2:45 The problem is not only with storage, the main problem is transmitting the block though the entire network. If we increase keep increasing the limit, we would need more and more bandwidth. So it will end up centralized. Instead of increasing the Blocksize, Bitcoin Core wants to implement Segwit which can increase the capacity to almost (2MB), implement Schnorr signatures and LN. Eventually we will need to increase the limit, but now right now.

This dude just explained the problem of blockchain scalability simply. Something that I watched several "experts" bumble about without really saying much. This makes me wonder about some guys in the Blockchain/crypto space. If they can't explain a concept plainly to a non technical person, they probably don't understand it well enough to be explaining. All in all, great job Sir... Hats off to you.

Thank you for all your videos, they are instructive than practically anything else I have found

The blocksize limit or 1MB has already been removed months ago. We now have a block weight limit of 4 million units or 4MB maximum with segwit activation.

As a new cryptocurrency investor, I REALLY appreciate you doing videos on this topic. You rock!

@computerphile Can you get Jimmy song to talk about side chains and drive chains?

Congrats on an approaching 1M subs!

Extra layer payment channels is definitely the solution for scaling. The 1st layer acts as the "clearing house" and is still the ultimate source of trust. Anyone arguing for bigger blocks is impatient and unnecessarily worried.

Feature matching video from Dr. Pound and I will love you forever and ever

This guy had 80 charges against him and wasn't scared, just annoyed that he got caught. Mustafa Al-Bassam is a legend!

Thanks for the video. I would like to see a video about Sidechains implementations. This is another way to scale bitcoin outside blockchain.

I think the problem of chain size bloat is due to no block is ever finalized (e.g., theoretically fork can happen on any block), a blockchain that has finalization property can simply keep only the merkle tree root of the finalized block, discarding the block content.

Please do a video on Iota/directed acyclic graphs!

Lightning is cool, but another scaling technology we can use is the RSK sidechain (which should launch in 2-3 weeks). We can also achieve scale (and better privacy) for bitcoin by using a mimblewimble blockchain.

What about hierarchical blockchains? When a specific blockchain gets full, it splits into two chains, with a shared parent chain to combine/reconcile the sub-chains. This will increase latency when users in separate chains need to process transactions, but will reduce latency between transactions within a given chain. End-users would migrate between chains as needed based on their evolving behavior, the goal being to minimize the number of cross-chain transactions. Another approach is for each user to have multiple accounts, where they may allocate their total balance between accounts in any way desired. Each account would be on separate blockchains (perhaps one per retailer, such as Amazon), and the total balance would be managed by a bank blockchain (perhaps your Credit Union or PayPal). This would separate "net" transactions (purchases) from "zero-sum" transactions (inter-account transfers). In essence, it would permit multiple independent blockchains to be federated from the perspective of individual users, much like independent GNU Social (Mastodon) instances are federated to form the global social network. Having multiple smaller blockchains would mean that "local" transactions would be blindingly fast, while chain-crossing transactions would take longer. This would push much of the overall management (And control) off the network of blockchains and onto each individual user. Miners could work as many or as few blockchains as desired, primarily limited by miner capacity and blockchain transaction rates. Hierarchical blockchains would also incentivize users to mine the chains they use, increasing the number of miners.

Please more blockchain videos!

You forgot these other big block issues 1) CPU time to validate more hashes in larger blocks 2) Network propagation. The blocks have to propagate across the world (and possibly LEO/GEO orbits) then reach consensus before the next block arriaves.. possibly from space or another hemisphere 3) Economics. Tragedy of the Commons. You want there to be an adequate workload for miners and economic incentives to secure the network. // btrash Lightning A) Payment hubs *can* be an issue..... but you have the *choice* to connect to them or not. B) Upfront capital. You can open a channel and do a 'push tx'. Which doesn't require any additional upfront cost // These issues are solved with software (via a lightning wallet)

Very interesting

Please do a video on Net Neutrality! If you don't do a video on it, it might be abolished!

I think the "bigger block size -> bigger chain -> less full node" argument is not very sound: another possibility is: "bigger the chain -> more people using Bitcoin -> more full nodes".

Side chains sound like you're pushing trust off the chain, doesn't seem like it makes sense to me. Why not just create a new blockchain and move value off that main chain to the side chain. That side chain has an address where the value transfers to.

With the latter solution, the "off-chain" solution. Don't you have a problem tracking everybody's balances if some ledges won't even know about certain balances? How do you verify that somebody has the funds?

Nice

So basically they want blockchains to have the same functionality as git, with branches, merging squashed commits, and push/pull with remotes

does bob get a fee for this "middle man" type action?

Hello hello. I would appreciate it if in every video you ask the guest to recommend a book for the viewers.

So basically it's kind of like NAT (Network Address Translation) for blockchains instead of network addresses. hmmm... nice!!!

You gotta call Jake Davis now

Most on-chain scaling people support off-chain scaling as well, they just accept a decent solution while a perfect one is not yet available. Limiting transactions in the current state hurts Bitcoin irreversably. Luckily many cryptocurrencies are available that all attempt different scaling methods.

How does segwit fit in to this? My understanding is that it is an on-chain solution that allows for increase in transaction capacity without having to increase block size.

It will all end in tears.

This might seem like a really naive solution, but what would happen if say you developed a solution whereby you get the nodes to create and agree on a data dump, essentially one big file that says that this person has x amount, this person has y. You could have it linked to the previous blocks in the same way, but then the average person doesn't need to store the entire blockchain and essentially start from scratch. If this type of thing is agreed in the same way as normal transactions, then there should be implicit trust that it is correct, right?

The conclusion of the main argument of this video seems to be that decentralized crypto-currency is unfeasible.The end. Can anyone tell me I'm wrong and why?

make the fork, but do it in such a way that allows for future forks/forking to not cause such a problem with between forks

Tflow! ;) all grown up and official =)

These scaling problems is one of the reasons why I find IOTA really interesting. No transaction fees is also a nice feature.

Major problem with Lightning: A network of payment channels will naturally have to be centralised

holoCurrency (Ħ) feat. AI-specific scaling/scoring of bloch-chain-sphere diagramming and cross block-chain interaction/tesserAction is the entangled future. A future with quantum computer/server, Internet/Neuronet, hosted holomedia such as geo-located holograms or interactive holo interfaces or holographic game content, holographic environment, or environmental holographic mApping. :: Ħ displaces $, or reassigns scanned $ or hardCurrency into replaced Ħ values.

Why not just have it make snapshots of the current balance and make it so that the full blockchain lives on the super nodes for verification, but the snapshot chains are used for day to day use?

Any solution that causes the system to take one or more steps towards centralization will be, by definition, undermining the very philosophy and point of the system: to a _decentralized_ system. Decentralization implies the users remain under control. The more centralized the system, the less control the individual user has. Consider how much control you _really_ have over your KZfaq, Google, Facebook, and/or any other "social media" account you have..

This guy is a hacker, phd student and only 22 wow

The problem is already solved Bitcoin Cash can scale. Off chain happens anyway as private agreements and private services it has nothing to do with bitcoin development.

You can check out ByteBall, a unique coin, based on DAG, programmed and distributed for free, had no crowdfunding / ICO.

What do we expect the ledger size to be in 2 years ? 100Gb ? 1Tb ? About the argument: raising the storage size would make it more difficult for EVERYBODY to be a node is kind of bad. Storage quite inexpensive and the base storage requiered already implies that your device has a high storage capacity, so doubling it doen't seem to be an issue.. unless double it doesn't solve the issue and i needs to triple... decuple to be useable..

Who was Alice and Bob and why are they so popular?

Every single penny MUST be tracked!!!

I guess the market will decide

Why would increasing the block size increase the size of the blockchain? If anything, with the same amount of transactions, the chain would be smaller as there is less overhead.

I think LTN is not very good because it requires the client to be online all the time, or the client has to tell some third-party service to be online for him - way too complex. And many other problem...

How safe are my bitcoins in Blockchain.info if I have all the security features turned on and I won't lose my passwords? Can the Blockchain.info servers be hacked?

Now they just to make it so full nodes can actually earn something, i have a microserver sitting at my feet doing nothing, it can't mine but it can run a full node, there just isn't any incentive for me to do this

Both on-chain and off-chain scaling cause centralization. Both in a different manner (miner centralization or payment channel centralization). Therefore we should have both a bit, so that neither type of centralization gets the overhand. Segwit2x should have been accepted, but now the payment channel centralization gets the overhand. I'm afraid that if this continues, somebody with lets say 1 BTC in his/her wallet can not move that 1 BTC anymore in the future, because it costs maybe 0.8 BTC in fees to get that transaction being picked up by miners. This would be really bad for many people! A few days ago I did a withdrawal from an exchange, and saw that the exchange payed more that 0.15BTC in fee! That transaction consisted of many transactions, combining >100BTC in value, so that's kind of ok, but nobody can compete with that if you want to transfer your bitcoin p2p to for example a friend. Bigger blocks must therefore happen in the future in my opinion. Its too bad it didn't happen today. Hummm... what if Bitcoin Cash is going to integrate Segwit? ;-p

Off chain transactions are called exchanges.

I agree with most of what was said but the blocks aren't really full unless the network is being spammed for political purposes. The great news is that spamming Bitcoin costs money unlike email spam so at least the attackers are paying something. We know these transactions are spam because the mempool sometimes empties to a couple thousand transactions and other times it spikes to 100k transactions. Where is this spike coming from? It's not like a large number of people suddenly started using Bitcoin at the same time.

This could hurt bitcoin if people want more widespread adoption. It might incentivize people that are using bitcoin for illegal purposes; where you would send just a large quantity of off blockchain transactions, so as to not have a trail. The argument could be made that It can be extremely difficult for you to trace bitcoin with the blockchain so it shouldn't matter. But this could lead bitcoin to not being a more largely adopted.

lightning network is more logical

i really like computerphile but i'm having a really hard time when it comes to bitcoins...

In my opinion they should accept Bitcoin as the gold equivalent that it is and not adapt it for microtransactions. If you want to buy a coffee, use Litecoin, IOTA or other more suitable cryptos.

INTJ with developed Te

on chain scaling is fine to a point, it's certainly better than doing nothing and risking bitcoin being displaced at the top due to being unusable - off chain solutions can be sorted later.

Only 2 days Late for my Distributed Systems Exam

Can the Bitcoin network be hacked?

Why can't you truncate the blockchain? Why include transactions from years ago?

Can't we just start a whole other set of "blocks".?? .....uhhhh BChain v2??

the fundamental problem of scaling through blocksize increase is that this will lead to a lack of a "fee market" which will result in a very insecure chain in the long run due to the fact that the block reward will go to 0 over time

Can you guys do a video on DAG-based cryptocurrencies like IOTA?

Ethereum

I still enjoy the LULZ

MORE VIDEOS ABOUT BITCOIN PLEASE !!!! YOU ARE THE BEST PEOPLE THAT KNOW SOMETHING ABOUT IT.

charli e

Sooo... basically this ‘marvelous amazing decentralised’ system is slowly moving to the exact same topology as normal banks? *slow clap*

Just like roads, increasing size will only make people fill it faster. I've seen statistics that show credit card transactions are upwards of 100 billion a year (more than likely more), which puts it at about 3200 transactions per second.....current bitcoin maximum is about 10 transactions per second. This means we would have to scale to 320MB per block just to meet the present demand of credit cards. The blockchain would go from 20GB to 6.2TB in a week for miners to be forced to keep on system for the merkle root pruning.....and it would increase to nearly 12TB easily. This is a highly inefficient method and bitcoin will eventually implode on itself.

Off-chain scaling is about creating external markets that can be monetized and regulated and controlled outside of the pure bitcoin blockchain itself. It defeats the purpose of the original ideas and goals of a decentralized currency - becasue while off-chain scaling might incorrectly seem decentralizsed to you, it is centralizing the control of the transactions into that off-chain third party that is ultimately NOT bitcoin.This is not good for bitcoin as a result

So WHO decided and insert changes in BITCOIN? arent it suppose to be decentralized?

FUD!

Just as Bitcoin passes £6000

Is it just me or this guy really looks kinda like Joey Tribbiani.

LULZ

Buy xtrabytes

Wtf?!?!?

Can you do a video about Bitcoin Gold?

@2:15 "The main argument against [bigger block size] ...is storage" WRONG. Bandwidth dummy. Storage is not the issue.

can you give subss!!!

Bitcoin is not decentralized...

dude we see that you're high

Can you put a beer into mustafa before the next video? he needs to relax ;)

he is so cute , come here to watch him :p

The more I learn about BItcoin, the less I trust it. I t really irritates me that I have to pay for some services in this currency whose exchange rate is more volatile than Zimbabwe dollars at the moment. All I wanna do is pay a damned bill not spin a roulette wheel.

Bitcoins were invented to provide an alternative to Banks, but now Banks (Payment hubs and Supernodes) have crept in Bitcoin itself.

First

Either you want to scale or you don't. Block(the)Stream's Core group is not scaling in any measurable way, and fees are off the charts. The currency is unusable - in a panic, people won't be able to move their Bitcoin. Get out now, the writing is on the wall.

Its called BUTTCOIN

Very disappointed with Computerphile for posting this without a counterbalancing view point!! This is not helpful. Please look at Craig Wright's explanations. The bigger blocks sizes were always envisioned. Lightning is controlled by banksters. Love to see this man debate Ryan X. Charles or Craig Wright. BCH works. Fast, cheap, secure. No segwit... EVER. Just look at nChain developments. On-chain scaling is the future. Despite the critics, they will build a global network of peer to peer value exchange.

I'm having trouble following you with your accent. It's like cockney with an Arabic accent.

WHO will decide and implement the size. It is decentralized!!!!

his speech is unclear. Just have someone who speaks well to explain things

IOTA wins all

Bitcoin is not something that academic funds should be poured into to fix it.. To fight it, perhaps.

Bitcoin is gambling technically

Sorry computerphile, but now you are back to featuring people whose accents are too thick for such public speaking. Only got 77 seconds into this before giving up trying to listen to some random unknown (not a star) trying to summarize something he didn't invent. You could have chosen lots of experts to explain this, no need to use someone who doesn't speak clearly.