I would regularly check coin.dance/blocks for updates on Segregated Witness and Bitcoin Unlimited popularity among miners. Another proposal called "Emergent Consensus" has appeared. Could someone explain this proposal? While we're at it, could you explain BIP100 and Extension blocks, too?
Emergent consensus is a protocol upgrade proposal championed by the Bitcoin Unlimited developers. The idea is for each node to broadcast its acceptable block size and to thusly establish a feedback mechanism by which the miners can determine the size of blocks to create. In parallel, rules governing block size would be largely lifted under the assumption that a free market would discover the correct block size. The proposal is marketed as a permanent solution to the block size debate and as a means to put the block size limit into the hands of the users.
However, this proposal has collected a fair share of criticism, such as:
- It's expected that it'll end up putting the block size limit into the hands of the miners instead of the users, as the out-of-band feedback mechanism is in no way binding to the miners.
- Larger miners might be able to abuse the mechanism to put smaller miners at a disadvantage, resulting in further incentives for hashrate to consolidate.
- Incentives for larger blocks overwhelm reasons to restrain block size growth, e.g. costs of larger blocks mostly burden node operators, while miners mostly benefit.
- Specific implementation issues with the mechanism of establishing bigger block sizes in the network.
- The necessity of a hard fork to activate the proposal.
For a more comprehensive overview, I'd recommend that you read the excellent three part series by Aaron van Wirdum that explores the proposal in detail:
- A Closer Look at Bitcoin Unlimited’s Configurable Block Size Proposal
- How Bitcoin Unlimited Users May End Up on Different Blockchains
- Why Bitcoin Unlimited’s “Emergent Consensus” Is a Gamble
Is a block size voting mechanism proposed by Jeff Garzik in 2015. It would establish a dynamic block size per vote of the miners. Each block would cast a vote for a maximum blocksize, and from the votes of the past 12,000 blocks the current limit would be derived by discarding the lowest 20% of the votes and picking the remaining minimum.
The proposal puts block size under control of the miners and requires a hard fork to activate.
Aaron van Wirdum covered it here: A Closer Look at BIP100: The Block Size Proposal Bitcoin Miners are Rallying Behind
Another proposal to improve capacity which suggests appending additional blockspace similarly to how segwit adds the space for the witness programs. Unupgraded nodes only see the stripped version of the block (just like in segwit). However, while segwit keeps all the transaction "instructions" in the stripped block and only partitions the one-time-use witness program, the extension block actually allows complete transactions in the added space. Unlike segwit, the extension block therefore establishes a second UTXO set only visible to upgraded nodes. While the proposal could be rolled out as a "softfork" due to the stripped blocks being backwards compatible, the partioning of the UTXO set would basically force everyone to upgrade immediately to the new version in order to be able to track all economic activity on the network. (Whereas with segwit even old nodes see all transactions, even though the nodes don't understand some restrictions that apply to the segwit formatted transactions.)