Hard fork Bitcoin: Controversy over increasing the bitcoin block size resurfaces

Hard fork Bitcoin

Gavin Andresen recently updated his Github account with code he is testing to increase the block size limit to 20 mb from 1 mb today. In typical fashion, the bitcoin community got very emotional and heated responses ensued on reddit. It is no surprise that Gavin is moving in this direction and has been very vocal about this plan for sometime now. In response to one reddit user’s comments, Gavin stated

“it is intended as ‘it is time to discuss this now.’ I will be writing a series of blog posts in the coming week or two responding to objections I’ve heard.”

This is a big deal as the change in code would require a hard fork, which if accepted would render miners who do not upgrade to be useless.

What is a hard fork?

A hard fork happens when a new version of the full node client accepts blocks as valid that previous versions would have rejected as invalid. The moment such a block gets submitted to the live network, the new clients will accept it and build on that chain, whereas the old clients will reject it and start building on a different block at that height, at which point the two networks are operating on different longest chains (hence fork). – From reddit user d4d5c4e5

The Bitcoin wiki has a page dedicated to keeping a “Hardfork Wishlist” to record changes to Bitcoin that might be desirable, but that will require a “hard” block-chain split (everybody must upgrade, old software will not accept blocks/transactions created with the new rules, considering them to be invalid blocks).

So it is safe to say hardfork discussions have been around for a while, but it seems many vocal members of the community look at hardforks with great suspicion.

Technom4ge comments from reddit

I understand your arguments. I don’t see Bitcoin necessarily needing to aim at Visa transaction counts directly either. However Bitcoin does need to reach the amounts of regular wire transfers / SEPA directly and currently it simply doesn’t cut it. The 1MB limit is not some holy grail Satoshi came up with, with good reasons. No, unlike many of Bitcoin’s features, this one was almost arbitrary and clearly meant as a temporary measure.

Summa summarum, I agree we shouldn’t try to solve all transaction use cases with Bitcoin direct. However even taking that into account, 1MB is not enough.

Personally I see the 20MB as a great compromise. It does not affect full nodes requirements in any significant way and this has already been proven. And it allows the base network itself to handle at least a decent amount of transactions.

Some common issues regarding the increase in block size

Transaction Fee Death Spiral

In common economics, when scarcity is high, so is demand. Therefore, many bitcoin enthusiasts believe it is essential to maintain the block size limit (limiting the transaction amount creates scarcity) in order to create demand and avoid transaction fees going to zero. However, the economic theory states that in a competitive market, supply, demand, and price will find an equilibrium.

Centralization Death Spiral

In order to run a fully-validating node client, a complete copy of the public ledger must be kept up-to-date. If the block limit size was increased to an arbitrarily large amount, only entities with access to large data centers will be able to support the network. We will inevitably be required to rely upon large institutions to support the network, resulting in Bitcoin’s biggest fear – centralization.

Block Subsidy, Fees, and Blockchain Security

The block size limit of 1MB currently supports approximately 7-transactions-per-second. The concern is if the transaction fees will draw enough mining power to secure the network as the block reward reduces over time. Those in favor of increasing the block size support the argument by saying transaction fees will be lower, but more transactions will generate more fees.