The thing that will destroy Bitcoin.

[BADecker]

The thing that will destroy Bitcoins is simple. It will be the bloated size of the blockchain. Soon the blockchain will be so large, that we will need centralized "housing" to hold it. We will have to trust "Central" to state what is in the blockchain every time we want to use Bitcoin. And, this is the thing we are trying to get away from - centralization.

The blockchain is almost 70 gigabytes large right now. Imagine what it would be like if people around the world traded something like the Forex on it.

Someone better figure out a way to make it a whole lot smaller, and fast. Why fast? Because when fiat collapses, a lot of people will jump into Bitcoin. The blockchain will surge in size. In a short time it will be too large to use.

One solution that I can see would be to keep the current blockchain in total, on several servers around the world, including any person who thinks he has enough hard drive space to store it on, and enough bandwidth to handle all the transactions that hit Bitcoin at once. At the same time, implement a new kind of blockchain that keeps only a record of the first use of a new address, and, say, the last two uses of it. Everything else is deleted. This will be a reasonable solution, I think, until a better kind of blockchain can be developed.

What does anybody think about this?

What ideas do you have for overcoming this problem?




Okay. From the comments we have more points. So far we have...

1. Blockchain size for storage and downloading;

2. Number of transactions causes impractical updating of the blockchain because of bandwidth limitations;

3. Impractical confirmation time without a fee;

4. Unreasonably long confirmation time even with a fee;

5, (more?).

[aarons6]

The thing that will destroy Bitcoins is simple. It will be the bloated size of the blockchain. Soon the blockchain will be so large, that we will need centralized "housing" to hold it. We will have to trust "Central" to state what is in the blockchain every time we want to use Bitcoin. And, this is the thing we are trying to get away from - centralization.

The blockchain is almost 70 gigabytes large right now. Imagine what it would be like if people around the world traded something like the Forex on it.

Someone better figure out a way to make it a whole lot smaller, and fast. Why fast? Because when fiat collapses, a lot of people will jump into Bitcoin. The blockchain will surge in size. In a short time it will be too large to use.

One solution that I can see would be to keep the current blockchain in total, on several servers around the world, including any person who thinks he has enough hard drive space to store it on, and enough bandwidth to handle all the transactions that hit Bitcoin at once. At the same time, implement a new kind of blockchain that keeps only a record of the first use of a new address, and, say, the last two uses of it. Everything else is deleted. This will be a reasonable solution, I think, until a better kind of blockchain can be developed.

What does anybody think about this?

What ideas do you have for overcoming this problem?

i think before that people are going to get tired of waiting days for transaction confirmations..

as it is now, if i want to sell something for bitcoin, i have to wait HOURS before it confirms.. unless you tack on a stupid high fee.. which nobody does..

[Hugroll]

The thing that will destroy Bitcoins is simple. It will be the bloated size of the blockchain. Soon the blockchain will be so large, that we will need centralized "housing" to hold it. We will have to trust "Central" to state what is in the blockchain every time we want to use Bitcoin. And, this is the thing we are trying to get away from - centralization.

The blockchain is almost 70 gigabytes large right now. Imagine what it would be like if people around the world traded something like the Forex on it

i think before that people are going to get tired of waiting days for transaction confirmations..

as it is now, if i want to sell something for bitcoin, i have to wait HOURS before it confirms.. unless you tack on a stupid high fee.. which nobody does..

i totally agree, whenever i send a transaction it takes around an hour for it to confirm, and it gets tedious when you have to stay in contact with your buyer for an hour to make sure it confirms.

[ranochigo]

-snip-

Okay. From the comments we have more points. So far we have...

1. Blockchain size for storage and downloading;

2. Number of transactions causes impractical updating of the blockchain because of bandwidth limitations;

3. Impractical confirmation time without a fee;

4. Unreasonably long confirmation time even with a fee;

5, (more?).

1. Blockchain size bloat isn't so much of an issue for storage space any longer. Pruning would be fully enabled by Bitcoin Core 0.12 and there isn't too much of an security risk of average users not having the full blockchain. Bandwidth is however, a problem though. With storage space and bandwidth becoming cheaper, there is little worry for this.
2. I don't get what you mean. If you meant that the blockchain is not able to handle a high TPS, a block size increase could eliminate this issue.
3. Bitcoin fees are relatively low. It makes no sense to not pay a fee.
4. It's a problem with varience, there is no fix about it.

[franky1]

blocksize bloat is not an issue.. raising it to 2mb is not a problem. if you think its a problem then you better also call Netflix and tell them that their 30million customers will need to be datacenters if they want to currently stream 500mb every 10 minutes watching HD movies. (oh wait its alread 500mb per 10 minutes HD shows.. and users are not datacenters..)

a 2terrebyte hard drive is just $100, even at 2mb blocks and the naive belief that every block is filled to 1.99mb always.. thats only 104gb a year, or the equivelent of 35 netflix HD tv shows. (about a weeks worth of viewing time)

which 104gb allows for 20 years of growth.

in four years (average timescale people upgrade their computers) maybe a 4tb hard drive will be $100 maybe 8tb.. this allows people in general to happily accept more data without care.
and internet speed.. well they are faster now than 4 years ago and are going to be faster again in the next 4 years.. so still not a problem
afterall we are not asking for the moon (20mb blocks (1tb per year) we are just asking for 2mb blocks and then a rise in the future..

also when calling netflix (if you disagree with me) please complain to activision because 'Call of Duty' is many GB per release. and no one plays it because of it.. (oh wait they do).

now onto what will distrupt bitcoin.
greedy miners ignoring transactions and using the blocklimit to extort people into paying higher fee's in some first class vs working class snobby debate that forces bitcoin into corporation control rather than easy utility for anyone.

transaction fee's should not be used as a bargaining tool for atleast a couple decades. the blockreward is sufficient to cover miner costs for that time, and the tx fee should only be used as a small disincentive for naferious actors to not want to spam the blockchain.

in short if transaction fee's go above 10cents each, regular people will lose interest. so to avoid that we need the 2mb so that greedy miners cant use the 1mb as an excuse to(analogy) play the first class ticket premium game by saying theres not enough seats on the train for everyone.
causing people to wait for hours. 10 minutes is fine, everyone expects to wait 0minutes per train.. but being told to not get on the next train, and then seeing 5 trains go by, all with snobby rich people onboard. will make regular people try something different

[ebliever]

I would also mention that altcoins are experimenting with a range of different approaches to blockchain bloat. If even one of them pans out, the software protocol of bitcoin can almost certainly be updated to implement the fix. (I would think sidechains would be a good solution to bloat as well.) People who declare that bitcoin is sure to die should first make sure their expected cause of death really is something intractable and unsolvable by changing the software. Bitcoin is not a fixed entity.

[NorrisK]

-snip-

Okay. From the comments we have more points. So far we have...

1. Blockchain size for storage and downloading;

2. Number of transactions causes impractical updating of the blockchain because of bandwidth limitations;

3. Impractical confirmation time without a fee;

4. Unreasonably long confirmation time even with a fee;

5, (more?).

1. Blockchain size bloat isn't so much of an issue for storage space any longer. Pruning would be fully enabled by Bitcoin Core 0.12 and there isn't too much of an security risk of average users not having the full blockchain. Bandwidth is however, a problem though. With storage space and bandwidth becoming cheaper, there is little worry for this.
2. I don't get what you mean. If you meant that the blockchain is not able to handle a high TPS, a block size increase could eliminate this issue.
3. Bitcoin fees are relatively low. It makes no sense to not pay a fee.
4. It's a problem with varience, there is no fix about it.

Bitcoin fees are low at the moment. If the price keeps rising and miners keep making empty blocks, the mining fee will rise. For some people the fees may really become a turnoff. Especially because they don't normally see the fees in their daily transactions. In bitcoin you are constantly reminded of the fees with every transaction.

1 and 2 are really not an issue. Storage space is growing faster than a blockchain could ever do, and bandwidth is improving around the world.

[lumeire]

I think bitcoin as it is is enough already, whats gonna destroy it is probably the lack of trust in the ecosystem, like the neverending hackings and even scams.

[owm123]

Nope it want. Storage is very cheap. Also, there are already talks about pruning blockchain.

[madjules007]

blocksize bloat is not an issue.. raising it to 2mb is not a problem. if you think its a problem then you better also call Netflix and tell them that their 30million customers will need to be datacenters if they want to currently stream 500mb every 10 minutes watching HD movies. (oh wait its alread 500mb per 10 minutes HD shows.. and users are not datacenters..)

Uh, so you're comparing a single download stream...to the upload bandwidth required for a full node to reach dozens (or ideally hundreds) of connections on a constant basis. That's not how it works.

In my case, my only options are (actually were) shitty capped cable or DSL. After a dispute with a my cable provider, I am limited to DSL. Running a node on a regular basis is now completely out of the question for me. I turn on my node and sync up only to validate my own transactions. In other words, I can only afford (and I don't mean financially) to be a leech on the system.

We're limited by internet infrastructure. And this basic misunderstanding of download vs. upload bandwidth (not to mention the level of redundancy that block propagation requires) isn't helping the discussion.

[crazyivan]

I also do not think blockchain size is an issue. Storage devices grow in capacity, year after year. Take a look what we considered a medium sized HDD two years ago and what we consider it to be now.

This block capacity issue is what worries me more. FFS, they cannot agree on it for 6 months.

[MedaR]

5.
Constant rise of hash rate and halving of block can influence demise of bitcoin in future..I just can't imagine bitcoin have price of 100k per btc. One day bitcoin will come to his boundaries , hashrate will stay the same or will fall and that can be very dangerous! Fall of hash rate can cause dumps and instability.

[Jstalk]

Maybe if bitcoin becomes reversible payment method like paypal,buyers will have more trust and will use it more often.Now it`s hard to buy stuff with btc and exchange btc for other currency because of the lack ot trust and many scam attempts.

[Patatas]

Simply deleting the previous transactions of the address and keeping only the two of the most recent won't be a suitable solution for the problem.In fact,this destroys the entire meaning of decentralization.Agree there is short of storage space.I think , like miners any other hardware should be introduced which comes along with mining equipment to store the address transactions of the blocks processed by that miner.This way,we can expect more storage space and it will be mandatory for all the miners.

[franky1]

Uh, so you're comparing a single download stream...to the upload bandwidth required for a full node to reach dozens (or ideally hundreds) of connections on a constant basis. That's not how it works.

In my case, my only options are (actually were) shitty capped cable or DSL. After a dispute with a my cable provider, I am limited to DSL. Running a node on a regular basis is now completely out of the question for me. I turn on my node and sync up only to validate my own transactions. In other words, I can only afford (and I don't mean financially) to be a leech on the system.

We're limited by internet infrastructure. And this basic misunderstanding of download vs. upload bandwidth (not to mention the level of redundancy that block propagation requires) isn't helping the discussion.

well for you, segwit is not an option either then.. as although they tell you its 1mb.. the real data is 2mb.. (if you want to be a full node).

but most people are able to get fibre.. so although you may have had an argument with your cable provider. i dont think that should be a good example of why bitcoin shouldnt grow.
maybe you can call your cable company back and reconcile. rather then trying to strangle bitcoin growth for everyone

[franky1]

Simply deleting the previous transactions of the address and keeping only the two of the most recent won't be a suitable solution for the problem.In fact,this destroys the entire meaning of decentralization.Agree there is short of storage space.I think , like miners any other hardware should be introduced which comes along with mining equipment to store the address transactions of the blocks processed by that miner.This way,we can expect more storage space and it will be mandatory for all the miners.

storage is not an issue.
the mining pools just buys a $100 2terrebyte hard drive..1 per pool, not per asic..

the pool stores the transaction data after all.. and the mining ASICs are just hashing small bits of data. of temporary data that they forget about after the average 10 minutes to work on new data.. (unscrew a asic, there is no hard drive) only the pool needs a $100 hard drive.. not each asic, so relax

[aarons6]

Simply deleting the previous transactions of the address and keeping only the two of the most recent won't be a suitable solution for the problem.In fact,this destroys the entire meaning of decentralization.Agree there is short of storage space.I think , like miners any other hardware should be introduced which comes along with mining equipment to store the address transactions of the blocks processed by that miner.This way,we can expect more storage space and it will be mandatory for all the miners.

storage is not an issue.
the mining pools just buys a $100 2terrebyte hard drive..1 per pool, not per asic..

the pool stores the transaction data after all.. and the mining ASICs are just hashing small bits of data. of temporary data that they forget about after the average 10 minutes to work on new data.. (unscrew a asic, there is no hard drive) only the pool needs a $100 hard drive.. not each asic, so relax

i get your debate.. but.. server hard drives are MUCH more expensive.. plus any decent pool would have more then one.. for backup, in case one fails..

bigger blocks would help confirmation times.. as most of the blocks will be able to hold ALL of the transactions.. but variance will still play a huge factor in how long you would have to wait to get confirmed.

[madjules007]

Uh, so you're comparing a single download stream...to the upload bandwidth required for a full node to reach dozens (or ideally hundreds) of connections on a constant basis. That's not how it works.

In my case, my only options are (actually were) shitty capped cable or DSL. After a dispute with a my cable provider, I am limited to DSL. Running a node on a regular basis is now completely out of the question for me. I turn on my node and sync up only to validate my own transactions. In other words, I can only afford (and I don't mean financially) to be a leech on the system.

We're limited by internet infrastructure. And this basic misunderstanding of download vs. upload bandwidth (not to mention the level of redundancy that block propagation requires) isn't helping the discussion.

well for you, segwit is not an option either then.. as although they tell you its 1mb.. the real data is 2mb.. (if you want to be a full node).

but most people are able to get fibre.. so although you may have had an argument with your cable provider. i dont think that should be a good example of why bitcoin shouldnt grow.
maybe you can call your cable company back and reconcile. rather then trying to strangle bitcoin growth for everyone

No, it's capped anyway. So I could never run a full time node. I'm in a major US city by the way. Not the middle of nowhere.

Do you have any evidence that "most people are able to get fiber"? According to broadbandnow.com/Fiber, 25% of the US has such coverage. And I'm willing to bet that it significantly skewed towards dense urban areas due to the prohibitive cost of building the infrastructure. (I don't think that we should limit bitcoin to those in major cities, for instance)

I'm saying the requirements to run a node are already prohibitive. Pushing for 2MB when all you can say is "2MB is fine, derp" is not really a good argument.

[Amph]

it's actually 62gb only, and anyway based on the moor law, we have no problem handling it for the future

also 62 giga is very tiny compared to the enormous 6T and beyond, that we are currently able to purchase on the market for cheap

[franky1]

No, it's capped anyway. So I could never run a full time node. I'm in a major US city by the way. Not the middle of nowhere.

Do you have any evidence that "most people are able to get fiber"? According to broadbandnow.com/Fiber, 25% of the US has such coverage. And I'm willing to bet that it significantly skewed towards dense urban areas due to the prohibitive cost of building the infrastructure. (I don't think that we should limit bitcoin to those in major cities, for instance)

I'm saying the requirements to run a node are already prohibitive. Pushing for 2MB when all you can say is "2MB is fine, derp" is not really a good argument.

so essentially your saying that out of 1 million people in bitcoin. the reason for only 5000 full nodes is due to a possibility that only 0.1% of the bitcoin community can get fibre..

ok..

ill call Activision and tell them that online gaming is impossible because 0.1% of population cant play online constantly.
then ill call blizzard

oh and ill call youtube and tell them that they need to turn off their video upload feature because people cant upload.
wait. next ill call skype as their 3mb a minute video calls (30mb per 10 minutes) are impossible.

anyone else you want me to call? as it seems there must be a problem.. and strangely these services have not been aware..