The inefficiencies in the Bitcoin protocol are pretty big by comparison; every full node has to maintain a copy of the entire blockchain, the proof of work scheme is terribly redundant and wasteful. Some partial solutions (peercoin, ripple/stellar) have arisen in the crypto space but they just don't have the popularity of Bitcoin, it'll be interesting to see where the field stands in even five years. The real world demand for decentralized digital currency is still questionable in my mind.
This is a key point. I believe that general people do not care about decentralized currency, what they want are easy payment solutions. If bitcoin doesn't provide easy payment solutions that are cheap, convenient, works everywhere, then it will be beat by something that does provide that.
Even if you could with huge engineering effort slowly move Bitcoin to cheap, easy, everywhere, it is likely going to be beat by non-decentralized solutions because those non-decentralized solutions are cheaper to create, deploy and maintain -- also easier to monetize.
I think that the future is to companies like Stripe, ApplePay -- technologies which focus on customer capture via ease of use and many options and being agnostic towards the underlying technologies.
Bitcoin is neat, just like Gnutella was, but being neat doesn't correlate with commercial success.
The ultimate reason why I have a bank account is so that my employers can deposit their payroll into it. And since my money is already there, I pay my bills and expenses from it.
I fully expect my bank to narc on me to the feds at every turn, and screw me hard if I ever have a real problem with them. I expect them to nickel-and-dime me with fees. I expect them to introduce seemingly arbitrary delays and paperwork requirements for their own convenience. I expect them to offer rates on deposits that cannot keep up with inflation and charge rates on loans and credit that would make even the serenest of divine avatars flip tables. And if the turds ever hit the turbines, I expect that making a trip to the nearest bank branch office would be the least useful thing I could possibly do.
I don't really have a lot of confidence in Bitcoin, but the bar they need to pass with respect to trust is ridiculously low--set there by the banking system. Where they fail is definitely convenience.
If my employer offered to make payroll in Bitcoin, and if my local utilities, gas stations, and grocery stores accepted payment in Bitcoin, I would be using it already.
But all those places prefer to trust human bankers rather than crypto algorithms running in a distributed computing network. I don't see that changing any time soon. If anything, an alt-currency backed by an alliance of megaretailers like Walmart, Target, and Amazon would have to crack that nut first. To drive adoption, there has to be support at the point-of-sale terminals.
And if it is illegal just because the actual money is somewhat anonymous (which Bitcoin isn't, really) doesn't mean that the change in material wealth is unnoticeable.
That's just silly. You can certainly think of things people have a legal right to do that they still wouldn't want any record of. If you're paying for an abortion it's perfectly reasonable to want the payment to be truly anonymous. The same goes for other types of medical treatment, any purchases of an embarrassing nature, donations to politically charged organizations like Planned Parenthood, charity by people who believe altruism is only satisfied when the donation absolutely cannot be attributed, etc. There is also the simple metadata concern that giving anyone a record of everything ever purchased by you (and everyone else) could be used against you or for evil purposes generally.
The ability to make anonymous payments has been the norm since money has existed. Cash is anonymous. It's only the last few decades with the rise of electronic payments that tracking them all has even been possible, and there is no law of the universe that says we have to accept the loss of privacy.
"BTC" is just the unit of account powering the blockchain. You can use a blockchain to make almost any form of digitized information into a public good. Case in point: https://www.blockscan.com/asset
The blockchain is the ultimate distributed database. No other distributed database works without a third party controlling and administering it. This is critically important for all kinds of financial markets, because otherwise the entity in control of the DB is able to manipulate outcomes, hide data, and play games with insider information, and yes this is really happening: http://deepcapture.com
And why does a purely online-only, digital content provider need your home address to send you a video file, anyway? What about a server host? Gaming platform? The credit card paradigm is obsolete for a huge amount of premium digital content. This isn't to say the US dollar is obsolete, because you can represent dollars, or euros, or gold, whatever, as auxiliary currency units on top of the Bitcoin blockchain, although you'd still need to own BTC to send it around, for many this is BTC's primary form of inherent value.
Finally, it's impossible for many people in developing nations to engage in online commerce at all, because their banking system is utter crap. Billions of people the world over are underserved by the banking system, or don't have access to a bank account.
I think most people would welcome a shift to more anonymous transactions, but they're not willing to give up the advantages they get from tracking. When my purchase is tracked by a the seller, I get volume and loyalty discounts and easy returns. When my purchase is tracked by a credit card company, I get the ability to dispute payments and better manage cash flow. Most people and businesses aren't willing to give up those conveniences for anonymity.
My feeling is that the interesting thing about Bitcoin as a currency is the pressure it puts on centralized payment systems to compete with it. The problem with banking is that it's so hard to innovate, primarily because of the regulatory burden both making it very expensive for anyone to enter the market and severely constraining what can be improved if they do.
For example, with credit cards the burden of fraud is put on the credit card company. That turns out to be one of their major costs, even though the credit card company is hardly the lowest cost avoider for credit card fraud (they're just a convenient deep pocket), which requires them to charge high transaction fees to cover the liability. Hence no micropayments; fees too high.
Now suppose that Bitcoin disabuses people of the notion that the payment processor should be the one who takes the hit for fraud. If you don't secure your secrets then you lose your money; you can buy insurance if you like. Suppose regulations get passed to allow this for Bitcoin. Now non-Bitcoin payment processors can make the argument that they should be able to offer a centralized service with the same rules, and we can finally have micropayments.
>(Netflix et al) are far more popular and commercially successful for watching online video
File sharing versions are not aimed at being commercially successful so i think that is an irrelevant comparison.
Also you I expect are from the USA, netflix/hulu etc arenot available everywhere so they are certainly not more popular than the alternatives in countries where they are not avaiable.
Popcorn Time was a free, decentralized Netflix alternative and it was hugely popular, only that the developers pulled the plug on it did it not become the defacto method of watching movies for free online.
Anyway not sure how these compare to bitcoin but I dont think your analogy holds up even when not compared to bitcoin.
When there are so many similar versions which one do you choose.
Also you probably don't want it to be popular anyway unless you don't mind the authorities knocking on your door.
I think the market will always decide which media consumption method it wants to use depending on how frictionless the experience is.
In the case of Bittorrent, as a mainstream consumer you'd have to know what program to download, where to find a torrent file yada, yada yada, while on Netflix you can just enter your credit card information and watch as many movies as you like.
I think recently, Facebook also realized this, which is why they are currently aiming towards bringing a great search engine to their customers.
Once that is in place sufficiently, homepages for small businesses will become obsolete, as people rather search for a company/product on Facebook than on the internet as it is a much easier to do.
This same principle again will probably apply to Bitcoin and digital payments in general. Right now, Apple is doing a great job reducing friction in payments, at least for mainstream consumers.
While this may or may not be true, am I the only one who is slightly frightened by the ramifications of this?
Facebook is fine, for what it is, but the idea that it could soon become a proxy for the WWW isn't my idea of optimum.
But I think the only reason Netflix doesn't use BT is that the content owners are super-skittish about people having a copy of the data on their computers instead of streaming it and throwing it away. So even if BT is a better technical solution, it's politically unworkable.
It indicates that BT didn't provide something that Netflix et al did require. I imagine that is something like maintaining control of all digital copies of a work as you went on to suggest. This is not a technical limitation but rather a business need. So BT may be technically better (I imagine in terms of resources required to provide content it is superior) however it didn't meet the business case as well as a centralised solution did hence BT doesn't get used.
I would disagree that the reason people use Netflix instead of bittorrent is not because Netflix is "less work" or that "there are costs to be paid by being decentralized". Your argument above is ignoring the legal issues with transmitting copyrighted data that deters ordinary citizens concerned about following the law. I would actually say that Netflix is more work than bittorret, espcially compared to the bittorrent clients nowadays that can stream bitorrented video. Plus in the past netflix always required a bloated plugin (silverlight).
What about skype, that was built upon supernodes, and was decentralised?
Skype accounts were centrally registered and a single company was responsible for network health, security, etc.
The Gnutella with super hubs was called G2 protocol. Gnutella also utilized GWebCaches for retrieving bootstrap information of other active nodes.
[1] https://www.wto.org/english/Tratop_e/dispu_e/cases_e/ds285_e...
But consider for a moment the popularity of Popcorn time or torrents in general. They're massively popular despite the fact everyone knows it's illegal and knows many countries have histories of punishing illegal downloaders, let alone the big uploaders. And it's not just that they're popular, they work really, really well for something completely free.
Does that mean we prefer using commercial solutions most of the time? It depends. Torrenting tech is monetized through ads but kept free, just like things like gmail or facebook. The difference is that gmail isn't illegal and thus can get venture capital investments, partnerships, build elaborate products (like gmail for business) that interface with online identities (like businesses, phone security) and infrastructures (banking & payment processing infrastructure), have customer service etc. While most torrenting services have to operate in the dark, anonymously, outside of infrastructure, with few partnerships, little investment etc.
In short, I don't think we should try to reason from analogy here. Centralized/decentralized doesn't inherently predict anything.
I think at the end of the day we have to look at the 'per person' scale. The average US citizen makes 2 transactions a day (one of the metrics I saw used in this slideshow, too.) The average bitcoin transaction is 500 bytes. For a global citizen it's likely slightly less than 2 on average considering the US one of the most consumerist societies, but let's say 2 for everyone.
Alright so 1 kilobyte of data per day for everyone in the world.
Is that a big deal? No. It's not. It's still 14 TB of data, daily, but it's manageable.
It requires an 82 mb/s connection to handle that much data. That's for the entire world populations (1 month old babies included!) to average as many transactions as the average US citizen.
Just put that into perspective. We're not talking about this kind of technology today, bitcoin has only a few million users. It's unlikely bitcoin will ever become popular enough to service 7 billion people in the first place, the internet's been around for decades and it's not even close, hell electricity or clean water doesn't even do that. But hypothetically if it happened, we'd be talking about some date that's at LEAST 20 years out from now to get to 7 billion and that in itself is already extremely optimistic. For perspective, it took the internet nearly 2 decades to get to 2 billion internet users.
So to say a 82 mb/s internet connection in 2035 is too big of a deal for what is likely going to be a node run by a professional business like Coinbase. (i.e. I believe bitcoin's infrastructure will be run more and more by professional businesses rather than home desktop PCs, but these businesses will exist in every country on the planet and be decentralized a bit like how ISPs are decentralized only much more because running a node isn't nearly as big a deal obviously, as it's one of many things running on the ISP's network).
Average home internet is about 5 megabytes per second today in OECD countries. And we know bandwidth, like CPU and storage, has grown roughly 50-60% in the past decades and it's likely to average something close to that (like 40-50%) in the next few decades. 50% yoy would be 3300x in 20 years. 5 MB/S would be 15 gb/s by 2035. Optimistic? Absolutely, it's the most optimistic metric based on past data. But let's just say it's extremely likely that reaching 1 gigabyte per second will easily happen by the time a bitcoin node needs 0.082 gigabytes per second. And that's for a standard, average, consumer, home connection. NOT the much better industrial connections that you expect bitcoin nodes will be hooked into by 2035 (if it indeed is so important that 7 billion people use it for 2 transactions a day).
In short, 1 kilobyte of data per person is so little that whether it's centralized or decentralized it doesn't really matter that much. It can be dealt with. Especially when you consider that many of these transactions can happen off-chain, on sidechains, treechains, less secure chains (e.g. to buy coffee or make a 1 cent payment to read a New York Times article automatically instead of being served a 10 second advertisement with a 1 cent revenue per view value, these transactions can happen on less secure channels).
Decentralized/centralized is an interesting discussion, sure, but at the 1 kilobyte per day per person scale, I don't think it's very relevant when it comes to technical scalability. It's relevant for other reasons definitely (like a decentralized digital global banking system being much less prone to anti-competitive banking monopolies and licensing than say a centralized one like Paypal) but I don't think technical scalability is the biggest problem.
Beyond that bitcoin is so nascent and small... I mean I quite like it and think it just might win, but I still feel it's a bit silly to talk about whether bitcoin can scale to a billion users all the time, when its growth rates are really not all that impressive. I mean don't get me wrong, scalability is important to talk about, but it commands such a disproportionate amount of talk. It's like wondering if 50% of the Mars debate was about whether Mars can support a city of a million humans, let's see if we can get 20 people living there first you know?
First there was Gnutella (as Napster had a centralized directory) - but it didn't scale, although it proved popular.
Then they added support to Gnutella for spoke/hub models, which had superpeers that reduced the load on most individuals. But in the end Gnutella never really took off.
What replaced Gnutella was bittorrent, which actually only did transfer via P2P but the searching and discovery was centralized again PirateBay and Trackers.
But for all bittorrent's success, Netflix (and others), which are centralized and highly controlled alternatives, are far more popular and commercially successful for watching online video -- it is just less work to use Netflix, even though it doesn't have the benefits of being free and open.
So for all the attraction of P2P decentralization, often centralized, commercial focused easy solutions win out. There are costs to be paid by being decentralized and often general people who are not ideological do not want to pay these costs.
But I am mostly an outsider with regards to Bitcoin so who knows.