From 89a0287199b62ff454ed6298c31942c286db26c2 Mon Sep 17 00:00:00 2001 From: nkohen Date: Tue, 29 Sep 2020 11:34:41 -0500 Subject: [PATCH 1/4] October meeting doc --- meeting/meeting_october_6_2020.md | 69 +++++++++++++++++++++++++++++++ 1 file changed, 69 insertions(+) create mode 100644 meeting/meeting_october_6_2020.md diff --git a/meeting/meeting_october_6_2020.md b/meeting/meeting_october_6_2020.md new file mode 100644 index 0000000..05cff66 --- /dev/null +++ b/meeting/meeting_october_6_2020.md @@ -0,0 +1,69 @@ +# October 6th (7 PM CST)/7th (9 AM JST) Meeting 2020 + +## Secp256k1 Branch Progress + +* schnorrsig merged into secp256k1 ([#558](https://github.com/bitcoin-core/secp256k1/pull/558))! +* Updated temp-everything branches called new-temp-everything + * [ecdsa adaptor sigs](https://github.com/nkohen/secp256k1/tree/new-temp-everything) + * [with JNI integration](https://github.com/nkohen/secp256k1/tree/new-temp-everything-with-jni) +* ZKP Branch built on top of updated schnorrsig containing ECDSA adaptor sigs - upcoming + * ZKP has new tooling allowing for merging upstream PRs, blockstream folks are in the process of updating ZKP to match upstream + * Once this is done, I will be opening an ECDSA adaptor sig branch which will be permanent (until merged) and replace `new-temp-everything` branch +* BIP 340 support for rust-secp256k1 + * https://github.com/rust-bitcoin/rust-secp256k1/pull/237 + +## Oracle Specifications + +* [URI format proposal](https://github.com/discreetlogcontracts/dlcspecs/pull/63) +* [Descriptor proposal](https://github.com/discreetlogcontracts/dlcspecs/pull/55) +* ["Oracle Address" proposal](https://github.com/discreetlogcontracts/dlcspecs/issues/99) + +## TLV and LN Message Format + +* Turns out these are two separate things and BOLT 1 is unanimously agreed to be vague and confusing +* Matt Corallo suggests we use fragmentation if we want to be LN [BOLT 8](https://github.com/lightningnetwork/lightning-rfc/blob/master/08-transport.md) compatible + * This will likely require specification at some point, but may be general enough to end up in the BOLTs + +## P2P Network Considerations + +* LN doesn't solve all of the P2P problems we had assumed it does (like Network Address Translation) +* Nicholas proposes we use existing TOR client infrastructure +* Alternative is to have users set up their own port forwarding +* Any other ideas? + +## Specification Writing + +* Initial TLV types and deterministic fee computation merged! + * https://github.com/discreetlogcontracts/dlcspecs/pull/81 +* Static test vector PR nearing completion + * https://github.com/discreetlogcontracts/dlcspecs/pull/100 +* Initial proposal for on-chain/non-interactive handling protocol + * ariard has received and responded to initial review + * https://github.com/discreetlogcontracts/dlcspecs/pull/87 +* Lloyd will write a doc like BIP 340 describing our variant of ECDSA Adaptor Signatures +* Oracle Specification + * What should be merged? +* Using SIGHASH_SINGLE (or other) for winner-take-all cases? + * https://github.com/discreetlogcontracts/dlcspecs/issues/91 +* At some point we will need to begin work on the following, any volunteers? + * Non-enumerated outcomes (multiple nonces) + * Multiple-oracle DLCs +* Anything else? + +## Miscellaneous TODOs + +* [TLV Streams](https://github.com/discreetlogcontracts/dlcspecs/issues/73) +* [Dust Limit Computation](#https://github.com/discreetlogcontracts/dlcspecs/issues/11) +* [Tx Input and Output Ordering (using `serial_id`s)](https://github.com/discreetlogcontracts/dlcspecs/issues/18) +* [Links between spec docs that reference each other](https://github.com/discreetlogcontracts/dlcspecs/issues/60) +* [Specify string encodings](https://github.com/discreetlogcontracts/dlcspecs/issues/89) +* [Pretty Pictures!](https://github.com/discreetlogcontracts/dlcspecs/issues/77) +* [Linter for Specs](https://github.com/discreetlogcontracts/dlcspecs/issues/85) +* [Non-prev_tx DLC version](https://github.com/discreetlogcontracts/dlcspecs/issues/98) + +## Lightning DLCs + +* We have begun trying to support more general outputs on commitment transactions in eclair +* Lloyd progress update on witness-asymmetric channels +* Antoine progress update on general outputs in rust-lightning +* Anyone else have anything to report? \ No newline at end of file From c9cfe4c7a1cf0c58b3f8661de3a4af46acbf6d87 Mon Sep 17 00:00:00 2001 From: nkohen Date: Tue, 29 Sep 2020 11:44:24 -0500 Subject: [PATCH 2/4] Made sighash comment less percise and hence less wrong --- meeting/meeting_october_6_2020.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meeting/meeting_october_6_2020.md b/meeting/meeting_october_6_2020.md index 05cff66..3939dfe 100644 --- a/meeting/meeting_october_6_2020.md +++ b/meeting/meeting_october_6_2020.md @@ -43,7 +43,7 @@ * Lloyd will write a doc like BIP 340 describing our variant of ECDSA Adaptor Signatures * Oracle Specification * What should be merged? -* Using SIGHASH_SINGLE (or other) for winner-take-all cases? +* Using SIGHASH_SINGLE (or other sighashes)? * https://github.com/discreetlogcontracts/dlcspecs/issues/91 * At some point we will need to begin work on the following, any volunteers? * Non-enumerated outcomes (multiple nonces) From 593576b917ded66feb6a866990eb01fd008b61c4 Mon Sep 17 00:00:00 2001 From: nkohen Date: Mon, 5 Oct 2020 11:41:31 -0500 Subject: [PATCH 3/4] Some updates --- meeting/meeting_october_6_2020.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/meeting/meeting_october_6_2020.md b/meeting/meeting_october_6_2020.md index 3939dfe..08cfc94 100644 --- a/meeting/meeting_october_6_2020.md +++ b/meeting/meeting_october_6_2020.md @@ -1,5 +1,9 @@ # October 6th (7 PM CST)/7th (9 AM JST) Meeting 2020 +## Housekeeping + +* Discuss the results of the form responses + ## Secp256k1 Branch Progress * schnorrsig merged into secp256k1 ([#558](https://github.com/bitcoin-core/secp256k1/pull/558))! @@ -17,6 +21,7 @@ * [URI format proposal](https://github.com/discreetlogcontracts/dlcspecs/pull/63) * [Descriptor proposal](https://github.com/discreetlogcontracts/dlcspecs/pull/55) * ["Oracle Address" proposal](https://github.com/discreetlogcontracts/dlcspecs/issues/99) +* [Oracle Key Rotation/Public Key Infrastructure discussion](https://github.com/discreetlogcontracts/dlcspecs/issues/93) ## TLV and LN Message Format @@ -45,6 +50,8 @@ * What should be merged? * Using SIGHASH_SINGLE (or other sighashes)? * https://github.com/discreetlogcontracts/dlcspecs/issues/91 +* Tibo wants Antoine to clarify whether change (anchor) outputs on funding tx is a MAY or a MUST + * That is, is it a security issue or a convenience or something else? * At some point we will need to begin work on the following, any volunteers? * Non-enumerated outcomes (multiple nonces) * Multiple-oracle DLCs From 4a18e72ea4a4f0c77c83f626943d020342982cea Mon Sep 17 00:00:00 2001 From: nkohen Date: Wed, 7 Oct 2020 10:07:58 -0500 Subject: [PATCH 4/4] Post meeting updates --- meeting/meeting_october_6_2020.md | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) diff --git a/meeting/meeting_october_6_2020.md b/meeting/meeting_october_6_2020.md index 08cfc94..ed299b4 100644 --- a/meeting/meeting_october_6_2020.md +++ b/meeting/meeting_october_6_2020.md @@ -2,7 +2,7 @@ ## Housekeeping -* Discuss the results of the form responses +* There is a google form asking for feedback about our meeting format ## Secp256k1 Branch Progress @@ -20,8 +20,14 @@ * [URI format proposal](https://github.com/discreetlogcontracts/dlcspecs/pull/63) * [Descriptor proposal](https://github.com/discreetlogcontracts/dlcspecs/pull/55) + * We have decided to move forward with the descriptor proposal where we will add a kind of event descriptor allowing for Lloyd's URI-based events * ["Oracle Address" proposal](https://github.com/discreetlogcontracts/dlcspecs/issues/99) + * We have decided we want a serialized blob which can be copy/pasted that the oracle signs for each event + * Nadav proposes we use TLV format to be inter-operable with P2P messaging, Ben will implement a proposal * [Oracle Key Rotation/Public Key Infrastructure discussion](https://github.com/discreetlogcontracts/dlcspecs/issues/93) + * Everyone agrees that oracles should not have too many events for a single signing key + * Everyone agrees that having multiple keys is part (or all) of the solution + * Not everyone agrees about the idea of rotating publicly known keys ## TLV and LN Message Format @@ -35,6 +41,10 @@ * Nicholas proposes we use existing TOR client infrastructure * Alternative is to have users set up their own port forwarding * Any other ideas? + * BIP 324? + * Impl not moving forward + * Maybe Lloyd wants to do something about this? + * Antoine doesn't like BIP 324 ## Specification Writing @@ -46,15 +56,21 @@ * ariard has received and responded to initial review * https://github.com/discreetlogcontracts/dlcspecs/pull/87 * Lloyd will write a doc like BIP 340 describing our variant of ECDSA Adaptor Signatures + * High level section (motivation, abstract) are higher priority because people ask for resources + * Actual variant specification is lower priority as everyone is using the same binaries right now * Oracle Specification - * What should be merged? + * See above * Using SIGHASH_SINGLE (or other sighashes)? * https://github.com/discreetlogcontracts/dlcspecs/issues/91 + * Contract flag bit? + * Double CETs? + * Discussion to be continued online * Tibo wants Antoine to clarify whether change (anchor) outputs on funding tx is a MAY or a MUST * That is, is it a security issue or a convenience or something else? * At some point we will need to begin work on the following, any volunteers? * Non-enumerated outcomes (multiple nonces) * Multiple-oracle DLCs + * Antoine volunteers to write a doc for [client-side oracle validation](https://github.com/discreetlogcontracts/dlcspecs/issues/97) * Anything else? ## Miscellaneous TODOs @@ -67,6 +83,7 @@ * [Pretty Pictures!](https://github.com/discreetlogcontracts/dlcspecs/issues/77) * [Linter for Specs](https://github.com/discreetlogcontracts/dlcspecs/issues/85) * [Non-prev_tx DLC version](https://github.com/discreetlogcontracts/dlcspecs/issues/98) +* [I've been hacked message](https://github.com/discreetlogcontracts/dlcspecs/issues/94) ## Lightning DLCs