wrote an outlook for the upcoming “Toccata” hard fork -- native L1 covenants, based zk apps, why the activation window moved, and what the road from feature freeze to mainnet looks like:
https://t.co/JFRkYp6Yd4
Happy 5th birthday Kaspa!
so young, and carrying so many hopes,
already rusty, yet smooth on your feet.
you’ve danced through crescendos, quickening (y)our beat,
then a toccata, a covenant bond --
our trust in your hands, your future in our care.
a knight drawing near, swift and hard to sway,
keeping time with the world, come what may.
whole worlds to build, and silver ink to dream them in --
im(.ag)ine all the songs still waiting to begin
evm smart contracts can obviously express such locking logic as well*. the tradeoff is probably that having such logic prevents also legit fast liquidity operations which users like binance opt for
* it’s possible that the covenant design space has some inherent safety advantages, or leads more naturally to safe patterns due to its partitioned local/custom logic, but i think the burden of proof is still on us (especially if you also want good UX and usability on non-black-swan days)
watching the timeline with awe. human beings in a world of ai superintelligence. how humbling. @hashdag, crypto and defi need fundamental redesigns, i think you should be leading with more determination / visibility.
the delicate part seems to be (at least in this particular binance case) centralized infra vulnerability (one cannot ignore the recent explode in infra zerodays found by llms and effecting nearly every other live server on the internet), followed by clever and fast onchain operations, entering “code is law” zones before humans can respond. one thought yonatan shared in the midchain panel this week was low cost small language models detecting anomalies onchain and activated either by contact or by consensus majority (the latter being not fully cryptographic and hence more dynamic and adaptive).
this is just one idea. i agree that broad thinking needs to be applied and that we all have a role here. i will definitely ponder on this as we go forward with programibility on kaspa
wow that’s an amazing speedup. also, formal verification for getting manual assembly back into the loop is beautiful. it almost sounds like we can soon have formal-verification-based compilers that will generate custom assembly for dedicated code segments (something like what ASICs are for hardware)
@hus_qy couldn’t agree more. it was truly exciting to see this appear out of nowhere today, and it’s at perfect timing when there’s so much tools to build on/with and innovate
@moxypixy15683 it’s possible to harness miners as attesters since Toccata (using kip21’s commitment to miner payloads). the pipes are there so to speak, but statistical protocols need to be defined over them + they need to be accessed through a based zk app