The problem is choice itself regarding blocksize

[AZwarel]

        First of all, we tend to forget how awesome is the fact, that we have a debate over the nature of the “block size limit solution”. Some are pro some, are contra, some are agnostic, and yet we fail to grasp how amazing is the very fact that we have an open debate about the economic-technical policy of our p2p money system!

Think about it for a moment, and imagine the same raving debate going over how the FED should or not raise interest, money supply etc. Isn't the debate itself is a proof how open Bitcoin is?
   I have tried to approach the subject from a less technical angle, focusing on the principal choices behind the Kbs. I may be wrong, but I think the problem is choice itself: whichever side “wins” majority, it will – probably irreversibly – set the Bitcoin network on a course of two mutually exclusive fate.

   An example of those two options are as follows.

   A “small”, elite, 1MB (or 8/20 as well in the near future) block size'd blockchain will render bitcoin's usage as “digital gold/ledger”, an asset literally held for store of value and proof of existence, open only for value transfers and economic activities requiring the waste amount of security and immutable nature of the blockchain. This version would be valued for its world class trust less consensus security, which is also very scarce for acquiring.
   Why? Because if free market allowed to work it's ways, than the insanely high demand of transacting directly on the blockchain will increase “miner fees” beyond what micro and “grocery” value transactions can afford, – those will literally die out from the blockchain, moving to semi centralized services, offchain payment processors, sidechain networks validating them only in bundles, or just utilizing a transaction focused altcoin for everyday, small value transfers.
   The “elitecoin” approach would make the Bitcoin network less focused – or able - on transaction speed, confirmation speed and used more as a long term savings account, a technically manageable fully validating node on every serious (value wise) bitcoin holders “home PC”, a digital gold-like depository, walled and secured on cold storage devices, paper, maybe in a bank vault for irony's sake.

   The other path to walk would be the “Bitcoin WesternMasterVisa”, enabling dynamic blocksizes – if that technically feasible to implement -, where, if major adoption happens, hundred thousands of transactions will happen every second, from 2 $cents tippings on reddit to 200M$ company sells on the “Blockchain Nasdaq”.
   This type of bitcoin would be very different from the first one. The enforced neutrality regarding the transacted amount of bitcoins in a sea of potentially several MB per second limits fees way below today's “dust limit”, reducing individual transaction cost to fractures of pennies, but a million of them for every block. This implementation has a high cost for individually validate the blockchain – which is not good for decentralization some may say, but to be fair, the average user – now the overwhelming majority of the users, assuming this hypothetical mainstream adoption - would not care.
   Let me repeat that: the average user will not run nodes anyway, as of today they are not running one, no matter the hardware or knowledge factor it requires!
   This approach would preserve the aspect of “one network for all transaction”, which we are familiar today, but with a cost: huge data centres as acting nodes servicing thousands, millions of SPV wallets around the world, effectively marginalizing individual user's voice on the network, delegation consensus decisions on the hands of a few powerful central point (“few” in the sense of versus millions; still thousands of organizations would have to collaborate!).

   No matter which path we choose with upgrading the protocol – we have to evolve for sure -, we will reach a “point of no return”, a size of adoption where drastic changes can not reach consensus any more, yet, we will have to take the first step now if we want to scale up the network. Unintended consequences all over the way.

   Hence, my opinion regarding the block size debate beyond technicalities is:
the problem is choice itself, for it will be (most likely) final for the fate of Bitcoin.

[oblivi]

Isnt 8MB increase every 2 years + LN the best solution? We have planed halvings, planed increases on the blocksize limit would be as convenient.



Technology will keep making storage cheaper and cheaper. But yes, i agree that no matter what, it's only the hardcore people that will run nodes. The average end user is going to use something like Electrum or whatever. We had tons of nodes early on because we only had Bitcoin QT.

[smiletyson]

Current blockchain is 40 GB on windows (Using bitcoin core).
Current block number is 362084.
If you assume 1 block is 1 MB then total blockchain size should be 350+ GB. But it doesn't work like that.
Block size doesn't mean increase of storage...

[AZwarel]

That's a nice graph:)

I am personally agnostic, i do not presume to know which path will be better in the far future. Although, the hardware rate (cpu speed, hdd storage, bandwidth) increased like 100x last 20 years, i see no
reason why wouldn't it in the future (imagine someone said to you you will be able to watch streamed full hd videos in just 10 years ago, was preposterous!).

It is feasible to see 1Petabyte home HDD's, cheap abundant RAM, and 1gigabit internet rampant by 2024, which will keep running the network cheap and decentralized.

[Lauda]

Current blockchain is 40 GB on windows (Using bitcoin core).
Current block number is 362084.
If you assume 1 block is 1 MB then total blockchain size should be 350+ GB. But it doesn't work like that.
Block size doesn't mean increase of storage...

Relax. This won't be a problem to average users. It might be a problem for nodes however. The pruning feature is going to be introduced with the release of 0.11.

Last 288 blocks: source ). You will probably have to change a setting or two that would enable pruning.

This enables you to only download a small part of the blockchain.

-snip-
It is feasible to see 1Petabyte home HDD's, cheap abundant RAM, and 1gigabit internet rampant by 2024, which will keep running the network cheap and decentralized

1 Petabyte HDDs? Not really. The problem isn't the speed of the internet, it is the bandwidth. If we assume that we have a 8 Mb block every 10 minutes, that is just 800 Kb of data per minute.

The global average connection speed was 4.6Mbps in the second quarter of 2014.

This is ~0.6 MB/s, meaning that someone could download the block in less than 20 seconds. However, some nodes might have a problem with the increased amount of upload bandwidth.

[johnyj]

It won't be used for daily spending unless the price volatility calms down. But that is impossible due to reward halving every 4 years: If FED cut money supply every 4 years, what would you expect on USD's value? Fiat money's value is stable because central banks constantly adjust the supply to meet the demand, but bitcoin is not working like that, supply have no flexibility like a stock, destined to be a bubble maker

So sooner or later people will focus on the investment property of bitcoin and try to get as many exchanges as possible in each country