Block Ladder

[blockladder]

been toying with ideas how to reduce orphan rate

this idea of exponential difficulty backoff comes to mind

in standing state the network quickly mines low difficulty blocks DIFF=1

when a fork of depth 1 is observed the DIFF=2 is raised and rate of block generation is halved
deeper forks than 1 cannot be healed.
the fork then can be merged by the successor block and potential doublespend conflicts are resolved
by a list of TX ID's to revert in successor block.

when fork is settled the block speed is slowly falls to maximum again
after two succesive blocks of DIFF=4, the difficulty is again halved to 2, then two more blocks to 1 etc.

obviously the fee doubles with difficulty also
larger blocks have 2x capacity and so on

malleability is fixed by seg witness or schnorr sig etc

https://github.com/blockladder/

opinions?

[blockladder]

I will be doing the technical part in my spare time
I'm pretty confident this project will be successful and by that I mean
It will either be mineable as an demo (possibly later as an altcoin or even real Bitcoin).
I also set myself the ideal (don't promise) to either write a paper (I'm not an academic tho)
or at least provide a huge amount of live test data from a working system.

by working system I don't mean an altcoin (too ambitious) or even deployed
on real Bitcoin (one can dream right).

I want to have the minimal viable product e.g to have the block creation
operating at desired level (think 1 block per second) and the orphans
resolved, doublespends reverted.

Very difficult part will be the visualisation of the results.

Then once ready, we can maybe ask a core dev to comment on this I think that would be a nice
conclusion of the project.

[monsterer]

when a fork of depth 1 is observed the DIFF=2 is raised and rate of block generation is halved
deeper forks than 1 cannot be healed.

So, as a DOS attacker, I create 1000 forks at depth 1 and the DIFF skyrockets freezing the chain?