Bitcoin Forum

Bitcoin relies on distributing a file with unlimited size to every single node in the network in order to function. Bitcoin therefore would theoretically succeed if the resources required for storing and transmitting such an unlimited file size were also unlimited.

However, its clear that for the foreseeable future, storage is not unlimited, and also bandwidth is not unlimited. The unlimited size of the blockchain; the fact that anybody can add transactions for no cost, eg. SDice. The overhead with distributing an unlimited sized file to every node.

Moore's law is cited in Satoshi's paper as being a fundamental tenet on how the bitcoin system works. Quoting from Satoshi's paper:


However, "Moore's law" is not a law at all. It was just an observation made 40 years ago that the price of storage would decrease for the foreseeable future - that we were entering an era of microcomputing. It cannot hold true forever. There is a hard limit to how much information can be stored on a chip.  Its just we haven't reached it.

Therefore the concept of distributing an unlimited sized file to every node on the network is fundamentally not sustainable or economical - READ: flawed. Storage will not be getting cheaper forever but the principle underlying bitcoin is for the filesize of the blockchain to be getting bigger forever.

Let's not also forget the bandwidth usage requirement as well, which will also be getting bigger forever.

The requirements of porn means the creation of greater storage will never cease.

I wrote in depth about Moore's law this year for English. And while it's true that it was "just an observation 40 years ago," people have done plot studies of exactly how accurate Moore's law has been (read: The Singularity is Near by Ray Kuzweil).
  In addition to the astounding accuracy of Moore's law, the general implications of the axiom hold true for nearly every technology from the past to the future. Much before Moore's law (or even Gordon Moore himself) existed, there were various paradigms of technology (read: vacuum tubes, transistors, etc.). Because of the obvious path for technological innovation, new technologies will arise and continue to follow Moore's law.

In fact, HP already uses platinum latch technology for printer cartridges and is working on moving that technology over to IC's. And at the same time, companies like Intel and IBM are putting a ton of work into carbon nanotubes. The future is literally unfathomable. Following your closed-minded notions, it's easily understandable why people in the 80's thought computers would always be room-sized.

Alright everyone, it's time to quit the experiment.

The OP has discovered something that nobody at any point during the last four years noticed before, and certainly haven't talked about ad nauseam on this forum as well as written wiki pages about.

This problem is clearly insolvable so we should just give up now. I can't believe we missed this the entire time.

Bitcoins biggest flaw is that it depends upon the existence of moors law? I'm ok with that im pretty sure moors law is real.

@OP:

https://bitcointalk.org/index.php?action=search

Parameters:
Search for: Block chain compression
Search in: Development & Technical Discussion (make sure to check all sub-boards)

Next time you find a flaw (Wild guess says the next thing you hit on will be deflation) Try searching for that, before you post.

Also look on here: https://en.bitcoin.it/wiki/Main_Page

You don't actually need a copy of the whole blockchain!

You only need the recent part of it that deals with the bitcoins you own or receiving or storing.

Right now, everyone stores the whole thing because the blockchain is still pretty small.

But as time goes on, it is likely that the average user won't hold the whole blockchain on only large servers or "bank"-like entities will keep the whole thing.

I agree with the flawed part, but not on the reason:
You can simply use a safe ewallet like the blockchain.info.

- Electrum servers
- Pruned clients (Bitcoin Wallet, Multibit...)
- Unspent coins (pruned blockchain)
.
.
.

Bitcoin isn't a flawed technology.
It is just not designed to function as a practical system to build a society on. It was meant as a proof of concept and was hijacked by a weird libertarian cargo cult.

Bitcoin is an implementation of crypto-currency that has shown that it can work. Bitcoin failed to forecast the consequences of wild speculation (or if it forecasted, then Satoshi is a real evil person) and other problems, but crypto-currencies are still a sound idea. In my opinion, unless all those  problems are solved bitcoin will fail, and other not yet existing crypto-currency, designed to solve those problems, will be able to succeed

That word has become meaningless in today's Jerry Springer world. This post-modern patriot plutocracy needs a new technology, flawed or not.

That's why the first non-experimental cryptocurrency which comes after Bitcoin has the greatest potential for long term success.
Essentially Bitcoin aims for the same kind of plutocracy we had in the past, that's why it will ultimately become irrelevant.

That word has become meaningless in today's quantum-relativistic pot-smoking Sagan-worshipping, post-modern neo-scientism world. Everything is an experiment. Nothing is black and white, right and wrong like they used to be in the good old days. One day people will look back at how good we had it with Bitcoin and troll the message boards about it.

Not in reality.
The academic societies of Economics, Computer Science and Mathematics don't really care about our sub-culture here. And the one where it comes to Business and Politics cares even less.

Right. Hijacked.

Ok maybe it was Bitcoin which hijacked libertarianism.  ;)
I mean there is a certain appeal of Anarchism and Cyberpunk for Libertarians out there, but that doesn't make it the same thing. (Even if you guys to claim you are Anarcho-Capitalists you are not)

It's wonderful you hold these societies in such high esteem. Surely they will bring your salvation when the barbarians are darkening your doorway as has happened to every civilization in history. Your politicians and economists have the answers you are looking for.

Either that or humanity will die out.

More likely you freedom lovers are choosing your own plutocrat.

It's always been the freedom-fighters and paradigm-shifting philosphers that have led the revolutions and evolution of every new social system. In fact, they are the ones that pay the highest price to defend humanity against itself. It's your politicians and economists that shy away from progress and lead toward austerity and a more Modest Proposal.

Those freedom fighters didn't hoard tokens to inherit their decedents for them to become the new elite.

Excelent, if this is correct, lets do it!

The whole thing could be made avalible as a download by the offical site and backed up elsewhere, maybe as well as being stored in a way that makes it's size more logarithmic over time.

But clients that are designed to exist all over the place, maybe even web clients like most exchanges, they need only store what they use as they use it.

Are you saying that politicians don't hoard money for their progeny?

I was under the impression that this label "Anarcho-Capitalist" applied to me. If that is infact not the case i would be interested in learning why. Would you be so kind as to explain your position with a bit more detail?

Are you saying they are freedom fighters?

I don't know, I am not really familiar with your views.

To elaborate: Libertarianism puts the concept of ownership and contracts under the hood of "freedoms" which is inherently flawed.
The crux of the matter is that both are subject to power and the use of force, and they cannot exist with out it. Anarchism threats this in an agnostic way while Libertarians are dependent to take them for granted or their ideology falls apart.

Yes. Bitcoiners are non-violent freedom fighters. Bitcoiners don't hoard like politicians, they don't brainwash populations like economists, and they don't build WMD's like scientists. Their battleground is virtual and their ammunition is cognitive dissonance.

I reject your concept of freedom.

also see (https://bitcointalk.org/index.php?topic=181109.msg1895235#msg1895235)

Gotta love when people admit they're speaking from a position of ignorance.

That was meant on a personal level, not meant for the association to Libertarianism or "AnaCap"
Yes, I think I do know your personal views well enough.

Why is this thread under Economics?

Flawed also:

Easy to fake/spam and limited:
- tcp/ip
- dns
- email

Inefficient:
- Ethernet protocol, invented in the 80s, considered as being obsolete a long time ago

Bitcoin is not going to get superseded by any "better" currency (good luck finding all that crypto power btc already has)

As with DNS, email and pretty much any other Internet protocol, services can work on top of what's already out there: dnssec, encrypted email, ip masquerading etc.

RE: BitCloud

I am just thinking about it logically, ignoring any historical precedents. "The past doesnt predict the future." Correct logic predicts the future. If someone invented a new mechanism of storage that was more efficient than magnetic or flash in memory storage, it still wouldnt change the logical premise that there is a hard limit to how much data can be stored on a chip or, lets just say, 'storage device', and therefore bitcoin is theoretically flawed, because the size of the blockchain will never stop growing over time. Of course there may be some 'out of this world' technology that comes along which cannot be even imagined right now, that makes storage unlimited and therefore bitcoin feasible. Maybe a bit like nuclear fusion supposedly makes energy unlimited. That is a big , crazy assumption.

Practically, if the blockchain keeps growing at the current rate, i think we're going to see some real problems with scalability of bitcoin in the much more closer future - before flash or magnetic storage is superseeded - i'm talking about in the next 6 months.

The tubes! They will be clogged by the blockchain!

Bitcoin isn't a standard nor a technology. It's a concrete implementation which was meant as a proof of concept.
This would be like if we were using cern httpd.

Public service ; User "Xavier" was on ignore list prior to this post.

ah yea deontological libertarins can be a little bit silly at times. To me property is nothing more than a useful construct that we as a society use to avoid conflict. You know since 2 people can not both eat the same apple, we can either chose to fight over it or we can establish some sort of property norm. To me the state is the former and i personally prefer the latter.

That is an absolutely absurd time line for the average user to run out of space for the block chain. I understand that you firmly believe the block chain growth will out pace the produced storage space (theoretically), but there already exists enough storage space on the average computer to hold the block chain for years to come (and you can quote me on that). In addition to that, storage will continue to increase at an increasing rate (read: exponential growth) through this paradigm and onto the next (it's true that the past doesn't predict the future, but it gives a damn good idea). Once a simple method for truncating the block chain has been distributed, the storage argument won't even be valid anymore.

They can both eat half an apple.

And there is force involved every time, alone the task of grasping the apple is a subject to potential conflict.

Not if they both need a whole apple to survive.

Then there is no other possible solution other than (deadly!) force.

Of course in nature there is never just one apple...

Back to topic, guys?  ???

Turns out Moore's Law is both correct and incorrect.

But in a way it's not so favorable to Bitcoin.

Single core processor speed as well as hard drive sizes have not significantly improved through the recent years.

That's because the desktop PC is dying. Everything goes smaller and mobile.

And these mobile devices are getting better and better. And the processors and storage media smaller and smaller.

It's only in this respect now that Moore's Law still holds true.

Either way, it's Satoshi himself who mentions that mining and blockchain storage will migrate to specialized data centers. And I'm afraid these specialists will turn out to be those corporations who don't care much if their servers cost $100,000 or a million.

That isn't really true afik.

Even so hard drive capacity grows faster than moores law. It is just that it has become unfeasible to use hard disks in the way they were used before.

In the past it was possible to re-write and re-read hard disk many times in their lifetime. But since transfer speeds are increasing one exponent below that, mostly linear this becomes a bottleneck.
The ultimate death of the hard drives will not be that they can't be produced with higher capacity but that they can't be filled and read back two times before they become unusable.

SSD memory grows even faster still, but it too has some issues in that regard. It grows to the power of three where bandwidth grows to the power of two, not as bad as the 2:1 relationship with hard drives, but still.
The other issue is that the rapid growth in capacity is facilitated by storing more information in a single capacitor which makes them less reliable.
All in all increasing capacity should outperform the growth of the blockchain for quite some time. And after that I doubt that bitcoin will still be relevant. It should still be around but exponential growth should be over.

And what sense does make a limited amount cryptocurrency make for sustained exponential growth? None, it has to either hit a ceiling or fade away.

Have you heard of a company called Intel? Yeah, they not only follow Moore's law (since Gordon Moore was a co-founder), but their products have shown upward movement on a log scale in terms of transistors per chip, size of transistors, power consumption (decreasing, obviously). Not to mention that every time they have released a product since the days of Core 2 Duo, there has ALWAYS been a 5-10% increase in single threaded performance. Please do a little research before spouting lies.

Any simple method will not be able to stop the problem, that the size of the blockchain will increase exponentially over time. It will just be able to optimize it, or to provide some temporary relief. The fundamental design problem with bitcoin will still hold true, its in the design of the system.



Also, if large corporations were to come in to fill the void when the blockchain model becomes unviable..... that will result in bitcoin loosing its USP and becoming just like any other previous currency out there that required a central body to operate.

The blockchain right now is about 5GB if I understand it? If bitcoin continues to experience exponential growth, providing trnasaction limit is lifted,  my intuition tells me it wont be long until storage & bandwidth requirements exceed whats currently available. Understand VISA processes 11,000 tx per second. I will do the math later tonight.

read this https://en.bitcoin.it/wiki/Scalability

we can reasonably expect it to scale to about the size of visa. If we really need more transactions than this than we can use alternative block-chains to avoid centralization in exchange for undesirable heterogeneity. I consider that a worthy trade off. This all assumes that someone doesn't find a decentralized way to scale the first blockchain up.

Only full nodes have to store the whole blockchain.. and for storage, I remember arround 16 years ago, I had to pay hundred of dollars for 8 Mb of RAM, last week I bought 8 Gb for less than 50 buck.. this is 2000 times cheaper than 16 years ago.. so, what will worth a 8 Tb memory chip in 15 years, I dont car if I have to buy few of the to stay a full node, I will for sure !

I just disagree on the OP point.. and many other option are possible in the software sphere !  No panic at all here.. Sustainable IMO !

Isn't capitalism great?

ElectricMucus has stuck it to yall so effectively I've decided to take him off ignore.

So ... SSD and bandwidth grow slower than bitcoin transactions? Or quicker?

To all those people talking about pruning or any such optimizations: what we are considering here is growth rate not file size so please stop talking about pruning. My arguement is that the blockchain will grow faster than bandwidth / storage, making the P2P model invalid.

Only people with valid arguements about growth should continue to post in this thread.

How has this managed to slpi past us???? gee whizz I'm glad you spotted that... doh! As opposed to the other flawless options that I missed before??  ::)

Of course it has flaws, everything does, even your head no matter beautiful or perfect it is. The only difference is with this flawed system is Im not getting taxed to death everytime I fart out a penny, any assosiated fees and charges that go with using any currency day to day are no where nearly outrageous scandalous as the the ones the bankers pirates and merchants thieves charge you on your human money.

Besides, BTC is more of a hobby that a currency. Its just so the community can jump on the rollorcoaster exch/rate and play pretend economy with having someone scold us if we have to much fun.

So stop being such a grouch, turn that frown upside down and jump on the band wagon with the rest of us!! :)

Data May be stored on your skin in the future. They are finding DNA has far better storage than any current technology.

https://theconversation.com/dna-data-storage-100-million-hours-of-hd-video-in-every-cup-11777

http://earthsky.org/human-world/scientists-successfully-store-data-in-dna

I wouldn't be trying to talk down bitcoin on the basis of technology not moving forward fast enough. We are heading into a totally different age with the way we do things.   

Your argument is just plain wrong though.
 
There is absolutely no way bitcoin will out pace the already exponential growth of technology and storage (and definitely not in 6 months).

What you fail to realize is that storage and technology have a 30 year head start. Try graphing  two exponential functions, and give one of those functions an exponent of (t+30).
No matter what you do (within reason, because we know the relative growth rates of the block chain), you WILL NOT get those graphs to converge within a reasonable timeframe.

Look, I understand there is a theoretical flaw in bitcoin. I won't argue that. But realize that we don't live in theory land where friction doesn't exist and physics all works perfectly. We live in the real world where theoretical problems don't become real problems until after the human species goes extinct.

The last time I checked it was lower, but that was a few months ago, before the rise in popularity in satoshi dice.
What really bugs me though is that storing the blockchain isn't rewarded in any way, while a hard drive costs about the same as a gpu.

All in all I do think that eventually Bitcoin will be superseeded by another cryptocurrency, not a fork like those altchains but a re-implementation from scratch,

It isn't drive or bandwidth size that is relevant, but available excess capacity.

If bandwidth doubles, but bandwidth demand quadruples, not good.

It just takes a few jokers like SDice to start adding transactions like crazy..... and things can get bad.

I agree- there needs to be some fee for storage.

For this reason anyway, I continue to think that bitcoin's scalability is an issue.

Well, transactions grow with usage.  You could presume that transactions ramp up in identical fashion to paypal over similar numbers of years.  Or presume that they ramp up to "all humans on earth" and some fraction of their transaction volume.  Let's take the latter case, as it is the limit, that after that volume is reached, then current transaction volume is limited by population size.

But storage capacity, processing speed and networks will continue to grow by compounding, ie, by an exponential factor.  You can argue that might be a weak or a strong exponential.

Thus in the relative long term, storage certainly does outpace transaction growth.  Could there be a momentary hiccup or two as a result of sudden surge in usage?  Yeah, but so what.

hahaha, u beat me to it

no, because bitcoin technology is profitable when used in a good way, almost the biggest country uses bitcoin, and many bitcoin users are almost all over the world, if bitcoin technology is wrong why bitcoin is made, because bitcoin is a lucrative job if you can use it.

I would agree.  blockchain is still growing and improving over time, and bitcoin is not flawed technology like mentioned about blockchain its growing and improving slowly but surely, it has its downs but every cryptocurrency does

Bitcoin is not and can never be a flawed technology, from the qoute it is very clear that he (Satoshi) made his estimate with the worst case scenario and the system was still going to function perfectly with the worst case scenario of systems shipping with 2GB ram. This just goes to show that the technology behind BTC isn't flawed and would remain around for many many years to come.