So that probably wasn't the best video to send (there's one by him about mimblewimble and scriptless scripts which is very good) but it does explain what it is. Also the 100MB range proofs are no longer that big, MW uses bulletproofs which were proven by Greg maxwell shortly after that video was created. But basically, you get the ability to guarantee a historical transaction is valid without needing any historical transactions at all. The entire blockchain is one giant coin join where all you need is the UTXO set (in original MW, in his change, you need those "excesses" he talks about, they're called transaction kernels, and they're needed to ensure things like multisig and threshold signatures and time locks and the like, if you don't care about hat they can always equal 0), no historical data whatsoever. What that means is that you don't even need a block size, the whole "we need a block size so the blockchain doesn't grow too big that nodes get centralized" point becomes moot. You can have as many transactions per second as you can send across the physical network and back per second within the block time at the speed of light. Latency and block time become your scaling bottleneck, not block size. A block time of ten minutes on a network that takes max 5 minutes for the slowest connection to ping from one end or the other can have unlimited TPS. Scaling becomes a solved problem, no second layer required. *Transactions don't need to be saved after they're spent.* Really think about what I just said and what that means. Space money. The block time can be the number of light minutes across a civilization is and the entire civilization can run on one base layer currency that scales to its size, fully decentralized. You don't want that on bitcoin? You'd rather have a thousand dollar toaster? Which, BTW, you can still have with space money. Additionally, this scheme has confidential senders, confidential amounts and confidential recipients, but no forward secrecy which is it's shortcoming, someone can just archive historical data instead of deleting it and build a transaction graph. Still better than bitcoin on the privacy front, not as good as Monero on that front but better on every other front.

Oh. And it is working in the wild. Grin. the problem with grin is that after the initial launch of a basically perfect network, every single thing about improvement gets decided by committee instead of people just building and publishing. As a result the entire ecosystem of wallet apps and stuff is just collapsing. It's a great network, just looking at the network itself, everything else, not so great unfortunately. The stewards of the community have their heads up their asses and as a result nobody is really interested anymore. I'm not, and I see the potential.

You explained it better than the dude... Were there debates? I wanna see two super knowledgeable dudes/dudettes debate it. I feel like that must've been more prevalent before I got here... I saw a debate between Voorhees (sp?) and someone and it was epic. Haven't seen one since... The entire blockchain being a coinjoin is interesting. It actually hits pretty close to an idea I've had for bitcoin... But this is actually what I'm saying - do the thing he said in the beginning where you merge it into bitcoin, but do it onto a hard fork. If its really so good, it will survive and win in the end. No bitcoiners can be mad because they get keys on that one too. Improve/attack bitcoin, either one is better than diverting into an altcoin.

That's a real shame. If there's something you can do to fix it, please do it... Otherwise, bring that energy into bitcoin. Write articles about it. Something...