Which BIP is winning?

Thursday 17 December 2015

TL;DR The great XT debate went quiet, but something has to happen soon

It’s all gone quiet since the summer, when the bitter debate around Bitcoin XT reached its peak - but that doesn't mean the issue has gone away.

To recap quickly, bitcoin is running out of elbow room. If the number of transactions continues to increase, there won’t be the space to include all of them in the blockchain. Already some transactions are delayed at times of peak activity and are pushed to the next block, or the one after.

Three major solutions have been proposed, which we'll call Tall, Grande and Venti for simplicity.

Tall. Some people - predominantly miners - claim that’s no problem; if people want a transaction included they will have to compete for it with higher fees. If we stick with the status quo of 1 MB blocks, bitcoin cannot scale. It might conceivably become a kind of digital gold, used for large transfers, but not mass adoption.


Like a trip to Starbucks, there are important decisions to be made, with potentially messy consequences

Venti. Bitcoin XT, also known as BIP-101 (Bitcoin Improvement Protocol), was the much-vaunted solution proposed by Gavin Andresen and Mike Hearn, two of bitcoin’s core developers. It involves:

  • A one-time fork
  • Blocks of 8 MB
  • Block size doubles every 2 years
  • After 20 years, at 32 GB blocks, there are no further increases.

XT was hugely controversial, mainly because it was seen as a unilateral attempt to hijack bitcoin by just two developers. The software was released, and gained a minority of support (at one point up to around 14 percent of blocks mined), but has declined in popularity over recent months.

Recently, Andresen has suggested he might take over development of XT, because Hearn has moved on to other things. Here’s what Hearn said on Reddit:

I will do releases to keep XT operational and may merge patches from time to time. It's there for when people want it (or just the BIP 101 patches).
The Bitcoin community needs to demonstrate a commitment to the blockchain it's actually built on, today. I suspect things might start to happen after the December conference, as every big player I talked to expressed huge frustration that nothing was happening, but given their behaviour so far all it'd take is for Blockstream to announce another conference and they'd go back to waiting for something to happen.

Clearly, he’s not a happy bunny.

Grande. The alternative to leaving things as they are and having transactions back up - basically killing bitcoin - is Jeff Garzik’s BIP-100 proposal. This involves a hardfork to remove the 1 MB block limit, instead allowing miners to vote for the block size they favour. The bottom 20 percent of votes are discarded and the lowest value left is the new block size. BIP-100 would only become active when it attracted 90 percent of hashing power, enabling a smooth transition.

There are other proposals, of course - everything from removing the block size limit altogether, to increasing block frequency. But few of these have attracted much support. As things currently stand, BIP-100 is supported by a fair but not spectacular percentage of mining power, but XT is down around 10%. The remaining hash power is voting for the status quo.

So whilst BIP-100 is popular, it hasn’t come close to the 90% threshold it needs to be activated. January 2016 will likely see renewed interest - as well as fierce debate, attacks and price volatility as a result.

comments powered by Disqus