller back, if they have 51%?)
i haven’t kept up with ENS to this deal. they moved from a multisig to a DAO. as GP claims, they also restricted the set of actions available to governance before this move. on the face of it then, it doesn’t look unlikely that the DAO will vote to further restrict its own abilities as time goes on. this exact mechanism has been used to progressively decentralize other protocols, and as we see with other “governance tokens” (e.g. Uniswap — where IIRC governance can tweak like one single boolean parameter), massively restricting governance abilities doesn’t destroy the viability of the token (which is to say, it’s not clear to me in which direction the finance incentive — if any — points with regards to such “ungovernance”).
^ this, except I'd add that as a large ENS delegate myself (as one of the founders of Metaphor), I and other large delegates don't factor in the price of the token into decisions.
We never intend to sell $ENS, it's a public good, not a profit-generating protocol. Think of it like an on-chain B-corp.
That said, if you really want to make sure you can't ever face the "egregious renewal fee" attack vector, you could just register it for a really long time. For most names, it costs like $500 to register it for 100 years.
This wouldn't happen as it's against the economic interests of the ENS DAO. Similar to a company maximizing the value of its stock, the DAO aims to maximize the value of the $ENS token by being a good steward of the project.
The whole point of a DAO like this is to try to create good incentives for the project without central control.
For example, a website hosting child pornography I think would be quite detrimental to the value of the project. Would the ENS holders seek to maximize their economic value, or "property rights"? Or is child porn good for business?
The renewal vector is the only way atm, yes. But it's easy to prevent this as a .eth holder: just register for a really long time. Costs like $500 to register for 100 years for most names.
As for the controller bit, even if you change the controller, we can't revoke the .eth names. We can only change new registration parameters (as you mention).
> ensuring that insiders don’t get to register names for free (since the fees go back to them)
Burning the fees would fix this. For something intended to be standard infrastructure, and whose ongoing costs are paid by the users, it'd be kinda nice if it were like that anyway.
The central criticism of ENS in this blog post - that the multisig (in the past) or the DAO (now) could take away your .ETH name - is not true. No one has the ability to take a .ETH name away from someone. The DAO (which launched Nov 8, not Nov 10 as it said), wasn't created to take over that power from the multisig because the multisig didn't have that power either. The multisig does manage the ENS root, but the .ETH TLD is permanently locked.
Re the fact the DAO controls the pricing and registration parameters of .ETH names, it's true that it would be possible for the DAO to change them to something that targets a specific user or name. The DAO cannot take away a .eth name that is registered (it could only make it difficult or practically impossible to extend or register anew). To guard against this, a user can pay ahead for as many years as they want, ensuring they have the name for that period (e.g. 100 years). Given that DAO proposals have required voting periods, it wouldn't be possible to surprise this on someone, and they'd be able to simply extend their registration for 100 years (or whatever period they want) before it went through. As you mention, the ENS constitution also prohibits it. I find this attack vector extremely unlikely to be attempted and near impossible to carry out successfully.
There's more I could say, but I think these were two main points.
I responded to this by e-mail, but I'll put it here as well:
> The central criticism of ENS in this blog post - that the multisig (in the past) or the DAO (now) could take away your .ETH name - is not true. No one has the ability to take a .ETH name away from someone. The DAO (which launched Nov 8, not Nov 10 as it said), wasn't created to take over that power from the multisig because the multisig didn't have that power either. The multisig does manage the ENS root, but the .ETH TLD is permanently locked.
But I do not, in fact, state this:
"ENS is censorship-resistant in the sense that nobody can directly seize your domain."
> Re the fact the DAO controls the pricing and registration parameters of .ETH names, it's true that it would be possible for the DAO to change them to something that targets a specific user or name. The DAO cannot take away a .eth name that is registered (it could only make it difficult or practically impossible to extend or register anew).
Now see, I would personally contend that "mak[ing] it difficult or practically impossible to extend or register" is a form of censorship.
> To guard against this, a user can pay ahead for as many years as they want, ensuring they have the name for that period (e.g. 100 years).
Well, if you assume that the registration fee never is increased above $5, sure. This also raises the question of why to even bother with expiration at all, other than as an opportunity to extract more rent.
Likewise, if people will just go out and buy the names for two thousand years whenever the renewal fees go up, this raises the question of why you would even bother and not just grandfather the old names in. (I mean, I suppose for throwaway names or trivial fee hikes, but it's not entirely obvious)
And I can still see some attacks about that business. For example, you start off by raising the fee to $6. Assuming this doesn't trigger them to buy it for 100 years, you raise it to $7, and so on. By the time the fee has reached let's say $100 p.a., hedging for 100 years would cost $10k, which is not exactly pocket money. And at that point, you can raise it even higher, etc.
Also, on an aesthetic note, having to register names for 100 years because the renewal can be pulled out from under your feet at any time seems like somewhat of a kludge.
> Given that DAO proposals have required voting periods, it wouldn't be possible to surprise this on someone, and they'd be able to simply extend their registration for 100 years (or whatever period they want) before it went through.
Are you suggesting people keep a close eye on the voting process of the infrastructure that controls their domain? If so, would you say that them feeling necessitated to do so indicates confidence in this infrastructure?
> As you mention, the ENS constitution also prohibits it.
Ah yes, but does it prevent it?
> I find this attack vector extremely unlikely to be attempted and near impossible to carry out successfully.
Well, a lot of people thought it would be unlikely for Ethereum to pull of a hardfork to override the steadfast iron will of unstoppable code, but here we are. In the end, it's not about what you think, but about what actually happens.
What happens when someone registers and actually starts using dailystormer.eth, causing a rash of bad h...
1. Author likes some cryptocurrencies and so takes on this single project with that lens rather than dismissive of all crypto
2. Links to some sources
3. Disclosure about financial interests at the top
4. These are valid concerns around censorship. I believe the ENS creators have thought about these but still when someone starts registering offensive names that will be the true test, or using ENS to point to scihub or something.
What to improve:
1. Seems like a long winded way to say "ENS isn't decentralized like Uniswap but it would be better if it was"
2. Title is overly clickbaity. Doesn't deliver. ENS team works hard from what I can tell. There's room for criticism but "clowns" is out of place.
3. Some of the exploits seems like they could be explained better. "3 ways to take down ENS are:". It's basically a penetration testing style analysis. But since you don't go to that level of depth it just comes across a bit sloppy.
Overall, better than 95% of the cryptocurrency content out there, unfortunately.
83 comments
[ 2.5 ms ] story [ 159 ms ] threadYou are 100% right. We need more of these specific critiques on web3 projects to separate the clowns from the scammers from the legit actors.
Pretty much everyone is going to flip from
"I don't like this, there is no use because I haven't seen one, I wish it would go away"
to
"I don't like this, here is how I can improve it"
What's good about it? Half the post is about Bitcoin. A large chunk is about the ENS token distribution.
There's no link to any of the contracts or code. There's no comparison with existing providers and the trade-offs. There's no discussion on how DAOs are still very immature and evolving. There's no acknowledgement of how ENS has behaved since inception.
Flash loan voting? Clueless.
Yes, because Bitcoin is an actually working, properly designed system used by people for real-world ends.
> A large chunk is about the ENS token distribution
That would be two lines just below "Which one is ENS?", plus the image text?
> There's no link to any of the contracts or code.
What is there you want me to link to, specifically? I mean, I suppose I could give a source for "they could crank up the fee for renewing only your name to $1,000,000,000, and then allocate it to whomever they please once it expires" if you insist. But this is not contentious, surely? Is there any specific contract or code you think the article would be improved by linking to?
> There's no comparison with existing providers
Unstoppable Domains is arguably even less serious, Handshake doesn't allow you to own domains at the second level but is otherwise solid, [project] is good but there is a conflict of interest associated with me promoting it, MoneroDNS and Emercoin seem all right but I haven't looked into them deeply.
> the trade-offs
The trade-offs only exist because ENS exists as a "crypto-asset," the governance of which has to be freely traded on the open market for the economic structure to work. If ENS wouldn't have the purpose of making a profit for the insiders, there wouldn't be any reason to make all of these strange trade-offs in the first place.
When you sacrifice censorship-resistance and decentralization, what is it that you get in exchange? What are you trading it for?
> There's no discussion on how DAOs are still very immature and evolving.
How do you expect them to "evolve" in the future? Is there a specific issue that will be solved in a certain way, or is it like Proof of Stake, where "new developments" at an unspecified point in the future are supposed to fix all the problems, so you shouldn't ask too much tricky questions today?
> There's no acknowledgement of how ENS has behaved since inception.
Well, I note that ENS made direct lies about their decentralization status prior to November 10, 2021, I note that ENS publicly announced their intention to censor if required, and I note that a few hundred keyholders is better than seven.
Is there something else you think I should acknowledge?
> Flash loan voting? Clueless.
Does ENS have some special precaution against this? Or is it simply outside of the threat model?
I can imagine raising the price of renewal on a domain. But how would "they" control allocating the domain when it expires?
I believe there's also more complex schemes, where you directly restrict who can register it, but I'm not sure about those.
Comments like that will kill web3 on it's own. No need for anti-web3 people to get involved.
From [0] as 'advertised':
> Launch censorship-resistant decentralised websites with ENS. Upload your website to IPFS and access it with your ENS name.
And from the evidence given here [1] and [2]:
> It takes 4 out of 7 geeks to bypass the blockchain in case of emergency, court orders or kneecaps.
Not even 7. What is the point of the multisig then? At least ICANN requires all keyholders, but this is less than the maximum holders to bypass the blockchain.
> “Moving forward, we want to be as responsible as we can. This includes possibly seeking to register .ETH through the normal ICANN process” - note that the “normal ICANN process” for gTLD regitration requires compliance with trademark law.
So ENS is not 'censorship resistant' and it is basically a blockchain version of GoDaddy selling .eth domains which you do not own and 'can' take your domain away (or redirect it) whenever they want.
Sounds like decentra-lies all over again and foreshadowing the web3 delusion, unless they take away the 'censorship-resistant' and 'decentralised websites' claims in [0].
[0] https://ens.domains/
[1] https://mailarchive.ietf.org/arch/msg/dnsop/-9zBqWpvNBlekGot...
[2] https://medium.com/the-ethereum-name-service/why-ens-doesnt-...
You also have the problems that it's all resolved through a centralized service, in practice. So if Infura gets a court order to NXDOMAIN thepiratebay.eth, that's pretty much good night.
Also want to read an analysis of how to take down ENS DAO with a flash loan of $ENS. Would need to be a higher level of detail than the parent article
This is a non-starter anyway. The new gTLD program has not been open for applications since 2012. Even if it were to reopen tomorrow, ICANN is highly unlikely to approve a gTLD that does not operate under a standard registry/registrar model. I'm not certain that they'd even be willing to consider it.
But it goes beyond just having those two distinct entities -- any ICANN TLD is expected to follow a set of operating policies (e.g. domain registration, renewal, transfer, expiry, and redemption; dispute resolution; WHOIS verification... etc) which is fundamentally incompatible with ENS.
I'm assuming ICANN approved all this gTLD nonsense because they like the money. This naturally raises the question: how much money would they need to approve this? And is it more or less than $0.9 billion dollars?
I think you forgot to divide by s.
see for yourself https://app.ens.domains/name/totallyrandom.eth/register
> Increase registration period to avoid paying gas every year.
I shouldn't have to. When I use my credit card, VISA charges the merchant (which can be passed to me), yes, but not at 5000% of the purchase.
The price of gas is not constant. Wait for lower demand.
In the meantime, I've generally spent about $60 or so on gas fees, and since it's $5 a year for the actual registration, I tend to offset the gas by registering for more years. If you spend $100 you get a domain for 8 years. That's as good as cheaper TLDs.
What? Ownership/power is distributed among all the miners, albeit a number much bigger than 7. It still exists and can (and will eventually) be consolidated by a majority cartel.
The article is correct about ENS, but implying that N=1000000 power shares (approx # of Bitcoin mining individuals, likely trending down as mining margins decrease) is different than N=7 power shares (ENS sigs) in the realm of security guarantees is just smoke and mirrors. Neither system abolishes power nor is immune to manipulation.
Not much bigger. Over the last month, 97% of blocks were mined by the top 10 mining pools.
https://btc.com/stats/pool
This is in contrast to ENS, where the equivalent of a 51% attack allows you to do just that - steal names, change registration parameters, basically anything.
Furthermore, in PoW coins, 51% attacks are fleeting by nature - they usually only run for a short time. Compare this to ENS, where if you get 51% control once, you can give yourself 100% control for perpetuity, because you can execute arbitrary code on the smart contract.
Maybe I don't understand crypto well enough, but since everything is in a public ledger, can't 'everyone' just decide to go on a hard fork right before the moment the bad actor took control, thus de facto invalidating all transactions after that point? If that is true, in a way the power over a crypto resides in wherever (rather, whatever fork) the majority of users believes it resides.
If it did happen, it would raise serious questions about the nature of the decentralization of Ethereum, and how suitable it is as a deterministic platform to build stuff on. Then again, it already happened and nobody seemed to really care.
If Ethereum is willing to fork over X, what's to say they're not willing to fork over Y? And the answer to this question is probably to do with "the big guys with the money at the top like X, but not Y". So much for decentralization.
https://levelup.gitconnected.com/how-ethereum-reversed-a-50-...
That's not quite true. To the extent that the Bitcoin network is 'owned', it is owned by the full node operators, of which there are many more than there are miners.
See the SegWit2x vote[0], for example, wherein many large exchanges and miners representing < 83% of the hashing power of the network voted to increase the block size, but it failed for lack of adoption among node operators.
Notably, anyone can operate a full node. There is no specialized hardware required.
[0] https://dcgco.medium.com/bitcoin-scaling-agreement-at-consen...
Great! The 51% cartel summons their crony Beff Jezos who runs a billion minimal & globally-distributed (compromised) BTC full nodes with his enormous compute resources. How does this affect the integrity of the network?
How does my new full node connect to the network? Are there hardcoded relay/coordinator IPs that it tries to connect to? Who controls those?
it doesn't. operating a node isn't a vote.
>How does my new full node connect to the network
there are boostrap nodes that the client initially connects to, but afterwards it connects to nodes based on previously seen nodes and nodes it has heard from other nodes.
It's related, yes. The main difference is that rather than your power being determined by how much bitcoins you hold, your power is determined by how much influence/power/contribution you have in the overall bitcoin ecososytem. A major merchant might not hold that many coins compared to the average whale, but probably holds more influence.
>so they require even more trust.
I wouldn't say it requires more trust. It's not like bitcoin holds an election every few years for determining which party holds the economic majority.
But it's at least possible, from the protocol, to determine the balance of each address, and for each address to vote (with a message on the blockchain) for or against a given proposal.
Compare that to trying to determine which address "probably holds more influence", which seems like something you'd have to trust someone else's (off-chain) judgement on.
As in software, a lack of determinism means a lack of reproducibility, and without reproducibility it's harder to have consensus.
In particular, if bitcoin makes a rule change (or refuses to implement a rule change), such a decision lacks the transparency and accountability needed for good governance, because the bad actor can just say "Well, that's just how things worked out that time. Blame the RNG.".
I don't think this is a theoretical concern when it comes to a bad actor corrupting the bitcoin protocol development process.
> Indeed, as long as a proposal gains more than 50% of the votes with a 1% quorum, it can execute abitrary code on behalf of the DAO - even such code that will give anyone dictatorial control over it in perpetuity.
Of course this is how it works. The DAO can do whatever it wants, and is controlled by the thousands of users who vote using their tokens. This is a huge improvement over a small multisig!
Detailed in the "Who Watches the Watchmen" paper:
https://www.frontiersin.org/articles/10.3389/fbloc.2020.5901...
What if I tell people that if they vote for my proposal, I'll raid the Community Treasury and bribe them with the money I get from that? Do they have skin in the game? And if so, which game?
If you successfully pass a proposal to raid the treasury, all the $ENS tokens in there would likely crash in value as the market reacts to the news. Those who already have skin-in-the-game (by holding ENS and voting) would likely be worse off overall, so they wouldn't agree.
These are great questions! DAOs are essentially a big experiment in game theory and new forms of organization. I think you'd genuinely enjoy Vitalik's discussion about DAOs: https://vitalik.ca/general/2021/03/23/legitimacy.html
Either way, there are some barriers. Each vote is subject to a 7-day period, so other delegates have time to react; votes must first reach some level of interest through Snapshot off-chain voting before moving on to the on-chain vote.
Ideally power should go to the best ideas but since each of us is limited in knowledge, the wisdom of the crowd is the next best option.
Is being a read-only user of ENS, zero skin in the game?
This might not feel as democratic as direct one vote-per-token, but that scheme tends to have much lower turnout (as seen with compound.finance and curve.fi voting).
[1] https://sybil.org/#/delegates/ens
I was having trouble forming examples of why I felt it important to be able to have full participation (e.g. running a full node for a given thing, or whatever else) before I involved myself in a "decentralized" thing
This is a well noted example of investigating how far someone can partake, and finding it woefully short, so thanks for this
> not seizing people’s names
This is false. Before control over the ENS protocol was handed over to the DAO, they made it impossible for the controller of the protocol to revoke any ".eth" name.
We never intend to sell $ENS, it's a public good, not a profit-generating protocol. Think of it like an on-chain B-corp.
That said, if you really want to make sure you can't ever face the "egregious renewal fee" attack vector, you could just register it for a really long time. For most names, it costs like $500 to register it for 100 years.
The whole point of a DAO like this is to try to create good incentives for the project without central control.
For example, a website hosting child pornography I think would be quite detrimental to the value of the project. Would the ENS holders seek to maximize their economic value, or "property rights"? Or is child porn good for business?
As for the controller bit, even if you change the controller, we can't revoke the .eth names. We can only change new registration parameters (as you mention).
Burning the fees would fix this. For something intended to be standard infrastructure, and whose ongoing costs are paid by the users, it'd be kinda nice if it were like that anyway.
Re the fact the DAO controls the pricing and registration parameters of .ETH names, it's true that it would be possible for the DAO to change them to something that targets a specific user or name. The DAO cannot take away a .eth name that is registered (it could only make it difficult or practically impossible to extend or register anew). To guard against this, a user can pay ahead for as many years as they want, ensuring they have the name for that period (e.g. 100 years). Given that DAO proposals have required voting periods, it wouldn't be possible to surprise this on someone, and they'd be able to simply extend their registration for 100 years (or whatever period they want) before it went through. As you mention, the ENS constitution also prohibits it. I find this attack vector extremely unlikely to be attempted and near impossible to carry out successfully.
There's more I could say, but I think these were two main points.
> The central criticism of ENS in this blog post - that the multisig (in the past) or the DAO (now) could take away your .ETH name - is not true. No one has the ability to take a .ETH name away from someone. The DAO (which launched Nov 8, not Nov 10 as it said), wasn't created to take over that power from the multisig because the multisig didn't have that power either. The multisig does manage the ENS root, but the .ETH TLD is permanently locked.
But I do not, in fact, state this:
"ENS is censorship-resistant in the sense that nobody can directly seize your domain."
> Re the fact the DAO controls the pricing and registration parameters of .ETH names, it's true that it would be possible for the DAO to change them to something that targets a specific user or name. The DAO cannot take away a .eth name that is registered (it could only make it difficult or practically impossible to extend or register anew).
Now see, I would personally contend that "mak[ing] it difficult or practically impossible to extend or register" is a form of censorship.
> To guard against this, a user can pay ahead for as many years as they want, ensuring they have the name for that period (e.g. 100 years).
Well, if you assume that the registration fee never is increased above $5, sure. This also raises the question of why to even bother with expiration at all, other than as an opportunity to extract more rent.
Likewise, if people will just go out and buy the names for two thousand years whenever the renewal fees go up, this raises the question of why you would even bother and not just grandfather the old names in. (I mean, I suppose for throwaway names or trivial fee hikes, but it's not entirely obvious)
And I can still see some attacks about that business. For example, you start off by raising the fee to $6. Assuming this doesn't trigger them to buy it for 100 years, you raise it to $7, and so on. By the time the fee has reached let's say $100 p.a., hedging for 100 years would cost $10k, which is not exactly pocket money. And at that point, you can raise it even higher, etc.
Also, on an aesthetic note, having to register names for 100 years because the renewal can be pulled out from under your feet at any time seems like somewhat of a kludge.
(All this assumes nothing gets migrated, as happened in <https://docs.ens.domains/ens-migration-february-2020/technic...>. Based on the high degree of centralization inherent to ENS, my money is on this being an easier process than it ought to be.)
> Given that DAO proposals have required voting periods, it wouldn't be possible to surprise this on someone, and they'd be able to simply extend their registration for 100 years (or whatever period they want) before it went through.
If I'm reading this right, the voting period is either 9 or 14 days: https://docs.ens.domains/v/governance/process
Are you suggesting people keep a close eye on the voting process of the infrastructure that controls their domain? If so, would you say that them feeling necessitated to do so indicates confidence in this infrastructure?
> As you mention, the ENS constitution also prohibits it.
Ah yes, but does it prevent it?
> I find this attack vector extremely unlikely to be attempted and near impossible to carry out successfully.
Well, a lot of people thought it would be unlikely for Ethereum to pull of a hardfork to override the steadfast iron will of unstoppable code, but here we are. In the end, it's not about what you think, but about what actually happens.
What happens when someone registers and actually starts using dailystormer.eth, causing a rash of bad h...
1. Author likes some cryptocurrencies and so takes on this single project with that lens rather than dismissive of all crypto
2. Links to some sources
3. Disclosure about financial interests at the top
4. These are valid concerns around censorship. I believe the ENS creators have thought about these but still when someone starts registering offensive names that will be the true test, or using ENS to point to scihub or something.
What to improve:
1. Seems like a long winded way to say "ENS isn't decentralized like Uniswap but it would be better if it was"
2. Title is overly clickbaity. Doesn't deliver. ENS team works hard from what I can tell. There's room for criticism but "clowns" is out of place.
3. Some of the exploits seems like they could be explained better. "3 ways to take down ENS are:". It's basically a penetration testing style analysis. But since you don't go to that level of depth it just comes across a bit sloppy.
Overall, better than 95% of the cryptocurrency content out there, unfortunately.