The 2-4-8 proposal would be disastrous for bitcoin

[Steve Sokolowski]

There exist a number of proposals to resolve the bitcoin blocksize debate. Perhaps the most well known is BIP 101, but BIP 100, BIP 103, and a number of others are being considered. However, there is one proposal that some see as a "compromise" or as a "default" - the so-called "2-4-8" proposal. While there would be significant drawbacks to some of the other proposals, 2-4-8 in particular would be disastrous for bitcoin, and the situation is particularly disturbing because the first "Scaling Bitcoin" conference resulted in some attendees becoming convinced of the necessity for 2-4-8.

2-4-8 is a solution where the maximum blocksize would increase to 2MB, then to 4MB, and then to 8MB. There would be a set time for each intermediate level. More than one version of this proposal has been mentioned, with different implementation times and different numbers of steps, like a "2-4" plan or a one-time increase to 2MB. However, all of them have one feature in common: after the blocksize reaches the last level, there are no further increases scheduled. Supporters see 2-4-8 as a stopgap measure that can be implemented immediately, with a "permanent" solution coming at a later time.

2-4-8 is flawed because it requires at least two more hard forks to be conducted in the future: the first to get rid of 2-4-8 in a few years, and the second before 2106 to resolve the unsigned Unix timestamp issue (the date in this sentence was changed after Peter Todd corrected a mistake here). By contrast, BIP 101 does not reach its maximum level for this first fork for 22 years in the future. Because 2-4-8 will "expire" in just a few years, the blocksize debate will likely resume starting the day after 2-4-8 would be adopted as people see that the crisis has simply been pushed briefly into the future.

The critical point to understand with any temporary solution is the uncertainty it would create. Uncertainty is catastrophic for financial markets. Those who follow stocks can easily recall times when the market is on a knife's edge the day of a Federal Reserve meeting. Traders actually watch television screens and try to interpret Greenspan's, or Bernanke's, or Yellen's body language to see whether interest rates will rise or fall. They play and replay the footage listening for word choice and even tonality to get any hint of the Fed's decision. If the chairwoman looks down at the wrong time, the Dow falls 2% because of "fears of action on interest rates."

Similar effects occur when government shutdowns loom. The military often preemptively suspends funding for programs even if it is likely they will be continued, simply as a contingency in the case that Cruz decides to play another game of chicken. When tax hikes appear on the horizon, businesses suspend hiring and wait to see what the legislature decides, because they are unsure if they will be able to make profit on the new employees. When the price of oil crashes, gas stations hesitate to lower their own prices because they are unsure whether their costs will remain low or rise back up (but when prices creep up, they raise prices immediately because there is no uncertainty about that).

When I wrote about BIP 101 adoption yesterday, a single quote that I happened to throw in as an afterthought, about not buying bitcoins because I have a "wait and see" attitude, made the rounds on many sites. I had thought that such a position would be obvious to most traders. While our small business matters little in the larger picture, there are almost certainly much larger firms working on "blockchain tech" waiting for certainty before investing in bitcoin.

What is important for developers to realize is that 2-4-8 does absolutely nothing to resolve the critical issue of uncertainty. So what if the crisis is pushed two or four years down the line? Venture capitalists expect that it will take five years or more to get a return on their investment. It took us 10,200 hours of effort for two people working 2 years to get a mining pool to a state we consider feature-complete. If we were starting out right now and had nothing, Chris and I agreed that it is completely out of the question that we would decide to put in that effort when we wouldn't be finished until 2-4-8 was starting to become a crisis yet again.

To make things worse, the adoption of 2-4-8 would set a precedent. While the adoption of BIP 101 would prove that the community has the potential to come together and solve problems, 2-4-8 would demonstrate that bitcoin faced a problem and failed to come to a consensus. People came up with many solutions and in the end, forces holding back progress managed to successfully stonewall development.

2-4-8's adoption would show that arguments over the blocksize limit are going to be a continuing issue that will plague bitcoin into the indefinite future, like the debt ceiling rears its head every few years and creates continual uncertainty over US bonds (and that is reflected in its credit rating). The precedent would be that every few years it is now time to start over with the censorship, criminal attacks, namecalling, and brinksmanship, and eventually recognize that since there are even more stakeholders this time the only solution is a one-time 12MB increase, which will be delayed an additional year because it costs so much for everyone to upgrade and test their software. And at some point, this pattern will wear out, because there will be so many ideas from so many new people that even 51% consensus on any idea is impossible. There is only one chance to come up with a permanent solution, and that time is now.

Like us, the people who run these companies have many other opportunities to make money. It doesn't make sense to them, or to us, to create new businesses in the bitcoin industry with 2-4-8 hanging over our heads. Instead of writing a new revolutionary bitcoin wallet, Chris would likely decide to move to Houston and use his chemistry background to design oil refineries - it's certain demand for gas will not go away, at least until solar takes over. And for those wondering why private blockchains are all the rage nowadays: instead of working on the bitcoin blockchain, venture capitalists would continue to put their money into new systems, ones where they can eliminate all the uncertainty themselves.