{
  "schemaVersion": 1,
  "videoId": "fd9DFOBbrP8",
  "videoUrl": "https://www.youtube.com/watch?v=fd9DFOBbrP8",
  "title": "Effectively exploring and testing state space of consensus protocols",
  "speakerCredit": "Aarnav Bos",
  "programmeTalkId": "dss26-e1ea70e96d138ab7",
  "language": "en",
  "captionType": "auto-generated",
  "humanAccuracyReviewed": false,
  "spokenAudioCompletenessVerified": false,
  "cueEndTimesAvailable": false,
  "source": {
    "provider": "YouTube",
    "method": "native browser transcript export",
    "sha256": "199ba4c8dfc6bc676d60ff2202110f51c09dd7acd42c07dcb851d51084c27aaf",
    "exportedAt": "2026-11-16T07:35:32.656950+00:00"
  },
  "qualityCaveats": [
    "Names, organizations, technical terms, code and numbers may be misrecognized.",
    "Source wording and disfluencies are preserved; no editorial corrections were applied.",
    "Export contains whole-second cue starts only; cue ends and subsecond timing are unavailable.",
    "A full caption-track export does not establish that every spoken word is represented.",
    "Automatic YouTube captions; not human-verified against the audio."
  ],
  "timingValidation": {
    "first_cue_start_seconds": 0,
    "last_cue_start_seconds": 330,
    "video_duration_seconds": 333,
    "seconds_after_last_cue_start": 3,
    "cue_start_gaps_over_30_seconds": [],
    "boundary_check": "pass",
    "criterion": "first cue <=15 seconds; last cue start within 30 seconds of player duration, allowing 1-second precision",
    "meaning": "Timing plausibility only; silence/applause or source caption omissions require audio review."
  },
  "segments": [
    {
      "index": 0,
      "start_seconds": 0,
      "source_timestamp": "0:00",
      "end_seconds": null,
      "text": "Um thanks for coming everybody. I'm"
    },
    {
      "index": 1,
      "start_seconds": 2,
      "source_timestamp": "0:02",
      "end_seconds": null,
      "text": "pretty excited to present um some of the"
    },
    {
      "index": 2,
      "start_seconds": 5,
      "source_timestamp": "0:05",
      "end_seconds": null,
      "text": "work my colleagues and I have been doing"
    },
    {
      "index": 3,
      "start_seconds": 7,
      "source_timestamp": "0:07",
      "end_seconds": null,
      "text": "while testing Alpenlow which is Salana's"
    },
    {
      "index": 4,
      "start_seconds": 10,
      "source_timestamp": "0:10",
      "end_seconds": null,
      "text": "new consensus protocol. I'll be"
    },
    {
      "index": 5,
      "start_seconds": 12,
      "source_timestamp": "0:12",
      "end_seconds": null,
      "text": "presenting two approaches bisfuzz and"
    },
    {
      "index": 6,
      "start_seconds": 14,
      "source_timestamp": "0:14",
      "end_seconds": null,
      "text": "forkraph and how we measured coverage."
    },
    {
      "index": 7,
      "start_seconds": 17,
      "source_timestamp": "0:17",
      "end_seconds": null,
      "text": "So state coverage for both of these"
    },
    {
      "index": 8,
      "start_seconds": 19,
      "source_timestamp": "0:19",
      "end_seconds": null,
      "text": "approaches. A quick overview on openlow."
    },
    {
      "index": 9,
      "start_seconds": 22,
      "source_timestamp": "0:22",
      "end_seconds": null,
      "text": "It's um yeah new consensus protocol"
    },
    {
      "index": 10,
      "start_seconds": 25,
      "source_timestamp": "0:25",
      "end_seconds": null,
      "text": "which has parallel processing of blocks"
    },
    {
      "index": 11,
      "start_seconds": 27,
      "source_timestamp": "0:27",
      "end_seconds": null,
      "text": "complex interactions between components"
    },
    {
      "index": 12,
      "start_seconds": 29,
      "source_timestamp": "0:29",
      "end_seconds": null,
      "text": "and is built with fast finality in mind."
    },
    {
      "index": 13,
      "start_seconds": 32,
      "source_timestamp": "0:32",
      "end_seconds": null,
      "text": "So while testing we have throughput"
    },
    {
      "index": 14,
      "start_seconds": 34,
      "source_timestamp": "0:34",
      "end_seconds": null,
      "text": "correctness and parallelism right this"
    },
    {
      "index": 15,
      "start_seconds": 37,
      "source_timestamp": "0:37",
      "end_seconds": null,
      "text": "is difficult to test effectively but"
    },
    {
      "index": 16,
      "start_seconds": 39,
      "source_timestamp": "0:39",
      "end_seconds": null,
      "text": "like we have to be very rigorous and an"
    },
    {
      "index": 17,
      "start_seconds": 42,
      "source_timestamp": "0:42",
      "end_seconds": null,
      "text": "overview on the assurance asymmetric"
    },
    {
      "index": 18,
      "start_seconds": 44,
      "source_timestamp": "0:44",
      "end_seconds": null,
      "text": "research is a security partner for the"
    },
    {
      "index": 19,
      "start_seconds": 47,
      "source_timestamp": "0:47",
      "end_seconds": null,
      "text": "Salana foundation and anza we do"
    },
    {
      "index": 20,
      "start_seconds": 49,
      "source_timestamp": "0:49",
      "end_seconds": null,
      "text": "continuous assessment of changes and"
    },
    {
      "index": 21,
      "start_seconds": 51,
      "source_timestamp": "0:51",
      "end_seconds": null,
      "text": "fixes for anza and we also did a"
    },
    {
      "index": 22,
      "start_seconds": 54,
      "source_timestamp": "0:54",
      "end_seconds": null,
      "text": "differential review with fire dancer"
    },
    {
      "index": 23,
      "start_seconds": 56,
      "source_timestamp": "0:56",
      "end_seconds": null,
      "text": "which is another client for Salana and"
    },
    {
      "index": 24,
      "start_seconds": 59,
      "source_timestamp": "0:59",
      "end_seconds": null,
      "text": "for this we needed to build reusable and"
    },
    {
      "index": 25,
      "start_seconds": 61,
      "source_timestamp": "1:01",
      "end_seconds": null,
      "text": "scalable tooling Right? Because you when"
    },
    {
      "index": 26,
      "start_seconds": 63,
      "source_timestamp": "1:03",
      "end_seconds": null,
      "text": "you do a continuous audit, you just you"
    },
    {
      "index": 27,
      "start_seconds": 65,
      "source_timestamp": "1:05",
      "end_seconds": null,
      "text": "don't want to duplicate efforts"
    },
    {
      "index": 28,
      "start_seconds": 67,
      "source_timestamp": "1:07",
      "end_seconds": null,
      "text": "essentially. Um I'd start with bisfuzz"
    },
    {
      "index": 29,
      "start_seconds": 70,
      "source_timestamp": "1:10",
      "end_seconds": null,
      "text": "which is um I think a well-known paper"
    },
    {
      "index": 30,
      "start_seconds": 72,
      "source_timestamp": "1:12",
      "end_seconds": null,
      "text": "in testing consensus protocols um where"
    },
    {
      "index": 31,
      "start_seconds": 75,
      "source_timestamp": "1:15",
      "end_seconds": null,
      "text": "you initialize the state of a network"
    },
    {
      "index": 32,
      "start_seconds": 77,
      "source_timestamp": "1:17",
      "end_seconds": null,
      "text": "and then you perform a set of actions on"
    },
    {
      "index": 33,
      "start_seconds": 79,
      "source_timestamp": "1:19",
      "end_seconds": null,
      "text": "it. And our assumption is that the bugs"
    },
    {
      "index": 34,
      "start_seconds": 83,
      "source_timestamp": "1:23",
      "end_seconds": null,
      "text": "lie in the interactions or in the"
    },
    {
      "index": 35,
      "start_seconds": 86,
      "source_timestamp": "1:26",
      "end_seconds": null,
      "text": "formation of the state, right? And as"
    },
    {
      "index": 36,
      "start_seconds": 88,
      "source_timestamp": "1:28",
      "end_seconds": null,
      "text": "you can see here, these are like all the"
    },
    {
      "index": 37,
      "start_seconds": 89,
      "source_timestamp": "1:29",
      "end_seconds": null,
      "text": "different kinds of configurations for"
    },
    {
      "index": 38,
      "start_seconds": 91,
      "source_timestamp": "1:31",
      "end_seconds": null,
      "text": "the network. And to cover these"
    },
    {
      "index": 39,
      "start_seconds": 93,
      "source_timestamp": "1:33",
      "end_seconds": null,
      "text": "exhaustively, we'd need like 1,024"
    },
    {
      "index": 40,
      "start_seconds": 96,
      "source_timestamp": "1:36",
      "end_seconds": null,
      "text": "tests, which honestly like takes around"
    },
    {
      "index": 41,
      "start_seconds": 99,
      "source_timestamp": "1:39",
      "end_seconds": null,
      "text": "3 hours considering that each test is 10"
    },
    {
      "index": 42,
      "start_seconds": 102,
      "source_timestamp": "1:42",
      "end_seconds": null,
      "text": "seconds because you have to run a"
    },
    {
      "index": 43,
      "start_seconds": 103,
      "source_timestamp": "1:43",
      "end_seconds": null,
      "text": "cluster, you have to run a node and so"
    },
    {
      "index": 44,
      "start_seconds": 105,
      "source_timestamp": "1:45",
      "end_seconds": null,
      "text": "on. Now this gets pretty difficult to do"
    },
    {
      "index": 45,
      "start_seconds": 107,
      "source_timestamp": "1:47",
      "end_seconds": null,
      "text": "in a continuous fashion, right? So what"
    },
    {
      "index": 46,
      "start_seconds": 110,
      "source_timestamp": "1:50",
      "end_seconds": null,
      "text": "you can then do is use something called"
    },
    {
      "index": 47,
      "start_seconds": 112,
      "source_timestamp": "1:52",
      "end_seconds": null,
      "text": "T-way coverage. Um this is a research um"
    },
    {
      "index": 48,
      "start_seconds": 116,
      "source_timestamp": "1:56",
      "end_seconds": null,
      "text": "under the umbrella of combinatorial"
    },
    {
      "index": 49,
      "start_seconds": 117,
      "source_timestamp": "1:57",
      "end_seconds": null,
      "text": "testing where you select a subset of"
    },
    {
      "index": 50,
      "start_seconds": 121,
      "source_timestamp": "2:01",
      "end_seconds": null,
      "text": "interactions. So since we believe that"
    },
    {
      "index": 51,
      "start_seconds": 123,
      "source_timestamp": "2:03",
      "end_seconds": null,
      "text": "the interactions are in the way the"
    },
    {
      "index": 52,
      "start_seconds": 125,
      "source_timestamp": "2:05",
      "end_seconds": null,
      "text": "state is formed or the bugs are um in"
    },
    {
      "index": 53,
      "start_seconds": 128,
      "source_timestamp": "2:08",
      "end_seconds": null,
      "text": "the way the state is formed, you can say"
    },
    {
      "index": 54,
      "start_seconds": 130,
      "source_timestamp": "2:10",
      "end_seconds": null,
      "text": "okay I have tested window start with a"
    },
    {
      "index": 55,
      "start_seconds": 133,
      "source_timestamp": "2:13",
      "end_seconds": null,
      "text": "50/50 partition where everyone receives"
    },
    {
      "index": 56,
      "start_seconds": 136,
      "source_timestamp": "2:16",
      "end_seconds": null,
      "text": "a corrupted message. I have done a"
    },
    {
      "index": 57,
      "start_seconds": 138,
      "source_timestamp": "2:18",
      "end_seconds": null,
      "text": "window start um where the first round is"
    },
    {
      "index": 58,
      "start_seconds": 143,
      "source_timestamp": "2:23",
      "end_seconds": null,
      "text": "mutated and again everyone gets the"
    },
    {
      "index": 59,
      "start_seconds": 145,
      "source_timestamp": "2:25",
      "end_seconds": null,
      "text": "message. So it's not about exhaustively"
    },
    {
      "index": 60,
      "start_seconds": 149,
      "source_timestamp": "2:29",
      "end_seconds": null,
      "text": "covering every feature with every um"
    },
    {
      "index": 61,
      "start_seconds": 153,
      "source_timestamp": "2:33",
      "end_seconds": null,
      "text": "parameter with every value sorry but"
    },
    {
      "index": 62,
      "start_seconds": 155,
      "source_timestamp": "2:35",
      "end_seconds": null,
      "text": "it's about a subset of parameters with"
    },
    {
      "index": 63,
      "start_seconds": 157,
      "source_timestamp": "2:37",
      "end_seconds": null,
      "text": "all values and this again reduces the"
    },
    {
      "index": 64,
      "start_seconds": 160,
      "source_timestamp": "2:40",
      "end_seconds": null,
      "text": "search space gives you some guarantee"
    },
    {
      "index": 65,
      "start_seconds": 162,
      "source_timestamp": "2:42",
      "end_seconds": null,
      "text": "and allows you to measure the state"
    },
    {
      "index": 66,
      "start_seconds": 164,
      "source_timestamp": "2:44",
      "end_seconds": null,
      "text": "space. Um, and this reduced our testing"
    },
    {
      "index": 67,
      "start_seconds": 167,
      "source_timestamp": "2:47",
      "end_seconds": null,
      "text": "time to 10 minutes, which was pretty"
    },
    {
      "index": 68,
      "start_seconds": 168,
      "source_timestamp": "2:48",
      "end_seconds": null,
      "text": "reasonable to do as part of continuous"
    },
    {
      "index": 69,
      "start_seconds": 170,
      "source_timestamp": "2:50",
      "end_seconds": null,
      "text": "integration, right? And it found a lot"
    },
    {
      "index": 70,
      "start_seconds": 172,
      "source_timestamp": "2:52",
      "end_seconds": null,
      "text": "of bugs and allowed us to be sure of"
    },
    {
      "index": 71,
      "start_seconds": 174,
      "source_timestamp": "2:54",
      "end_seconds": null,
      "text": "what we had tested. Um, and I think one"
    },
    {
      "index": 72,
      "start_seconds": 178,
      "source_timestamp": "2:58",
      "end_seconds": null,
      "text": "thing that's important to mention is"
    },
    {
      "index": 73,
      "start_seconds": 179,
      "source_timestamp": "2:59",
      "end_seconds": null,
      "text": "generating these um, parameters and"
    },
    {
      "index": 74,
      "start_seconds": 182,
      "source_timestamp": "3:02",
      "end_seconds": null,
      "text": "values can be done by a fuzzer, they can"
    },
    {
      "index": 75,
      "start_seconds": 184,
      "source_timestamp": "3:04",
      "end_seconds": null,
      "text": "be done randomly, they can be done using"
    },
    {
      "index": 76,
      "start_seconds": 185,
      "source_timestamp": "3:05",
      "end_seconds": null,
      "text": "genetic algorithms. It's really up to"
    },
    {
      "index": 77,
      "start_seconds": 187,
      "source_timestamp": "3:07",
      "end_seconds": null,
      "text": "you. This is just one way to measure it."
    },
    {
      "index": 78,
      "start_seconds": 191,
      "source_timestamp": "3:11",
      "end_seconds": null,
      "text": "Um another uh approach that we took to"
    },
    {
      "index": 79,
      "start_seconds": 194,
      "source_timestamp": "3:14",
      "end_seconds": null,
      "text": "test was using the classic fork graph."
    },
    {
      "index": 80,
      "start_seconds": 197,
      "source_timestamp": "3:17",
      "end_seconds": null,
      "text": "Um since we have parallelism um our"
    },
    {
      "index": 81,
      "start_seconds": 199,
      "source_timestamp": "3:19",
      "end_seconds": null,
      "text": "assumption is that the bugs lie in the"
    },
    {
      "index": 82,
      "start_seconds": 202,
      "source_timestamp": "3:22",
      "end_seconds": null,
      "text": "order of events and since you know over"
    },
    {
      "index": 83,
      "start_seconds": 205,
      "source_timestamp": "3:25",
      "end_seconds": null,
      "text": "the network there's no guarantee in the"
    },
    {
      "index": 84,
      "start_seconds": 206,
      "source_timestamp": "3:26",
      "end_seconds": null,
      "text": "order. Um we want to test state after a"
    },
    {
      "index": 85,
      "start_seconds": 211,
      "source_timestamp": "3:31",
      "end_seconds": null,
      "text": "series of actions. Now this is pretty"
    },
    {
      "index": 86,
      "start_seconds": 212,
      "source_timestamp": "3:32",
      "end_seconds": null,
      "text": "common with interpreters where executing"
    },
    {
      "index": 87,
      "start_seconds": 214,
      "source_timestamp": "3:34",
      "end_seconds": null,
      "text": "bite code or any sort of state machine,"
    },
    {
      "index": 88,
      "start_seconds": 216,
      "source_timestamp": "3:36",
      "end_seconds": null,
      "text": "right? So sequence coverage can be used"
    },
    {
      "index": 89,
      "start_seconds": 219,
      "source_timestamp": "3:39",
      "end_seconds": null,
      "text": "for that. And to give you an example,"
    },
    {
      "index": 90,
      "start_seconds": 221,
      "source_timestamp": "3:41",
      "end_seconds": null,
      "text": "let's say we have eight slots with two"
    },
    {
      "index": 91,
      "start_seconds": 223,
      "source_timestamp": "3:43",
      "end_seconds": null,
      "text": "blocks each cuz everyone loves"
    },
    {
      "index": 92,
      "start_seconds": 225,
      "source_timestamp": "3:45",
      "end_seconds": null,
      "text": "equivocation. Um, and five different"
    },
    {
      "index": 93,
      "start_seconds": 228,
      "source_timestamp": "3:48",
      "end_seconds": null,
      "text": "resolutions of how a block can end. You"
    },
    {
      "index": 94,
      "start_seconds": 230,
      "source_timestamp": "3:50",
      "end_seconds": null,
      "text": "have like a finalized block, you have a"
    },
    {
      "index": 95,
      "start_seconds": 231,
      "source_timestamp": "3:51",
      "end_seconds": null,
      "text": "skip block, you have a block with a"
    },
    {
      "index": 96,
      "start_seconds": 233,
      "source_timestamp": "3:53",
      "end_seconds": null,
      "text": "fallback, and maybe just a block that"
    },
    {
      "index": 97,
      "start_seconds": 235,
      "source_timestamp": "3:55",
      "end_seconds": null,
      "text": "was proposed, but no one actually voted"
    },
    {
      "index": 98,
      "start_seconds": 236,
      "source_timestamp": "3:56",
      "end_seconds": null,
      "text": "on it. This is like an insanely large"
    },
    {
      "index": 99,
      "start_seconds": 239,
      "source_timestamp": "3:59",
      "end_seconds": null,
      "text": "number to test exhaustively, right? So"
    },
    {
      "index": 100,
      "start_seconds": 242,
      "source_timestamp": "4:02",
      "end_seconds": null,
      "text": "what you can then do is use something"
    },
    {
      "index": 101,
      "start_seconds": 243,
      "source_timestamp": "4:03",
      "end_seconds": null,
      "text": "called sequence coverage where you say"
    },
    {
      "index": 102,
      "start_seconds": 246,
      "source_timestamp": "4:06",
      "end_seconds": null,
      "text": "okay I don't care about the adjacency of"
    },
    {
      "index": 103,
      "start_seconds": 249,
      "source_timestamp": "4:09",
      "end_seconds": null,
      "text": "the events but I care that they happen"
    },
    {
      "index": 104,
      "start_seconds": 250,
      "source_timestamp": "4:10",
      "end_seconds": null,
      "text": "in order. So suppose for the same block"
    },
    {
      "index": 105,
      "start_seconds": 254,
      "source_timestamp": "4:14",
      "end_seconds": null,
      "text": "one uh for the same slots sorry you get"
    },
    {
      "index": 106,
      "start_seconds": 256,
      "source_timestamp": "4:16",
      "end_seconds": null,
      "text": "block 1 B and 1 A while 1 B is complete"
    },
    {
      "index": 107,
      "start_seconds": 261,
      "source_timestamp": "4:21",
      "end_seconds": null,
      "text": "but 1 A is finalized. So you should"
    },
    {
      "index": 108,
      "start_seconds": 264,
      "source_timestamp": "4:24",
      "end_seconds": null,
      "text": "still be able to finalize after you get"
    },
    {
      "index": 109,
      "start_seconds": 266,
      "source_timestamp": "4:26",
      "end_seconds": null,
      "text": "a competing block first slot. Um yeah,"
    },
    {
      "index": 110,
      "start_seconds": 270,
      "source_timestamp": "4:30",
      "end_seconds": null,
      "text": "in general. So this tests the order of"
    },
    {
      "index": 111,
      "start_seconds": 272,
      "source_timestamp": "4:32",
      "end_seconds": null,
      "text": "events but not adjacency. Um and again"
    },
    {
      "index": 112,
      "start_seconds": 274,
      "source_timestamp": "4:34",
      "end_seconds": null,
      "text": "allows you to reduce the amount of tests"
    },
    {
      "index": 113,
      "start_seconds": 276,
      "source_timestamp": "4:36",
      "end_seconds": null,
      "text": "that you need to do and allows you to"
    },
    {
      "index": 114,
      "start_seconds": 278,
      "source_timestamp": "4:38",
      "end_seconds": null,
      "text": "measure what you've tested."
    },
    {
      "index": 115,
      "start_seconds": 281,
      "source_timestamp": "4:41",
      "end_seconds": null,
      "text": "Um so to summarize, if you want to test"
    },
    {
      "index": 116,
      "start_seconds": 284,
      "source_timestamp": "4:44",
      "end_seconds": null,
      "text": "actions after state, so you want to"
    },
    {
      "index": 117,
      "start_seconds": 286,
      "source_timestamp": "4:46",
      "end_seconds": null,
      "text": "measure the diversity of your state, you"
    },
    {
      "index": 118,
      "start_seconds": 288,
      "source_timestamp": "4:48",
      "end_seconds": null,
      "text": "use TVA coverage. And if you want to"
    },
    {
      "index": 119,
      "start_seconds": 290,
      "source_timestamp": "4:50",
      "end_seconds": null,
      "text": "measure um or if you want to test after"
    },
    {
      "index": 120,
      "start_seconds": 294,
      "source_timestamp": "4:54",
      "end_seconds": null,
      "text": "um if you want to test your state after"
    },
    {
      "index": 121,
      "start_seconds": 296,
      "source_timestamp": "4:56",
      "end_seconds": null,
      "text": "a series of actions, you would use"
    },
    {
      "index": 122,
      "start_seconds": 297,
      "source_timestamp": "4:57",
      "end_seconds": null,
      "text": "sequence coverage."
    },
    {
      "index": 123,
      "start_seconds": 300,
      "source_timestamp": "5:00",
      "end_seconds": null,
      "text": "Um yeah, one more thing to mention is"
    },
    {
      "index": 124,
      "start_seconds": 303,
      "source_timestamp": "5:03",
      "end_seconds": null,
      "text": "that this suffers the same problem as"
    },
    {
      "index": 125,
      "start_seconds": 306,
      "source_timestamp": "5:06",
      "end_seconds": null,
      "text": "fuzzing does, right? Where you need to"
    },
    {
      "index": 126,
      "start_seconds": 307,
      "source_timestamp": "5:07",
      "end_seconds": null,
      "text": "model your input space and you still are"
    },
    {
      "index": 127,
      "start_seconds": 310,
      "source_timestamp": "5:10",
      "end_seconds": null,
      "text": "constrained by time. But I think this"
    },
    {
      "index": 128,
      "start_seconds": 313,
      "source_timestamp": "5:13",
      "end_seconds": null,
      "text": "offers a pretty good compromise on"
    },
    {
      "index": 129,
      "start_seconds": 317,
      "source_timestamp": "5:17",
      "end_seconds": null,
      "text": "knowing what you're testing, right? cuz"
    },
    {
      "index": 130,
      "start_seconds": 318,
      "source_timestamp": "5:18",
      "end_seconds": null,
      "text": "if you just have like a series of"
    },
    {
      "index": 131,
      "start_seconds": 320,
      "source_timestamp": "5:20",
      "end_seconds": null,
      "text": "events, you don't really know how much"
    },
    {
      "index": 132,
      "start_seconds": 322,
      "source_timestamp": "5:22",
      "end_seconds": null,
      "text": "of the state space you've tested. Thank"
    },
    {
      "index": 133,
      "start_seconds": 325,
      "source_timestamp": "5:25",
      "end_seconds": null,
      "text": "you. And happy to chat more after the"
    },
    {
      "index": 134,
      "start_seconds": 328,
      "source_timestamp": "5:28",
      "end_seconds": null,
      "text": "talk if anybody has any questions about"
    },
    {
      "index": 135,
      "start_seconds": 330,
      "source_timestamp": "5:30",
      "end_seconds": null,
      "text": "applicability or whatever."
    }
  ]
}
