Hacker News new | comments | show | ask | jobs | submit login
The Decentralized Future (ycombinator.com)
195 points by rrecuero 5 months ago | hide | past | web | favorite | 141 comments



Or it's like all the other services we have, it evolves from decentralized to centralized: email, search, forums,...The case could be made even for governments. The only limit was technical, now you can control the cabs in all the cities from one place. The future is definitely more and more centralized, just look at the money distribution.

My guess is that we're having our geeky dreams without thinking what's the reality happening around us. It's very comfortable to be utopist when thinking of bitcoin, instead of seeing that it's still the same winner-takes-it-all scheme.


This is just an example of technical change aligning itself with real world human want.

People will ALWAYS put convenience first. Even those that say they won’t, it’s typically an isolated action (like, email is harder when not using google apps, but it’s worth not being spied on. Walk in their house, Echos and smart devices everywhere).

The hope for Blockchain is we can now, finally, begin building true decentralized things without sacrificing this innate human want for convenience. It’s technology advancing towards multiple mutual goals (a technical one and a human one).


Lightning networks to handle scaling limitations are the first step on the slow re-centralization of bitcoin. Eventually an overwhelming majority mining will be controlled by a centralized/federated oligopoly.

Be prepared to rebel and re-re-decentralize in 20 years.


Personally I'm much more bullish on Bitcoin Cash. It's following the same path that got us to this point today, and they're not opposed to Layer 2. But they have a clear understanding that a currency is only as good as it's utility. If they can maintain that balance of decentralization, utility, and security, while picking up the pieces that Bitcoin Core dropped it's likely we'll see a flippening in the coming year.


But isn't Lightning and segwit supposed to make btc faster and cheaper?


It's a different kind of centralization, if you can call it that at all, since you don't have to trust your counterparty in a lightning network style payment channel


I think I’d probably agree with you. But Blockchain technology (the backbone of a decentralized movement) is the parent to bitcoin.

I’m speaking on Blockchain, not bitcoin.


Even viewing it more generically, the same scaling issues apply, and the same inevitable re-decentralization awaits it. Centralized systems trade robustness for efficiency. The drive to lower costs will always result in re-centralization.


The scaling problems aren't necessarily only a symptom of decentralization. Proof of stake scales much better than proof of work, for instance.


Your right, they can be at your secrity module cost.

I currently (as I understand the state of pos) see pos useful for other the other blochain ESC stuff and pow for a safe currency (BTC)


we can change the world. some people always don't see the vision or they don't believe in it. we should not count on them for society progression regardless if we fail or not. otherwise let's all make a big corp dystopia our dream and go sleep.


Or we could attempt to see the bigger picture (money distribution) and fix that instead, that was my point.


I don’t see any actual practical use cases described here that make mainstream adoption of blockchain tech make any sense. Or any concrete arguments about how it improves current systems. And the weaknesses of blockchain are glossed over or ignored. And “it’s never been hacked” is the frankly pretty toothless argument in favor of its security. Basically this is a weak article for Y Combinator to be publishing. If this is the best they can come up with, then blockchain has worse prospects than I thought.


Blockchains boil down to one thing:

Where before you had to trust one server, now you can have many. Multiple writers, multiple readers. That's all.

They are a drop-in replacement for having to trust admins.

On the other hand, backups and replicas could prevent other things, such as: https://www.youtube.com/watch?v=E1d5VvCa8Fo


> Where before you had to trust one server, now you can have many. Multiple writers, multiple readers. That's all.

Those multiple writers, multiple readers are not free. They waste billions in server/electricity costs over traditional architecture designs. The most prominent blockchain bitcoin costs 2.5 billion dollars a year [0] to process a miniscule fraction of what Visa can handle.

[0]: https://digiconomist.net/bitcoin-energy-consumption


YES, we know! If there truly isn't an alternative to PoW, yes then we are pretty much fucked. Quite a few more efficient alternatives are currently being tested (and I guess a lot more are to come), and it looks very promising.


Of the top 8 coins on coinmarketcap I think four don't use proof of work - ripple, cardano, stellar and neo. So there seem to be alternatives.


you have a lot of hatred in you like others. did you stop to think why? the article is nice and describes use cases. "the best they can come up with"? sounds like they are trying to convince you. that's not happening. they are describing the endless possibilities offered by a new technology. it's you that needs to keep up.


PG/YC started a revolution: They suggested a way through which programmers came into money.

Now, with Bitcoin and Cryptocurrencies: Programmers are creating money itself.

This article makes a bull-case very articulately, with sprinklings of realism. I am looking forward to this series. Being a blockchain cynic is the norm these days, and I appreciate when someone takes the trouble to lay down a well-reasoned and well-researched argument - great job rrecuero.

Clearly, there is irrational optimism/scams abound, but we can often forget that the mass-irrationality can be decoupled with the 'actual' promise of the idea. And the fundamental promise of the idea (Bitcoin, Cryptocurrencies, Blockchain etc) is for us to rethink 'how things work the way they work and why' and attempt to improve it. The Bitcoin paper should atleast get us to think about what is money. It matters less whether these "blockchain things" ultimately succeed or fail -- the fact that something NEW is being attempted, has to be applauded (Ofcourse, it would be great to figure out a way to do this without harming people using their credit cards to buy shit coins).

I love reading about blockchain ideas/projects. Like many, I too am jaded by the scams - but I think it is worth maintaining some optimism through this.

This article doesn't even touch upon some very important use-cases like: moving money across borders and money being seizure resistant etc. I think there is was, and has always been a market for a product that allows us to store value in a way that is cross-border, govt resistant and anonymous (whether it is a $500B market or $5B market is less interesting than appreciating that a product was invented to fulfil this use case). There are plenty of interesting use-cases to explore here.

In the section on Tokens, the article says: "contributors can transfer their assets instantly and easily to other people (pending regulation)" while referring to securities. IF this becomes a reality, it would be huge.

I see a lot of upside if even a fraction of the promise of this tech is realized.


Thank you for your kind words. My goal was exactly that; to show that there is something beyond the irrational optimism, extreme cynicism, scams and bubble talks that dominate conversations.

As you mentioned, asset tokenization (securitization) is one of the biggest markets. If companies like Harbor achieve their vision, it would be huge.


Harbor is promising. Their paper and code is still light on the details, but I am rooting for them. There are some open questions around compliant secondary trading, real world enforcement etc - but very promising direction. Do you know other projects that are attempting real asset security tokens?


There are others like Polymath but Harbor is the most promising right now imo


I’ve reacently read the Burnskie book about crypto assets in the hope that I might come away more enlightened about the market and how to assess it. The opposite was the case. After reading the book and looking into it more, it’s very hard to see these things as assets at the moment, or to see much utility in the technology. Don’t get me wrong, there certainly is some and it is certainly ‘cool’, but right now blockhain feels like a solution looking for a problem.


The problem is not because cryptoassets are bad, it's because you read an investment book instead of technology book. (I read the cryptoassets book too and I think it's a great book, it's just that it's not a book to solve the specific problem you're talking about)

I can confirm from my own experience that reading pop-sci books about blockchains--although they sound really cool and do provide motivation to look further into the tech--doesn't give you any confidence about the technology itself.

There's no shortcut, if you want to be more "enlightened", you should stop reading medium blog posts or pop-sci books and start looking into the protocol itself, run a node yourself, and try stuff out. When you do so, you'll get the "enlightment" you were looking for.


> When you do so, you'll get the "enlightment" you were looking for.

Which is what? That the blockchain is a solution in search of a problem? There is very, very little reason to use a cumbersome data structure when better ones exist (eg: just a regular database, or something like git). Worse, the only way the blockchain gets all it's supposed valuable attributes is when you provide a financial incentive to mine. Aka -- bolt a "currency" on top of it and then go market it to tons of rubes so you can convert all of the generated tokens into cold, hard fiat to pay your bills. Without the "currency", who will pay miners to trustlessly verify all the transactions? The parties running the service? If so... since you seem to trust all them to pay the bills.... why not just use a database instead and save some substantial fiat?

The whole bitcoin/blockchain "movement" is just pure hype. It's people at the top of the pyramid hyping a worthless technology in order to cash out their tokens before the whole thing collapses in on itself...


> There is very, very little reason to use a cumbersome data structure when better ones exist

Do you not see any value in decentralization? In theory with blockchains, you could replace giant companies with open-source programs that pay for their own server costs with transaction fees, without taking any profit margin off the top. That would be very interesting for stuff like cloud storage and social networks. You can argue it's overly idealistic and may not work out, but you can't pretend there's no value in decentralization


> In theory with blockchains, you could replace giant companies with open-source programs that pay for their own server costs with transaction fees, without taking any profit margin off the top

Do you think these applications are viable if the token has significant monetary/speculative value attached to it?


Absolutely - in some cases the monetary value is what makes the applications viable in the first place. It's just like any other market; in theory the price of the token should reach equilibrium to reflect the value it provides, even if there's some speculation along the way


Internet was a "solution in search of a problem" too.

That said, you're free to believe what you want to believe.

Just one thing though, everything you said above I heard hundreds of times before, and there are good explanations for them. I just don't have time to argue here because these are things you can find out on your own if you put your mind into it. (To be honest I don't think I can convince you no matter how I try unless you open up your mind, so I won't try)

Anyway, if you don't care, that's fine too. It's not like not believing in cryptocurrency will doom your future. You'll do fine. And even if the whole thing fails, I'll do fine too. But it's always good to have an open mind about things you don't yet completely understand and try to learn. ("Thinking that you understand completely" doesn't count as actually completely understanding)


> Internet was a "solution in search of a problem" too

Eh, the history of ARPANET involved practical use cases (i.e. communication between specific institutions and ballistic missile defense coordination) from the start [1]. If one includes circuit switches, i.e. the telegraph/telephone system, as a blueprint Internet then the practical application reaches back earlier.

[1] https://en.wikipedia.org/wiki/ARPANET#Creation


And Bitcoin doesn't have a practical use case?


Please elaborate?


> Internet was a "solution in search of a problem" too.

No, no it wasn't. Close to light speed communication provided about 10 orders of magnitude faster/cheaper communication than before. I cannot even fathom how someone would not immediately realize how that massive increase in communication would not be useful.

Contrast this with blockchains which are practically 10 orders of magnitude more COSTLY since every node has to process every transaction when you could just do that on one server (hence why bitcoin transactions cost 2.5 Billion dollars a year to process a fraction of what Visa can process).


> I cannot even fathom how someone would not immediately realize how that massive increase in communication would not be useful.

Just like you cannot fathom how Bitcoin can be useful in February of 2018, 99.9% of the world population couldn't fathom how Internet would take off back then when it launched. That was why it was not mainstream for a long time until it took off.


> 99.9% of the world population couldn't fathom how Internet would take off back then when it launched.

The dotcom bubble was fueled by speculation in how useful the internet would be. Granted, it busted but many companies survived it and some are now the biggest companies in the world.

> That was why it was not mainstream for a long time until it took off.

Internet growth has been incredibly fast: https://en.wikipedia.org/wiki/Global_Internet_usage

I can see some utility in blockchains. However, my main point is that the economics of them make them terrible for 99.99% of problems.

Internet: 10 orders of magnitude faster/cheaper communication than before.

blockchains: 10 orders of magnitude more COSTLY.


> I can see some utility in blockchains. However, my main point is that the economics of them make them terrible for 99.99% of problems.

Then we're talking about different things. My comment was specifically in response to "Internet was a solution searching for a problem".

Your point actually supports my comment because like you said it's terrible for 99.99% of problems, which means it may be great for 0.01%, which means it's a solution that's solving only a small problem for now, and looking for more problems.


Lol, appreciate the patronising explanation, as if I'm some sort of idiot just waiting to find the road to Damascus.

I understand the technology perfectly well. I read the Cryptoassets book to try and undsetand exactly that: the value of them as 'assets'. I came away exceedingly underwhelmed.


If LOL'ing and making snarky comments about how you understand but it's still underwhelming makes you feel better, good for you.

But I have never seen a single person who truly "understand the technology perfectly well" say it's underwhelming.

Most of those who actually dug deep enough to "truly understand" do see the limitations, but because by that point they understand the potential, they see the limitations as problems to solve and opportunities to take.

It is those who only scraped the surface that say it's "uninteresting", "underwhelming", etc.

Ancient people thought they "understood the world perfectly well" and thought it was flat. Suit yourself.


>never seen a single person who truly "understand the technology perfectly well" say it's underwhelming

I'd say I've noticed an inverse correlation - people who don't understand the technology see it as this amazing thing, the kind of people who know their hash functions are a bit skeptical.


“But I have never seen a single person who truly "understand the technology perfectly well" say it's underwhelming.”

This is tautological, so i hope you’re aware of how logically useless it is. Either you haven’t seen much, or you define away any disagreers as people who don’t understand.


In what way is your comment different from 'No True Scotsman'?


It isn’t. Apparently I’m just not devout enough for crypto....


Money transfer is still very difficult. As in right now, short of paypal, I cannot send you money easily without leaving my chair in under 1 minute regardless of where you are in the world. Send me your bitcoin address and the problem is solved.

The biggest problems with Bitcoin (specifically) are: 1) it is not necessarily a store of value and 2) it fluctuates too much.

If someone can solve these two problems then we will have an extremely powerful cryptocurrency.


You'll be doing well to get a bitcoin payment through in less than a minute. Half an hour plus is typical. There are faster cryptos though.


> As in right now, short of paypal, I cannot send you money easily without leaving my chair in under 1 minute regardless of where you are in the world

Venmo. Zelle. During banking hours: e-mail my banker, wire transferred within minutes.


> Venmo. Zelle.

If you accept the limitation that the receiver can be anywhere in the world, but must be from the US or have an US bank account...


Transferwise


> Don’t get me wrong, there certainly is some and it is certainly ‘cool’, but right now blockhain feels like a solution looking for a problem.

"Blockchain" is probably fine. I suspect that it will eventually get some nice implementations where you want to be able to attest to integrity over time/space where multiple trusted parties evolve and change. A "distributed digital notary" if you will.

Cryptocurrencies as currently implemented (bitcoin and ilk) are eventually going to die for lack of a use case.

Problem 1: Most people like centralized authority of money. Most people like being able to get back their money when things get "stolen", for example. And people like reputation systems so that they don't get scammed (ie. illegal drug market).

Problem 2: Fees suck, but they're invisible to most people. Merchants hate fees, but you need users before that gets moving. We're also seeing that cryptocurrencies have non-trivial fees, as well. Funny that.

To be fair, traditional fees keep falling and some of that is probably due to cryptocurrencies. But some of that is simply trying to ward off the rise of a single digital purchase standard which would disrupt all of the incumbents. The payments space is a no-holds barred fight to the death and the incumbents would rather die than let anybody else get a foothold.

Problem 3: Cryptocurrencies aren't sufficiently anonymous. There is a use case for genuine anonymity. However, the current cryptocurrencies fail to be truly anonymous at blockchain as well as reality. Keeping your operational security sufficient to stay truly anonymous is damn difficult and having to go through an exchange doesn't help. Cryptocurrencies really need a decentralized exchange, but I'm not sure how that would even work. And, even if it did, since nobody could charge fees, nobody would have any incentive.

So, cryptocurrencies are mostly useful for people who fear central authority, are possibly concerned about fees, and only need to remain "semi"-anonymous. That pretty much describes petty criminals and mid-tier corrupt party bosses and ... not much else.


Nano (ticker: XRB) has no-fee transactions. Send 3 XRB, receive 3 XRB. I'd say it's biggest wart yet is the volatility, but there are already half-way solutions like instant fiat resolution. These kinds of services take their own fees unfortunately, but I am sure some sort of peer-to-peer crypto<->fiat will pop up eventually.

Monero is the go-to privacy coin. Unfortunately, I am not as experienced with privacy/security, but it appears to be good enough for many. Perhaps someone else can speak to its efficacy?

Finally, there are decentralized crypto exchanges. None that take fiat in (AFAIK) but there are definitely decentralized exchanges. Don't know of any off the top of my head, but I'm sure they're just a quick google search away.

Those are the things that stuck out to me about your post.


"Attack of the Fifty Foot Blockchain" and the associated blog is worth a read


I read that book too, and while it does provide some good insights, it's too much biased the other way around to the point that I couldn't stand the amount of snark.

I think it's good to take a look at both sides of the story but the only way to find the truth is going deeper and learning the protocol yourself and form your own "sovereign" opinion, which is what I did after getting frustrated with tons of conflicting opinions online, which by the way turned out to be mostly propaganda to preserve their own crypto-wealth as I started learning more about the landscape.


It was an eye opener for me. Since then I approach the whole field with more caution and skepticism.


How is trustless digital programmable peer-to-peer monetary transaction not a great solution to the antiquated, corrupt status quo?


Just because you put the words antiquated and corrupt in the question doesn’t make those two things true.

Based on my somewhat brief experience as a crypto investor it hasn’t given me great hope of it solving corruption...


It's democratizing corruption.

Immune systems will develop I believe.


The nature of single things is decentralized. Therefore every new era starts with decentralization. The internet started off quite decentralized, for instance.

But do we really have a new era here? I'm not sure if with 32 years of age I'm already too old to recognize the New and grog it. I don't feel it. I feel there's a lot of buzz around blockchain, sure. But it looks like it would be all. Bitcoin is just as centralized as banks are, for instance.

And here is where the other stuff starts. When single things start systems these systems tend to be quite centralized. That is the natural order of systems. E.g., people could manage their own bitcoins on their laptop, but they choose to have an Exchange take that effort of their hands.

Seeing centralization take a foothold and not seeing anything real besides buzz, I'd argue the bet is still against.

On the other hand I'm seeing Amazong completely automating logistics in more and more areas of life. Shouldn't that be something to look into?


I think the best way to think about this to think about the tcp/ip protocol. It's a standard protocol but it's decentralized still to this day.

Blockchain takes it further and conceptually adds history to the TCP/IP protocol. This has extreme value as it allow data to behave like physical properties.


> Bitcoin is just as centralized as banks are, for instance.

That's a rather odd statement. Can you justify that?


I've put it in the second example. People can manage their bitcoins by themselves, but instead they let exchanges manage it for them. So all the problems of centralization (exploitation of users, market manipulation, big time theft, etc) can be found in the bitcoin ecosystem as well.

You see the same history repeating, just a lot quicker. Therefore I assume whatever you do with blockchains will become a centralized system as well.

PS: If you think about it, as technology cash is even more decentralized as bitcoin. Yet banks happened.



https://www.bloomberg.com/news/articles/2017-12-08/the-bitco...

A handful of actors with a controlling stake isn't exactly the same thing as "centralized" but it's close.


It's also inherently "centralised throughout liquidity". Because it's both deflationary and risky, the amount of liquidity available at any given moment is small, so a willing actor with some money at hand can affect the market disproportionately. Look at the way Tether manipulated the price.


Not really. They're pretty different things.


The phenomenon of decentralization won't slow until the wildly inefficient businesses of Washington, Hollywood, New York, and Silicon Valley have been broken into thousands of pieces.

Bitcoin and the rest are coming for New York/Silicon Valley money managers, which only exist for gatekeeping and rent seeking. The finance industry will be eliminated by new exchanges and equity crowdfunding.


I feel like proof of work blockchains don't actually solve the problem of "who to trust". Nodes just automatically assign trust to whoever mined the best hash that they know of, which turns out to be whoever holds the most mining power.

But not even that: If you're in China and the original blockchain is censored and all your node will ever see is the StateChain, you have misplaced your trust.


In a proof-of-work chain, the fundamental trust assumption is that no malicious mining collusion holds more than 50% of the hash power. If you can make this assumption, your security holds as long as you manage to obtain the most recent block with some frequency (usually at least every 6 blocks ~ 1 hour). Of course, if someone manages to MITM attack you, they can mine blocks and feed them to you in order to double spend on you, for example. One way to accomplish this would be to engineer an Eclipse attack [1].

However, as long as you have any source for the most recent block that the MITM does not control, they cannot trick you into following a shorter chain. In other words, as long as any side-channel (malicious or otherwise) posts blocks, you can always follow the longest chain. The Blockstream Satellite [2] was created for this reason -- so that nodes always have a reliable side-channel separate from the bitcoin protocol to verify blocks coming from the bitcoin network. Alternatively, some nodes check Tor mirrors, which should be difficult for adversaries to control.

[1] https://eprint.iacr.org/2015/263.pdf [2] https://blockstream.com/satellite/


> However, right now you need to install browser extensions like Metamask to interact with websites that enable “crypto functionality”. They are still really clunky and the average user does not know they even exist.

This is not true, you can already build (and people are) dApps that run in the browser (no extension) using the latest native Web Crypto API that we've made easy to use:

https://hackernoon.com/so-you-want-to-build-a-p2p-twitter-wi...

Dominic Tarr and the Beaker Browser guys with SSB also have even webasembly-ified libsodium which is highly rated ( #4 ranked on GitHub https://github.com/topics/cryptography ).

It is just that a lot of investors are caught up in hyped-up vaporware scams, rather than real technology that already exists and works and is available for developers to use today.


By "crypto functionality" the author means talking to a blockchain, not doing cryptographic operations. Metamask, the example given, is a browser extension that provides access to the Ethereum blockchain by injecting a JS object into the web page. It provides a wallet and transaction signing and broadcasting. Web crypto API's are quite different.


Both are possible, if you read through more.


> traditional banking that takes days and charges high fees

Only in the USA. And that is hopefully changing in the next month or two with the upgrade in the ACH system.


I just checked transferwise for £1000 to about us$1400. The cost compared with google's estimate of the interbank rate was $6 and they said it would take one day. I've also recently done £ to Thai baht and most currencies are there. It's quite quick and cheap such that it would be hard for crypto transfers to compete for most stuff.


> It's quite quick

Really? I use it often and it always takes a few days end-to-end


Wow. This blog post literally parallels what my colleagues and I have been writing and working on for the past 6 months at Intercoin.

I mean, look at the video right here: https://intercoin.org/

It even uses the same terms, such as "Power to the People". https://qbix.com/blog/index.php/2017/12/power-to-the-people/

I am glad that these challenges in crypto are finally being recognized and publicly written about at YC. Perhaps we should apply!


> Chris Burniske and Jack Tatar suggest that crypto assets have low/negative correlation with traditional asset classes

The DJIA and Bitcoin charts seem well correlated, at least over the past month: DJIA: https://www.marketwatch.com/investing/index/djia Bitcoin:https://bitcoincharts.com/charts/bitstampUSD#rg30ztgCzm1g10z...

In my opinion people treat it as they would any speculative asset.


Huh? You can’t cherry pick your data and claim something is “well correlated”. Well researched data shows bitcoin has no correlation to any exising asset class which is why it’s a great risk reducer to add to a portfolio.


Your retort would be more convincing if you linked to some sources.



It's not a new asset class. Most cryptocurrencies are junk bonds. Those have been around for ages.

If the coiners (what do you call this tribe?) accomplish "Permissionless Innovation" though, they will have did something great.


It's sort of an new asset class - there wasn't anyone transferring crypto in Shakespeare's day. They don't have an iffy promise to pay you money like junk bonds. In traditional terms they are a bit more like assets that can cost something to mine. Whether they end up valued like gold or like sea water we shall see.


Happy to answer any follow-up questions here


Great article!

Bringing financial inclusion to the billions of unbaked it's the biggest unfulfilled promise of cryptocurrencies.

For it to work it must be engineered on the protocol level for this use case, I think Cardano is trying to solve this problem.

The store of value and savings part it's more or less easy to solve, the big problem is the lending without a stable coin it will be very difficult to implement, big Fiat currencies like the Dollar and the Euro have the powerful backing of a central bank and ultimately the collateralization of all the taxpayers as a protection against high volatility and black swan events, this will be very difficult to implement in crypto, MakerDAO is tring to do it but I don't see it working on a big scale.


> It’s not a coincidence that the whitepaper was published after the 2008 subprime mortgage crisis.

Is there any evidence for this? Perhaps a statement from Satoshi that they were motivated to publish by the crisis?


In the Genesis Block he explicitly references the post financial crisis bailouts.


Yep. He noted a headline in that day's paper

"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks"

Source: https://en.bitcoin.it/wiki/Genesis_block (or you can verify on the blockchain as well)


Thanks for this writeup! You've done a great job at organizing many of the points that were discussed in this Bitcoin thread[1] from last month. I'll be using this as a starting point for discussion with friends that are interested in cryptocurrency.

[1] https://news.ycombinator.com/item?id=16158463


Thank you. That was the intention, to provide a primer for discussion.


It's not quite a complete discussion without mentioning the transaction rate/scaling limitations and also Ethereum's sharding approach to this.


sharding is in the same boat as proof-of-stake, nobody knows if a system can be built that preserves the level of security of current monolithic and proof-of-work systems, it is being studied on an academic an implementation level by several teams including Cardano, ethereum and ethereum classic.

On the scalability issue if you sacrifice the decentralization is theoretically possible for Bitcoin to process on-chain, Visa level of transactions and with current energy consumption levels or even less.

You would probably get 3 to 6 physical data center locations probably on the same continent connected with high-speed links with 1 Terabyte 10min blocks.


Proof of steak sounds delicious and would be enough to tempt me off the sidelines into investing.


>Proof of steak sounds delicious and would be enough to tempt me off the sidelines into investing.

LOL Proof of steak https://youtu.be/U3OEk2dNYcU?t=44


I tried to mention that scalability is a moving target. As you mentioned Plasma, Raiden and sharding are different approaches to solve it


Raiden is a poor and incomplete implementation of Lightning. Bitcoin has 3 independent implementations and a developing ecosystem of software around it.

Proof of stake is a pipe dream. Such a system can’t reliably achieve consensus due to the nothing-at-stake problem. Additionally, randomly selecting a winner in a distributed yet unpredictable way continues to be an unsolved problem.

Sharding has the cost of weakening security.

Bitcoin is addressing scaling better than any other blockchain. It’s making transactions more efficient through technologies like Segwit and Schnorr. It’s adding robust smart contracts and instant payments using second layer sidechains like rootstock and lightning.

Ethereum is already imploding under its own weight — it’s impossible to even sync a full node on basic hardware. In a few years it’ll be dead.


> Raiden is a poor and incomplete implementation

How so?

> Sharding has the cost of weakening security.

Why?


>>Raiden is a poor and incomplete implementation of Lightning.

Raiden is a comprehensive implementation that includes Lightning Networks for ERC20 tokens, something that Bitcoin doesn't even have.

>>Proof of stake is a pipe dream.

Your information is incorrect. Ethereum's proof of stake is close to complete conceptually and the first version, a hybrid PoW/PoS protocol called Casper the Friendly Finality Gadget, is currently under development and will be implemented in Ethereum soon.

>>Such a system can’t reliably achieve consensus due to the nothing-at-stake problem.

Your criticism is long out of date. Vitalik Buterin addressed how the Nothing at Stake problem can be solved in 2014:

https://blog.ethereum.org/2014/11/25/proof-stake-learned-lov...

>>Bitcoin is addressing scaling better than any other blockchain. It’s making transactions more efficient through technologies like Segwit and Schnorr. It’s adding robust smart contracts and instant payments using second layer sidechains like rootstock and lightning.

Bitcoin's lack of Turing Completeness at the base layer is a critical weakness that limits the range of second layer solutions that are possible. Ethereum has far more sophisticated and numerous scaling solutions in development as a result.

>>Sharding has the cost of weakening security.

That is not certain to be true. There are a lot security impacting variables at play that sharding affects, and some of them positively. Even if security were reduced, it wouldn't mean the trade off for more scalability wouldn't be worth it. Sharding enables massive scaling while preserving the ability for consumer grade nodes to contribute to validation. That seems like a good trade off.

>>Ethereum is already imploding under its own weight — it’s impossible to even sync a full node on basic hardware.

That is highly misleading. SSDs have been syncing fine. HDDs have been having syncing problems, but that was solved in the most recent Geth release.

>>In a few years it’ll be dead.

Disingenuous FUD.

Ethereum is the primary Dapp platform, with 30X more developers working on it than the next most active platform. It has a multipronged scaling strategy that is in an advanced state of development and dwarfs that of any other blockchain. It is now the most widely used blockchain in the world, processing 3X more transactions per day than #2 Bitcoin. And in all of these categories, its momentum is growing.


> Raiden is a comprehensive implementation

Is the lone implementation which is still on testnet and no ecosystem surrounding it.

> ERC20 tokens, something that Bitcoin doesn't even have.

Bitcoin has no interest in implementing an equivalent as a layer-1 feature. This should be a surprise to no one. Off-chain scaling is the official direction of the Bitcoin community. There's no interest in adding bloat to the layer 1 protocol. Ethereum is the "everything including the kitchen sink" approach. Anyways, saying Bitcoin does'nt have ERC20 is factually incorrect: Rootstock implements ERC20 tokens as a second layer network.

> Your information is incorrect. Ethereum's proof of stake is close to complete conceptually and the first version

It's not any where near complete, which is why the difficulty bomb was rolled back (which in and of itself is hilarious -- Ethereum only pretends to be decentralized).

> Vitalik Buterin addressed how the Nothing at Stake problem can be solved in 2014

He described how to solve it 4 years ago, yet they still can't release even a hybrid PoS solution. Slasher does not solve the nothing at stake problem, it merely obfuscates it. Long range attacks are still possible.

I noticed you completely glossed over how PoS systems will agree on randomness. Good luck. I'm sure it's just another 4 years away.

> Bitcoin's lack of Turing Completeness at the base layer is a critical weakness that limits the range of second layer solutions that are possible.

Lack of Turing Completeness is a feature and intentional. If you want a general purpose distributed computer use rootstock. Which, by the way, is 100% compatible with the Eth VM.

> Sharding enables massive scaling while preserving the ability for consumer grade nodes to contribute to validation.

That's cute. Ethereum doesn't even have any "consumer grade" nodes today. Anyways, sounds like you're taking the long way around to agreeing with me: Sharding sacrifices security.

> That is highly misleading. SSDs have been syncing fine. HDDs have been having syncing problems,

Again, taking the long way around to agreeing with me. A full archive node currently requires a 400GB SSD! You're only option is to run a light node, which further contributes to your networks centralization.

> Disingenuous FUD.

Eh, you'll see. The monolithic approach of Ethereum is very naive.

> with 30X more developers working on it than the next most active platform

Easily proved false, anyone reading can take a look at the respective github repos. Bitcoin currently has over 550 contributors, Ethereum only has 200. Bitcoin has over 16k commits. Ethereum has only 9k. This doesn't even take into account the larger ecosystem of software Bitcoin has.

> Ethereum is the primary Dapp platform,

A little early to make any of these claims...the only popular Dapp in recent memory is cryptokitties and it brought the Ethereum network to its knees. Anyways, like I've said, Bitcoin has a fully compatible Ethereum VM running as a layer 2 network.

> It has a multipronged scaling strategy that is in an advanced state of development

I wouldn't call it multipronged -- Vitalik has it all hinging on PoS and that has been officially delayed until late 2018 at best.

> dwarfs that of any other blockchain.

Well, until you look at actual commit and contributor count. Then it's clear Bitcoin has far more development happening.

> It is now the most widely used blockchain in the world, processing 3X more transactions per day than #2 Bitcoin

I mean, bitcoin is operating at capacity. I'm sure Ethereum has more spam transactions though, I agree. Saying Ethereum is the most widely used blockchain in the world his hilarious. People know Bitcoin. A small fraction of those people know Ethereum.

> And in all of these categories, its momentum is growing.

As its network continues to crumble. Best of luck. We'll reconvene when Casper finally launches...hopefully while we're still young.


> A full archive node currently requires a 400GB SSD! You're only option is to run a light node,

Uh, what about running a full archive node? The bitcoin txo set is also much larger the the utxo set


Bitcoin doesn’t need an ssd to sync. Additionally, a bitcoin full node requires just 149gb — and it’s been around much longer than Ethereum! Ethereum is broken.


Do you mean full node or archive node? (ie is the 149gb a requirement for storing the txo set or the utxo set)


149gb for the txo set, the entire history.


>>Is the lone implementation which is still on testnet and no ecosystem surrounding it.

You're right that it's incomplete, in the sense that it's not completely finished, but not much more incomplete than the LN implementations on Bitcoin. They're on mainnet, but the developers say it's just there for testing purposes, and large amounts should be transacted on it. Its ecosystem is limited to buying sweaters from Blockstream and buying a VPN subscription.

To claim that the Raiden implementation is significantly behind the Bitcoin ones is misleading.

>>Bitcoin has no interest in implementing an equivalent as a layer-1 feature.

Which is a disastrous technical decision based on absurd self-defeating dogma. Ethereum already has 92% of the token market capitalization, up from 73% in July 2017. All the while, the token market grew from $4 billion to $60 billion.

Network effects matter, and the destructive dogma you're promoting resulted in Bitcoin letting Ethereum becoming the dominant crypto-asset platform, with a growing share of a rapidly growing market.

In any case, the fact that Raiden has ERC20-compatibility is another example of why your characterization of it as being "a poor and incomplete implementation of Lightning" is absurd. It's a comprehensive, full-featured implementation that is very nearly complete.

>>It's not any where near complete, which is why the difficulty bomb was rolled back (which in and of itself is hilarious -- Ethereum only pretends to be decentralized).

I didn't say it is "anywhere near complete". I said that Ethereum's proof of stake is close to complete conceptually and the first version, a hybrid PoW/PoS protocol called Casper the Friendly Finality Gadget, is currently under development and will be implemented in Ethereum soon.

>>Easily proved false, anyone reading can take a look at the respective github repos. Bitcoin currently has over 550 contributors, Ethereum only has 200. Bitcoin has over 16k commits. Ethereum has only 9k. This doesn't even take into account the larger ecosystem of software Bitcoin has.

Ethereum developers are working on all of the Dapps being built on top of Ethereum, each of which has its own set of repositories.

Only comparing the repositories for the Bitcoin and Ethereum official projects, which are centred only around the main clients, is an absolutely ridiculous measure of total development activity with respect to the platform.

And even this comparison is extremely disingenuous. You're comparing ONE repository hosted by the Ethereum project to one repository hosted by the Bitcoin project, when the Ethereum project has its work divided across a far larger number of repositories (151 vs 4):

https://github.com/ethereum

https://github.com/bitcoin/

It seems like every single argument you've made is misleading.

>>He described how to solve it 4 years ago, yet they still can't release even a hybrid PoS solution.

Good development takes time. The fact that the implementation has taken time to develop is not an indictment of the solution provided by VB to the Nothing at Stake problem.

>>Slasher does not solve the nothing at stake problem, it merely obfuscates it.

Ethereum is not developing Slasher, which was a very early proposal that does not contain the solution to the Nothing at Stake problem that VB proposed in the linked article.

At least learn the basic facts before confidently criticizing Ethereum.

>>Lack of Turing Completeness is a feature and intentional.

The lack of Turing Completeness limits the range and sophistication of second layer solutions that can be built on top of a blockchain. Ethereum's range of second layer solutions is far larger than Bitcoin as a result of having it. Turing-Completeness is an incredible benefit, and it's very shortsighted (or malicious) to discounts its value.

>>That's cute. Ethereum doesn't even have any "consumer grade" nodes today.

Sure it does. SSDs are consumer grade hardware, and the recent Geth update allows HDDs to sync as well. Sharding allows massive scaling while still preserving the ability of nodes to run consumer grade hardware. You have no response to this point except your snarky "that's cute".

>>Anyways, sounds like you're taking the long way around to agreeing with me: Sharding sacrifices security.

I explicitly disagreed with you. You can't even respond honestly to my comments at a basic level.

>>Again, taking the long way around to agreeing with me.

I just disagreed with you!

Here I'll past the comment again:

"HDDs have been having syncing problems, but that was solved in the most recent Geth release."

Translation: HDDs work fine now. You do not need an SSD. So I disagreed with you. Please stop lying so much.

>>A full archive node currently requires a 400GB SSD!

A full archive node is not necessary for full validation. Full validation only requires 30 GB, and like I just explained, it's now possible with HDDs as well.

>>A little early to make any of these claims...the only popular Dapp in recent memory is cryptokitties

It is currently the most widely used Dapp platform. That's factually true. And the development happening on new Dapps dwarfs anything happening on any other platform.

>>it brought the Ethereum network to its knees.

Even at its most congested, Ethereum did not get anywhere near as congested as Bitcoin, so you are implicitly arguing that Bitcoin is regularly brought to its knees, to a much more severe degree than Ethereum, and while processing one-third as many transactions.


Have you looked into things like Synapse.ai ? They are building exactly what you're talking about in terms of data/AI/ML/currency/ownership.

Here is their founder giving a presentation about it at the Decentralized AI Summit: https://www.youtube.com/watch?v=9QMgVsbHu98

It ties together things like the "Who owns the future" book by Jaron Lanier and AI/Bots/APIs for new economies.


Thanks for sharing. Reminds me a bit of Numerai


Yeah numerai is great, they were one of the first ones in the financial/trading space.


I am really looking forward to "The Future of Organizations". I think decentralized governance and consensus is one of the most exciting aspects of these technologies, but are being totally drowned out by the fintech applications in the meantime. Hopefully this article will address some of these possibilities.


It will. One of the great companies in the space will write it.

I also recommend this article from Fred Ehrsam https://medium.com/@FEhrsam/blockchain-governance-programmin...


Are ICOs illegal in the USA?


Is working for a crypto, ICO, or even DAICO launched in a different country legal? Is launching one legal?

I've skimmed the SEC documents Reg A, Reg S, Reg D, Reg CF they are all nonsensical when it comes to raising capital with crypto.


It is illegal to sell securities to non-accredited investors


This is not true without qualification.

Non accredited investors buy securities all the time through their brokers.

Advertising to general non accredited public for unregistered securities may get you in trouble with sec.

Selling unregistered securities to non accredited investors happens all the time through private placements.


I think a good argument can be made that a law restricting the sale of non-registered securities to wealthy ("accredited") investors and excluding poor people violates the Equal Protection clause of the 14th Amendment. (And, although the 14th Amendment applies only to the states, the Supreme Court has held in Bolling v. Sharpe (1954) that the same principle operates against the federal government via the Due Process clause of the 5th Amendment.)


I hope and think tokens will democratize access to investments


Do tokens fall under that category ?


I just recently saw the SEC secretary say that ico’s are not illegal if they were to be done roughly behind closed doors, so to speak.

Meaning, don’t spam to the world your ico address and be diligent as to who you’re talking to about potential investments.

The SEC secretary wants a fine line distinction between a “private” and “public” raising of money. He’s right to point that all icos (besides filecoin) are trying to be considered as “private” while publicly share their address and don’t discern who sends in funds.

Here’s a recent SEC release: https://www.sec.gov/news/public-statement/statement-clayton-...


It depends on how they are used, but frequently, yes. Even so, you can sell to accredited investors totally legally. A few ICOs have worked this way.


Is a cryptokitty illegal for non accredited investors to buy? Or is it just a “normal” good.


No it’s just as trivial as a Pokémon card


Cryotpkitty is under the current existing Ethereum network. Like how you can still buy bitcoin. ICO are initial coin offerings, so it's an investment before the actual coin is released and on the market.


Blockchain is a real innovation, sure, but I don't think it will bring evolution to how money is used. There are many other uses.

Not surprising that an investor is seeing it as the future. Investing is regulated and libertarians often see bitcoin as a good thing.

The reality is that bitcoin is volatile, and nobody wants an economy that is built on instability.


> I think the same is true with Bitcoin: traditional banking that takes days and charges high fees can be easily implemented on Bitcoin, but the opposite is not true.

"Easily" here seems like major exaggeration. If that were true, projects like Lightning wouldn't exist.


This was true at different moments in time. Now, as you mentioned you would need to use lightning. You could do it with LTC right now


Once LN is established, I expect that this is going to be opaque for an end user. Example: Today, you don't need to specify http:// in order to use it in a browser.


"nobody can destroy or seize your coins"

"keeping your investment safe is not simple"

Hmm...


One of the great ironies of Bitcoin is that it was marketed as "the internet of money".... except one of the best way to secure it is to avoid exposing it to the internet and by printing its keys out on sheets of paper.


I think this means:

- nobody can destroy or seize your coins unless they have your private keys

- keeping your private keys safe is not simple


Compare with the Business Insider headline

"People are making a fortune buying government-seized bitcoins" http://www.businessinsider.com/bitcoin-price-government-auct...


A government or entity can only seize your coins if they have or find the private keys


As others have commented, this sounds like more power to central authorities, with higher levels of fidelity than ever before.

The first and second point are inherently at odds with each other.

The first point concludes that we need simpler gatekeepers/service providers to make it easier for normal people to use the Asset class

The second point concludes that bitcoin makes it easier to be decentralized.

But I think these sections are the light weight part of the article. I think the other points raised are more material and impactful.


Source 5 is a broken link. It should be:

https://www.barrons.com/articles/the-bitcoin-consensus-yes-i...


Thanks. Fixed


re: "The question is not whether [databases] are going to be hacked or not, the question is when. These are natural honeypots for hackers."

This is not a good argument for cryptocurrency.

Just like a regular wallet attracts pickpockets, a Bitcoin wallet is a natural honeypot for hackers.

Just like a bank attracts bank robbers, a crypto exchange is a natural honeypot for hackers.

Really, the issue here is money. Money attracts thieves. The money is the honey, and the computer is the pot.

So far there's no evidence that cryptocurrency reduces theft, and quite a lot of evidence that it encourages theft.


The difference here is that if I fail to secure my wallet only I loose my money, but if the bank fails to secure their database, all its customers loose their money.


Do you know an example of a bank that lost its customers' money due to being hacked?

On the other hand, plenty of crypto exchanges have been hacked.


A centralized crypto exchange is as centralized as a bank. Both have the same security risks. What cannot get hacked in the same way are decentralized exchanges like RadarRelay


I guess, but looking at RadarRelay, apparently you can't trade anything except Ethereum-based tokens, so that's barely an exchange.


92 of the top 100 tokens and something like 40 of the top 100 crypto assets (tokens and coins), by market capitalization, are Ethereum-based tokens, so that is a substantial potential market.


That's okay as far as it goes but it's very limiting. Everything that really matters in life and nearly all possible investments are not on any blockchain. Even for crypto coins, the most important trades are for things not on a blockchain.


> When Satoshi Nakamoto released the Bitcoin Whitepaper he defined it as a peer to peer electronic cash system.

How do we know Satoshi is a he?


Of course we don't know for sure whether Satoshi is really a man, a woman, or a group of people. But for ease of conversation, people use the gender and number of the pseudonym (masculine, singular).


That is an embarrassingly obvious answer, thank you. Someone had told me Satoshi was a made-up name without a clear gender, and I had never bothered to check that fact.


Because it's a (typically) masculine Japanese name. If the chosen pseudonym was "Steve", would it be unreasonable to assume it's a male?


He put his gender as male on his p2pfoundation profile.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: