Captain log

[nicola]

We are at week 21 of the year. The idea is to post the log of what is happening every week!

This is my first attempt about running this log - suggestions welcome

Week 21 Updates

🎓 research log

• 🎓 (new) #102 : Investigate Sonic
• 🎓 (new) #101 : Using CommA instead of listing CommRs in PoSt
• 🎓 (new) #100 : PoRep: How do ZigZag graphs inherit the expansion property of bipartite expanders?
• 🎓 (new) #98 : Challenge sampling
• 🎓 (new) #96 : Improving on the best known attacks for Depth Robust Graphs
• 🎓 (new) #95 : Speeding up Proof-of-Replication (via space-tradeoffs)
• 🎓 #92 : Better SNARKs
• 🎓 #63 : PoSt Stamps: proposal for removing VDF
• 🎓 #61 : PoSt Committee: Proposal for PoSTs - New challenge methods & removing VDFs
• 🎓 #53 : Avoiding forks and workarounds for known fixes for zkcrypto libraries
• 🎓 #1 : Signature Aggregation

closed this week

• 🎉 #43 : Make SEALing maliciously fast

---

🔭 specs log

• 🔭 (new) #206 : Evaluate order of cids in parent field
• 🔭 (new) #204 : Consider using compact serialization instead of CBOR RPC
• 🔭 (new) #203 : Consider usage of types: attofil vs integer in spec
• 🔭 (new) #202 : Evaluate need for signature on client deal proposal
• 🔭 (new) #201 : Decide on adding a data transfer protocol specification to the storage deal protocol
• 🔭 (new) #200 : Decide on including a starting block height in a deal proposal
• 🔭 (new) #199 : How do miners signal upgrade readiness? Other things?
• 🔭 (new) #198 : Use protocol versions during sync
• 🔭 (new) #197 : Protocol Versions
• 🔭 (new) #189 : Talking about height in EC (+ tickets, null blocks, etc)
• 🔭 (new) #187 : Spec clarification: Deadlines for PoSt proofs in the presence of null blocks
• 🔭 (new) #186 : Spec clarification: adding gen attack threshold in the glossary
• 🔭 (new) #185 : [tickets] Tie breaking the tie breaker
• 🔭 (new) #182 : Deal acceptance to commitment timeouts
• 🔭 (new) #181 : Proving Period Adjustments on Faults
• 🔭 (new) #180 : Filecoin <-> Other Blockchain Ideas
• 🔭 #176 : Modify spec so we can compute slashing conditions in state machine
• 🔭 #175 : (Proposal) Multiple Sector Size
• 🔭 #174 : The diagrams issue
• 🔭 #173 : Investigate https://github.com/tabatkins/bikeshed/
• 🔭 #159 : On burning funds and rewarding slashers

PRs to review

• 🔎 (new) #207 : Simpler spec for zigzag
• 🔎 (new) #196 : Bootstrap Spec
• 🔎 (new) #195 : ZigZag Offline circuit spec
• 🔎 (new) #194 : Proofs parameter design notes
• 🔎 (new) #193 : StorageDeal: transfer mechanisms and assurances
• 🔎 (new) #192 : add PoStProof description and note size of arrays
• 🔎 (new) #191 : added soft block delay
• 🔎 (new) #190 : added a tie-breaker to equal weighted, same min ticket tipsets
• 🔎 (new) #188 : feat: basic support for bikeshed

done this week

• 🎉 #208 : Fix some links
• 🎉 #205 : not an apostrophe issue
• 🎉 #179 : review feedback for actors.md
• 🎉 #169 : Init actor sketch
• 🎉 #164 : Refactor Signature Spec
• 🎉 #162 : update slashing rules

---

🎲 proofs log

active issues

• 🐣 (new) #566 : verify_post or generate_post should produce a validation error if provided a CommR which cannot be transformed into Fr
• 🐣 (new) #565 : Investigate possible performance improvements for belllman
• 🐣 (new) #564 : Investigate multicore scaling of ZigZag
• 🐣 (new) #563 : Apex Commitment
• 🐣 (new) #553 : a panic should never unwind the stack past the FFI
• 🐣 (new) #552 : Add sample config file and document.
• 🐣 (new) #551 : non-fatal panic appears in paramcache logs
• 🐣 (new) #550 : Populate the ZigZag parent's cache in advance for both forward and backward directions.
• 🐣 (new) #541 : add groth param check for CI
• 🐣 (new) #536 : create standard bencher config we can use for performance comparison/regression testing
• 🐣 (new) #531 : Sealing with Blake2s (was: Hybrid Merkle Trees)
• 🎲 #530 : Write benchmark on how long it takes to encode a single node
• 🎲 #519 : Improve compile times
• 🎲 #516 : define PR merge topology strategy policy
• 🎲 #504 : Avoid last data duplication in transform_and_replicate_layers
• 🎲 #500 : Improve zigzag example
• 🎲 #481 : disk-backed merkle trees
• 🎲 #460 : fr32: load more than one byte at a time before shifting
• 🎲 #421 : Integrate Piece Inclusion Proof
• 🎲 #396 : Write a spec document that describes current hardware requirements
• 🎲 #354 : Write spec for Symmetric Proof of Replication construction
• 🎲 #160 : Optimizations (maybe) worth doing
• 🎲 #157 : How to make circuits practical with 200 challenges.
• 🎲 #77 : ZigZag PoRep

PRs to review

• 🔎 (new) #571 : Delete the outdated link
• 🔎 (new) #569 : feat: switch to the latest pairing
• 🔎 (new) #567 : [WIP] disk-backed merkle trees
• 🔎 (new) #562 : Apex merkle commitment
• 🔎 (new) #534 : refactor(layered-drgporep): restructure parallelized proving
• 🔎 #529 : use MmapStore for MerkleTree generation

done this week

• 🏆 #570 : fix misspellings of "verifiable" and "verify"
• 🏆 #568 : fix spelling of log entry at verify_post finish
• 🏆 #561 : spawn additional loggers from single root logger
• 🏆 #560 : zigzag example shows log messages clobbering
• 🏆 #559 : Fix some links
• 🏆 #558 : shortflag for --verbose should be -v, not -a
• 🏆 #557 : Feat/gf1586 integrate real post
• 🏆 #556 : Duplicate flag in paramfetch
• 🏆 #543 : feat(storage-proofs): zizag, drgporep and por abstract over the hasher in circuits
• 🏆 #528 : [WIP] deps: update merkle_light::merkle API
• 🏆 #487 : [WIP] measure memory usage with gperftools
• 🏆 #485 : paramfetch fails from Filecoin install
• 🏆 #458 : Distributed Params - CID pinning
• 🏆 #455 : Cache (expansion?) parents
• 🏆 #446 : [WIP] Implement BBF-VC construction

[porcuquine]

@nicola These links all go to json, which is not really conducive to reading. How are people meant to consume this?

[nicola]

will update, sorry! 🤦‍♂️ It is now fixed