WEBVTT Kind: captions Language: en 00:00:14.379 --> 00:00:18.670 align:start position:0% hmm<00:00:15.379> hey<00:00:16.039> everyone<00:00:17.170> so<00:00:18.170> this<00:00:18.200> time<00:00:18.440> I'm<00:00:18.560> gonna 00:00:18.670 --> 00:00:18.680 align:start position:0% hmm hey everyone so this time I'm gonna 00:00:18.680 --> 00:00:21.730 align:start position:0% hmm hey everyone so this time I'm gonna start<00:00:19.250> my<00:00:19.400> talk<00:00:19.580> for<00:00:19.790> real<00:00:20.020> no<00:00:21.020> no<00:00:21.410> kidding 00:00:21.730 --> 00:00:21.740 align:start position:0% start my talk for real no no kidding 00:00:21.740 --> 00:00:23.620 align:start position:0% start my talk for real no no kidding around<00:00:21.830> we're<00:00:22.670> gonna<00:00:22.730> hear<00:00:22.910> about<00:00:22.970> a<00:00:23.120> crack<00:00:23.540> by 00:00:23.620 --> 00:00:23.630 align:start position:0% around we're gonna hear about a crack by 00:00:23.630 --> 00:00:24.970 align:start position:0% around we're gonna hear about a crack by construction<00:00:23.960> block<00:00:24.230> chain<00:00:24.410> consensus 00:00:24.970 --> 00:00:24.980 align:start position:0% construction block chain consensus 00:00:24.980 --> 00:00:27.340 align:start position:0% construction block chain consensus protocol<00:00:25.310> and<00:00:25.580> you<00:00:26.480> know<00:00:26.570> I<00:00:26.810> hope<00:00:26.990> you're<00:00:27.170> all 00:00:27.340 --> 00:00:27.350 align:start position:0% protocol and you know I hope you're all 00:00:27.350 --> 00:00:29.670 align:start position:0% protocol and you know I hope you're all excited<00:00:28.220> to<00:00:28.490> be<00:00:28.580> here<00:00:28.760> and<00:00:28.970> I<00:00:29.060> hope<00:00:29.150> you're<00:00:29.360> all 00:00:29.670 --> 00:00:29.680 align:start position:0% excited to be here and I hope you're all 00:00:29.680 --> 00:00:32.080 align:start position:0% excited to be here and I hope you're all aware<00:00:30.680> that<00:00:30.950> I'm<00:00:31.070> starting<00:00:31.370> to<00:00:31.400> speak<00:00:31.700> and<00:00:31.940> so 00:00:32.080 --> 00:00:32.090 align:start position:0% aware that I'm starting to speak and so 00:00:32.090 --> 00:00:35.950 align:start position:0% aware that I'm starting to speak and so are<00:00:32.239> coming<00:00:32.570> into<00:00:32.689> the<00:00:32.840> room<00:00:34.510> but<00:00:35.510> I<00:00:35.750> don't 00:00:35.950 --> 00:00:35.960 align:start position:0% are coming into the room but I don't 00:00:35.960 --> 00:00:39.540 align:start position:0% are coming into the room but I don't have<00:00:36.050> time<00:00:36.230> to<00:00:36.380> wait<00:00:36.530> so<00:00:37.250> here<00:00:37.820> we<00:00:37.910> go<00:00:38.050> so 00:00:39.540 --> 00:00:39.550 align:start position:0% have time to wait so here we go so 00:00:39.550 --> 00:00:41.800 align:start position:0% have time to wait so here we go so firstly<00:00:40.550> let's<00:00:40.790> do<00:00:41.180> a<00:00:41.210> basic<00:00:41.390> introduction 00:00:41.800 --> 00:00:41.810 align:start position:0% firstly let's do a basic introduction 00:00:41.810 --> 00:00:43.390 align:start position:0% firstly let's do a basic introduction like<00:00:42.290> consensus<00:00:42.710> protocols<00:00:43.100> are<00:00:43.190> kind<00:00:43.340> of 00:00:43.390 --> 00:00:43.400 align:start position:0% like consensus protocols are kind of 00:00:43.400 --> 00:00:47.020 align:start position:0% like consensus protocols are kind of fundamentally<00:00:44.360> about<00:00:45.370> deciding<00:00:46.370> like<00:00:46.790> notes 00:00:47.020 --> 00:00:47.030 align:start position:0% fundamentally about deciding like notes 00:00:47.030 --> 00:00:48.639 align:start position:0% fundamentally about deciding like notes on<00:00:47.180> a<00:00:47.210> distributed<00:00:47.660> network<00:00:47.870> making<00:00:48.290> the<00:00:48.470> same 00:00:48.639 --> 00:00:48.649 align:start position:0% on a distributed network making the same 00:00:48.649 --> 00:00:50.110 align:start position:0% on a distributed network making the same decisions<00:00:49.070> even<00:00:49.820> though<00:00:49.880> they<00:00:49.970> might<00:00:50.060> have 00:00:50.110 --> 00:00:50.120 align:start position:0% decisions even though they might have 00:00:50.120 --> 00:00:52.080 align:start position:0% decisions even though they might have different<00:00:50.420> views<00:00:50.780> of<00:00:50.899> the<00:00:50.960> network<00:00:51.260> right<00:00:51.649> and 00:00:52.080 --> 00:00:52.090 align:start position:0% different views of the network right and 00:00:52.090 --> 00:00:54.250 align:start position:0% different views of the network right and those<00:00:53.090> decisions<00:00:53.450> you<00:00:53.809> think<00:00:54.020> of<00:00:54.079> them<00:00:54.170> as 00:00:54.250 --> 00:00:54.260 align:start position:0% those decisions you think of them as 00:00:54.260 --> 00:00:56.170 align:start position:0% those decisions you think of them as being<00:00:54.350> on<00:00:54.530> values<00:00:54.920> but<00:00:55.280> normally<00:00:55.700> you<00:00:55.850> know<00:00:55.970> we 00:00:56.170 --> 00:00:56.180 align:start position:0% being on values but normally you know we 00:00:56.180 --> 00:00:58.600 align:start position:0% being on values but normally you know we think<00:00:56.300> about<00:00:56.390> in<00:00:56.840> our<00:00:57.410> context<00:00:57.890> replicating<00:00:58.520> a 00:00:58.600 --> 00:00:58.610 align:start position:0% think about in our context replicating a 00:00:58.610 --> 00:00:59.889 align:start position:0% think about in our context replicating a virtual<00:00:58.880> machine<00:00:59.000> or<00:00:59.120> some<00:00:59.510> kind<00:00:59.660> of<00:00:59.719> state 00:00:59.889 --> 00:00:59.899 align:start position:0% virtual machine or some kind of state 00:00:59.899 --> 00:01:05.350 align:start position:0% virtual machine or some kind of state machine<00:01:01.000> so<00:01:04.030> traditional<00:01:05.030> consensus 00:01:05.350 --> 00:01:05.360 align:start position:0% machine so traditional consensus 00:01:05.360 --> 00:01:07.840 align:start position:0% machine so traditional consensus protocols<00:01:05.750> decide<00:01:06.229> on<00:01:06.409> one<00:01:07.190> block<00:01:07.490> of 00:01:07.840 --> 00:01:07.850 align:start position:0% protocols decide on one block of 00:01:07.850 --> 00:01:12.840 align:start position:0% protocols decide on one block of transactions<00:01:08.360> at<00:01:08.479> a<00:01:08.510> time<00:01:09.340> whereas 00:01:12.840 --> 00:01:12.850 align:start position:0% 00:01:12.850 --> 00:01:14.980 align:start position:0% and<00:01:13.850> they<00:01:14.150> do<00:01:14.270> it<00:01:14.510> with<00:01:14.659> irreversible 00:01:14.980 --> 00:01:14.990 align:start position:0% and they do it with irreversible 00:01:14.990 --> 00:01:17.440 align:start position:0% and they do it with irreversible finality<00:01:15.890> with<00:01:16.100> every<00:01:16.520> single<00:01:16.940> block<00:01:17.119> right 00:01:17.440 --> 00:01:17.450 align:start position:0% finality with every single block right 00:01:17.450 --> 00:01:19.120 align:start position:0% finality with every single block right you<00:01:17.840> come<00:01:18.080> to<00:01:18.170> a<00:01:18.200> Byzantine<00:01:18.530> quorum<00:01:18.830> you<00:01:18.920> sign 00:01:19.120 --> 00:01:19.130 align:start position:0% you come to a Byzantine quorum you sign 00:01:19.130 --> 00:01:21.160 align:start position:0% you come to a Byzantine quorum you sign the<00:01:19.250> block<00:01:19.430> and<00:01:19.640> you<00:01:19.880> and<00:01:20.090> every<00:01:20.450> block<00:01:20.630> is 00:01:21.160 --> 00:01:21.170 align:start position:0% the block and you and every block is 00:01:21.170 --> 00:01:24.180 align:start position:0% the block and you and every block is created<00:01:21.560> and<00:01:21.650> there's<00:01:21.860> never<00:01:21.979> any<00:01:22.159> Forks<00:01:22.460> and 00:01:24.180 --> 00:01:24.190 align:start position:0% created and there's never any Forks and 00:01:24.190 --> 00:01:26.649 align:start position:0% created and there's never any Forks and this<00:01:25.190> usually<00:01:25.580> requires<00:01:25.909> on<00:01:26.240> the<00:01:26.390> order<00:01:26.630> of 00:01:26.649 --> 00:01:26.659 align:start position:0% this usually requires on the order of 00:01:26.659 --> 00:01:28.600 align:start position:0% this usually requires on the order of the<00:01:26.750> number<00:01:26.930> of<00:01:27.020> validators<00:01:27.530> messages<00:01:28.400> for 00:01:28.600 --> 00:01:28.610 align:start position:0% the number of validators messages for 00:01:28.610 --> 00:01:30.490 align:start position:0% the number of validators messages for every<00:01:28.850> block<00:01:29.060> to<00:01:29.930> accomplish<00:01:30.229> this<00:01:30.320> right 00:01:30.490 --> 00:01:30.500 align:start position:0% every block to accomplish this right 00:01:30.500 --> 00:01:31.630 align:start position:0% every block to accomplish this right because<00:01:30.650> you<00:01:30.710> have<00:01:30.799> to<00:01:30.860> come<00:01:31.010> to<00:01:31.189> consensus 00:01:31.630 --> 00:01:31.640 align:start position:0% because you have to come to consensus 00:01:31.640 --> 00:01:33.510 align:start position:0% because you have to come to consensus before<00:01:31.820> you<00:01:32.090> make<00:01:32.210> every<00:01:32.390> block<00:01:32.600> essentially 00:01:33.510 --> 00:01:33.520 align:start position:0% before you make every block essentially 00:01:33.520 --> 00:01:35.500 align:start position:0% before you make every block essentially improve<00:01:34.520> award<00:01:34.820> consensus<00:01:35.210> on<00:01:35.330> the<00:01:35.390> other 00:01:35.500 --> 00:01:35.510 align:start position:0% improve award consensus on the other 00:01:35.510 --> 00:01:38.380 align:start position:0% improve award consensus on the other hand<00:01:35.860> we<00:01:36.860> don't<00:01:37.400> come<00:01:37.700> to<00:01:37.790> consensus<00:01:38.119> on<00:01:38.210> every 00:01:38.380 --> 00:01:38.390 align:start position:0% hand we don't come to consensus on every 00:01:38.390 --> 00:01:40.030 align:start position:0% hand we don't come to consensus on every block<00:01:38.540> before<00:01:38.689> we<00:01:38.900> make<00:01:39.020> it<00:01:39.049> instead<00:01:39.619> we<00:01:39.860> have 00:01:40.030 --> 00:01:40.040 align:start position:0% block before we make it instead we have 00:01:40.040 --> 00:01:42.429 align:start position:0% block before we make it instead we have this<00:01:40.250> kind<00:01:40.460> of<00:01:40.490> forking<00:01:41.210> phenomena<00:01:41.750> where<00:01:42.020> we 00:01:42.429 --> 00:01:42.439 align:start position:0% this kind of forking phenomena where we 00:01:42.439 --> 00:01:44.050 align:start position:0% this kind of forking phenomena where we just<00:01:42.710> try<00:01:43.010> to<00:01:43.070> put<00:01:43.280> the<00:01:43.369> block<00:01:43.549> on<00:01:43.729> there<00:01:43.909> but 00:01:44.050 --> 00:01:44.060 align:start position:0% just try to put the block on there but 00:01:44.060 --> 00:01:47.200 align:start position:0% just try to put the block on there but if<00:01:44.150> we<00:01:44.270> fail<00:01:44.509> it<00:01:44.840> might<00:01:45.080> leave<00:01:45.860> but<00:01:46.159> somehow<00:01:46.340> if 00:01:47.200 --> 00:01:47.210 align:start position:0% if we fail it might leave but somehow if 00:01:47.210 --> 00:01:50.170 align:start position:0% if we fail it might leave but somehow if we<00:01:47.420> wait<00:01:47.630> long<00:01:47.840> enough<00:01:47.869> we<00:01:48.430> feel<00:01:49.430> like<00:01:49.580> oh<00:01:49.759> it's 00:01:50.170 --> 00:01:50.180 align:start position:0% we wait long enough we feel like oh it's 00:01:50.180 --> 00:01:52.990 align:start position:0% we wait long enough we feel like oh it's never<00:01:50.330> gonna<00:01:50.479> leave<00:01:50.659> the<00:01:50.810> block<00:01:50.960> chain<00:01:52.000> and 00:01:52.990 --> 00:01:53.000 align:start position:0% never gonna leave the block chain and 00:01:53.000 --> 00:01:54.370 align:start position:0% never gonna leave the block chain and these<00:01:53.180> protocols<00:01:53.509> have<00:01:53.869> a<00:01:53.900> much<00:01:54.170> lower 00:01:54.370 --> 00:01:54.380 align:start position:0% these protocols have a much lower 00:01:54.380 --> 00:01:56.649 align:start position:0% these protocols have a much lower overhead<00:01:54.890> so<00:01:55.759> basically<00:01:56.090> the<00:01:56.240> thing<00:01:56.479> that<00:01:56.600> I'm 00:01:56.649 --> 00:01:56.659 align:start position:0% overhead so basically the thing that I'm 00:01:56.659 --> 00:01:57.670 align:start position:0% overhead so basically the thing that I'm gonna<00:01:56.780> show<00:01:56.840> you<00:01:56.900> today<00:01:57.020> is<00:01:57.170> consensus 00:01:57.670 --> 00:01:57.680 align:start position:0% gonna show you today is consensus 00:01:57.680 --> 00:01:58.780 align:start position:0% gonna show you today is consensus protocol<00:01:57.979> cast<00:01:58.340> with<00:01:58.460> a<00:01:58.520> Friendly<00:01:58.729> Ghost 00:01:58.780 --> 00:01:58.790 align:start position:0% protocol cast with a Friendly Ghost 00:01:58.790 --> 00:02:00.730 align:start position:0% protocol cast with a Friendly Ghost which<00:01:59.450> finalized<00:02:00.020> every<00:02:00.290> block<00:02:00.500> with 00:02:00.730 --> 00:02:00.740 align:start position:0% which finalized every block with 00:02:00.740 --> 00:02:02.289 align:start position:0% which finalized every block with asynchronous<00:02:01.130> Byzantine<00:02:01.670> fault<00:02:01.880> tolerant 00:02:02.289 --> 00:02:02.299 align:start position:0% asynchronous Byzantine fault tolerant 00:02:02.299 --> 00:02:04.120 align:start position:0% asynchronous Byzantine fault tolerant consensus<00:02:03.079> safety<00:02:03.350> the<00:02:03.829> way<00:02:03.950> that 00:02:04.120 --> 00:02:04.130 align:start position:0% consensus safety the way that 00:02:04.130 --> 00:02:06.370 align:start position:0% consensus safety the way that traditional<00:02:04.220> sense<00:02:04.640> protocols<00:02:05.000> do<00:02:05.570> but<00:02:06.259> is 00:02:06.370 --> 00:02:06.380 align:start position:0% traditional sense protocols do but is 00:02:06.380 --> 00:02:08.460 align:start position:0% traditional sense protocols do but is able<00:02:06.500> to<00:02:06.619> do<00:02:06.799> this<00:02:06.920> with<00:02:07.159> the<00:02:07.400> overhead<00:02:07.820> of 00:02:08.460 --> 00:02:08.470 align:start position:0% able to do this with the overhead of 00:02:08.470 --> 00:02:12.910 align:start position:0% able to do this with the overhead of proof<00:02:09.470> work<00:02:09.679> law<00:02:09.830> chains<00:02:10.099> so<00:02:11.650> you<00:02:12.650> know<00:02:12.770> it 00:02:12.910 --> 00:02:12.920 align:start position:0% proof work law chains so you know it 00:02:12.920 --> 00:02:14.530 align:start position:0% proof work law chains so you know it kind<00:02:13.250> of<00:02:13.280> represents<00:02:13.819> some<00:02:14.120> of<00:02:14.150> the<00:02:14.239> best<00:02:14.300> of 00:02:14.530 --> 00:02:14.540 align:start position:0% kind of represents some of the best of 00:02:14.540 --> 00:02:15.309 align:start position:0% kind of represents some of the best of both<00:02:14.599> worlds<00:02:14.720> right 00:02:15.309 --> 00:02:15.319 align:start position:0% both worlds right 00:02:15.319 --> 00:02:18.130 align:start position:0% both worlds right both<00:02:15.409> finalize<00:02:16.010> decisions<00:02:16.489> and<00:02:16.700> also<00:02:17.390> hyper 00:02:18.130 --> 00:02:18.140 align:start position:0% both finalize decisions and also hyper 00:02:18.140 --> 00:02:20.949 align:start position:0% both finalize decisions and also hyper low<00:02:18.909> network<00:02:19.909> overhead<00:02:20.150> and<00:02:20.450> the<00:02:20.659> reason<00:02:20.870> why 00:02:20.949 --> 00:02:20.959 align:start position:0% low network overhead and the reason why 00:02:20.959 --> 00:02:22.030 align:start position:0% low network overhead and the reason why this<00:02:21.019> is<00:02:21.109> cool<00:02:21.379> is<00:02:21.500> because<00:02:21.739> it<00:02:21.829> kind<00:02:22.010> of 00:02:22.030 --> 00:02:22.040 align:start position:0% this is cool is because it kind of 00:02:22.040 --> 00:02:24.160 align:start position:0% this is cool is because it kind of provides<00:02:22.760> us<00:02:23.030> a<00:02:23.150> unified<00:02:23.599> way<00:02:23.780> for<00:02:23.989> thinking 00:02:24.160 --> 00:02:24.170 align:start position:0% provides us a unified way for thinking 00:02:24.170 --> 00:02:26.170 align:start position:0% provides us a unified way for thinking about<00:02:24.230> both<00:02:24.469> of<00:02:24.500> these<00:02:24.709> things<00:02:24.950> so<00:02:25.939> I'm<00:02:26.030> gonna 00:02:26.170 --> 00:02:26.180 align:start position:0% about both of these things so I'm gonna 00:02:26.180 --> 00:02:30.099 align:start position:0% about both of these things so I'm gonna give<00:02:26.390> a<00:02:26.780> little<00:02:27.439> overview<00:02:28.069> of<00:02:28.670> my<00:02:29.659> talk<00:02:29.689> and 00:02:30.099 --> 00:02:30.109 align:start position:0% give a little overview of my talk and 00:02:30.109 --> 00:02:31.599 align:start position:0% give a little overview of my talk and then<00:02:30.230> I'm<00:02:30.290> gonna<00:02:30.409> go<00:02:30.560> right<00:02:30.739> into<00:02:30.890> it<00:02:30.980> so<00:02:31.069> the 00:02:31.599 --> 00:02:31.609 align:start position:0% then I'm gonna go right into it so the 00:02:31.609 --> 00:02:33.970 align:start position:0% then I'm gonna go right into it so the consensus<00:02:32.390> safety<00:02:32.629> proof<00:02:32.870> is<00:02:33.079> this<00:02:33.379> thing 00:02:33.970 --> 00:02:33.980 align:start position:0% consensus safety proof is this thing 00:02:33.980 --> 00:02:35.679 align:start position:0% consensus safety proof is this thing that<00:02:34.010> all<00:02:34.579> of<00:02:34.849> the<00:02:34.909> consensus<00:02:35.239> protocols<00:02:35.599> that 00:02:35.679 --> 00:02:35.689 align:start position:0% that all of the consensus protocols that 00:02:35.689 --> 00:02:37.000 align:start position:0% that all of the consensus protocols that are<00:02:35.750> correct<00:02:36.170> by<00:02:36.319> construction<00:02:36.620> process 00:02:37.000 --> 00:02:37.010 align:start position:0% are correct by construction process 00:02:37.010 --> 00:02:41.589 align:start position:0% are correct by construction process satisfying<00:02:38.140> and<00:02:39.140> then<00:02:40.090> I'm<00:02:41.090> gonna<00:02:41.269> specify 00:02:41.589 --> 00:02:41.599 align:start position:0% satisfying and then I'm gonna specify 00:02:41.599 --> 00:02:44.440 align:start position:0% satisfying and then I'm gonna specify some<00:02:42.319> of<00:02:42.409> these<00:02:42.470> things<00:02:43.090> and<00:02:44.090> then<00:02:44.269> I'm<00:02:44.329> gonna 00:02:44.440 --> 00:02:44.450 align:start position:0% some of these things and then I'm gonna 00:02:44.450 --> 00:02:47.110 align:start position:0% some of these things and then I'm gonna talk<00:02:44.629> about<00:02:45.370> maybe<00:02:46.370> live<00:02:46.609> this<00:02:46.730> a<00:02:46.790> little<00:02:46.939> bit 00:02:47.110 --> 00:02:47.120 align:start position:0% talk about maybe live this a little bit 00:02:47.120 --> 00:02:48.580 align:start position:0% talk about maybe live this a little bit and<00:02:47.239> then<00:02:47.420> give<00:02:47.629> some<00:02:47.659> experimental 00:02:48.580 --> 00:02:48.590 align:start position:0% and then give some experimental 00:02:48.590 --> 00:02:50.170 align:start position:0% and then give some experimental observations<00:02:49.159> where<00:02:49.489> we're<00:02:49.579> gonna<00:02:49.639> see<00:02:49.970> this 00:02:50.170 --> 00:02:50.180 align:start position:0% observations where we're gonna see this 00:02:50.180 --> 00:02:51.339 align:start position:0% observations where we're gonna see this kind<00:02:50.389> of<00:02:50.420> network<00:02:50.629> efficiency<00:02:50.930> that<00:02:51.230> I'm 00:02:51.339 --> 00:02:51.349 align:start position:0% kind of network efficiency that I'm 00:02:51.349 --> 00:02:53.949 align:start position:0% kind of network efficiency that I'm talking<00:02:51.500> about<00:02:52.299> so<00:02:53.299> let's<00:02:53.480> get<00:02:53.569> started<00:02:53.900> so 00:02:53.949 --> 00:02:53.959 align:start position:0% talking about so let's get started so 00:02:53.959 --> 00:02:56.949 align:start position:0% talking about so let's get started so the<00:02:54.409> consensus<00:02:54.829> safety<00:02:55.069> proof<00:02:55.340> is<00:02:55.959> probably 00:02:56.949 --> 00:02:56.959 align:start position:0% the consensus safety proof is probably 00:02:56.959 --> 00:02:58.000 align:start position:0% the consensus safety proof is probably the<00:02:57.079> most<00:02:57.200> important<00:02:57.620> thing<00:02:57.769> that<00:02:57.799> you 00:02:58.000 --> 00:02:58.010 align:start position:0% the most important thing that you 00:02:58.010 --> 00:03:01.059 align:start position:0% the most important thing that you understand<00:02:58.430> in<00:02:58.669> order<00:02:59.090> to<00:02:59.439> grab<00:03:00.439> grok<00:03:00.859> this 00:03:01.059 --> 00:03:01.069 align:start position:0% understand in order to grab grok this 00:03:01.069 --> 00:03:04.000 align:start position:0% understand in order to grab grok this whole<00:03:01.250> process<00:03:01.519> so<00:03:02.530> we<00:03:03.530> need<00:03:03.680> a<00:03:03.709> few<00:03:03.950> things 00:03:04.000 --> 00:03:04.010 align:start position:0% whole process so we need a few things 00:03:04.010 --> 00:03:05.649 align:start position:0% whole process so we need a few things for<00:03:04.879> this<00:03:05.000> thing<00:03:05.090> to<00:03:05.150> be<00:03:05.239> fine<00:03:05.389> we<00:03:05.510> have 00:03:05.649 --> 00:03:05.659 align:start position:0% for this thing to be fine we have 00:03:05.659 --> 00:03:08.339 align:start position:0% for this thing to be fine we have protocol<00:03:06.169> States<00:03:06.349> and<00:03:06.790> state<00:03:07.790> transitions 00:03:08.339 --> 00:03:08.349 align:start position:0% protocol States and state transitions 00:03:08.349 --> 00:03:10.960 align:start position:0% protocol States and state transitions and<00:03:09.349> we<00:03:09.560> have<00:03:09.680> an<00:03:09.769> estimator<00:03:10.159> and<00:03:10.430> this<00:03:10.849> thing 00:03:10.960 --> 00:03:10.970 align:start position:0% and we have an estimator and this thing 00:03:10.970 --> 00:03:11.890 align:start position:0% and we have an estimator and this thing called<00:03:11.090> ask<00:03:11.209> me<00:03:11.299> safe<00:03:11.450> team<00:03:11.599> you<00:03:11.689> can<00:03:11.780> find 00:03:11.890 --> 00:03:11.900 align:start position:0% called ask me safe team you can find 00:03:11.900 --> 00:03:13.300 align:start position:0% called ask me safe team you can find them<00:03:11.989> in<00:03:12.049> a<00:03:12.079> second<00:03:12.319> and<00:03:12.500> then<00:03:13.040> we<00:03:13.099> have<00:03:13.220> this 00:03:13.300 --> 00:03:13.310 align:start position:0% them in a second and then we have this 00:03:13.310 --> 00:03:16.809 align:start position:0% them in a second and then we have this Theron<00:03:13.970> this<00:03:14.060> result<00:03:14.329> that<00:03:14.569> if<00:03:14.689> nodes<00:03:15.079> have<00:03:16.010> a 00:03:16.809 --> 00:03:16.819 align:start position:0% Theron this result that if nodes have a 00:03:16.819 --> 00:03:19.990 align:start position:0% Theron this result that if nodes have a common<00:03:17.659> future<00:03:18.109> state<00:03:18.500> in<00:03:18.739> the<00:03:18.919> protocol<00:03:19.250> then 00:03:19.990 --> 00:03:20.000 align:start position:0% common future state in the protocol then 00:03:20.000 --> 00:03:20.949 align:start position:0% common future state in the protocol then they<00:03:20.150> have<00:03:20.299> because<00:03:20.510> then<00:03:20.780> they<00:03:20.930> have 00:03:20.949 --> 00:03:20.959 align:start position:0% they have because then they have 00:03:20.959 --> 00:03:22.960 align:start position:0% they have because then they have consensus<00:03:21.560> safety<00:03:21.859> on<00:03:22.010> decisions<00:03:22.760> that<00:03:22.819> are 00:03:22.960 --> 00:03:22.970 align:start position:0% consensus safety on decisions that are 00:03:22.970 --> 00:03:24.969 align:start position:0% consensus safety on decisions that are that<00:03:23.389> have<00:03:23.449> estimate<00:03:24.019> safety<00:03:24.319> so<00:03:24.500> if<00:03:24.590> I<00:03:24.650> decide 00:03:24.969 --> 00:03:24.979 align:start position:0% that have estimate safety so if I decide 00:03:24.979 --> 00:03:26.289 align:start position:0% that have estimate safety so if I decide on<00:03:25.010> something<00:03:25.400> with<00:03:25.489> estimate<00:03:25.790> safety<00:03:26.090> then 00:03:26.289 --> 00:03:26.299 align:start position:0% on something with estimate safety then 00:03:26.299 --> 00:03:27.610 align:start position:0% on something with estimate safety then I'm<00:03:26.389> gonna<00:03:26.479> get<00:03:26.629> consensus<00:03:27.079> safety<00:03:27.290> for<00:03:27.440> free 00:03:27.610 --> 00:03:27.620 align:start position:0% I'm gonna get consensus safety for free 00:03:27.620 --> 00:03:29.229 align:start position:0% I'm gonna get consensus safety for free if<00:03:27.919> we<00:03:28.489> have<00:03:28.609> a<00:03:28.639> common<00:03:28.940> future 00:03:29.229 --> 00:03:29.239 align:start position:0% if we have a common future 00:03:29.239 --> 00:03:35.080 align:start position:0% if we have a common future and<00:03:29.479> so<00:03:30.349> basically<00:03:33.190> protocol<00:03:34.190> states<00:03:34.459> are 00:03:35.080 --> 00:03:35.090 align:start position:0% and so basically protocol states are 00:03:35.090 --> 00:03:36.399 align:start position:0% and so basically protocol states are just<00:03:35.329> like<00:03:35.449> things<00:03:35.750> in<00:03:35.959> a<00:03:36.049> group<00:03:36.229> and 00:03:36.399 --> 00:03:36.409 align:start position:0% just like things in a group and 00:03:36.409 --> 00:03:37.750 align:start position:0% just like things in a group and purchased<00:03:36.859> a<00:03:36.919> transitions<00:03:37.310> are<00:03:37.549> kind<00:03:37.729> of 00:03:37.750 --> 00:03:37.760 align:start position:0% purchased a transitions are kind of 00:03:37.760 --> 00:03:39.789 align:start position:0% purchased a transitions are kind of paths<00:03:37.970> between<00:03:38.359> them<00:03:38.510> right<00:03:38.720> an<00:03:38.989> estimator 00:03:39.789 --> 00:03:39.799 align:start position:0% paths between them right an estimator 00:03:39.799 --> 00:03:41.140 align:start position:0% paths between them right an estimator takes<00:03:40.069> these<00:03:40.220> protocol<00:03:40.549> states<00:03:40.729> and<00:03:40.940> maps 00:03:41.140 --> 00:03:41.150 align:start position:0% takes these protocol states and maps 00:03:41.150 --> 00:03:42.789 align:start position:0% takes these protocol states and maps them<00:03:41.359> to<00:03:41.510> in<00:03:41.690> the<00:03:41.750> binary<00:03:41.930> consensus<00:03:42.379> like<00:03:42.500> a<00:03:42.530> 0 00:03:42.789 --> 00:03:42.799 align:start position:0% them to in the binary consensus like a 0 00:03:42.799 --> 00:03:44.800 align:start position:0% them to in the binary consensus like a 0 or<00:03:42.829> 1<00:03:42.919> and<00:03:43.280> the<00:03:43.549> blockchain<00:03:43.760> consensus<00:03:44.419> like<00:03:44.780> a 00:03:44.800 --> 00:03:44.810 align:start position:0% or 1 and the blockchain consensus like a 00:03:44.810 --> 00:03:46.689 align:start position:0% or 1 and the blockchain consensus like a whole<00:03:45.019> blockchain<00:03:45.409> so<00:03:45.979> it's<00:03:46.190> kind<00:03:46.370> of<00:03:46.400> our 00:03:46.689 --> 00:03:46.699 align:start position:0% whole blockchain so it's kind of our 00:03:46.699 --> 00:03:48.699 align:start position:0% whole blockchain so it's kind of our fork<00:03:47.090> choice<00:03:47.299> rule<00:03:47.629> mechanic<00:03:48.229> right<00:03:48.409> a<00:03:48.470> little 00:03:48.699 --> 00:03:48.709 align:start position:0% fork choice rule mechanic right a little 00:03:48.709 --> 00:03:50.349 align:start position:0% fork choice rule mechanic right a little guess<00:03:48.889> and<00:03:49.370> estimator<00:03:49.699> is<00:03:49.760> not<00:03:49.879> a<00:03:49.909> decision<00:03:50.239> it 00:03:50.349 --> 00:03:50.359 align:start position:0% guess and estimator is not a decision it 00:03:50.359 --> 00:04:04.630 align:start position:0% guess and estimator is not a decision it just<00:03:50.389> guesses<00:03:56.949> looks<00:03:57.949> like<00:03:58.040> we<00:03:58.129> have 00:04:04.630 --> 00:04:04.640 align:start position:0% 00:04:04.640 --> 00:04:07.459 align:start position:0% so<00:04:05.640> an<00:04:05.819> estimate<00:04:06.090> is<00:04:06.180> safe<00:04:06.420> basically<00:04:07.170> if<00:04:07.319> the 00:04:07.459 --> 00:04:07.469 align:start position:0% so an estimate is safe basically if the 00:04:07.469 --> 00:04:11.149 align:start position:0% so an estimate is safe basically if the estimator<00:04:08.209> for<00:04:09.209> all<00:04:09.420> future<00:04:10.290> possible 00:04:11.149 --> 00:04:11.159 align:start position:0% estimator for all future possible 00:04:11.159 --> 00:04:13.099 align:start position:0% estimator for all future possible protocol<00:04:11.549> states<00:04:11.819> that<00:04:12.390> can<00:04:12.510> be<00:04:12.629> access<00:04:13.079> from 00:04:13.099 --> 00:04:13.109 align:start position:0% protocol states that can be access from 00:04:13.109 --> 00:04:15.949 align:start position:0% protocol states that can be access from that<00:04:13.230> from<00:04:13.590> that<00:04:13.620> state<00:04:14.090> the<00:04:15.090> the<00:04:15.569> estimator 00:04:15.949 --> 00:04:15.959 align:start position:0% that from that state the the estimator 00:04:15.959 --> 00:04:18.020 align:start position:0% that from that state the the estimator returns<00:04:16.470> this<00:04:16.590> returns<00:04:17.070> a<00:04:17.130> value<00:04:17.250> like<00:04:17.549> 1<00:04:17.759> or 00:04:18.020 --> 00:04:18.030 align:start position:0% returns this returns a value like 1 or 00:04:18.030 --> 00:04:24.140 align:start position:0% returns this returns a value like 1 or has<00:04:18.449> a<00:04:18.479> blockchain<00:04:18.810> that<00:04:19.199> also<00:04:20.190> is<00:04:22.639> that<00:04:23.639> that 00:04:24.140 --> 00:04:24.150 align:start position:0% has a blockchain that also is that that 00:04:24.150 --> 00:04:25.999 align:start position:0% has a blockchain that also is that that will<00:04:24.300> always<00:04:24.509> have<00:04:24.900> that<00:04:25.259> block<00:04:25.560> in<00:04:25.740> the<00:04:25.800> chain 00:04:25.999 --> 00:04:26.009 align:start position:0% will always have that block in the chain 00:04:26.009 --> 00:04:27.499 align:start position:0% will always have that block in the chain so<00:04:26.280> block<00:04:26.550> is<00:04:26.699> safe<00:04:26.880> if<00:04:27.090> all<00:04:27.300> future 00:04:27.499 --> 00:04:27.509 align:start position:0% so block is safe if all future 00:04:27.509 --> 00:04:29.689 align:start position:0% so block is safe if all future blockchain<00:04:27.990> Fork<00:04:28.410> choice<00:04:28.590> rules<00:04:28.889> for<00:04:29.310> choices 00:04:29.689 --> 00:04:29.699 align:start position:0% blockchain Fork choice rules for choices 00:04:29.699 --> 00:04:31.219 align:start position:0% blockchain Fork choice rules for choices have<00:04:30.150> that<00:04:30.330> block<00:04:30.479> in<00:04:30.600> the<00:04:30.660> chain<00:04:30.840> and<00:04:30.990> a<00:04:31.080> bit 00:04:31.219 --> 00:04:31.229 align:start position:0% have that block in the chain and a bit 00:04:31.229 --> 00:04:32.900 align:start position:0% have that block in the chain and a bit is<00:04:31.350> safe<00:04:31.500> if<00:04:31.710> all<00:04:31.949> future<00:04:32.250> estimators<00:04:32.729> have 00:04:32.900 --> 00:04:32.910 align:start position:0% is safe if all future estimators have 00:04:32.910 --> 00:04:37.400 align:start position:0% is safe if all future estimators have that<00:04:33.030> bit<00:04:34.520> so<00:04:35.520> if<00:04:35.940> two<00:04:36.270> estimates<00:04:36.720> are<00:04:37.110> safe<00:04:37.259> at 00:04:37.400 --> 00:04:37.410 align:start position:0% that bit so if two estimates are safe at 00:04:37.410 --> 00:04:39.650 align:start position:0% that bit so if two estimates are safe at two<00:04:37.560> States<00:04:37.860> and<00:04:38.099> those<00:04:38.639> states<00:04:39.000> have<00:04:39.210> a<00:04:39.270> state 00:04:39.650 --> 00:04:39.660 align:start position:0% two States and those states have a state 00:04:39.660 --> 00:04:42.589 align:start position:0% two States and those states have a state in<00:04:39.750> common<00:04:40.110> then<00:04:40.889> because<00:04:41.610> those<00:04:41.910> things<00:04:42.449> are 00:04:42.589 --> 00:04:42.599 align:start position:0% in common then because those things are 00:04:42.599 --> 00:04:44.689 align:start position:0% in common then because those things are the<00:04:43.410> estimates<00:04:43.650> hold<00:04:43.860> for<00:04:44.039> all<00:04:44.130> future<00:04:44.460> states 00:04:44.689 --> 00:04:44.699 align:start position:0% the estimates hold for all future states 00:04:44.699 --> 00:04:46.850 align:start position:0% the estimates hold for all future states and<00:04:44.970> they<00:04:45.270> have<00:04:45.389> one<00:04:45.539> in<00:04:45.630> common<00:04:45.960> they'll<00:04:46.620> both 00:04:46.850 --> 00:04:46.860 align:start position:0% and they have one in common they'll both 00:04:46.860 --> 00:04:49.580 align:start position:0% and they have one in common they'll both hold<00:04:47.190> from<00:04:47.820> in<00:04:48.240> that<00:04:48.720> one<00:04:48.990> state<00:04:49.380> that<00:04:49.530> they 00:04:49.580 --> 00:04:49.590 align:start position:0% hold from in that one state that they 00:04:49.590 --> 00:04:50.150 align:start position:0% hold from in that one state that they have<00:04:49.710> in<00:04:49.800> common 00:04:50.150 --> 00:04:50.160 align:start position:0% have in common 00:04:50.160 --> 00:04:51.350 align:start position:0% have in common and<00:04:50.250> that<00:04:50.880> means<00:04:51.000> they<00:04:51.120> have<00:04:51.240> to<00:04:51.300> be 00:04:51.350 --> 00:04:51.360 align:start position:0% and that means they have to be 00:04:51.360 --> 00:04:53.120 align:start position:0% and that means they have to be consistent<00:04:51.870> because<00:04:52.139> you<00:04:52.410> can't<00:04:52.560> have<00:04:52.710> you 00:04:53.120 --> 00:04:53.130 align:start position:0% consistent because you can't have you 00:04:53.130 --> 00:04:55.400 align:start position:0% consistent because you can't have you know<00:04:53.220> the<00:04:53.580> block<00:04:53.849> be<00:04:54.270> in<00:04:54.479> the<00:04:54.660> fork<00:04:54.870> and<00:04:55.020> not<00:04:55.199> at 00:04:55.400 --> 00:04:55.410 align:start position:0% know the block be in the fork and not at 00:04:55.410 --> 00:04:59.689 align:start position:0% know the block be in the fork and not at the<00:04:55.440> same<00:04:55.680> at<00:04:55.889> the<00:04:55.949> same<00:04:56.190> thing<00:04:56.400> so<00:04:57.860> in<00:04:58.860> that 00:04:59.689 --> 00:04:59.699 align:start position:0% the same at the same thing so in that 00:04:59.699 --> 00:05:00.860 align:start position:0% the same at the same thing so in that was<00:04:59.820> come<00:04:59.940> the<00:04:59.970> first<00:05:00.300> part<00:05:00.539> of<00:05:00.570> the<00:05:00.660> proof 00:05:00.860 --> 00:05:00.870 align:start position:0% was come the first part of the proof 00:05:00.870 --> 00:05:02.659 align:start position:0% was come the first part of the proof basically<00:05:01.259> it<00:05:01.349> says<00:05:01.470> if<00:05:01.500> noes<00:05:02.130> have<00:05:02.310> a<00:05:02.400> common 00:05:02.659 --> 00:05:02.669 align:start position:0% basically it says if noes have a common 00:05:02.669 --> 00:05:04.159 align:start position:0% basically it says if noes have a common feature<00:05:02.820> then<00:05:03.330> they<00:05:03.419> have<00:05:03.539> consensus<00:05:03.930> safety 00:05:04.159 --> 00:05:04.169 align:start position:0% feature then they have consensus safety 00:05:04.169 --> 00:05:06.260 align:start position:0% feature then they have consensus safety the<00:05:04.830> next<00:05:05.099> part<00:05:05.310> basically<00:05:05.880> is<00:05:06.060> a 00:05:06.260 --> 00:05:06.270 align:start position:0% the next part basically is a 00:05:06.270 --> 00:05:07.760 align:start position:0% the next part basically is a construction<00:05:06.690> to<00:05:06.930> make<00:05:06.960> sure<00:05:07.320> that<00:05:07.590> they<00:05:07.650> will 00:05:07.760 --> 00:05:07.770 align:start position:0% construction to make sure that they will 00:05:07.770 --> 00:05:09.020 align:start position:0% construction to make sure that they will have<00:05:07.919> a<00:05:07.949> common<00:05:08.220> future<00:05:08.370> unless<00:05:08.699> there's<00:05:08.909> more 00:05:09.020 --> 00:05:09.030 align:start position:0% have a common future unless there's more 00:05:09.030 --> 00:05:10.310 align:start position:0% have a common future unless there's more than<00:05:09.180> some<00:05:09.300> number<00:05:09.330> of<00:05:09.570> visit<00:05:09.780> defaults<00:05:10.169> and 00:05:10.310 --> 00:05:10.320 align:start position:0% than some number of visit defaults and 00:05:10.320 --> 00:05:11.689 align:start position:0% than some number of visit defaults and the<00:05:10.770> way<00:05:10.860> this<00:05:10.949> works<00:05:11.159> is<00:05:11.280> basically<00:05:11.520> have 00:05:11.689 --> 00:05:11.699 align:start position:0% the way this works is basically have 00:05:11.699 --> 00:05:14.060 align:start position:0% the way this works is basically have protocol<00:05:12.120> states<00:05:12.389> which<00:05:12.570> are<00:05:12.690> messages<00:05:13.169> we 00:05:14.060 --> 00:05:14.070 align:start position:0% protocol states which are messages we 00:05:14.070 --> 00:05:15.770 align:start position:0% protocol states which are messages we have<00:05:14.250> state<00:05:14.580> transitions<00:05:15.000> which<00:05:15.150> corresponds 00:05:15.770 --> 00:05:15.780 align:start position:0% have state transitions which corresponds 00:05:15.780 --> 00:05:18.080 align:start position:0% have state transitions which corresponds to<00:05:15.840> receiving<00:05:16.680> messages<00:05:16.800> or<00:05:17.159> increasing<00:05:18.060> your 00:05:18.080 --> 00:05:18.090 align:start position:0% to receiving messages or increasing your 00:05:18.090 --> 00:05:19.939 align:start position:0% to receiving messages or increasing your the<00:05:18.510> saw<00:05:18.690> like<00:05:18.870> you<00:05:18.990> know<00:05:19.139> going<00:05:19.530> to<00:05:19.650> super 00:05:19.939 --> 00:05:19.949 align:start position:0% the saw like you know going to super 00:05:19.949 --> 00:05:22.430 align:start position:0% the saw like you know going to super sets<00:05:20.130> of<00:05:20.550> protocol<00:05:20.880> messages<00:05:21.229> and<00:05:22.229> if<00:05:22.349> we 00:05:22.430 --> 00:05:22.440 align:start position:0% sets of protocol messages and if we 00:05:22.440 --> 00:05:24.469 align:start position:0% sets of protocol messages and if we count<00:05:22.590> Byzantine<00:05:23.010> faults<00:05:23.310> and<00:05:23.520> exclude<00:05:24.270> the 00:05:24.469 --> 00:05:24.479 align:start position:0% count Byzantine faults and exclude the 00:05:24.479 --> 00:05:25.909 align:start position:0% count Byzantine faults and exclude the sets<00:05:24.659> of<00:05:24.720> messages<00:05:24.870> I<00:05:25.169> have<00:05:25.199> too<00:05:25.440> many<00:05:25.530> faults 00:05:25.909 --> 00:05:25.919 align:start position:0% sets of messages I have too many faults 00:05:25.919 --> 00:05:30.409 align:start position:0% sets of messages I have too many faults then<00:05:26.820> we<00:05:27.479> can<00:05:29.030> guarantee<00:05:30.030> that<00:05:30.060> they<00:05:30.270> have<00:05:30.389> a 00:05:30.409 --> 00:05:30.419 align:start position:0% then we can guarantee that they have a 00:05:30.419 --> 00:05:31.550 align:start position:0% then we can guarantee that they have a common<00:05:30.570> future<00:05:30.810> is<00:05:30.990> unless<00:05:31.229> there's<00:05:31.409> that 00:05:31.550 --> 00:05:31.560 align:start position:0% common future is unless there's that 00:05:31.560 --> 00:05:33.409 align:start position:0% common future is unless there's that number<00:05:31.770> of<00:05:31.860> faults<00:05:32.099> basically<00:05:32.669> because<00:05:32.940> first 00:05:33.409 --> 00:05:33.419 align:start position:0% number of faults basically because first 00:05:33.419 --> 00:05:34.730 align:start position:0% number of faults basically because first we<00:05:33.599> allow<00:05:33.780> all<00:05:34.050> of<00:05:34.169> them<00:05:34.289> to<00:05:34.409> have<00:05:34.500> a<00:05:34.530> common 00:05:34.730 --> 00:05:34.740 align:start position:0% we allow all of them to have a common 00:05:34.740 --> 00:05:37.040 align:start position:0% we allow all of them to have a common feature<00:05:34.860> no<00:05:35.009> matter<00:05:35.159> what<00:05:35.419> then<00:05:36.419> we<00:05:36.690> remove 00:05:37.040 --> 00:05:37.050 align:start position:0% feature no matter what then we remove 00:05:37.050 --> 00:05:38.510 align:start position:0% feature no matter what then we remove the<00:05:37.080> states<00:05:37.470> that<00:05:37.650> have<00:05:37.800> too<00:05:38.009> many<00:05:38.099> visiting 00:05:38.510 --> 00:05:38.520 align:start position:0% the states that have too many visiting 00:05:38.520 --> 00:05:41.600 align:start position:0% the states that have too many visiting faults<00:05:38.789> and<00:05:39.349> so<00:05:40.349> two<00:05:41.039> states<00:05:41.310> will<00:05:41.460> have<00:05:41.490> a 00:05:41.600 --> 00:05:41.610 align:start position:0% faults and so two states will have a 00:05:41.610 --> 00:05:42.710 align:start position:0% faults and so two states will have a common<00:05:41.849> future<00:05:42.000> unless<00:05:42.180> there's<00:05:42.570> too<00:05:42.659> many 00:05:42.710 --> 00:05:42.720 align:start position:0% common future unless there's too many 00:05:42.720 --> 00:05:44.629 align:start position:0% common future unless there's too many Byzantine<00:05:43.110> faults<00:05:43.320> and<00:05:43.530> that<00:05:43.919> gives<00:05:44.250> us<00:05:44.370> kind 00:05:44.629 --> 00:05:44.639 align:start position:0% Byzantine faults and that gives us kind 00:05:44.639 --> 00:05:47.210 align:start position:0% Byzantine faults and that gives us kind of<00:05:44.669> our<00:05:44.789> key<00:05:45.419> result<00:05:45.750> here<00:05:45.960> which<00:05:46.650> is<00:05:46.800> that<00:05:47.009> as 00:05:47.210 --> 00:05:47.220 align:start position:0% of our key result here which is that as 00:05:47.220 --> 00:05:49.129 align:start position:0% of our key result here which is that as long<00:05:47.909> as<00:05:47.940> there's<00:05:48.270> not<00:05:48.479> too<00:05:48.690> many<00:05:48.780> Byzantine 00:05:49.129 --> 00:05:49.139 align:start position:0% long as there's not too many Byzantine 00:05:49.139 --> 00:05:50.560 align:start position:0% long as there's not too many Byzantine fault<00:05:49.320> so<00:05:49.440> they<00:05:49.530> have<00:05:49.650> a<00:05:49.680> common<00:05:49.979> future<00:05:50.099> and 00:05:50.560 --> 00:05:50.570 align:start position:0% fault so they have a common future and 00:05:50.570 --> 00:05:52.460 align:start position:0% fault so they have a common future and if<00:05:51.570> a<00:05:51.630> common<00:05:51.960> future<00:05:52.050> then<00:05:52.289> they<00:05:52.349> have 00:05:52.460 --> 00:05:52.470 align:start position:0% if a common future then they have 00:05:52.470 --> 00:05:54.260 align:start position:0% if a common future then they have consensus<00:05:52.860> safety<00:05:53.130> and<00:05:53.400> so<00:05:53.639> this<00:05:53.759> this 00:05:54.260 --> 00:05:54.270 align:start position:0% consensus safety and so this this 00:05:54.270 --> 00:05:55.580 align:start position:0% consensus safety and so this this consensus<00:05:54.720> safety<00:05:54.930> proof<00:05:55.139> basically<00:05:55.349> says 00:05:55.580 --> 00:05:55.590 align:start position:0% consensus safety proof basically says 00:05:55.590 --> 00:05:56.930 align:start position:0% consensus safety proof basically says look<00:05:55.620> I've<00:05:55.889> always<00:05:56.039> thought<00:05:56.310> too<00:05:56.520> many 00:05:56.930 --> 00:05:56.940 align:start position:0% look I've always thought too many 00:05:56.940 --> 00:05:58.490 align:start position:0% look I've always thought too many Byzantine<00:05:57.240> faults<00:05:57.509> then<00:05:57.990> we<00:05:58.050> have<00:05:58.139> consensus 00:05:58.490 --> 00:05:58.500 align:start position:0% Byzantine faults then we have consensus 00:05:58.500 --> 00:06:00.170 align:start position:0% Byzantine faults then we have consensus safety<00:05:58.740> and<00:05:58.919> the<00:05:59.460> cool<00:05:59.759> thing<00:05:59.789> about<00:05:59.940> this<00:06:00.120> is 00:06:00.170 --> 00:06:00.180 align:start position:0% safety and the cool thing about this is 00:06:00.180 --> 00:06:02.210 align:start position:0% safety and the cool thing about this is I<00:06:00.330> had<00:06:00.599> I<00:06:00.990> had<00:06:01.139> to<00:06:01.169> say<00:06:01.349> almost<00:06:01.680> nothing<00:06:02.159> about 00:06:02.210 --> 00:06:02.220 align:start position:0% I had I had to say almost nothing about 00:06:02.220 --> 00:06:03.830 align:start position:0% I had I had to say almost nothing about the<00:06:02.400> consensus<00:06:02.729> protocol<00:06:03.060> right<00:06:03.210> all<00:06:03.630> I<00:06:03.720> had 00:06:03.830 --> 00:06:03.840 align:start position:0% the consensus protocol right all I had 00:06:03.840 --> 00:06:07.480 align:start position:0% the consensus protocol right all I had to<00:06:03.930> do<00:06:04.020> was<00:06:04.169> say<00:06:04.409> it<00:06:05.190> has<00:06:05.310> protocol<00:06:05.789> states<00:06:06.030> and 00:06:07.480 --> 00:06:07.490 align:start position:0% to do was say it has protocol states and 00:06:07.490 --> 00:06:09.800 align:start position:0% to do was say it has protocol states and and<00:06:08.490> I<00:06:08.729> mean<00:06:09.030> I<00:06:09.120> said<00:06:09.240> nothing<00:06:09.539> about<00:06:09.570> the 00:06:09.800 --> 00:06:09.810 align:start position:0% and I mean I said nothing about the 00:06:09.810 --> 00:06:12.290 align:start position:0% and I mean I said nothing about the nature<00:06:10.199> of<00:06:10.229> the<00:06:10.320> estimator<00:06:10.710> right<00:06:11.219> and<00:06:11.460> so<00:06:12.270> now 00:06:12.290 --> 00:06:12.300 align:start position:0% nature of the estimator right and so now 00:06:12.300 --> 00:06:13.550 align:start position:0% nature of the estimator right and so now we're<00:06:12.479> gonna<00:06:12.570> kind<00:06:12.779> of<00:06:12.810> see<00:06:13.020> this<00:06:13.169> we're<00:06:13.500> gonna 00:06:13.550 --> 00:06:13.560 align:start position:0% we're gonna kind of see this we're gonna 00:06:13.560 --> 00:06:14.690 align:start position:0% we're gonna kind of see this we're gonna see<00:06:13.740> this<00:06:13.830> thing<00:06:13.950> happen 00:06:14.690 --> 00:06:14.700 align:start position:0% see this thing happen 00:06:14.700 --> 00:06:18.350 align:start position:0% see this thing happen so<00:06:15.420> I'm<00:06:15.630> going<00:06:15.780> to<00:06:15.810> define<00:06:16.940> protocol<00:06:17.940> messages 00:06:18.350 --> 00:06:18.360 align:start position:0% so I'm going to define protocol messages 00:06:18.360 --> 00:06:20.420 align:start position:0% so I'm going to define protocol messages and<00:06:18.600> then<00:06:18.870> the<00:06:18.960> estimator<00:06:19.380> isn't<00:06:20.100> seen<00:06:20.190> fault 00:06:20.420 --> 00:06:20.430 align:start position:0% and then the estimator isn't seen fault 00:06:20.430 --> 00:06:22.610 align:start position:0% and then the estimator isn't seen fault detection<00:06:20.900> States<00:06:21.900> and<00:06:22.080> state<00:06:22.200> transitions 00:06:22.610 --> 00:06:22.620 align:start position:0% detection States and state transitions 00:06:22.620 --> 00:06:24.590 align:start position:0% detection States and state transitions and<00:06:22.770> then<00:06:22.830> the<00:06:22.980> an<00:06:23.190> estimate<00:06:23.670> safety<00:06:23.940> for<00:06:24.540> the 00:06:24.590 --> 00:06:24.600 align:start position:0% and then the an estimate safety for the 00:06:24.600 --> 00:06:26.210 align:start position:0% and then the an estimate safety for the binary<00:06:24.750> consensus<00:06:25.260> and<00:06:25.410> then<00:06:25.740> we're<00:06:25.860> gonna<00:06:25.950> do 00:06:26.210 --> 00:06:26.220 align:start position:0% binary consensus and then we're gonna do 00:06:26.220 --> 00:06:27.800 align:start position:0% binary consensus and then we're gonna do a<00:06:26.250> tiny<00:06:26.580> change<00:06:26.940> and<00:06:27.150> generate<00:06:27.450> the<00:06:27.540> walk<00:06:27.660> team 00:06:27.800 --> 00:06:27.810 align:start position:0% a tiny change and generate the walk team 00:06:27.810 --> 00:06:29.030 align:start position:0% a tiny change and generate the walk team consensus<00:06:28.140> and<00:06:28.230> a<00:06:28.290> tiny<00:06:28.470> change<00:06:28.710> and<00:06:28.980> to 00:06:29.030 --> 00:06:29.040 align:start position:0% consensus and a tiny change and to 00:06:29.040 --> 00:06:30.470 align:start position:0% consensus and a tiny change and to validate<00:06:29.220> rotation<00:06:29.730> a<00:06:29.820> tiny<00:06:30.030> change<00:06:30.270> and 00:06:30.470 --> 00:06:30.480 align:start position:0% validate rotation a tiny change and 00:06:30.480 --> 00:06:32.690 align:start position:0% validate rotation a tiny change and remove<00:06:31.440> fault<00:06:32.100> tolerance<00:06:32.190> threshold<00:06:32.430> from 00:06:32.690 --> 00:06:32.700 align:start position:0% remove fault tolerance threshold from 00:06:32.700 --> 00:06:35.390 align:start position:0% remove fault tolerance threshold from the<00:06:32.790> protocol<00:06:33.410> and<00:06:34.410> that'll<00:06:35.250> kind<00:06:35.370> of 00:06:35.390 --> 00:06:35.400 align:start position:0% the protocol and that'll kind of 00:06:35.400 --> 00:06:37.460 align:start position:0% the protocol and that'll kind of demonstrate<00:06:35.940> the<00:06:36.300> the<00:06:36.630> the<00:06:37.050> power<00:06:37.290> of<00:06:37.380> the 00:06:37.460 --> 00:06:37.470 align:start position:0% demonstrate the the the power of the 00:06:37.470 --> 00:06:39.290 align:start position:0% demonstrate the the the power of the approach<00:06:37.740> here<00:06:37.950> so<00:06:38.670> for<00:06:39.060> the<00:06:39.150> binary 00:06:39.290 --> 00:06:39.300 align:start position:0% approach here so for the binary 00:06:39.300 --> 00:06:41.030 align:start position:0% approach here so for the binary consensus<00:06:39.840> we<00:06:39.990> have<00:06:40.140> protocol<00:06:40.470> messages<00:06:40.890> that 00:06:41.030 --> 00:06:41.040 align:start position:0% consensus we have protocol messages that 00:06:41.040 --> 00:06:42.680 align:start position:0% consensus we have protocol messages that are<00:06:41.160> bits<00:06:41.340> they<00:06:41.850> sorry<00:06:42.240> they<00:06:42.360> have<00:06:42.480> three 00:06:42.680 --> 00:06:42.690 align:start position:0% are bits they sorry they have three 00:06:42.690 --> 00:06:45.310 align:start position:0% are bits they sorry they have three parts<00:06:42.960> a<00:06:43.050> bit<00:06:43.560> for<00:06:44.250> the<00:06:44.340> for<00:06:44.640> the<00:06:44.760> estimate 00:06:45.310 --> 00:06:45.320 align:start position:0% parts a bit for the for the estimate 00:06:45.320 --> 00:06:47.540 align:start position:0% parts a bit for the for the estimate sender<00:06:46.320> which<00:06:46.470> is<00:06:46.590> like<00:06:46.710> the<00:06:46.800> node<00:06:47.100> assented 00:06:47.540 --> 00:06:47.550 align:start position:0% sender which is like the node assented 00:06:47.550 --> 00:06:49.610 align:start position:0% sender which is like the node assented and<00:06:47.640> a<00:06:48.240> justification<00:06:48.870> which<00:06:49.020> is<00:06:49.140> a<00:06:49.170> set<00:06:49.530> of 00:06:49.610 --> 00:06:49.620 align:start position:0% and a justification which is a set of 00:06:49.620 --> 00:06:54.380 align:start position:0% and a justification which is a set of messages<00:06:50.780> the<00:06:53.300> protocol<00:06:54.300> messages 00:06:54.380 --> 00:06:54.390 align:start position:0% messages the protocol messages 00:06:54.390 --> 00:06:56.480 align:start position:0% messages the protocol messages dependency<00:06:54.960> of<00:06:54.990> the<00:06:55.140> other<00:06:55.290> if<00:06:55.590> it's<00:06:55.770> like<00:06:55.950> in 00:06:56.480 --> 00:06:56.490 align:start position:0% dependency of the other if it's like in 00:06:56.490 --> 00:06:58.250 align:start position:0% dependency of the other if it's like in the<00:06:56.850> justification<00:06:57.570> or<00:06:57.840> in<00:06:58.080> those 00:06:58.250 --> 00:06:58.260 align:start position:0% the justification or in those 00:06:58.260 --> 00:06:59.540 align:start position:0% the justification or in those justifications<00:06:58.590> basically<00:06:59.130> if<00:06:59.190> there's<00:06:59.340> pre 00:06:59.540 --> 00:06:59.550 align:start position:0% justifications basically if there's pre 00:06:59.550 --> 00:07:01.430 align:start position:0% justifications basically if there's pre block<00:06:59.790> pointers<00:07:00.210> or<00:07:00.360> pre<00:07:00.720> patch<00:07:00.900> pointers<00:07:01.260> to 00:07:01.430 --> 00:07:01.440 align:start position:0% block pointers or pre patch pointers to 00:07:01.440 --> 00:07:03.950 align:start position:0% block pointers or pre patch pointers to it<00:07:01.590> and<00:07:01.770> then<00:07:01.860> it's<00:07:01.980> a<00:07:02.010> dependency<00:07:02.960> latest 00:07:03.950 --> 00:07:03.960 align:start position:0% it and then it's a dependency latest 00:07:03.960 --> 00:07:08.030 align:start position:0% it and then it's a dependency latest message<00:07:04.380> is<00:07:04.590> one<00:07:05.430> which<00:07:06.260> there's<00:07:07.260> no<00:07:07.530> there's 00:07:08.030 --> 00:07:08.040 align:start position:0% message is one which there's no there's 00:07:08.040 --> 00:07:09.470 align:start position:0% message is one which there's no there's there's<00:07:08.370> nothing<00:07:08.730> earlier<00:07:09.030> right<00:07:09.330> so 00:07:09.470 --> 00:07:09.480 align:start position:0% there's nothing earlier right so 00:07:09.480 --> 00:07:12.530 align:start position:0% there's nothing earlier right so dependencies<00:07:10.110> earlier<00:07:11.060> if<00:07:12.060> you're<00:07:12.390> not 00:07:12.530 --> 00:07:12.540 align:start position:0% dependencies earlier if you're not 00:07:12.540 --> 00:07:14.270 align:start position:0% dependencies earlier if you're not depending<00:07:12.900> than<00:07:13.020> later<00:07:13.350> and<00:07:13.620> a<00:07:13.950> latest 00:07:14.270 --> 00:07:14.280 align:start position:0% depending than later and a latest 00:07:14.280 --> 00:07:17.300 align:start position:0% depending than later and a latest message<00:07:14.900> is<00:07:15.900> it's<00:07:16.230> kind<00:07:16.590> of<00:07:16.650> extremum<00:07:17.130> there 00:07:17.300 --> 00:07:17.310 align:start position:0% message is it's kind of extremum there 00:07:17.310 --> 00:07:19.520 align:start position:0% message is it's kind of extremum there so<00:07:18.090> the<00:07:18.480> way<00:07:18.630> that<00:07:18.660> the<00:07:18.900> binary<00:07:19.080> comes<00:07:19.380> as<00:07:19.470> this 00:07:19.520 --> 00:07:19.530 align:start position:0% so the way that the binary comes as this 00:07:19.530 --> 00:07:21.110 align:start position:0% so the way that the binary comes as this works<00:07:19.770> is<00:07:19.980> you<00:07:20.160> look<00:07:20.370> at<00:07:20.460> all<00:07:20.580> the<00:07:20.730> validators 00:07:21.110 --> 00:07:21.120 align:start position:0% works is you look at all the validators 00:07:21.120 --> 00:07:23.060 align:start position:0% works is you look at all the validators latest<00:07:21.420> messages<00:07:21.720> if<00:07:21.990> more<00:07:22.650> of<00:07:22.740> them<00:07:22.890> by 00:07:23.060 --> 00:07:23.070 align:start position:0% latest messages if more of them by 00:07:23.070 --> 00:07:25.280 align:start position:0% latest messages if more of them by weight<00:07:23.280> have<00:07:23.520> zero<00:07:23.820> then<00:07:24.450> you<00:07:24.570> pick<00:07:24.810> zero<00:07:25.050> and 00:07:25.280 --> 00:07:25.290 align:start position:0% weight have zero then you pick zero and 00:07:25.290 --> 00:07:26.720 align:start position:0% weight have zero then you pick zero and if<00:07:25.350> more<00:07:25.500> than<00:07:25.530> by<00:07:25.740> weight<00:07:25.950> have<00:07:26.130> one<00:07:26.370> then<00:07:26.700> you 00:07:26.720 --> 00:07:26.730 align:start position:0% if more than by weight have one then you 00:07:26.730 --> 00:07:27.470 align:start position:0% if more than by weight have one then you pick<00:07:26.910> zero 00:07:27.470 --> 00:07:27.480 align:start position:0% pick zero 00:07:27.480 --> 00:07:30.680 align:start position:0% pick zero they<00:07:27.720> pick<00:07:27.900> one<00:07:28.080> so<00:07:28.410> basically<00:07:28.680> if<00:07:29.150> this<00:07:30.150> sort 00:07:30.680 --> 00:07:30.690 align:start position:0% they pick one so basically if this sort 00:07:30.690 --> 00:07:32.570 align:start position:0% they pick one so basically if this sort of<00:07:30.720> score<00:07:31.020> of<00:07:31.140> an<00:07:31.260> estimate<00:07:31.620> is<00:07:31.830> the<00:07:32.160> sum<00:07:32.550> of 00:07:32.570 --> 00:07:32.580 align:start position:0% of score of an estimate is the sum of 00:07:32.580 --> 00:07:34.850 align:start position:0% of score of an estimate is the sum of the<00:07:32.730> validators<00:07:33.300> who<00:07:33.720> have<00:07:33.900> latest<00:07:34.380> messages 00:07:34.850 --> 00:07:34.860 align:start position:0% the validators who have latest messages 00:07:34.860 --> 00:07:38.840 align:start position:0% the validators who have latest messages who<00:07:35.550> have<00:07:36.270> that<00:07:36.510> estimate<00:07:36.900> and<00:07:37.400> if<00:07:38.400> the<00:07:38.580> score 00:07:38.840 --> 00:07:38.850 align:start position:0% who have that estimate and if the score 00:07:38.850 --> 00:07:40.460 align:start position:0% who have that estimate and if the score on<00:07:39.180> zero<00:07:39.420> is<00:07:39.600> more<00:07:39.720> than<00:07:39.810> sierra<00:07:40.020> one<00:07:40.230> the 00:07:40.460 --> 00:07:40.470 align:start position:0% on zero is more than sierra one the 00:07:40.470 --> 00:07:42.200 align:start position:0% on zero is more than sierra one the estimator<00:07:40.770> returns<00:07:41.220> zero<00:07:41.430> otherwise<00:07:41.880> it 00:07:42.200 --> 00:07:42.210 align:start position:0% estimator returns zero otherwise it 00:07:42.210 --> 00:07:44.570 align:start position:0% estimator returns zero otherwise it returns<00:07:42.330> one<00:07:42.750> and<00:07:43.530> otherwise<00:07:43.770> and<00:07:44.100> and<00:07:44.520> if 00:07:44.570 --> 00:07:44.580 align:start position:0% returns one and otherwise and and if 00:07:44.580 --> 00:07:45.830 align:start position:0% returns one and otherwise and and if it's<00:07:44.730> a<00:07:44.760> tie<00:07:44.940> then<00:07:45.210> it<00:07:45.300> returns<00:07:45.540> it's<00:07:45.660> kind<00:07:45.780> of 00:07:45.830 --> 00:07:45.840 align:start position:0% it's a tie then it returns it's kind of 00:07:45.840 --> 00:07:51.050 align:start position:0% it's a tie then it returns it's kind of exception<00:07:47.000> so<00:07:49.430> so<00:07:50.430> so<00:07:50.640> that's<00:07:50.790> kind<00:07:50.940> of<00:07:50.970> a 00:07:51.050 --> 00:07:51.060 align:start position:0% exception so so so that's kind of a 00:07:51.060 --> 00:07:53.540 align:start position:0% exception so so so that's kind of a definition<00:07:51.330> of<00:07:51.450> estimator<00:07:51.900> and<00:07:52.770> and<00:07:53.190> that's 00:07:53.540 --> 00:07:53.550 align:start position:0% definition of estimator and and that's 00:07:53.550 --> 00:07:55.520 align:start position:0% definition of estimator and and that's that<00:07:53.880> that's<00:07:54.270> kind<00:07:54.510> of<00:07:54.570> gonna<00:07:55.230> be<00:07:55.350> necessary 00:07:55.520 --> 00:07:55.530 align:start position:0% that that's kind of gonna be necessary 00:07:55.530 --> 00:07:57.650 align:start position:0% that that's kind of gonna be necessary for<00:07:55.860> us<00:07:56.070> to<00:07:56.190> have<00:07:56.790> the<00:07:57.000> concept<00:07:57.300> of<00:07:57.360> deciding 00:07:57.650 --> 00:07:57.660 align:start position:0% for us to have the concept of deciding 00:07:57.660 --> 00:07:59.660 align:start position:0% for us to have the concept of deciding on<00:07:57.750> safe<00:07:57.930> estimates<00:07:58.290> so<00:07:58.770> Byzantine<00:07:59.370> faults 00:07:59.660 --> 00:07:59.670 align:start position:0% on safe estimates so Byzantine faults 00:07:59.670 --> 00:08:00.860 align:start position:0% on safe estimates so Byzantine faults are<00:07:59.760> kind<00:07:59.910> of<00:07:59.970> this<00:08:00.060> interesting<00:08:00.720> and<00:08:00.810> tricky 00:08:00.860 --> 00:08:00.870 align:start position:0% are kind of this interesting and tricky 00:08:00.870 --> 00:08:03.560 align:start position:0% are kind of this interesting and tricky thing<00:08:01.490> basically<00:08:02.490> each<00:08:02.700> of<00:08:02.850> these<00:08:02.910> nodes<00:08:03.180> have 00:08:03.560 --> 00:08:03.570 align:start position:0% thing basically each of these nodes have 00:08:03.570 --> 00:08:05.420 align:start position:0% thing basically each of these nodes have are<00:08:04.080> making<00:08:04.290> messages<00:08:04.650> where<00:08:04.920> they<00:08:05.040> store 00:08:05.420 --> 00:08:05.430 align:start position:0% are making messages where they store 00:08:05.430 --> 00:08:06.680 align:start position:0% are making messages where they store they<00:08:05.520> detest<00:08:05.910> what<00:08:06.120> they've<00:08:06.240> seen<00:08:06.420> from<00:08:06.570> other 00:08:06.680 --> 00:08:06.690 align:start position:0% they detest what they've seen from other 00:08:06.690 --> 00:08:09.230 align:start position:0% they detest what they've seen from other nodes<00:08:06.930> right<00:08:07.230> and<00:08:07.500> if<00:08:08.040> it's<00:08:08.700> ever<00:08:08.910> the<00:08:09.060> case 00:08:09.230 --> 00:08:09.240 align:start position:0% nodes right and if it's ever the case 00:08:09.240 --> 00:08:10.580 align:start position:0% nodes right and if it's ever the case that<00:08:09.510> I<00:08:09.540> have<00:08:09.660> a<00:08:09.690> set<00:08:09.870> of<00:08:09.900> messages<00:08:10.320> where 00:08:10.580 --> 00:08:10.590 align:start position:0% that I have a set of messages where 00:08:10.590 --> 00:08:13.820 align:start position:0% that I have a set of messages where different<00:08:11.550> nodes<00:08:11.730> attest<00:08:12.270> to<00:08:12.590> having<00:08:13.590> seen 00:08:13.820 --> 00:08:13.830 align:start position:0% different nodes attest to having seen 00:08:13.830 --> 00:08:16.430 align:start position:0% different nodes attest to having seen some<00:08:14.280> node<00:08:14.520> make<00:08:15.240> messages<00:08:15.690> that<00:08:15.840> could<00:08:16.200> never 00:08:16.430 --> 00:08:16.440 align:start position:0% some node make messages that could never 00:08:16.440 --> 00:08:19.450 align:start position:0% some node make messages that could never have<00:08:16.650> possibly<00:08:17.040> been<00:08:17.130> made<00:08:17.460> by<00:08:17.640> a<00:08:17.670> single<00:08:18.090> node 00:08:19.450 --> 00:08:19.460 align:start position:0% have possibly been made by a single node 00:08:19.460 --> 00:08:22.700 align:start position:0% have possibly been made by a single node then<00:08:20.460> we<00:08:20.850> have<00:08:21.480> a<00:08:21.510> Byzantine<00:08:22.080> fault<00:08:22.320> because 00:08:22.700 --> 00:08:22.710 align:start position:0% then we have a Byzantine fault because 00:08:22.710 --> 00:08:23.900 align:start position:0% then we have a Byzantine fault because basically<00:08:22.890> a<00:08:23.040> single<00:08:23.370> nodes<00:08:23.550> with<00:08:23.700> the<00:08:23.820> way 00:08:23.900 --> 00:08:23.910 align:start position:0% basically a single nodes with the way 00:08:23.910 --> 00:08:25.310 align:start position:0% basically a single nodes with the way they're<00:08:24.120> characterized<00:08:24.570> is<00:08:24.780> that<00:08:25.230> they 00:08:25.310 --> 00:08:25.320 align:start position:0% they're characterized is that they 00:08:25.320 --> 00:08:29.249 align:start position:0% they're characterized is that they follow<00:08:25.710> protocol<00:08:26.370> executions<00:08:27.320> according<00:08:28.320> to 00:08:29.249 --> 00:08:29.259 align:start position:0% follow protocol executions according to 00:08:29.259 --> 00:08:32.279 align:start position:0% follow protocol executions according to the<00:08:29.530> like<00:08:29.830> the<00:08:29.979> protocol<00:08:30.550> executions<00:08:31.090> in<00:08:31.990> in 00:08:32.279 --> 00:08:32.289 align:start position:0% the like the protocol executions in in 00:08:32.289 --> 00:08:35.909 align:start position:0% the like the protocol executions in in the<00:08:32.440> state<00:08:32.680> so<00:08:32.919> so<00:08:33.310> here<00:08:34.710> equivocation<00:08:35.710> is 00:08:35.909 --> 00:08:35.919 align:start position:0% the state so so here equivocation is 00:08:35.919 --> 00:08:38.820 align:start position:0% the state so so here equivocation is this<00:08:36.490> is<00:08:36.789> this<00:08:37.510> property<00:08:37.930> that<00:08:37.960> to<00:08:38.169> knows<00:08:38.349> that 00:08:38.820 --> 00:08:38.830 align:start position:0% this is this property that to knows that 00:08:38.830 --> 00:08:40.139 align:start position:0% this is this property that to knows that a<00:08:38.890> data<00:08:39.130> that<00:08:39.279> it's<00:08:39.399> that<00:08:39.550> a<00:08:39.580> note<00:08:39.820> isn't 00:08:40.139 --> 00:08:40.149 align:start position:0% a data that it's that a note isn't 00:08:40.149 --> 00:08:42.300 align:start position:0% a data that it's that a note isn't actually<00:08:40.659> a<00:08:40.690> single-threaded<00:08:41.140> and<00:08:41.830> it<00:08:42.190> means 00:08:42.300 --> 00:08:42.310 align:start position:0% actually a single-threaded and it means 00:08:42.310 --> 00:08:44.490 align:start position:0% actually a single-threaded and it means that<00:08:42.370> they<00:08:42.490> made<00:08:42.640> two<00:08:42.880> messages<00:08:43.330> neither<00:08:44.320> of 00:08:44.490 --> 00:08:44.500 align:start position:0% that they made two messages neither of 00:08:44.500 --> 00:08:45.780 align:start position:0% that they made two messages neither of which<00:08:44.560> are<00:08:44.770> in<00:08:44.860> the<00:08:44.890> dependency<00:08:45.310> of<00:08:45.339> the<00:08:45.430> other 00:08:45.780 --> 00:08:45.790 align:start position:0% which are in the dependency of the other 00:08:45.790 --> 00:08:47.430 align:start position:0% which are in the dependency of the other so<00:08:46.209> rather<00:08:46.390> than<00:08:46.480> including<00:08:46.959> their<00:08:47.080> messages 00:08:47.430 --> 00:08:47.440 align:start position:0% so rather than including their messages 00:08:47.440 --> 00:08:48.570 align:start position:0% so rather than including their messages in<00:08:47.529> each<00:08:47.620> other<00:08:47.770> and<00:08:47.950> having<00:08:48.130> a<00:08:48.160> single<00:08:48.399> thread 00:08:48.570 --> 00:08:48.580 align:start position:0% in each other and having a single thread 00:08:48.580 --> 00:08:50.400 align:start position:0% in each other and having a single thread they<00:08:48.850> have<00:08:48.970> them<00:08:49.089> all<00:08:49.240> it's<00:08:49.839> multi-threaded 00:08:50.400 --> 00:08:50.410 align:start position:0% they have them all it's multi-threaded 00:08:50.410 --> 00:08:53.550 align:start position:0% they have them all it's multi-threaded and<00:08:50.880> we<00:08:51.880> can<00:08:52.000> basically<00:08:52.420> are<00:08:52.810> gonna<00:08:52.959> be<00:08:53.500> able 00:08:53.550 --> 00:08:53.560 align:start position:0% and we can basically are gonna be able 00:08:53.560 --> 00:08:55.350 align:start position:0% and we can basically are gonna be able to<00:08:53.680> identify<00:08:54.190> Byzantine<00:08:54.760> nodes<00:08:54.940> by<00:08:55.180> their 00:08:55.350 --> 00:08:55.360 align:start position:0% to identify Byzantine nodes by their 00:08:55.360 --> 00:08:57.269 align:start position:0% to identify Byzantine nodes by their equivocations<00:08:55.930> and<00:08:56.230> we<00:08:56.709> can<00:08:56.860> like<00:08:56.980> you<00:08:57.160> know 00:08:57.269 --> 00:08:57.279 align:start position:0% equivocations and we can like you know 00:08:57.279 --> 00:08:59.610 align:start position:0% equivocations and we can like you know like<00:08:57.700> count<00:08:58.000> them<00:08:58.120> all<00:08:58.209> up<00:08:58.360> we<00:08:59.320> can<00:08:59.470> measure 00:08:59.610 --> 00:08:59.620 align:start position:0% like count them all up we can measure 00:08:59.620 --> 00:09:01.800 align:start position:0% like count them all up we can measure their<00:08:59.950> weight<00:09:00.160> and<00:09:00.610> then<00:09:01.510> we<00:09:01.600> can<00:09:01.720> define 00:09:01.800 --> 00:09:01.810 align:start position:0% their weight and then we can define 00:09:01.810 --> 00:09:03.690 align:start position:0% their weight and then we can define protocol<00:09:02.290> States<00:09:02.529> as<00:09:02.740> all<00:09:03.190> of<00:09:03.370> the<00:09:03.459> sets<00:09:03.640> of 00:09:03.690 --> 00:09:03.700 align:start position:0% protocol States as all of the sets of 00:09:03.700 --> 00:09:04.829 align:start position:0% protocol States as all of the sets of messages<00:09:04.060> that<00:09:04.089> evidence<00:09:04.570> less<00:09:04.720> than 00:09:04.829 --> 00:09:04.839 align:start position:0% messages that evidence less than 00:09:04.839 --> 00:09:06.900 align:start position:0% messages that evidence less than something<00:09:04.990> were<00:09:05.170> faults<00:09:05.470> then<00:09:06.459> we<00:09:06.610> defined 00:09:06.900 --> 00:09:06.910 align:start position:0% something were faults then we defined 00:09:06.910 --> 00:09:09.210 align:start position:0% something were faults then we defined protocol<00:09:07.209> executions<00:09:07.750> just<00:09:08.020> as<00:09:08.140> supersets<00:09:08.649> in 00:09:09.210 --> 00:09:09.220 align:start position:0% protocol executions just as supersets in 00:09:09.220 --> 00:09:11.550 align:start position:0% protocol executions just as supersets in those<00:09:09.459> things<00:09:09.700> and<00:09:09.910> then<00:09:10.750> we're<00:09:11.230> basically 00:09:11.550 --> 00:09:11.560 align:start position:0% those things and then we're basically 00:09:11.560 --> 00:09:15.300 align:start position:0% those things and then we're basically done<00:09:11.860> the<00:09:12.640> binary<00:09:14.010> estimate<00:09:15.010> safety 00:09:15.300 --> 00:09:15.310 align:start position:0% done the binary estimate safety 00:09:15.310 --> 00:09:17.160 align:start position:0% done the binary estimate safety basically<00:09:15.610> means<00:09:15.760> that<00:09:15.880> the<00:09:16.240> so<00:09:17.050> it's 00:09:17.160 --> 00:09:17.170 align:start position:0% basically means that the so it's 00:09:17.170 --> 00:09:19.410 align:start position:0% basically means that the so it's estimate<00:09:18.040> is<00:09:18.130> safe<00:09:18.370> give<00:09:19.060> us<00:09:19.120> some<00:09:19.240> fault 00:09:19.410 --> 00:09:19.420 align:start position:0% estimate is safe give us some fault 00:09:19.420 --> 00:09:21.000 align:start position:0% estimate is safe give us some fault tolerance<00:09:19.810> in<00:09:19.959> some<00:09:20.110> state<00:09:20.320> if<00:09:20.560> for<00:09:20.860> all 00:09:21.000 --> 00:09:21.010 align:start position:0% tolerance in some state if for all 00:09:21.010 --> 00:09:25.740 align:start position:0% tolerance in some state if for all future<00:09:21.370> states<00:09:22.560> that<00:09:23.820> estimate<00:09:24.820> is<00:09:24.940> also<00:09:25.630> the 00:09:25.740 --> 00:09:25.750 align:start position:0% future states that estimate is also the 00:09:25.750 --> 00:09:26.940 align:start position:0% future states that estimate is also the estimator<00:09:26.050> in<00:09:26.140> future<00:09:26.380> states<00:09:26.589> so<00:09:26.800> there's<00:09:26.920> I 00:09:26.940 --> 00:09:26.950 align:start position:0% estimator in future states so there's I 00:09:26.950 --> 00:09:28.290 align:start position:0% estimator in future states so there's I get<00:09:27.070> this<00:09:27.160> invariance<00:09:27.640> right<00:09:27.790> all<00:09:28.029> future 00:09:28.290 --> 00:09:28.300 align:start position:0% get this invariance right all future 00:09:28.300 --> 00:09:29.460 align:start position:0% get this invariance right all future states<00:09:28.480> will<00:09:28.600> return<00:09:28.810> zero<00:09:29.080> all<00:09:29.290> future 00:09:29.460 --> 00:09:29.470 align:start position:0% states will return zero all future 00:09:29.470 --> 00:09:30.750 align:start position:0% states will return zero all future states<00:09:29.709> tomorrow 00:09:30.750 --> 00:09:30.760 align:start position:0% states tomorrow 00:09:30.760 --> 00:09:32.579 align:start position:0% states tomorrow and<00:09:30.930> because<00:09:31.930> of<00:09:31.959> the<00:09:32.050> way<00:09:32.110> we<00:09:32.230> constructed 00:09:32.579 --> 00:09:32.589 align:start position:0% and because of the way we constructed 00:09:32.589 --> 00:09:34.290 align:start position:0% and because of the way we constructed this<00:09:32.709> this<00:09:33.190> satisfies<00:09:33.610> our<00:09:33.850> safety<00:09:34.029> proof 00:09:34.290 --> 00:09:34.300 align:start position:0% this this satisfies our safety proof 00:09:34.300 --> 00:09:35.579 align:start position:0% this this satisfies our safety proof they<00:09:34.570> all<00:09:34.690> have<00:09:34.810> future<00:09:35.110> states<00:09:35.410> unless 00:09:35.579 --> 00:09:35.589 align:start position:0% they all have future states unless 00:09:35.589 --> 00:09:36.990 align:start position:0% they all have future states unless there's<00:09:35.800> T<00:09:35.920> faults<00:09:36.250> if<00:09:36.610> they<00:09:36.760> have<00:09:36.850> future 00:09:36.990 --> 00:09:37.000 align:start position:0% there's T faults if they have future 00:09:37.000 --> 00:09:38.100 align:start position:0% there's T faults if they have future states<00:09:37.300> they<00:09:37.420> have<00:09:37.450> good<00:09:37.660> sense<00:09:37.839> of<00:09:37.930> safety 00:09:38.100 --> 00:09:38.110 align:start position:0% states they have good sense of safety 00:09:38.110 --> 00:09:40.170 align:start position:0% states they have good sense of safety and<00:09:38.410> so<00:09:38.950> we<00:09:39.160> have<00:09:39.310> a<00:09:39.339> binary<00:09:39.670> consensus 00:09:40.170 --> 00:09:40.180 align:start position:0% and so we have a binary consensus 00:09:40.180 --> 00:09:44.010 align:start position:0% and so we have a binary consensus protocol<00:09:40.510> with<00:09:40.810> safety<00:09:41.140> and<00:09:42.450> you<00:09:43.450> know<00:09:43.570> this 00:09:44.010 --> 00:09:44.020 align:start position:0% protocol with safety and you know this 00:09:44.020 --> 00:09:45.240 align:start position:0% protocol with safety and you know this is<00:09:44.170> kind<00:09:44.380> of<00:09:44.410> where<00:09:44.589> we<00:09:44.620> can<00:09:44.800> relax<00:09:45.040> because 00:09:45.240 --> 00:09:45.250 align:start position:0% is kind of where we can relax because 00:09:45.250 --> 00:09:46.769 align:start position:0% is kind of where we can relax because this<00:09:45.430> was<00:09:45.640> that<00:09:46.000> was<00:09:46.120> all<00:09:46.240> the<00:09:46.330> heart<00:09:46.480> that<00:09:46.660> was 00:09:46.769 --> 00:09:46.779 align:start position:0% this was that was all the heart that was 00:09:46.779 --> 00:09:50.220 align:start position:0% this was that was all the heart that was most<00:09:47.350> of<00:09:47.529> the<00:09:47.589> hard<00:09:47.740> work<00:09:48.959> we're<00:09:49.959> never<00:09:50.080> gonna 00:09:50.220 --> 00:09:50.230 align:start position:0% most of the hard work we're never gonna 00:09:50.230 --> 00:09:51.449 align:start position:0% most of the hard work we're never gonna find<00:09:50.410> the<00:09:50.560> block<00:09:50.680> tanks<00:09:50.950> in<00:09:51.070> this<00:09:51.130> protocol 00:09:51.449 --> 00:09:51.459 align:start position:0% find the block tanks in this protocol 00:09:51.459 --> 00:09:52.699 align:start position:0% find the block tanks in this protocol and<00:09:51.520> it's<00:09:51.820> really<00:09:52.180> gonna<00:09:52.300> be<00:09:52.450> a<00:09:52.480> lot<00:09:52.660> easier 00:09:52.699 --> 00:09:52.709 align:start position:0% and it's really gonna be a lot easier 00:09:52.709 --> 00:09:54.720 align:start position:0% and it's really gonna be a lot easier because<00:09:53.709> we've<00:09:53.950> already<00:09:54.130> done<00:09:54.370> most<00:09:54.520> of<00:09:54.640> the 00:09:54.720 --> 00:09:54.730 align:start position:0% because we've already done most of the 00:09:54.730 --> 00:09:59.490 align:start position:0% because we've already done most of the work<00:09:54.760> so<00:09:55.890> the<00:09:56.890> blocks<00:09:57.640> kind<00:09:58.390> of<00:09:58.540> look<00:09:59.290> very 00:09:59.490 --> 00:09:59.500 align:start position:0% work so the blocks kind of look very 00:09:59.500 --> 00:10:01.350 align:start position:0% work so the blocks kind of look very similar<00:09:59.860> to<00:10:00.040> the<00:10:00.220> messages<00:10:01.060> for<00:10:01.120> the<00:10:01.240> binary 00:10:01.350 --> 00:10:01.360 align:start position:0% similar to the messages for the binary 00:10:01.360 --> 00:10:02.730 align:start position:0% similar to the messages for the binary consensus<00:10:01.870> only<00:10:02.080> instead<00:10:02.230> of<00:10:02.320> having<00:10:02.440> a<00:10:02.529> bit 00:10:02.730 --> 00:10:02.740 align:start position:0% consensus only instead of having a bit 00:10:02.740 --> 00:10:05.280 align:start position:0% consensus only instead of having a bit there's<00:10:03.730> a<00:10:03.820> there's<00:10:04.240> another<00:10:04.450> block<00:10:04.779> in<00:10:05.020> the 00:10:05.280 --> 00:10:05.290 align:start position:0% there's a there's another block in the 00:10:05.290 --> 00:10:07.920 align:start position:0% there's a there's another block in the first<00:10:05.529> in<00:10:05.770> the<00:10:06.010> in<00:10:06.160> the<00:10:06.279> estimate<00:10:06.670> spot<00:10:06.930> so 00:10:07.920 --> 00:10:07.930 align:start position:0% first in the in the estimate spot so 00:10:07.930 --> 00:10:09.840 align:start position:0% first in the in the estimate spot so every<00:10:08.290> block<00:10:08.529> kind<00:10:08.830> of<00:10:08.890> has<00:10:08.980> a<00:10:09.010> pre<00:10:09.310> every 00:10:09.840 --> 00:10:09.850 align:start position:0% every block kind of has a pre every 00:10:09.850 --> 00:10:11.640 align:start position:0% every block kind of has a pre every block<00:10:10.089> kind<00:10:10.330> of<00:10:10.390> has<00:10:10.480> a<00:10:10.510> pre<00:10:10.779> block<00:10:11.020> a<00:10:11.200> sender 00:10:11.640 --> 00:10:11.650 align:start position:0% block kind of has a pre block a sender 00:10:11.650 --> 00:10:15.030 align:start position:0% block kind of has a pre block a sender and<00:10:11.800> justification<00:10:13.470> and<00:10:14.470> there's<00:10:14.860> this<00:10:14.980> thing 00:10:15.030 --> 00:10:15.040 align:start position:0% and justification and there's this thing 00:10:15.040 --> 00:10:17.430 align:start position:0% and justification and there's this thing called<00:10:15.190> the<00:10:15.400> Genesis<00:10:15.790> block<00:10:15.910> and<00:10:16.420> like<00:10:17.290> the 00:10:17.430 --> 00:10:17.440 align:start position:0% called the Genesis block and like the 00:10:17.440 --> 00:10:19.290 align:start position:0% called the Genesis block and like the and<00:10:17.980> and<00:10:18.160> and<00:10:18.310> this<00:10:18.670> is<00:10:18.760> kind<00:10:18.910> of<00:10:18.940> a<00:10:19.029> Heights 00:10:19.290 --> 00:10:19.300 align:start position:0% and and and this is kind of a Heights 00:10:19.300 --> 00:10:23.819 align:start position:0% and and and this is kind of a Heights the<00:10:19.870> hot<00:10:20.079> the<00:10:20.230> heights<00:10:20.470> of<00:10:20.650> the<00:10:20.740> blocks<00:10:20.980> so<00:10:22.829> you 00:10:23.819 --> 00:10:23.829 align:start position:0% the hot the heights of the blocks so you 00:10:23.829 --> 00:10:25.230 align:start position:0% the hot the heights of the blocks so you know<00:10:23.920> this<00:10:24.070> is<00:10:24.130> me<00:10:24.250> defining<00:10:24.640> like<00:10:24.760> all<00:10:24.940> of<00:10:25.149> the 00:10:25.230 --> 00:10:25.240 align:start position:0% know this is me defining like all of the 00:10:25.240 --> 00:10:26.430 align:start position:0% know this is me defining like all of the blocks<00:10:25.450> being<00:10:25.660> all<00:10:25.810> the<00:10:25.930> blocks<00:10:26.140> at<00:10:26.290> all<00:10:26.350> the 00:10:26.430 --> 00:10:26.440 align:start position:0% blocks being all the blocks at all the 00:10:26.440 --> 00:10:30.870 align:start position:0% blocks being all the blocks at all the Heights<00:10:28.529> so<00:10:29.529> there's<00:10:30.459> a<00:10:30.550> couple<00:10:30.700> more 00:10:30.870 --> 00:10:30.880 align:start position:0% Heights so there's a couple more 00:10:30.880 --> 00:10:32.130 align:start position:0% Heights so there's a couple more definition<00:10:31.270> that<00:10:31.329> we<00:10:31.420> need<00:10:31.570> basically 00:10:32.130 --> 00:10:32.140 align:start position:0% definition that we need basically 00:10:32.140 --> 00:10:35.040 align:start position:0% definition that we need basically instead<00:10:33.010> of<00:10:33.070> having<00:10:33.310> an<00:10:33.520> estimate<00:10:34.209> just<00:10:34.690> be 00:10:35.040 --> 00:10:35.050 align:start position:0% instead of having an estimate just be 00:10:35.050 --> 00:10:36.540 align:start position:0% instead of having an estimate just be the<00:10:35.230> same<00:10:35.470> we're<00:10:35.980> gonna<00:10:36.070> have<00:10:36.220> the<00:10:36.279> estimate 00:10:36.540 --> 00:10:36.550 align:start position:0% the same we're gonna have the estimate 00:10:36.550 --> 00:10:37.980 align:start position:0% the same we're gonna have the estimate be<00:10:36.640> a<00:10:36.670> member<00:10:37.000> of<00:10:37.089> the<00:10:37.209> blockchain<00:10:37.360> and<00:10:37.779> that's 00:10:37.980 --> 00:10:37.990 align:start position:0% be a member of the blockchain and that's 00:10:37.990 --> 00:10:40.350 align:start position:0% be a member of the blockchain and that's what<00:10:38.079> this<00:10:38.170> represents<00:10:38.410> like<00:10:38.940> block<00:10:39.940> one<00:10:40.150> is 00:10:40.350 --> 00:10:40.360 align:start position:0% what this represents like block one is 00:10:40.360 --> 00:10:42.390 align:start position:0% what this represents like block one is in<00:10:40.810> the<00:10:40.900> blockchain<00:10:41.079> a<00:10:41.230> block<00:10:41.410> two 00:10:42.390 --> 00:10:42.400 align:start position:0% in the blockchain a block two 00:10:42.400 --> 00:10:45.660 align:start position:0% in the blockchain a block two and<00:10:42.490> then<00:10:43.090> the<00:10:43.390> score<00:10:43.720> of<00:10:43.960> a<00:10:44.050> block<00:10:44.290> is<00:10:44.670> the 00:10:45.660 --> 00:10:45.670 align:start position:0% and then the score of a block is the 00:10:45.670 --> 00:10:47.280 align:start position:0% and then the score of a block is the weight<00:10:45.940> of<00:10:46.150> validators<00:10:46.780> whose<00:10:46.990> latest 00:10:47.280 --> 00:10:47.290 align:start position:0% weight of validators whose latest 00:10:47.290 --> 00:10:55.140 align:start position:0% weight of validators whose latest messages<00:10:49.110> are<00:10:50.700> have<00:10:51.700> the<00:10:51.940> Block<00:10:53.730> in<00:10:54.730> their 00:10:55.140 --> 00:10:55.150 align:start position:0% messages are have the Block in their 00:10:55.150 --> 00:10:58.260 align:start position:0% messages are have the Block in their block<00:10:55.360> chains<00:10:55.630> so<00:10:56.110> if<00:10:56.230> if<00:10:56.890> five<00:10:57.400> long<00:10:57.640> if<00:10:57.790> like 00:10:58.260 --> 00:10:58.270 align:start position:0% block chains so if if five long if like 00:10:58.270 --> 00:11:00.300 align:start position:0% block chains so if if five long if like five<00:10:58.570> way<00:10:58.810> to<00:10:58.870> the<00:10:59.020> validators<00:10:59.440> have<00:10:59.800> block<00:11:00.100> a 00:11:00.300 --> 00:11:00.310 align:start position:0% five way to the validators have block a 00:11:00.310 --> 00:11:03.270 align:start position:0% five way to the validators have block a in<00:11:00.940> their<00:11:01.540> latest<00:11:01.870> block<00:11:02.050> chains<00:11:02.260> and<00:11:02.440> three 00:11:03.270 --> 00:11:03.280 align:start position:0% in their latest block chains and three 00:11:03.280 --> 00:11:04.500 align:start position:0% in their latest block chains and three of<00:11:03.460> the<00:11:03.550> weight<00:11:03.670> have<00:11:03.940> locked<00:11:04.150> be<00:11:04.330> in<00:11:04.450> the 00:11:04.500 --> 00:11:04.510 align:start position:0% of the weight have locked be in the 00:11:04.510 --> 00:11:06.540 align:start position:0% of the weight have locked be in the latest<00:11:04.750> block<00:11:05.530> chains<00:11:05.740> and<00:11:05.890> block<00:11:06.070> a<00:11:06.250> will 00:11:06.540 --> 00:11:06.550 align:start position:0% latest block chains and block a will 00:11:06.550 --> 00:11:09.090 align:start position:0% latest block chains and block a will have<00:11:06.700> a<00:11:06.730> higher<00:11:06.880> score<00:11:07.380> so<00:11:08.380> then<00:11:08.530> we<00:11:08.920> can<00:11:09.040> have 00:11:09.090 --> 00:11:09.100 align:start position:0% have a higher score so then we can have 00:11:09.100 --> 00:11:11.040 align:start position:0% have a higher score so then we can have this<00:11:09.370> fork<00:11:10.000> choice<00:11:10.180> rule<00:11:10.420> it's<00:11:10.720> called<00:11:10.960> the 00:11:11.040 --> 00:11:11.050 align:start position:0% this fork choice rule it's called the 00:11:11.050 --> 00:11:12.510 align:start position:0% this fork choice rule it's called the greedy<00:11:11.260> heaviest<00:11:11.620> observer<00:11:11.920> subtree<00:11:12.310> you 00:11:12.510 --> 00:11:12.520 align:start position:0% greedy heaviest observer subtree you 00:11:12.520 --> 00:11:14.100 align:start position:0% greedy heaviest observer subtree you start<00:11:12.730> at<00:11:12.790> the<00:11:12.850> Genesis<00:11:13.150> block<00:11:13.270> you<00:11:13.840> look<00:11:14.020> at 00:11:14.100 --> 00:11:14.110 align:start position:0% start at the Genesis block you look at 00:11:14.110 --> 00:11:15.270 align:start position:0% start at the Genesis block you look at the<00:11:14.230> children<00:11:14.560> of<00:11:14.650> the<00:11:14.710> Genesis<00:11:14.980> block<00:11:15.130> which 00:11:15.270 --> 00:11:15.280 align:start position:0% the children of the Genesis block which 00:11:15.280 --> 00:11:16.590 align:start position:0% the children of the Genesis block which are<00:11:15.400> the<00:11:15.460> blocks<00:11:15.640> that<00:11:15.760> have<00:11:16.030> a<00:11:16.210> free<00:11:16.420> block 00:11:16.590 --> 00:11:16.600 align:start position:0% are the blocks that have a free block 00:11:16.600 --> 00:11:18.480 align:start position:0% are the blocks that have a free block pointer<00:11:16.900> to<00:11:17.020> the<00:11:17.110> Genesis<00:11:17.410> block<00:11:17.560> and<00:11:17.800> you<00:11:18.280> ask 00:11:18.480 --> 00:11:18.490 align:start position:0% pointer to the Genesis block and you ask 00:11:18.490 --> 00:11:20.400 align:start position:0% pointer to the Genesis block and you ask what<00:11:19.060> are<00:11:19.090> their<00:11:19.240> scores<00:11:19.480> and<00:11:19.750> their<00:11:20.200> scores 00:11:20.400 --> 00:11:20.410 align:start position:0% what are their scores and their scores 00:11:20.410 --> 00:11:22.500 align:start position:0% what are their scores and their scores are<00:11:20.560> given<00:11:20.650> by<00:11:20.860> whether<00:11:21.610> or<00:11:21.730> not<00:11:21.910> or<00:11:22.150> how<00:11:22.330> much 00:11:22.500 --> 00:11:22.510 align:start position:0% are given by whether or not or how much 00:11:22.510 --> 00:11:23.820 align:start position:0% are given by whether or not or how much of<00:11:22.630> the<00:11:22.690> weight<00:11:22.810> of<00:11:22.930> the<00:11:22.990> validators<00:11:23.440> latest 00:11:23.820 --> 00:11:23.830 align:start position:0% of the weight of the validators latest 00:11:23.830 --> 00:11:25.320 align:start position:0% of the weight of the validators latest messages<00:11:24.190> are<00:11:24.370> on<00:11:24.550> those<00:11:24.700> block<00:11:24.880> chains<00:11:25.120> and 00:11:25.320 --> 00:11:25.330 align:start position:0% messages are on those block chains and 00:11:25.330 --> 00:11:27.780 align:start position:0% messages are on those block chains and then<00:11:26.170> you<00:11:26.350> pick<00:11:26.530> the<00:11:26.650> best<00:11:26.770> child<00:11:27.070> and<00:11:27.340> then 00:11:27.780 --> 00:11:27.790 align:start position:0% then you pick the best child and then 00:11:27.790 --> 00:11:29.580 align:start position:0% then you pick the best child and then you<00:11:27.880> kind<00:11:28.030> of<00:11:28.090> repeat<00:11:28.390> the<00:11:28.480> process<00:11:28.630> where 00:11:29.580 --> 00:11:29.590 align:start position:0% you kind of repeat the process where 00:11:29.590 --> 00:11:31.680 align:start position:0% you kind of repeat the process where again<00:11:30.340> you<00:11:30.550> calculate<00:11:30.910> the<00:11:31.030> scores<00:11:31.360> for<00:11:31.660> each 00:11:31.680 --> 00:11:31.690 align:start position:0% again you calculate the scores for each 00:11:31.690 --> 00:11:34.200 align:start position:0% again you calculate the scores for each of<00:11:31.840> the<00:11:31.900> blocks<00:11:32.580> by<00:11:33.580> looking<00:11:33.910> at<00:11:34.000> the<00:11:34.060> latest 00:11:34.200 --> 00:11:34.210 align:start position:0% of the blocks by looking at the latest 00:11:34.210 --> 00:11:36.690 align:start position:0% of the blocks by looking at the latest messages<00:11:34.630> and<00:11:34.780> seeing<00:11:35.020> if<00:11:35.260> who's<00:11:36.190> validators 00:11:36.690 --> 00:11:36.700 align:start position:0% messages and seeing if who's validators 00:11:36.700 --> 00:11:38.040 align:start position:0% messages and seeing if who's validators latest<00:11:36.940> messages<00:11:37.180> are<00:11:37.300> on<00:11:37.390> those<00:11:37.510> blocks<00:11:37.720> and 00:11:38.040 --> 00:11:38.050 align:start position:0% latest messages are on those blocks and 00:11:38.050 --> 00:11:40.560 align:start position:0% latest messages are on those blocks and then<00:11:38.770> you<00:11:39.070> pick<00:11:39.460> the<00:11:39.490> block<00:11:40.060> with<00:11:40.270> the<00:11:40.300> highest 00:11:40.560 --> 00:11:40.570 align:start position:0% then you pick the block with the highest 00:11:40.570 --> 00:11:42.990 align:start position:0% then you pick the block with the highest score<00:11:40.900> until<00:11:41.680> at<00:11:41.770> some<00:11:41.830> point<00:11:42.190> you<00:11:42.760> get<00:11:42.850> to<00:11:42.970> a 00:11:42.990 --> 00:11:43.000 align:start position:0% score until at some point you get to a 00:11:43.000 --> 00:11:44.400 align:start position:0% score until at some point you get to a block<00:11:43.210> with<00:11:43.330> no<00:11:43.360> children<00:11:43.660> and<00:11:43.960> that's<00:11:44.230> kind 00:11:44.400 --> 00:11:44.410 align:start position:0% block with no children and that's kind 00:11:44.410 --> 00:11:48.030 align:start position:0% block with no children and that's kind of<00:11:44.470> represents<00:11:44.950> the<00:11:45.100> hole<00:11:45.520> for<00:11:45.760> choice<00:11:47.040> the 00:11:48.030 --> 00:11:48.040 align:start position:0% of represents the hole for choice the 00:11:48.040 --> 00:11:49.470 align:start position:0% of represents the hole for choice the cool<00:11:48.280> thing<00:11:48.460> here<00:11:48.490> is<00:11:48.670> that<00:11:48.760> protocol<00:11:49.270> states 00:11:49.470 --> 00:11:49.480 align:start position:0% cool thing here is that protocol states 00:11:49.480 --> 00:11:50.430 align:start position:0% cool thing here is that protocol states the<00:11:49.600> state<00:11:49.750> transitions<00:11:50.110> are<00:11:50.200> defined 00:11:50.430 --> 00:11:50.440 align:start position:0% the state transitions are defined 00:11:50.440 --> 00:11:53.160 align:start position:0% the state transitions are defined exactly<00:11:50.830> as<00:11:50.950> before<00:11:51.250> there<00:11:52.150> is<00:11:52.390> literally<00:11:52.990> no 00:11:53.160 --> 00:11:53.170 align:start position:0% exactly as before there is literally no 00:11:53.170 --> 00:11:54.870 align:start position:0% exactly as before there is literally no change<00:11:53.500> right<00:11:54.040> we<00:11:54.160> just<00:11:54.280> like<00:11:54.460> the<00:11:54.730> tech 00:11:54.870 --> 00:11:54.880 align:start position:0% change right we just like the tech 00:11:54.880 --> 00:11:56.340 align:start position:0% change right we just like the tech Byzantine<00:11:55.210> Falls<00:11:55.360> is<00:11:55.510> exactly<00:11:55.900> the<00:11:56.050> same<00:11:56.200> way 00:11:56.340 --> 00:11:56.350 align:start position:0% Byzantine Falls is exactly the same way 00:11:56.350 --> 00:11:58.380 align:start position:0% Byzantine Falls is exactly the same way exclude<00:11:57.070> defined<00:11:57.700> protocol<00:11:57.970> states<00:11:58.150> and<00:11:58.300> set 00:11:58.380 --> 00:11:58.390 align:start position:0% exclude defined protocol states and set 00:11:58.390 --> 00:11:59.550 align:start position:0% exclude defined protocol states and set some<00:11:58.510> messages<00:11:58.870> it<00:11:59.050> with<00:11:59.140> less<00:11:59.320> than<00:11:59.410> some 00:11:59.550 --> 00:11:59.560 align:start position:0% some messages it with less than some 00:11:59.560 --> 00:12:01.530 align:start position:0% some messages it with less than some faults<00:11:59.860> and<00:12:00.010> state<00:12:00.760> transitions<00:12:01.150> as<00:12:01.270> super 00:12:01.530 --> 00:12:01.540 align:start position:0% faults and state transitions as super 00:12:01.540 --> 00:12:03.630 align:start position:0% faults and state transitions as super sets<00:12:01.750> and<00:12:02.050> then<00:12:02.680> we<00:12:02.740> can<00:12:02.890> define<00:12:02.980> estimate 00:12:03.630 --> 00:12:03.640 align:start position:0% sets and then we can define estimate 00:12:03.640 --> 00:12:05.790 align:start position:0% sets and then we can define estimate safety<00:12:03.940> for<00:12:04.180> the<00:12:04.450> block<00:12:04.720> block<00:12:05.230> chain<00:12:05.440> and 00:12:05.790 --> 00:12:05.800 align:start position:0% safety for the block block chain and 00:12:05.800 --> 00:12:07.260 align:start position:0% safety for the block block chain and basically<00:12:06.220> is<00:12:06.340> very<00:12:06.370> similar<00:12:06.790> and<00:12:07.120> ease<00:12:07.240> that 00:12:07.260 --> 00:12:07.270 align:start position:0% basically is very similar and ease that 00:12:07.270 --> 00:12:10.620 align:start position:0% basically is very similar and ease that for<00:12:07.540> all<00:12:07.800> future<00:12:08.800> protocol<00:12:09.280> states<00:12:09.630> this 00:12:10.620 --> 00:12:10.630 align:start position:0% for all future protocol states this 00:12:10.630 --> 00:12:13.680 align:start position:0% for all future protocol states this block<00:12:10.930> is<00:12:11.170> in<00:12:11.350> the<00:12:11.560> blockchain<00:12:12.600> that<00:12:13.600> we're 00:12:13.680 --> 00:12:13.690 align:start position:0% block is in the blockchain that we're 00:12:13.690 --> 00:12:14.790 align:start position:0% block is in the blockchain that we're going<00:12:13.780> to<00:12:13.840> choose<00:12:14.080> of<00:12:14.260> those<00:12:14.350> future<00:12:14.590> states 00:12:14.790 --> 00:12:14.800 align:start position:0% going to choose of those future states 00:12:14.800 --> 00:12:17.430 align:start position:0% going to choose of those future states so<00:12:15.160> a<00:12:15.550> block<00:12:16.090> is<00:12:16.240> safe<00:12:16.420> if<00:12:16.630> at<00:12:16.780> all<00:12:16.930> future<00:12:17.260> for 00:12:17.430 --> 00:12:17.440 align:start position:0% so a block is safe if at all future for 00:12:17.440 --> 00:12:19.410 align:start position:0% so a block is safe if at all future for choice<00:12:17.590> rules<00:12:17.860> that<00:12:18.010> you<00:12:18.070> make<00:12:18.280> that<00:12:19.120> block 00:12:19.410 --> 00:12:19.420 align:start position:0% choice rules that you make that block 00:12:19.420 --> 00:12:23.640 align:start position:0% choice rules that you make that block will<00:12:19.630> be<00:12:19.750> in<00:12:19.870> there<00:12:21.690> so<00:12:22.690> that<00:12:23.110> and<00:12:23.380> that<00:12:23.530> and 00:12:23.640 --> 00:12:23.650 align:start position:0% will be in there so that and that and 00:12:23.650 --> 00:12:24.960 align:start position:0% will be in there so that and that and that<00:12:23.710> was<00:12:23.890> actually<00:12:24.010> it<00:12:24.280> right<00:12:24.580> because<00:12:24.760> all 00:12:24.960 --> 00:12:24.970 align:start position:0% that was actually it right because all 00:12:24.970 --> 00:12:26.700 align:start position:0% that was actually it right because all the<00:12:25.150> other<00:12:25.240> stuff<00:12:25.510> and<00:12:25.840> then<00:12:26.140> the<00:12:26.290> the 00:12:26.700 --> 00:12:26.710 align:start position:0% the other stuff and then the the 00:12:26.710 --> 00:12:27.780 align:start position:0% the other stuff and then the the specification<00:12:27.220> is<00:12:27.310> really<00:12:27.520> kind<00:12:27.700> of<00:12:27.730> like 00:12:27.780 --> 00:12:27.790 align:start position:0% specification is really kind of like 00:12:27.790 --> 00:12:31.050 align:start position:0% specification is really kind of like done<00:12:28.150> there<00:12:28.390> because<00:12:29.310> and<00:12:30.310> we<00:12:30.430> got<00:12:30.550> to<00:12:30.640> benefit 00:12:31.050 --> 00:12:31.060 align:start position:0% done there because and we got to benefit 00:12:31.060 --> 00:12:32.910 align:start position:0% done there because and we got to benefit from<00:12:31.270> all<00:12:31.630> of<00:12:31.900> the<00:12:32.050> work<00:12:32.200> we<00:12:32.320> did<00:12:32.470> earlier<00:12:32.740> from 00:12:32.910 --> 00:12:32.920 align:start position:0% from all of the work we did earlier from 00:12:32.920 --> 00:12:35.310 align:start position:0% from all of the work we did earlier from the<00:12:33.100> safety<00:12:33.400> proof<00:12:33.610> and<00:12:33.820> the<00:12:33.880> binary<00:12:34.320> pretty 00:12:35.310 --> 00:12:35.320 align:start position:0% the safety proof and the binary pretty 00:12:35.320 --> 00:12:37.950 align:start position:0% the safety proof and the binary pretty much<00:12:35.440> reuse<00:12:36.010> most<00:12:36.880> of<00:12:37.090> those<00:12:37.210> most<00:12:37.690> of<00:12:37.840> those 00:12:37.950 --> 00:12:37.960 align:start position:0% much reuse most of those most of those 00:12:37.960 --> 00:12:40.110 align:start position:0% much reuse most of those most of those most<00:12:38.620> of<00:12:38.770> that<00:12:38.890> work<00:12:39.010> and<00:12:39.250> only<00:12:39.910> had<00:12:40.030> to 00:12:40.110 --> 00:12:40.120 align:start position:0% most of that work and only had to 00:12:40.120 --> 00:12:43.710 align:start position:0% most of that work and only had to redefine<00:12:41.400> you<00:12:42.400> only<00:12:42.700> had<00:12:42.850> to<00:12:42.940> define<00:12:43.210> the 00:12:43.710 --> 00:12:43.720 align:start position:0% redefine you only had to define the 00:12:43.720 --> 00:12:48.060 align:start position:0% redefine you only had to define the message<00:12:44.350> type<00:12:45.030> the<00:12:46.030> score<00:12:46.450> of<00:12:46.630> blocks<00:12:47.070> the 00:12:48.060 --> 00:12:48.070 align:start position:0% message type the score of blocks the 00:12:48.070 --> 00:12:49.900 align:start position:0% message type the score of blocks the fork<00:12:48.460> choice<00:12:48.700> rule 00:12:49.900 --> 00:12:49.910 align:start position:0% fork choice rule 00:12:49.910 --> 00:12:53.410 align:start position:0% fork choice rule and<00:12:50.440> safety<00:12:51.640> so<00:12:52.640> that's<00:12:52.850> pretty<00:12:53.000> cool 00:12:53.410 --> 00:12:53.420 align:start position:0% and safety so that's pretty cool 00:12:53.420 --> 00:12:54.670 align:start position:0% and safety so that's pretty cool we<00:12:53.660> were<00:12:53.810> able<00:12:53.900> to<00:12:54.110> go<00:12:54.320> from<00:12:54.440> the<00:12:54.530> binary 00:12:54.670 --> 00:12:54.680 align:start position:0% we were able to go from the binary 00:12:54.680 --> 00:12:55.900 align:start position:0% we were able to go from the binary consensus<00:12:55.130> to<00:12:55.190> the<00:12:55.280> blockchain<00:12:55.460> consensus 00:12:55.900 --> 00:12:55.910 align:start position:0% consensus to the blockchain consensus 00:12:55.910 --> 00:12:59.620 align:start position:0% consensus to the blockchain consensus with<00:12:56.030> a<00:12:56.060> minimum<00:12:56.450> change<00:12:58.270> safety<00:12:59.270> Oracle's 00:12:59.620 --> 00:12:59.630 align:start position:0% with a minimum change safety Oracle's 00:12:59.630 --> 00:13:00.670 align:start position:0% with a minimum change safety Oracle's this<00:12:59.750> is<00:12:59.840> kind<00:12:59.990> of<00:13:00.020> important<00:13:00.380> it's<00:13:00.530> kind<00:13:00.650> of 00:13:00.670 --> 00:13:00.680 align:start position:0% this is kind of important it's kind of 00:13:00.680 --> 00:13:02.740 align:start position:0% this is kind of important it's kind of important<00:13:01.010> that<00:13:01.040> we<00:13:01.340> are<00:13:01.700> able<00:13:02.270> to<00:13:02.390> detect 00:13:02.740 --> 00:13:02.750 align:start position:0% important that we are able to detect 00:13:02.750 --> 00:13:04.800 align:start position:0% important that we are able to detect safety<00:13:03.170> I'm<00:13:03.770> not<00:13:03.890> sure<00:13:03.980> happened<00:13:04.250> this<00:13:04.340> slide 00:13:04.800 --> 00:13:04.810 align:start position:0% safety I'm not sure happened this slide 00:13:04.810 --> 00:13:10.690 align:start position:0% safety I'm not sure happened this slide so<00:13:06.970> and<00:13:08.410> no<00:13:09.410> like<00:13:09.770> what<00:13:10.100> I've<00:13:10.190> shown<00:13:10.400> so<00:13:10.520> far<00:13:10.550> is 00:13:10.690 --> 00:13:10.700 align:start position:0% so and no like what I've shown so far is 00:13:10.700 --> 00:13:12.970 align:start position:0% so and no like what I've shown so far is if<00:13:10.970> you<00:13:11.120> were<00:13:11.420> to<00:13:11.720> make<00:13:11.840> estimate<00:13:12.440> decisions 00:13:12.970 --> 00:13:12.980 align:start position:0% if you were to make estimate decisions 00:13:12.980 --> 00:13:14.380 align:start position:0% if you were to make estimate decisions on<00:13:13.040> safe<00:13:13.220> estimates<00:13:13.610> then<00:13:14.060> it<00:13:14.180> would<00:13:14.270> be 00:13:14.380 --> 00:13:14.390 align:start position:0% on safe estimates then it would be 00:13:14.390 --> 00:13:17.050 align:start position:0% on safe estimates then it would be consensus<00:13:14.780> safe<00:13:14.930> but<00:13:15.920> you<00:13:16.580> might<00:13:16.760> you<00:13:16.880> need<00:13:17.000> a 00:13:17.050 --> 00:13:17.060 align:start position:0% consensus safe but you might you need a 00:13:17.060 --> 00:13:18.330 align:start position:0% consensus safe but you might you need a way<00:13:17.210> we<00:13:17.420> might<00:13:17.570> actually<00:13:17.780> need<00:13:17.960> a<00:13:18.050> way<00:13:18.200> to 00:13:18.330 --> 00:13:18.340 align:start position:0% way we might actually need a way to 00:13:18.340 --> 00:13:20.800 align:start position:0% way we might actually need a way to figure<00:13:19.340> out<00:13:19.460> if<00:13:19.580> that's<00:13:19.790> the<00:13:19.880> case<00:13:20.030> so<00:13:20.630> we<00:13:20.690> have 00:13:20.800 --> 00:13:20.810 align:start position:0% figure out if that's the case so we have 00:13:20.810 --> 00:13:23.050 align:start position:0% figure out if that's the case so we have a<00:13:20.840> couple<00:13:20.990> of<00:13:21.080> definitions<00:13:21.560> here<00:13:21.790> and<00:13:22.790> this 00:13:23.050 --> 00:13:23.060 align:start position:0% a couple of definitions here and this 00:13:23.060 --> 00:13:24.790 align:start position:0% a couple of definitions here and this I'm<00:13:23.300> gonna<00:13:23.420> give<00:13:23.870> like<00:13:23.960> a<00:13:24.080> simple<00:13:24.440> way<00:13:24.530> to<00:13:24.590> do 00:13:24.790 --> 00:13:24.800 align:start position:0% I'm gonna give like a simple way to do 00:13:24.800 --> 00:13:27.340 align:start position:0% I'm gonna give like a simple way to do this<00:13:24.920> so<00:13:25.100> a<00:13:25.610> validator<00:13:26.390> sees<00:13:26.900> another 00:13:27.340 --> 00:13:27.350 align:start position:0% this so a validator sees another 00:13:27.350 --> 00:13:31.510 align:start position:0% this so a validator sees another validator<00:13:27.830> agreeing<00:13:28.070> on<00:13:28.220> an<00:13:28.370> estimate<00:13:28.790> if<00:13:30.520> the 00:13:31.510 --> 00:13:31.520 align:start position:0% validator agreeing on an estimate if the 00:13:31.520 --> 00:13:33.370 align:start position:0% validator agreeing on an estimate if the latest<00:13:31.910> in<00:13:32.330> the<00:13:32.690> latest<00:13:32.960> message<00:13:33.110> of 00:13:33.370 --> 00:13:33.380 align:start position:0% latest in the latest message of 00:13:33.380 --> 00:13:35.830 align:start position:0% latest in the latest message of validator<00:13:33.860> I<00:13:34.190> in<00:13:34.820> their<00:13:35.120> justification<00:13:35.750> the 00:13:35.830 --> 00:13:35.840 align:start position:0% validator I in their justification the 00:13:35.840 --> 00:13:37.480 align:start position:0% validator I in their justification the latest<00:13:36.110> message<00:13:36.230> Missy<00:13:36.560> from<00:13:36.650> validator<00:13:37.040> J 00:13:37.480 --> 00:13:37.490 align:start position:0% latest message Missy from validator J 00:13:37.490 --> 00:13:40.540 align:start position:0% latest message Missy from validator J has<00:13:38.150> an<00:13:38.510> estimate<00:13:38.900> that<00:13:38.930> agrees<00:13:39.440> with<00:13:40.400> the 00:13:40.540 --> 00:13:40.550 align:start position:0% has an estimate that agrees with the 00:13:40.550 --> 00:13:42.090 align:start position:0% has an estimate that agrees with the estimate<00:13:40.820> that<00:13:40.940> we're<00:13:41.060> trying<00:13:41.270> to<00:13:41.390> say<00:13:41.660> is 00:13:42.090 --> 00:13:42.100 align:start position:0% estimate that we're trying to say is 00:13:42.100 --> 00:13:44.980 align:start position:0% estimate that we're trying to say is safe<00:13:43.240> an<00:13:44.240> agreement<00:13:44.540> in<00:13:44.780> the<00:13:44.810> binary 00:13:44.980 --> 00:13:44.990 align:start position:0% safe an agreement in the binary 00:13:44.990 --> 00:13:47.230 align:start position:0% safe an agreement in the binary consensus<00:13:45.530> means<00:13:45.680> that<00:13:45.950> they<00:13:46.130> are<00:13:46.430> the<00:13:47.090> same 00:13:47.230 --> 00:13:47.240 align:start position:0% consensus means that they are the same 00:13:47.240 --> 00:13:48.730 align:start position:0% consensus means that they are the same bit<00:13:47.450> in<00:13:47.660> the<00:13:47.870> blockchain<00:13:48.050> consensus<00:13:48.590> means 00:13:48.730 --> 00:13:48.740 align:start position:0% bit in the blockchain consensus means 00:13:48.740 --> 00:13:51.250 align:start position:0% bit in the blockchain consensus means that<00:13:48.920> they<00:13:49.070> are<00:13:49.310> in<00:13:49.730> the<00:13:50.600> Block<00:13:50.900> in<00:13:51.110> the 00:13:51.250 --> 00:13:51.260 align:start position:0% that they are in the Block in the 00:13:51.260 --> 00:13:54.580 align:start position:0% that they are in the Block in the blockchain<00:13:51.470> of<00:13:51.890> the<00:13:52.520> other<00:13:52.640> block<00:13:52.900> so<00:13:53.900> so<00:13:54.290> VI 00:13:54.580 --> 00:13:54.590 align:start position:0% blockchain of the other block so so VI 00:13:54.590 --> 00:13:56.950 align:start position:0% blockchain of the other block so so VI sees<00:13:55.040> BJ<00:13:55.400> means<00:13:55.640> that<00:13:55.790> the<00:13:56.120> validator<00:13:56.510> I<00:13:56.600> see 00:13:56.950 --> 00:13:56.960 align:start position:0% sees BJ means that the validator I see 00:13:56.960 --> 00:13:58.420 align:start position:0% sees BJ means that the validator I see the<00:13:57.020> latest<00:13:57.320> message<00:13:57.560> from<00:13:57.650> validator<00:13:58.040> J<00:13:58.250> that 00:13:58.420 --> 00:13:58.430 align:start position:0% the latest message from validator J that 00:13:58.430 --> 00:14:00.550 align:start position:0% the latest message from validator J that agrees<00:13:58.810> and<00:13:59.810> there's<00:14:00.080> another<00:14:00.200> definition 00:14:00.550 --> 00:14:00.560 align:start position:0% agrees and there's another definition 00:14:00.560 --> 00:14:01.870 align:start position:0% agrees and there's another definition you<00:14:00.620> know<00:14:00.680> that's<00:14:00.830> cropped<00:14:01.220> out<00:14:01.370> and<00:14:01.670> it 00:14:01.870 --> 00:14:01.880 align:start position:0% you know that's cropped out and it 00:14:01.880 --> 00:14:04.000 align:start position:0% you know that's cropped out and it basically<00:14:01.940> says<00:14:02.210> the<00:14:02.300> VI<00:14:02.450> can't<00:14:03.050> see<00:14:03.470> VJ 00:14:04.000 --> 00:14:04.010 align:start position:0% basically says the VI can't see VJ 00:14:04.010 --> 00:14:06.580 align:start position:0% basically says the VI can't see VJ disagree<00:14:04.930> which<00:14:05.930> means<00:14:06.110> that<00:14:06.230> they<00:14:06.320> there's 00:14:06.580 --> 00:14:06.590 align:start position:0% disagree which means that they there's 00:14:06.590 --> 00:14:08.350 align:start position:0% disagree which means that they there's no<00:14:06.710> new<00:14:07.190> latest<00:14:07.370> message<00:14:07.670> from<00:14:07.790> GJ<00:14:08.120> that<00:14:08.270> they 00:14:08.350 --> 00:14:08.360 align:start position:0% no new latest message from GJ that they 00:14:08.360 --> 00:14:09.490 align:start position:0% no new latest message from GJ that they could<00:14:08.510> see<00:14:08.630> that<00:14:08.780> doesn't<00:14:09.140> agree<00:14:09.260> with<00:14:09.350> the 00:14:09.490 --> 00:14:09.500 align:start position:0% could see that doesn't agree with the 00:14:09.500 --> 00:14:12.580 align:start position:0% could see that doesn't agree with the estimate<00:14:09.770> and<00:14:10.040> so<00:14:10.550> basically<00:14:10.850> if<00:14:11.500> there's<00:14:12.500> a 00:14:12.580 --> 00:14:12.590 align:start position:0% estimate and so basically if there's a 00:14:12.590 --> 00:14:15.040 align:start position:0% estimate and so basically if there's a set<00:14:12.770> of<00:14:12.860> validators<00:14:13.310> who<00:14:13.670> can't<00:14:14.090> who<00:14:14.780> see<00:14:14.990> each 00:14:15.040 --> 00:14:15.050 align:start position:0% set of validators who can't who see each 00:14:15.050 --> 00:14:16.690 align:start position:0% set of validators who can't who see each other<00:14:15.230> agree<00:14:15.440> and<00:14:15.710> can't<00:14:16.100> see<00:14:16.400> each<00:14:16.520> other 00:14:16.690 --> 00:14:16.700 align:start position:0% other agree and can't see each other 00:14:16.700 --> 00:14:22.240 align:start position:0% other agree and can't see each other disagree<00:14:17.740> it<00:14:18.740> turns<00:14:18.980> out<00:14:19.190> that<00:14:20.380> if<00:14:21.380> they<00:14:22.040> have 00:14:22.240 --> 00:14:22.250 align:start position:0% disagree it turns out that if they have 00:14:22.250 --> 00:14:24.370 align:start position:0% disagree it turns out that if they have some<00:14:23.030> number<00:14:23.330> of<00:14:23.360> wait<00:14:23.630> some<00:14:24.140> amount<00:14:24.320> of 00:14:24.370 --> 00:14:24.380 align:start position:0% some number of wait some amount of 00:14:24.380 --> 00:14:29.740 align:start position:0% some number of wait some amount of weight<00:14:24.730> and<00:14:25.730> that<00:14:25.910> weight<00:14:25.970> is<00:14:27.250> more<00:14:28.250> than<00:14:28.750> more 00:14:29.740 --> 00:14:29.750 align:start position:0% weight and that weight is more than more 00:14:29.750 --> 00:14:31.780 align:start position:0% weight and that weight is more than more than<00:14:29.900> half<00:14:30.050> the<00:14:30.230> weight<00:14:30.380> then<00:14:30.800> you<00:14:31.130> then<00:14:31.550> you 00:14:31.780 --> 00:14:31.790 align:start position:0% than half the weight then you then you 00:14:31.790 --> 00:14:33.370 align:start position:0% than half the weight then you then you can<00:14:31.910> actually<00:14:32.060> show<00:14:32.660> that<00:14:32.930> there<00:14:33.080> are<00:14:33.170> some 00:14:33.370 --> 00:14:33.380 align:start position:0% can actually show that there are some 00:14:33.380 --> 00:14:35.320 align:start position:0% can actually show that there are some amount<00:14:33.620> of<00:14:33.650> fault<00:14:33.830> tolerance<00:14:34.220> but<00:14:34.700> our<00:14:35.120> safety 00:14:35.320 --> 00:14:35.330 align:start position:0% amount of fault tolerance but our safety 00:14:35.330 --> 00:14:37.210 align:start position:0% amount of fault tolerance but our safety definition<00:14:35.750> requires<00:14:36.050> that<00:14:36.290> we<00:14:36.410> have<00:14:36.560> t<00:14:36.860> fault 00:14:37.210 --> 00:14:37.220 align:start position:0% definition requires that we have t fault 00:14:37.220 --> 00:14:38.710 align:start position:0% definition requires that we have t fault although<00:14:37.580> we<00:14:37.700> can<00:14:37.790> tolerate<00:14:37.910> up<00:14:38.150> to<00:14:38.240> T<00:14:38.360> faults 00:14:38.710 --> 00:14:38.720 align:start position:0% although we can tolerate up to T faults 00:14:38.720 --> 00:14:43.750 align:start position:0% although we can tolerate up to T faults however<00:14:42.100> there<00:14:43.100> might<00:14:43.220> already<00:14:43.370> be<00:14:43.490> some 00:14:43.750 --> 00:14:43.760 align:start position:0% however there might already be some 00:14:43.760 --> 00:14:45.400 align:start position:0% however there might already be some initial<00:14:43.970> faults<00:14:44.390> in<00:14:44.480> our<00:14:44.570> view<00:14:44.810> so<00:14:45.110> we<00:14:45.200> kind<00:14:45.350> of 00:14:45.400 --> 00:14:45.410 align:start position:0% initial faults in our view so we kind of 00:14:45.410 --> 00:14:47.580 align:start position:0% initial faults in our view so we kind of subtract<00:14:45.770> that<00:14:45.920> so<00:14:46.160> basically<00:14:46.250> as<00:14:46.700> long<00:14:47.210> as 00:14:47.580 --> 00:14:47.590 align:start position:0% subtract that so basically as long as 00:14:47.590 --> 00:14:51.970 align:start position:0% subtract that so basically as long as the<00:14:48.970> the<00:14:49.970> the<00:14:50.810> this<00:14:51.320> is<00:14:51.440> this<00:14:51.500> is<00:14:51.620> kind<00:14:51.800> of<00:14:51.830> the 00:14:51.970 --> 00:14:51.980 align:start position:0% the the the this is this is kind of the 00:14:51.980 --> 00:14:55.540 align:start position:0% the the the this is this is kind of the overlap<00:14:52.520> between<00:14:52.550> two<00:14:54.070> sets<00:14:55.070> of<00:14:55.190> Valladares 00:14:55.540 --> 00:14:55.550 align:start position:0% overlap between two sets of Valladares 00:14:55.550 --> 00:14:58.060 align:start position:0% overlap between two sets of Valladares that<00:14:55.760> would<00:14:56.390> like<00:14:57.050> the<00:14:57.290> overlap<00:14:57.680> and<00:14:57.830> weight 00:14:58.060 --> 00:14:58.070 align:start position:0% that would like the overlap and weight 00:14:58.070 --> 00:15:01.060 align:start position:0% that would like the overlap and weight of<00:14:58.280> the<00:14:58.940> validators<00:14:59.630> in<00:14:59.720> the<00:14:59.780> klique<00:15:00.070> if 00:15:01.060 --> 00:15:01.070 align:start position:0% of the validators in the klique if 00:15:01.070 --> 00:15:03.430 align:start position:0% of the validators in the klique if that's<00:15:01.339> more<00:15:01.700> than<00:15:01.940> the<00:15:02.089> fault<00:15:02.330> tolerance 00:15:03.430 --> 00:15:03.440 align:start position:0% that's more than the fault tolerance 00:15:03.440 --> 00:15:05.800 align:start position:0% that's more than the fault tolerance then<00:15:04.100> we're<00:15:04.399> then<00:15:04.639> we<00:15:04.730> have<00:15:04.850> safety<00:15:05.209> it's<00:15:05.600> very 00:15:05.800 --> 00:15:05.810 align:start position:0% then we're then we have safety it's very 00:15:05.810 --> 00:15:06.790 align:start position:0% then we're then we have safety it's very similar<00:15:05.990> to<00:15:06.199> a<00:15:06.259> traditional<00:15:06.290> kind<00:15:06.680> of 00:15:06.790 --> 00:15:06.800 align:start position:0% similar to a traditional kind of 00:15:06.800 --> 00:15:09.160 align:start position:0% similar to a traditional kind of consensus<00:15:07.129> safety<00:15:07.339> proof<00:15:07.550> only<00:15:07.910> it's<00:15:08.660> working 00:15:09.160 --> 00:15:09.170 align:start position:0% consensus safety proof only it's working 00:15:09.170 --> 00:15:10.420 align:start position:0% consensus safety proof only it's working on<00:15:09.259> these<00:15:09.350> estimates<00:15:09.709> instead<00:15:10.040> and<00:15:10.279> the 00:15:10.420 --> 00:15:10.430 align:start position:0% on these estimates instead and the 00:15:10.430 --> 00:15:13.090 align:start position:0% on these estimates instead and the intuition<00:15:10.639> is<00:15:10.759> quite<00:15:10.970> clear<00:15:11.240> basically<00:15:11.839> if<00:15:12.529> in 00:15:13.090 --> 00:15:13.100 align:start position:0% intuition is quite clear basically if in 00:15:13.100 --> 00:15:14.199 align:start position:0% intuition is quite clear basically if in the<00:15:13.220> clique<00:15:13.399> they<00:15:13.550> can't<00:15:13.730> can't<00:15:14.029> convince 00:15:14.199 --> 00:15:14.209 align:start position:0% the clique they can't can't convince 00:15:14.209 --> 00:15:15.129 align:start position:0% the clique they can't can't convince each<00:15:14.360> other<00:15:14.449> that<00:15:14.569> they've<00:15:14.810> seen<00:15:14.990> something 00:15:15.129 --> 00:15:15.139 align:start position:0% each other that they've seen something 00:15:15.139 --> 00:15:17.079 align:start position:0% each other that they've seen something else<00:15:15.290> and<00:15:15.560> if<00:15:16.279> they<00:15:16.430> have<00:15:16.550> more<00:15:16.670> than<00:15:16.730> half<00:15:16.970> the 00:15:17.079 --> 00:15:17.089 align:start position:0% else and if they have more than half the 00:15:17.089 --> 00:15:18.790 align:start position:0% else and if they have more than half the weight<00:15:17.209> then<00:15:17.690> whoever<00:15:18.139> can<00:15:18.350> convince<00:15:18.680> them 00:15:18.790 --> 00:15:18.800 align:start position:0% weight then whoever can convince them 00:15:18.800 --> 00:15:20.019 align:start position:0% weight then whoever can convince them from<00:15:18.920> outside<00:15:19.250> the<00:15:19.370> clique<00:15:19.550> won't<00:15:19.879> have 00:15:20.019 --> 00:15:20.029 align:start position:0% from outside the clique won't have 00:15:20.029 --> 00:15:21.329 align:start position:0% from outside the clique won't have enough<00:15:20.209> weight<00:15:20.360> to<00:15:20.480> actually<00:15:20.720> contribute 00:15:21.329 --> 00:15:21.339 align:start position:0% enough weight to actually contribute 00:15:21.339 --> 00:15:23.829 align:start position:0% enough weight to actually contribute great<00:15:22.339> so<00:15:22.850> that<00:15:23.000> actually<00:15:23.300> was<00:15:23.420> a<00:15:23.480> hard<00:15:23.689> part 00:15:23.829 --> 00:15:23.839 align:start position:0% great so that actually was a hard part 00:15:23.839 --> 00:15:26.110 align:start position:0% great so that actually was a hard part also<00:15:23.990> but<00:15:24.519> we're<00:15:25.519> getting<00:15:25.730> through<00:15:25.879> it<00:15:25.970> we're 00:15:26.110 --> 00:15:26.120 align:start position:0% also but we're getting through it we're 00:15:26.120 --> 00:15:28.300 align:start position:0% also but we're getting through it we're almost<00:15:26.209> done<00:15:26.449> so<00:15:26.920> subjective<00:15:27.920> fault 00:15:28.300 --> 00:15:28.310 align:start position:0% almost done so subjective fault 00:15:28.310 --> 00:15:29.499 align:start position:0% almost done so subjective fault tolerance<00:15:28.430> thresholds<00:15:28.910> is<00:15:29.029> a<00:15:29.120> pretty<00:15:29.360> cool 00:15:29.499 --> 00:15:29.509 align:start position:0% tolerance thresholds is a pretty cool 00:15:29.509 --> 00:15:31.420 align:start position:0% tolerance thresholds is a pretty cool thing<00:15:29.540> basically<00:15:30.500> if<00:15:30.740> you<00:15:31.040> have<00:15:31.220> two 00:15:31.420 --> 00:15:31.430 align:start position:0% thing basically if you have two 00:15:31.430 --> 00:15:33.579 align:start position:0% thing basically if you have two protocols<00:15:31.759> which<00:15:32.240> have<00:15:32.629> the<00:15:32.750> same<00:15:32.779> words<00:15:33.470> are 00:15:33.579 --> 00:15:33.589 align:start position:0% protocols which have the same words are 00:15:33.589 --> 00:15:34.780 align:start position:0% protocols which have the same words are the<00:15:33.709> same<00:15:33.740> except<00:15:34.189> for<00:15:34.220> the<00:15:34.370> fact<00:15:34.550> that<00:15:34.639> they 00:15:34.780 --> 00:15:34.790 align:start position:0% the same except for the fact that they 00:15:34.790 --> 00:15:35.679 align:start position:0% the same except for the fact that they operate<00:15:34.910> two<00:15:35.209> different<00:15:35.420> fault<00:15:35.569> tolerance 00:15:35.679 --> 00:15:35.689 align:start position:0% operate two different fault tolerance 00:15:35.689 --> 00:15:37.059 align:start position:0% operate two different fault tolerance threshold<00:15:35.930> and<00:15:36.470> one<00:15:36.620> of<00:15:36.680> them<00:15:36.769> is<00:15:36.860> small<00:15:37.009> on 00:15:37.059 --> 00:15:37.069 align:start position:0% threshold and one of them is small on 00:15:37.069 --> 00:15:40.269 align:start position:0% threshold and one of them is small on the<00:15:37.160> other<00:15:37.279> then<00:15:37.930> consensus<00:15:38.930> then<00:15:39.470> no<00:15:39.980> to<00:15:40.040> make 00:15:40.269 --> 00:15:40.279 align:start position:0% the other then consensus then no to make 00:15:40.279 --> 00:15:41.290 align:start position:0% the other then consensus then no to make decisions<00:15:40.610> in<00:15:40.670> these<00:15:40.790> protocols<00:15:40.970> will 00:15:41.290 --> 00:15:41.300 align:start position:0% decisions in these protocols will 00:15:41.300 --> 00:15:43.629 align:start position:0% decisions in these protocols will actually<00:15:41.629> have<00:15:41.720> consensus<00:15:42.110> safety<00:15:42.379> as<00:15:43.279> long 00:15:43.629 --> 00:15:43.639 align:start position:0% actually have consensus safety as long 00:15:43.639 --> 00:15:44.920 align:start position:0% actually have consensus safety as long as<00:15:43.759> the<00:15:43.850> number<00:15:44.029> of<00:15:44.149> Falls<00:15:44.329> exhibited<00:15:44.810> in 00:15:44.920 --> 00:15:44.930 align:start position:0% as the number of Falls exhibited in 00:15:44.930 --> 00:15:46.449 align:start position:0% as the number of Falls exhibited in their<00:15:45.050> union<00:15:45.259> of<00:15:45.500> their<00:15:45.589> views<00:15:45.829> is<00:15:46.040> less<00:15:46.279> than 00:15:46.449 --> 00:15:46.459 align:start position:0% their union of their views is less than 00:15:46.459 --> 00:15:47.920 align:start position:0% their union of their views is less than the<00:15:46.550> smallest<00:15:46.910> of<00:15:47.180> their<00:15:47.329> fault<00:15:47.540> tolerance 00:15:47.920 --> 00:15:47.930 align:start position:0% the smallest of their fault tolerance 00:15:47.930 --> 00:15:49.689 align:start position:0% the smallest of their fault tolerance thresholds<00:15:48.319> and<00:15:48.560> it's<00:15:49.129> really<00:15:49.310> cool<00:15:49.519> because 00:15:49.689 --> 00:15:49.699 align:start position:0% thresholds and it's really cool because 00:15:49.699 --> 00:15:51.429 align:start position:0% thresholds and it's really cool because having<00:15:50.480> fault<00:15:50.750> tolerance<00:15:50.899> partials<00:15:51.290> in<00:15:51.379> the 00:15:51.429 --> 00:15:51.439 align:start position:0% having fault tolerance partials in the 00:15:51.439 --> 00:15:54.040 align:start position:0% having fault tolerance partials in the protocol<00:15:51.879> makes<00:15:52.879> it<00:15:53.029> easier<00:15:53.329> to<00:15:53.360> attack<00:15:53.750> and 00:15:54.040 --> 00:15:54.050 align:start position:0% protocol makes it easier to attack and 00:15:54.050 --> 00:15:55.360 align:start position:0% protocol makes it easier to attack and pose<00:15:54.230> a<00:15:54.259> bunch<00:15:54.529> of<00:15:54.560> economic<00:15:55.009> problems<00:15:55.310> that 00:15:55.360 --> 00:15:55.370 align:start position:0% pose a bunch of economic problems that 00:15:55.370 --> 00:15:58.059 align:start position:0% pose a bunch of economic problems that are<00:15:55.730> out<00:15:56.180> of<00:15:56.509> scope<00:15:56.689> for<00:15:56.750> this<00:15:56.899> talk<00:15:57.170> but<00:15:57.470> stuff 00:15:58.059 --> 00:15:58.069 align:start position:0% are out of scope for this talk but stuff 00:15:58.069 --> 00:15:59.879 align:start position:0% are out of scope for this talk but stuff about<00:15:58.220> cartelization<00:15:58.730> and<00:15:59.089> focal<00:15:59.329> points 00:15:59.879 --> 00:15:59.889 align:start position:0% about cartelization and focal points 00:15:59.889 --> 00:16:01.900 align:start position:0% about cartelization and focal points validate<00:16:00.889> a<00:16:00.920> rotation<00:16:01.310> this<00:16:01.430> is<00:16:01.579> an<00:16:01.670> exciting 00:16:01.900 --> 00:16:01.910 align:start position:0% validate a rotation this is an exciting 00:16:01.910 --> 00:16:02.199 align:start position:0% validate a rotation this is an exciting thing 00:16:02.199 --> 00:16:02.209 align:start position:0% thing 00:16:02.209 --> 00:16:04.960 align:start position:0% thing so<00:16:03.170> I<00:16:03.199> mean<00:16:03.350> I've<00:16:03.470> changed<00:16:04.279> the<00:16:04.430> structure<00:16:04.759> of 00:16:04.960 --> 00:16:04.970 align:start position:0% so I mean I've changed the structure of 00:16:04.970 --> 00:16:06.610 align:start position:0% so I mean I've changed the structure of the<00:16:05.029> blockchain<00:16:05.269> a<00:16:05.750> little<00:16:05.990> bit<00:16:06.079> to<00:16:06.500> add 00:16:06.610 --> 00:16:06.620 align:start position:0% the blockchain a little bit to add 00:16:06.620 --> 00:16:09.069 align:start position:0% the blockchain a little bit to add weights<00:16:07.220> to<00:16:07.399> the<00:16:07.730> to<00:16:07.970> the<00:16:08.089> block<00:16:08.329> right<00:16:08.750> before 00:16:09.069 --> 00:16:09.079 align:start position:0% weights to the to the block right before 00:16:09.079 --> 00:16:10.569 align:start position:0% weights to the to the block right before we<00:16:09.170> had<00:16:09.259> validated<00:16:09.800> weights<00:16:10.040> kind<00:16:10.160> of<00:16:10.279> this 00:16:10.569 --> 00:16:10.579 align:start position:0% we had validated weights kind of this 00:16:10.579 --> 00:16:12.249 align:start position:0% we had validated weights kind of this global<00:16:10.939> thing<00:16:11.120> now<00:16:11.660> we're<00:16:11.750> gonna<00:16:11.839> add<00:16:12.019> weights 00:16:12.249 --> 00:16:12.259 align:start position:0% global thing now we're gonna add weights 00:16:12.259 --> 00:16:14.800 align:start position:0% global thing now we're gonna add weights to<00:16:12.319> the<00:16:12.439> justification<00:16:13.040> though<00:16:13.880> so<00:16:14.060> the<00:16:14.509> you 00:16:14.800 --> 00:16:14.810 align:start position:0% to the justification though so the you 00:16:14.810 --> 00:16:15.819 align:start position:0% to the justification though so the you know<00:16:14.899> blocks<00:16:15.170> give<00:16:15.350> me<00:16:15.410> the<00:16:15.500> same<00:16:15.620> as<00:16:15.709> before 00:16:15.819 --> 00:16:15.829 align:start position:0% know blocks give me the same as before 00:16:15.829 --> 00:16:17.829 align:start position:0% know blocks give me the same as before like<00:16:16.250> you<00:16:16.430> know<00:16:16.550> they<00:16:16.639> have<00:16:16.910> a<00:16:16.939> block<00:16:17.209> ascender 00:16:17.829 --> 00:16:17.839 align:start position:0% like you know they have a block ascender 00:16:17.839 --> 00:16:20.319 align:start position:0% like you know they have a block ascender justification<00:16:18.740> and<00:16:18.920> the<00:16:19.610> justification<00:16:19.939> now 00:16:20.319 --> 00:16:20.329 align:start position:0% justification and the justification now 00:16:20.329 --> 00:16:21.579 align:start position:0% justification and the justification now has<00:16:20.509> this<00:16:20.600> additional<00:16:21.050> thing<00:16:21.230> called<00:16:21.529> the 00:16:21.579 --> 00:16:21.589 align:start position:0% has this additional thing called the 00:16:21.589 --> 00:16:25.059 align:start position:0% has this additional thing called the weights<00:16:21.769> and<00:16:22.630> then<00:16:23.630> we<00:16:23.899> can<00:16:24.110> we<00:16:24.800> need<00:16:24.920> to<00:16:24.980> be 00:16:25.059 --> 00:16:25.069 align:start position:0% weights and then we can we need to be 00:16:25.069 --> 00:16:27.790 align:start position:0% weights and then we can we need to be define<00:16:25.339> the<00:16:25.610> score<00:16:26.329> of<00:16:26.480> a<00:16:26.540> block<00:16:26.779> to<00:16:27.139> use<00:16:27.500> the 00:16:27.790 --> 00:16:27.800 align:start position:0% define the score of a block to use the 00:16:27.800 --> 00:16:30.670 align:start position:0% define the score of a block to use the weights<00:16:28.100> of<00:16:28.459> the<00:16:29.120> parents<00:16:29.689> block<00:16:29.930> in<00:16:30.319> order<00:16:30.589> to 00:16:30.670 --> 00:16:30.680 align:start position:0% weights of the parents block in order to 00:16:30.680 --> 00:16:33.490 align:start position:0% weights of the parents block in order to actually<00:16:30.829> choose<00:16:31.130> that<00:16:31.160> block<00:16:31.459> so<00:16:31.670> if<00:16:32.290> block<00:16:33.290> a 00:16:33.490 --> 00:16:33.500 align:start position:0% actually choose that block so if block a 00:16:33.500 --> 00:16:36.069 align:start position:0% actually choose that block so if block a has<00:16:33.980> weights<00:16:34.399> has<00:16:34.939> a<00:16:35.240> certain<00:16:35.689> set<00:16:35.839> of<00:16:35.870> weights 00:16:36.069 --> 00:16:36.079 align:start position:0% has weights has a certain set of weights 00:16:36.079 --> 00:16:37.749 align:start position:0% has weights has a certain set of weights then<00:16:36.410> the<00:16:36.529> children<00:16:37.009> walk<00:16:37.250> a<00:16:37.399> which<00:16:37.579> are<00:16:37.670> the 00:16:37.749 --> 00:16:37.759 align:start position:0% then the children walk a which are the 00:16:37.759 --> 00:16:39.579 align:start position:0% then the children walk a which are the blocks<00:16:38.209> that<00:16:38.329> have<00:16:38.689> a<00:16:38.720> pre-populated<00:16:39.139> block<00:16:39.560> a 00:16:39.579 --> 00:16:39.589 align:start position:0% blocks that have a pre-populated block a 00:16:39.589 --> 00:16:41.079 align:start position:0% blocks that have a pre-populated block a need<00:16:40.220> to<00:16:40.310> be<00:16:40.370> chosen<00:16:40.550> according<00:16:40.819> those 00:16:41.079 --> 00:16:41.089 align:start position:0% need to be chosen according those 00:16:41.089 --> 00:16:43.329 align:start position:0% need to be chosen according those weights<00:16:41.300> and<00:16:41.630> not<00:16:42.589> a<00:16:42.620> single<00:16:43.069> one<00:16:43.160> of<00:16:43.250> the 00:16:43.329 --> 00:16:43.339 align:start position:0% weights and not a single one of the 00:16:43.339 --> 00:16:44.920 align:start position:0% weights and not a single one of the other<00:16:43.430> definitions<00:16:43.880> need<00:16:44.000> to<00:16:44.089> change<00:16:44.300> and<00:16:44.540> so 00:16:44.920 --> 00:16:44.930 align:start position:0% other definitions need to change and so 00:16:44.930 --> 00:16:46.840 align:start position:0% other definitions need to change and so this<00:16:45.079> is<00:16:45.230> like<00:16:45.800> really<00:16:46.040> like<00:16:46.279> all<00:16:46.459> that<00:16:46.730> we 00:16:46.840 --> 00:16:46.850 align:start position:0% this is like really like all that we 00:16:46.850 --> 00:16:49.240 align:start position:0% this is like really like all that we needed<00:16:47.000> to<00:16:47.120> do<00:16:47.300> to<00:16:47.559> do<00:16:48.559> consensus<00:16:48.980> safety 00:16:49.240 --> 00:16:49.250 align:start position:0% needed to do to do consensus safety 00:16:49.250 --> 00:16:51.009 align:start position:0% needed to do to do consensus safety sorry<00:16:49.910> to<00:16:50.000> do<00:16:50.149> validate<00:16:50.509> rotation<00:16:50.839> with 00:16:51.009 --> 00:16:51.019 align:start position:0% sorry to do validate rotation with 00:16:51.019 --> 00:16:53.350 align:start position:0% sorry to do validate rotation with consent<00:16:51.319> the<00:16:51.380> safety<00:16:51.910> basically<00:16:52.910> because<00:16:53.149> all 00:16:53.350 --> 00:16:53.360 align:start position:0% consent the safety basically because all 00:16:53.360 --> 00:16:55.809 align:start position:0% consent the safety basically because all of<00:16:53.389> the<00:16:53.600> definitions<00:16:54.110> satisfy<00:16:54.680> the<00:16:54.740> the<00:16:55.339> car 00:16:55.809 --> 00:16:55.819 align:start position:0% of the definitions satisfy the the car 00:16:55.819 --> 00:16:57.370 align:start position:0% of the definitions satisfy the the car crackers<00:16:56.120> like<00:16:56.329> the<00:16:56.509> the<00:16:56.779> abstract<00:16:57.139> safety 00:16:57.370 --> 00:16:57.380 align:start position:0% crackers like the the abstract safety 00:16:57.380 --> 00:17:02.650 align:start position:0% crackers like the the abstract safety proof<00:16:59.110> you<00:17:00.110> know<00:17:00.230> because<00:17:01.220> like<00:17:01.540> because<00:17:02.540> they 00:17:02.650 --> 00:17:02.660 align:start position:0% proof you know because like because they 00:17:02.660 --> 00:17:04.720 align:start position:0% proof you know because like because they they<00:17:02.839> you<00:17:03.319> know<00:17:03.410> they're<00:17:03.559> they<00:17:04.189> have<00:17:04.220> a<00:17:04.429> common 00:17:04.720 --> 00:17:04.730 align:start position:0% they you know they're they have a common 00:17:04.730 --> 00:17:08.079 align:start position:0% they you know they're they have a common feature<00:17:04.880> unless<00:17:05.089> there's<00:17:05.270> t<00:17:05.390> faults<00:17:06.250> and<00:17:07.250> they 00:17:08.079 --> 00:17:08.089 align:start position:0% feature unless there's t faults and they 00:17:08.089 --> 00:17:09.309 align:start position:0% feature unless there's t faults and they only<00:17:08.240> decide<00:17:08.480> on<00:17:08.600> safe<00:17:08.780> estimates<00:17:09.110> so 00:17:09.309 --> 00:17:09.319 align:start position:0% only decide on safe estimates so 00:17:09.319 --> 00:17:11.559 align:start position:0% only decide on safe estimates so liveness<00:17:09.500> is<00:17:09.799> an<00:17:09.890> interesting<00:17:10.100> thing<00:17:10.569> you 00:17:11.559 --> 00:17:11.569 align:start position:0% liveness is an interesting thing you 00:17:11.569 --> 00:17:13.149 align:start position:0% liveness is an interesting thing you know<00:17:11.689> liveness<00:17:11.959> means<00:17:12.679> and<00:17:12.890> knows<00:17:12.980> like 00:17:13.149 --> 00:17:13.159 align:start position:0% know liveness means and knows like 00:17:13.159 --> 00:17:14.409 align:start position:0% know liveness means and knows like events<00:17:13.490> are<00:17:13.669> guaranteed<00:17:13.970> to<00:17:14.030> eventually<00:17:14.329> make 00:17:14.409 --> 00:17:14.419 align:start position:0% events are guaranteed to eventually make 00:17:14.419 --> 00:17:16.240 align:start position:0% events are guaranteed to eventually make a<00:17:14.480> decision<00:17:14.870> there's<00:17:15.319> a<00:17:15.380> result<00:17:15.649> in<00:17:15.890> consensus 00:17:16.240 --> 00:17:16.250 align:start position:0% a decision there's a result in consensus 00:17:16.250 --> 00:17:17.260 align:start position:0% a decision there's a result in consensus literature<00:17:16.520> called<00:17:16.640> FLP<00:17:17.000> and<00:17:17.089> poss 00:17:17.260 --> 00:17:17.270 align:start position:0% literature called FLP and poss 00:17:17.270 --> 00:17:18.579 align:start position:0% literature called FLP and poss ability<00:17:17.570> that<00:17:17.720> means<00:17:17.870> that<00:17:18.020> you<00:17:18.440> know<00:17:18.530> you 00:17:18.579 --> 00:17:18.589 align:start position:0% ability that means that you know you 00:17:18.589 --> 00:17:20.710 align:start position:0% ability that means that you know you can't<00:17:18.800> really<00:17:18.980> be<00:17:19.220> live<00:17:19.550> and<00:17:19.880> safe<00:17:20.420> in<00:17:20.630> an 00:17:20.710 --> 00:17:20.720 align:start position:0% can't really be live and safe in an 00:17:20.720 --> 00:17:22.810 align:start position:0% can't really be live and safe in an asynchronous<00:17:20.959> network<00:17:21.470> without<00:17:21.650> and<00:17:22.070> without 00:17:22.810 --> 00:17:22.820 align:start position:0% asynchronous network without and without 00:17:22.820 --> 00:17:26.590 align:start position:0% asynchronous network without and without using<00:17:23.120> entropy<00:17:24.080> or<00:17:24.260> oh<00:17:24.709> no<00:17:25.240> without<00:17:26.240> using 00:17:26.590 --> 00:17:26.600 align:start position:0% using entropy or oh no without using 00:17:26.600 --> 00:17:31.270 align:start position:0% using entropy or oh no without using entropy<00:17:26.930> or<00:17:28.060> cryptography<00:17:29.980> but<00:17:30.980> actually<00:17:31.220> we 00:17:31.270 --> 00:17:31.280 align:start position:0% entropy or cryptography but actually we 00:17:31.280 --> 00:17:33.310 align:start position:0% entropy or cryptography but actually we have<00:17:31.430> we<00:17:31.610> have<00:17:31.760> kind<00:17:32.540> of<00:17:32.600> a<00:17:32.720> bigger<00:17:32.900> issue<00:17:33.050> than 00:17:33.310 --> 00:17:33.320 align:start position:0% have we have kind of a bigger issue than 00:17:33.320 --> 00:17:34.630 align:start position:0% have we have kind of a bigger issue than just<00:17:33.500> have<00:17:33.620> up<00:17:33.710> being<00:17:33.860> possibility<00:17:34.370> which<00:17:34.550> is 00:17:34.630 --> 00:17:34.640 align:start position:0% just have up being possibility which is 00:17:34.640 --> 00:17:36.100 align:start position:0% just have up being possibility which is that<00:17:34.760> we<00:17:34.820> never<00:17:34.940> said<00:17:35.270> anything<00:17:35.540> about<00:17:35.720> when 00:17:36.100 --> 00:17:36.110 align:start position:0% that we never said anything about when 00:17:36.110 --> 00:17:37.480 align:start position:0% that we never said anything about when validators<00:17:36.680> were<00:17:36.770> actually<00:17:37.010> meant<00:17:37.220> to<00:17:37.340> send 00:17:37.480 --> 00:17:37.490 align:start position:0% validators were actually meant to send 00:17:37.490 --> 00:17:43.330 align:start position:0% validators were actually meant to send blocks<00:17:38.590> they<00:17:39.590> actually<00:17:42.010> we<00:17:43.010> just<00:17:43.040> haven't 00:17:43.330 --> 00:17:43.340 align:start position:0% blocks they actually we just haven't 00:17:43.340 --> 00:17:44.500 align:start position:0% blocks they actually we just haven't specified<00:17:43.520> that<00:17:43.850> yet<00:17:43.970> cuz<00:17:44.120> all<00:17:44.210> we<00:17:44.300> really 00:17:44.500 --> 00:17:44.510 align:start position:0% specified that yet cuz all we really 00:17:44.510 --> 00:17:48.040 align:start position:0% specified that yet cuz all we really talked<00:17:44.720> about<00:17:44.900> is<00:17:45.020> safety<00:17:45.350> and<00:17:45.650> so<00:17:47.050> however 00:17:48.040 --> 00:17:48.050 align:start position:0% talked about is safety and so however 00:17:48.050 --> 00:17:50.260 align:start position:0% talked about is safety and so however we're<00:17:48.320> gonna<00:17:48.440> see<00:17:48.860> in<00:17:49.190> our<00:17:49.430> experimental 00:17:50.260 --> 00:17:50.270 align:start position:0% we're gonna see in our experimental 00:17:50.270 --> 00:17:53.820 align:start position:0% we're gonna see in our experimental results<00:17:50.810> in<00:17:50.900> just<00:17:51.050> a<00:17:51.140> second<00:17:51.790> that<00:17:52.790> there<00:17:53.000> are 00:17:53.820 --> 00:17:53.830 align:start position:0% results in just a second that there are 00:17:53.830 --> 00:17:56.080 align:start position:0% results in just a second that there are state<00:17:54.830> sets<00:17:55.220> of<00:17:55.340> messages<00:17:55.460> that<00:17:55.820> they<00:17:55.880> could 00:17:56.080 --> 00:17:56.090 align:start position:0% state sets of messages that they could 00:17:56.090 --> 00:17:59.020 align:start position:0% state sets of messages that they could get<00:17:56.270> to<00:17:56.540> which<00:17:56.960> are<00:17:57.170> safe<00:17:57.670> and<00:17:58.670> that<00:17:58.730> means 00:17:59.020 --> 00:17:59.030 align:start position:0% get to which are safe and that means 00:17:59.030 --> 00:18:00.910 align:start position:0% get to which are safe and that means that<00:17:59.240> as<00:17:59.840> long<00:18:00.020> as<00:18:00.110> we<00:18:00.170> can<00:18:00.290> make<00:18:00.440> those<00:18:00.650> shapes 00:18:00.910 --> 00:18:00.920 align:start position:0% that as long as we can make those shapes 00:18:00.920 --> 00:18:04.270 align:start position:0% that as long as we can make those shapes then<00:18:01.460> we<00:18:01.580> can<00:18:01.700> have<00:18:01.820> liveness<00:18:02.060> but<00:18:02.920> and<00:18:03.920> and<00:18:04.190> we 00:18:04.270 --> 00:18:04.280 align:start position:0% then we can have liveness but and and we 00:18:04.280 --> 00:18:08.290 align:start position:0% then we can have liveness but and and we kind<00:18:04.460> of<00:18:04.520> know<00:18:05.290> that<00:18:06.290> in<00:18:07.130> an<00:18:07.310> in<00:18:07.490> a<00:18:07.790> synchronous 00:18:08.290 --> 00:18:08.300 align:start position:0% kind of know that in an in a synchronous 00:18:08.300 --> 00:18:10.300 align:start position:0% kind of know that in an in a synchronous network<00:18:08.660> or<00:18:08.870> in<00:18:09.380> a<00:18:09.440> partially<00:18:09.890> synchronous 00:18:10.300 --> 00:18:10.310 align:start position:0% network or in a partially synchronous 00:18:10.310 --> 00:18:13.120 align:start position:0% network or in a partially synchronous network<00:18:10.580> nodes<00:18:11.120> can<00:18:11.240> use<00:18:11.360> timeouts<00:18:11.660> to<00:18:12.440> make 00:18:13.120 --> 00:18:13.130 align:start position:0% network nodes can use timeouts to make 00:18:13.130 --> 00:18:15.610 align:start position:0% network nodes can use timeouts to make this<00:18:13.370> kind<00:18:13.520> of<00:18:13.610> message<00:18:14.420> shape<00:18:14.810> you<00:18:15.440> know<00:18:15.530> up 00:18:15.610 --> 00:18:15.620 align:start position:0% this kind of message shape you know up 00:18:15.620 --> 00:18:19.120 align:start position:0% this kind of message shape you know up to<00:18:15.710> some<00:18:15.830> number<00:18:15.950> of<00:18:16.070> faults<00:18:17.350> and<00:18:18.350> so<00:18:18.800> and<00:18:19.070> so 00:18:19.120 --> 00:18:19.130 align:start position:0% to some number of faults and so and so 00:18:19.130 --> 00:18:22.000 align:start position:0% to some number of faults and so and so if<00:18:19.250> we<00:18:19.340> were<00:18:19.430> to<00:18:19.550> pick<00:18:19.760> one<00:18:19.940> of<00:18:20.060> these<00:18:20.860> then<00:18:21.860> we 00:18:22.000 --> 00:18:22.010 align:start position:0% if we were to pick one of these then we 00:18:22.010 --> 00:18:24.310 align:start position:0% if we were to pick one of these then we could<00:18:22.280> we<00:18:22.790> could<00:18:23.060> give<00:18:23.330> a<00:18:23.360> we<00:18:23.840> could<00:18:24.080> give<00:18:24.290> a 00:18:24.310 --> 00:18:24.320 align:start position:0% could we could give a we could give a 00:18:24.320 --> 00:18:27.100 align:start position:0% could we could give a we could give a lightness<00:18:24.740> proof<00:18:24.950> for<00:18:25.550> some<00:18:25.820> strategy<00:18:26.360> that 00:18:27.100 --> 00:18:27.110 align:start position:0% lightness proof for some strategy that 00:18:27.110 --> 00:18:28.810 align:start position:0% lightness proof for some strategy that says<00:18:27.350> oh<00:18:27.620> look<00:18:27.650> you<00:18:28.070> guys<00:18:28.250> have<00:18:28.430> to<00:18:28.520> figure<00:18:28.550> out 00:18:28.810 --> 00:18:28.820 align:start position:0% says oh look you guys have to figure out 00:18:28.820 --> 00:18:30.430 align:start position:0% says oh look you guys have to figure out how<00:18:28.880> to<00:18:28.940> make<00:18:29.120> a<00:18:29.150> dag<00:18:29.390> and<00:18:29.570> this<00:18:29.660> messages<00:18:30.350> in 00:18:30.430 --> 00:18:30.440 align:start position:0% how to make a dag and this messages in 00:18:30.440 --> 00:18:35.230 align:start position:0% how to make a dag and this messages in this<00:18:30.500> in<00:18:30.680> this<00:18:30.770> form<00:18:33.100> but<00:18:34.100> you<00:18:34.700> know<00:18:34.820> we're<00:18:35.090> not 00:18:35.230 --> 00:18:35.240 align:start position:0% this in this form but you know we're not 00:18:35.240 --> 00:18:37.440 align:start position:0% this in this form but you know we're not really<00:18:35.450> gonna<00:18:35.600> do<00:18:35.780> that<00:18:35.930> we're<00:18:36.800> just<00:18:36.830> gonna 00:18:37.440 --> 00:18:37.450 align:start position:0% really gonna do that we're just gonna 00:18:37.450 --> 00:18:40.710 align:start position:0% really gonna do that we're just gonna leave<00:18:38.450> liveness<00:18:38.750> out<00:18:39.050> of<00:18:39.140> the<00:18:39.200> story<00:18:39.440> for<00:18:39.620> now 00:18:40.710 --> 00:18:40.720 align:start position:0% leave liveness out of the story for now 00:18:40.720 --> 00:18:43.060 align:start position:0% leave liveness out of the story for now so<00:18:41.720> I'm<00:18:41.810> gonna<00:18:41.930> I<00:18:42.020> mean<00:18:42.320> this<00:18:42.500> is<00:18:42.650> kind<00:18:42.860> of<00:18:42.890> the 00:18:43.060 --> 00:18:43.070 align:start position:0% so I'm gonna I mean this is kind of the 00:18:43.070 --> 00:18:44.650 align:start position:0% so I'm gonna I mean this is kind of the maybe<00:18:43.610> the<00:18:43.730> more<00:18:43.880> exciting<00:18:44.210> part<00:18:44.360> of<00:18:44.390> the<00:18:44.480> talk 00:18:44.650 --> 00:18:44.660 align:start position:0% maybe the more exciting part of the talk 00:18:44.660 --> 00:18:46.630 align:start position:0% maybe the more exciting part of the talk so<00:18:45.110> this<00:18:45.260> is<00:18:45.500> we're<00:18:45.920> gonna<00:18:46.010> look<00:18:46.190> at<00:18:46.370> actually 00:18:46.630 --> 00:18:46.640 align:start position:0% so this is we're gonna look at actually 00:18:46.640 --> 00:18:48.730 align:start position:0% so this is we're gonna look at actually and<00:18:46.850> try<00:18:46.970> to<00:18:47.030> see<00:18:47.750> what<00:18:47.900> this<00:18:47.990> looks<00:18:48.050> like<00:18:48.380> so 00:18:48.730 --> 00:18:48.740 align:start position:0% and try to see what this looks like so 00:18:48.740 --> 00:18:50.950 align:start position:0% and try to see what this looks like so these<00:18:49.730> are<00:18:49.910> three<00:18:50.150> validators<00:18:50.660> running<00:18:50.870> the 00:18:50.950 --> 00:18:50.960 align:start position:0% these are three validators running the 00:18:50.960 --> 00:18:54.970 align:start position:0% these are three validators running the binary<00:18:51.260> consensus<00:18:51.730> and<00:18:52.730> they<00:18:53.330> the<00:18:54.290> color<00:18:54.530> here 00:18:54.970 --> 00:18:54.980 align:start position:0% binary consensus and they the color here 00:18:54.980 --> 00:18:56.860 align:start position:0% binary consensus and they the color here means<00:18:55.400> that<00:18:55.490> they've<00:18:55.670> reached<00:18:55.910> safety<00:18:56.720> with 00:18:56.860 --> 00:18:56.870 align:start position:0% means that they've reached safety with 00:18:56.870 --> 00:18:57.790 align:start position:0% means that they've reached safety with some<00:18:56.990> of<00:18:57.050> them<00:18:57.170> some<00:18:57.410> amount<00:18:57.560> of<00:18:57.620> fault 00:18:57.790 --> 00:18:57.800 align:start position:0% some of them some amount of fault 00:18:57.800 --> 00:19:00.850 align:start position:0% some of them some amount of fault tolerance<00:18:58.480> and<00:18:59.480> kind<00:18:59.930> of<00:18:59.990> they<00:19:00.530> they're 00:19:00.850 --> 00:19:00.860 align:start position:0% tolerance and kind of they they're 00:19:00.860 --> 00:19:02.950 align:start position:0% tolerance and kind of they they're they're<00:19:01.070> the<00:19:01.340> the<00:19:01.820> numbers<00:19:02.210> in<00:19:02.390> here<00:19:02.630> you<00:19:02.780> see 00:19:02.950 --> 00:19:02.960 align:start position:0% they're the the numbers in here you see 00:19:02.960 --> 00:19:05.290 align:start position:0% they're the the numbers in here you see there's<00:19:03.380> ones<00:19:03.680> and<00:19:04.040> that<00:19:04.520> one's<00:19:04.730> a<00:19:04.850> zero<00:19:05.090> it 00:19:05.290 --> 00:19:05.300 align:start position:0% there's ones and that one's a zero it 00:19:05.300 --> 00:19:07.390 align:start position:0% there's ones and that one's a zero it can<00:19:05.420> be<00:19:05.480> a<00:19:05.540> little<00:19:05.630> bit<00:19:05.780> hard<00:19:05.930> to<00:19:05.960> see<00:19:06.400> the 00:19:07.390 --> 00:19:07.400 align:start position:0% can be a little bit hard to see the 00:19:07.400 --> 00:19:09.100 align:start position:0% can be a little bit hard to see the numbers<00:19:07.430> are<00:19:08.060> their<00:19:08.390> estimates<00:19:08.810> and<00:19:08.990> they 00:19:09.100 --> 00:19:09.110 align:start position:0% numbers are their estimates and they 00:19:09.110 --> 00:19:10.900 align:start position:0% numbers are their estimates and they need<00:19:09.260> to<00:19:09.350> have<00:19:09.740> the<00:19:09.950> estimate<00:19:10.340> be<00:19:10.490> the<00:19:10.670> max 00:19:10.900 --> 00:19:10.910 align:start position:0% need to have the estimate be the max 00:19:10.910 --> 00:19:12.250 align:start position:0% need to have the estimate be the max weight<00:19:11.210> of<00:19:11.480> the<00:19:11.570> latest<00:19:11.840> messages<00:19:12.110> they've 00:19:12.250 --> 00:19:12.260 align:start position:0% weight of the latest messages they've 00:19:12.260 --> 00:19:14.730 align:start position:0% weight of the latest messages they've seen<00:19:12.440> in<00:19:12.530> the<00:19:12.590> justifications<00:19:13.130> these<00:19:13.610> arrows 00:19:14.730 --> 00:19:14.740 align:start position:0% seen in the justifications these arrows 00:19:14.740 --> 00:19:17.940 align:start position:0% seen in the justifications these arrows these<00:19:15.740> lines<00:19:16.040> are<00:19:16.370> the<00:19:16.580> met<00:19:16.760> point<00:19:17.240> to<00:19:17.540> the 00:19:17.940 --> 00:19:17.950 align:start position:0% these lines are the met point to the 00:19:17.950 --> 00:19:19.600 align:start position:0% these lines are the met point to the messages<00:19:18.950> that<00:19:18.980> are<00:19:19.070> included<00:19:19.490> in<00:19:19.550> the 00:19:19.600 --> 00:19:19.610 align:start position:0% messages that are included in the 00:19:19.610 --> 00:19:21.520 align:start position:0% messages that are included in the justification<00:19:20.210> remember<00:19:20.480> the<00:19:20.690> the<00:19:21.140> data 00:19:21.520 --> 00:19:21.530 align:start position:0% justification remember the the data 00:19:21.530 --> 00:19:23.350 align:start position:0% justification remember the the data structure<00:19:21.890> is<00:19:22.010> estimate<00:19:22.910> which<00:19:23.090> is<00:19:23.120> a<00:19:23.180> bit 00:19:23.350 --> 00:19:23.360 align:start position:0% structure is estimate which is a bit 00:19:23.360 --> 00:19:25.180 align:start position:0% structure is estimate which is a bit sender<00:19:24.050> which<00:19:24.260> is<00:19:24.380> like<00:19:24.500> you<00:19:24.680> know<00:19:24.770> validator 00:19:25.180 --> 00:19:25.190 align:start position:0% sender which is like you know validator 00:19:25.190 --> 00:19:27.130 align:start position:0% sender which is like you know validator one<00:19:25.400> validator<00:19:25.910> to<00:19:25.940> about<00:19:26.240> a<00:19:26.270> three<00:19:26.570> and<00:19:26.720> the 00:19:27.130 --> 00:19:27.140 align:start position:0% one validator to about a three and the 00:19:27.140 --> 00:19:28.630 align:start position:0% one validator to about a three and the justification<00:19:27.470> which<00:19:27.890> is<00:19:28.010> encoded<00:19:28.340> with<00:19:28.400> this 00:19:28.630 --> 00:19:28.640 align:start position:0% justification which is encoded with this 00:19:28.640 --> 00:19:29.950 align:start position:0% justification which is encoded with this with<00:19:28.880> this<00:19:28.970> structure 00:19:29.950 --> 00:19:29.960 align:start position:0% with this structure 00:19:29.960 --> 00:19:32.440 align:start position:0% with this structure showing<00:19:30.960> you<00:19:31.290> know<00:19:31.410> a<00:19:31.440> few<00:19:31.710> more<00:19:31.860> nodes 00:19:32.440 --> 00:19:32.450 align:start position:0% showing you know a few more nodes 00:19:32.450 --> 00:19:36.110 align:start position:0% showing you know a few more nodes company<00:19:33.450> consents<00:19:33.810> on<00:19:33.900> a<00:19:33.930> bit<00:19:34.140> here<00:19:34.350> so<00:19:35.250> the 00:19:36.110 --> 00:19:36.120 align:start position:0% company consents on a bit here so the 00:19:36.120 --> 00:19:38.450 align:start position:0% company consents on a bit here so the darker<00:19:36.510> color<00:19:36.750> means<00:19:37.110> that<00:19:37.380> they<00:19:37.950> have<00:19:38.130> more 00:19:38.450 --> 00:19:38.460 align:start position:0% darker color means that they have more 00:19:38.460 --> 00:19:39.620 align:start position:0% darker color means that they have more fault<00:19:39.120> tolerance 00:19:39.620 --> 00:19:39.630 align:start position:0% fault tolerance 00:19:39.630 --> 00:19:41.330 align:start position:0% fault tolerance this<00:19:40.320> is<00:19:40.530> kind<00:19:40.740> of<00:19:40.800> cool<00:19:40.980> because<00:19:41.190> it 00:19:41.330 --> 00:19:41.340 align:start position:0% this is kind of cool because it 00:19:41.340 --> 00:19:43.310 align:start position:0% this is kind of cool because it illustrates<00:19:41.430> the<00:19:41.880> the<00:19:42.570> flexibility<00:19:43.170> that<00:19:43.260> we 00:19:43.310 --> 00:19:43.320 align:start position:0% illustrates the the flexibility that we 00:19:43.320 --> 00:19:45.320 align:start position:0% illustrates the the flexibility that we have<00:19:43.530> about<00:19:44.130> the<00:19:44.250> order<00:19:44.730> of<00:19:44.820> message<00:19:45.090> passing 00:19:45.320 --> 00:19:45.330 align:start position:0% have about the order of message passing 00:19:45.330 --> 00:19:47.240 align:start position:0% have about the order of message passing because<00:19:46.230> we<00:19:46.290> haven't<00:19:46.470> said<00:19:46.680> really<00:19:46.980> anything 00:19:47.240 --> 00:19:47.250 align:start position:0% because we haven't said really anything 00:19:47.250 --> 00:19:50.450 align:start position:0% because we haven't said really anything at<00:19:47.430> all<00:19:47.520> actually<00:19:48.180> about<00:19:48.330> that<00:19:49.400> now<00:19:50.400> the 00:19:50.450 --> 00:19:50.460 align:start position:0% at all actually about that now the 00:19:50.460 --> 00:19:51.620 align:start position:0% at all actually about that now the blockchain<00:19:50.760> consensus<00:19:51.240> protocol<00:19:51.480> is<00:19:51.570> a 00:19:51.620 --> 00:19:51.630 align:start position:0% blockchain consensus protocol is a 00:19:51.630 --> 00:19:52.670 align:start position:0% blockchain consensus protocol is a little<00:19:51.840> bit<00:19:51.930> more<00:19:52.080> complicated<00:19:52.320> to<00:19:52.530> think 00:19:52.670 --> 00:19:52.680 align:start position:0% little bit more complicated to think 00:19:52.680 --> 00:19:53.990 align:start position:0% little bit more complicated to think about<00:19:52.860> because<00:19:53.160> it's<00:19:53.400> got<00:19:53.610> these<00:19:53.790> things 00:19:53.990 --> 00:19:54.000 align:start position:0% about because it's got these things 00:19:54.000 --> 00:19:55.730 align:start position:0% about because it's got these things called<00:19:54.120> pre<00:19:54.510> block<00:19:54.750> pointers<00:19:55.110> right<00:19:55.320> so 00:19:55.730 --> 00:19:55.740 align:start position:0% called pre block pointers right so 00:19:55.740 --> 00:19:57.560 align:start position:0% called pre block pointers right so there's<00:19:55.950> dotted<00:19:56.310> lines<00:19:56.550> these<00:19:56.820> thin<00:19:57.180> ones<00:19:57.390> are 00:19:57.560 --> 00:19:57.570 align:start position:0% there's dotted lines these thin ones are 00:19:57.570 --> 00:19:59.540 align:start position:0% there's dotted lines these thin ones are justification<00:19:58.200> pointers<00:19:58.500> these<00:19:59.100> are<00:19:59.280> free 00:19:59.540 --> 00:19:59.550 align:start position:0% justification pointers these are free 00:19:59.550 --> 00:20:01.610 align:start position:0% justification pointers these are free block<00:19:59.790> pointers<00:20:00.000> the<00:20:00.570> blue<00:20:00.870> ones<00:20:01.110> are<00:20:01.380> the 00:20:01.610 --> 00:20:01.620 align:start position:0% block pointers the blue ones are the 00:20:01.620 --> 00:20:05.540 align:start position:0% block pointers the blue ones are the latest<00:20:02.270> for<00:20:03.270> Khalid<00:20:04.200> ators<00:20:04.440> and<00:20:04.650> the<00:20:05.160> red<00:20:05.370> one 00:20:05.540 --> 00:20:05.550 align:start position:0% latest for Khalid ators and the red one 00:20:05.550 --> 00:20:07.190 align:start position:0% latest for Khalid ators and the red one is<00:20:05.670> our<00:20:05.820> fork<00:20:06.060> choice<00:20:06.240> so<00:20:06.390> basically<00:20:06.600> like<00:20:06.780> you 00:20:07.190 --> 00:20:07.200 align:start position:0% is our fork choice so basically like you 00:20:07.200 --> 00:20:08.960 align:start position:0% is our fork choice so basically like you know<00:20:07.290> the<00:20:07.830> way<00:20:08.130> the<00:20:08.250> fork<00:20:08.400> choice<00:20:08.550> works<00:20:08.820> is 00:20:08.960 --> 00:20:08.970 align:start position:0% know the way the fork choice works is 00:20:08.970 --> 00:20:11.270 align:start position:0% know the way the fork choice works is okay<00:20:09.210> you<00:20:09.270> know<00:20:09.540> you<00:20:10.260> so<00:20:10.560> validator<00:20:10.920> wanna<00:20:11.100> has 00:20:11.270 --> 00:20:11.280 align:start position:0% okay you know you so validator wanna has 00:20:11.280 --> 00:20:12.980 align:start position:0% okay you know you so validator wanna has the<00:20:11.370> most<00:20:11.550> weight<00:20:12.000> so<00:20:12.480> from<00:20:12.660> the<00:20:12.750> Genesis 00:20:12.980 --> 00:20:12.990 align:start position:0% the most weight so from the Genesis 00:20:12.990 --> 00:20:15.740 align:start position:0% the most weight so from the Genesis block<00:20:13.050> we<00:20:13.230> go<00:20:13.320> to<00:20:13.380> balladeer<00:20:13.710> one<00:20:13.890> and<00:20:14.750> here 00:20:15.740 --> 00:20:15.750 align:start position:0% block we go to balladeer one and here 00:20:15.750 --> 00:20:18.500 align:start position:0% block we go to balladeer one and here there's<00:20:15.930> only<00:20:16.040> has<00:20:17.040> one<00:20:17.250> parent<00:20:17.640> and<00:20:18.180> here<00:20:18.390> we 00:20:18.500 --> 00:20:18.510 align:start position:0% there's only has one parent and here we 00:20:18.510 --> 00:20:19.580 align:start position:0% there's only has one parent and here we could<00:20:18.690> we<00:20:18.840> could<00:20:18.900> have<00:20:19.020> went<00:20:19.110> that<00:20:19.320> way<00:20:19.470> or 00:20:19.580 --> 00:20:19.590 align:start position:0% could we could have went that way or 00:20:19.590 --> 00:20:21.530 align:start position:0% could we could have went that way or this<00:20:19.620> way<00:20:19.770> but<00:20:20.340> actually<00:20:20.460> the<00:20:21.270> latest 00:20:21.530 --> 00:20:21.540 align:start position:0% this way but actually the latest 00:20:21.540 --> 00:20:23.000 align:start position:0% this way but actually the latest messages<00:20:21.900> are<00:20:22.020> here<00:20:22.260> there<00:20:22.470> and<00:20:22.560> there<00:20:22.650> and 00:20:23.000 --> 00:20:23.010 align:start position:0% messages are here there and there and 00:20:23.010 --> 00:20:25.760 align:start position:0% messages are here there and there and all<00:20:23.190> of<00:20:23.400> them<00:20:23.550> have<00:20:23.990> their<00:20:24.990> weight<00:20:25.260> here<00:20:25.620> and 00:20:25.760 --> 00:20:25.770 align:start position:0% all of them have their weight here and 00:20:25.770 --> 00:20:27.590 align:start position:0% all of them have their weight here and so<00:20:26.040> the<00:20:26.190> weight<00:20:26.310> is<00:20:26.460> all<00:20:26.670> on<00:20:26.880> this<00:20:27.060> fork<00:20:27.420> and 00:20:27.590 --> 00:20:27.600 align:start position:0% so the weight is all on this fork and 00:20:27.600 --> 00:20:29.420 align:start position:0% so the weight is all on this fork and not<00:20:27.720> on<00:20:27.840> that<00:20:27.960> one<00:20:27.990> and<00:20:28.320> so<00:20:28.530> we<00:20:28.620> pull<00:20:28.830> left<00:20:29.070> and 00:20:29.420 --> 00:20:29.430 align:start position:0% not on that one and so we pull left and 00:20:29.430 --> 00:20:31.910 align:start position:0% not on that one and so we pull left and then<00:20:30.210> here<00:20:30.600> you<00:20:30.720> know<00:20:30.810> there's<00:20:31.500> a<00:20:31.590> pre<00:20:31.740> block 00:20:31.910 --> 00:20:31.920 align:start position:0% then here you know there's a pre block 00:20:31.920 --> 00:20:33.800 align:start position:0% then here you know there's a pre block pointer<00:20:32.190> here<00:20:32.580> and<00:20:33.000> also<00:20:33.210> another<00:20:33.330> one<00:20:33.510> there 00:20:33.800 --> 00:20:33.810 align:start position:0% pointer here and also another one there 00:20:33.810 --> 00:20:37.240 align:start position:0% pointer here and also another one there and<00:20:34.200> if<00:20:34.500> you<00:20:34.560> look<00:20:34.680> at<00:20:34.770> the<00:20:34.830> latest<00:20:34.980> messages 00:20:37.240 --> 00:20:37.250 align:start position:0% and if you look at the latest messages 00:20:37.250 --> 00:20:41.120 align:start position:0% and if you look at the latest messages there's<00:20:38.250> more<00:20:38.580> of<00:20:38.700> them<00:20:38.790> here<00:20:39.590> and<00:20:40.590> so<00:20:40.800> and<00:20:41.040> so 00:20:41.120 --> 00:20:41.130 align:start position:0% there's more of them here and so and so 00:20:41.130 --> 00:20:44.600 align:start position:0% there's more of them here and so and so that's<00:20:41.280> why<00:20:41.370> we're<00:20:41.490> gonna<00:20:41.580> go<00:20:41.820> there<00:20:42.620> and<00:20:43.620> and 00:20:44.600 --> 00:20:44.610 align:start position:0% that's why we're gonna go there and and 00:20:44.610 --> 00:20:46.010 align:start position:0% that's why we're gonna go there and and and<00:20:44.730> and<00:20:45.000> and<00:20:45.240> what<00:20:45.390> this<00:20:45.480> kind<00:20:45.660> of<00:20:45.720> shows<00:20:45.900> you 00:20:46.010 --> 00:20:46.020 align:start position:0% and and and what this kind of shows you 00:20:46.020 --> 00:20:53.570 align:start position:0% and and and what this kind of shows you here<00:20:46.200> is<00:20:46.260> that<00:20:46.350> validator<00:20:46.910> 3<00:20:51.560> has<00:20:52.560> more<00:20:53.400> weight 00:20:53.570 --> 00:20:53.580 align:start position:0% here is that validator 3 has more weight 00:20:53.580 --> 00:20:59.090 align:start position:0% here is that validator 3 has more weight than<00:20:53.700> validator<00:20:54.120> 1<00:20:57.170> right<00:20:58.170> so<00:20:58.320> the<00:20:58.410> reason<00:20:59.010> why 00:20:59.090 --> 00:20:59.100 align:start position:0% than validator 1 right so the reason why 00:20:59.100 --> 00:21:00.440 align:start position:0% than validator 1 right so the reason why we<00:20:59.190> initially<00:20:59.280> chose<00:20:59.640> this<00:20:59.910> actually<00:21:00.330> wasn't 00:21:00.440 --> 00:21:00.450 align:start position:0% we initially chose this actually wasn't 00:21:00.450 --> 00:21:01.670 align:start position:0% we initially chose this actually wasn't because<00:21:00.660> validator<00:21:00.990> 0<00:21:01.170> had<00:21:01.290> more<00:21:01.440> weight<00:21:01.560> is 00:21:01.670 --> 00:21:01.680 align:start position:0% because validator 0 had more weight is 00:21:01.680 --> 00:21:03.260 align:start position:0% because validator 0 had more weight is because<00:21:01.860> all<00:21:02.190> of<00:21:02.280> their<00:21:02.370> latest<00:21:02.580> messages<00:21:03.090> are 00:21:03.260 --> 00:21:03.270 align:start position:0% because all of their latest messages are 00:21:03.270 --> 00:21:10.310 align:start position:0% because all of their latest messages are on<00:21:03.600> this<00:21:04.110> one<00:21:05.030> so<00:21:08.930> here<00:21:09.930> we<00:21:10.020> go 00:21:10.310 --> 00:21:10.320 align:start position:0% on this one so here we go 00:21:10.320 --> 00:21:11.810 align:start position:0% on this one so here we go so<00:21:10.470> this<00:21:10.680> so<00:21:11.220> this<00:21:11.340> is<00:21:11.400> kind<00:21:11.670> of<00:21:11.700> the 00:21:11.810 --> 00:21:11.820 align:start position:0% so this so this is kind of the 00:21:11.820 --> 00:21:13.370 align:start position:0% so this so this is kind of the blockchain<00:21:12.300> consensus<00:21:12.810> kind<00:21:12.990> of<00:21:13.050> an<00:21:13.140> action 00:21:13.370 --> 00:21:13.380 align:start position:0% blockchain consensus kind of an action 00:21:13.380 --> 00:21:16.310 align:start position:0% blockchain consensus kind of an action for<00:21:13.560> three<00:21:13.710> notes<00:21:14.510> there<00:21:15.510> they're<00:21:15.840> gonna<00:21:15.930> like 00:21:16.310 --> 00:21:16.320 align:start position:0% for three notes there they're gonna like 00:21:16.320 --> 00:21:18.410 align:start position:0% for three notes there they're gonna like use<00:21:17.100> greedy<00:21:17.460> heaviest<00:21:17.790> observed<00:21:18.060> for<00:21:18.240> choice 00:21:18.410 --> 00:21:18.420 align:start position:0% use greedy heaviest observed for choice 00:21:18.420 --> 00:21:20.810 align:start position:0% use greedy heaviest observed for choice rule<00:21:18.660> where<00:21:18.840> every<00:21:19.100> every<00:21:20.100> choice<00:21:20.310> in<00:21:20.670> the 00:21:20.810 --> 00:21:20.820 align:start position:0% rule where every every choice in the 00:21:20.820 --> 00:21:22.640 align:start position:0% rule where every every choice in the road<00:21:21.000> that<00:21:21.270> but<00:21:21.870> first<00:21:22.080> we<00:21:22.230> went<00:21:22.320> that<00:21:22.440> way<00:21:22.590> and 00:21:22.640 --> 00:21:22.650 align:start position:0% road that but first we went that way and 00:21:22.650 --> 00:21:23.480 align:start position:0% road that but first we went that way and then<00:21:22.770> we<00:21:22.800> went<00:21:22.920> that<00:21:23.040> way<00:21:23.130> because<00:21:23.250> more 00:21:23.480 --> 00:21:23.490 align:start position:0% then we went that way because more 00:21:23.490 --> 00:21:25.550 align:start position:0% then we went that way because more validators<00:21:24.000> made<00:21:24.150> walks<00:21:24.390> there<00:21:24.630> here<00:21:25.290> we<00:21:25.470> were 00:21:25.550 --> 00:21:25.560 align:start position:0% validators made walks there here we were 00:21:25.560 --> 00:21:27.470 align:start position:0% validators made walks there here we were only<00:21:25.980> fork<00:21:26.280> right<00:21:26.460> there's<00:21:26.670> this<00:21:26.850> forking<00:21:27.270> on 00:21:27.470 --> 00:21:27.480 align:start position:0% only fork right there's this forking on 00:21:27.480 --> 00:21:29.270 align:start position:0% only fork right there's this forking on the<00:21:27.630> top<00:21:27.840> and<00:21:28.110> then<00:21:28.530> here<00:21:28.650> we<00:21:28.740> achieve<00:21:29.040> safety 00:21:29.270 --> 00:21:29.280 align:start position:0% the top and then here we achieve safety 00:21:29.280 --> 00:21:31.340 align:start position:0% the top and then here we achieve safety right<00:21:29.940> this<00:21:30.150> kind<00:21:30.330> of<00:21:30.360> say<00:21:30.570> this<00:21:30.600> is<00:21:30.810> one<00:21:31.260> of 00:21:31.340 --> 00:21:31.350 align:start position:0% right this kind of say this is one of 00:21:31.350 --> 00:21:32.540 align:start position:0% right this kind of say this is one of these<00:21:31.410> things<00:21:31.560> get<00:21:31.770> colored<00:21:32.040> to<00:21:32.400> have<00:21:32.520> a 00:21:32.540 --> 00:21:32.550 align:start position:0% these things get colored to have a 00:21:32.550 --> 00:21:33.740 align:start position:0% these things get colored to have a synchronous<00:21:33.090> Byzantine<00:21:33.450> fault<00:21:33.630> tolerance 00:21:33.740 --> 00:21:33.750 align:start position:0% synchronous Byzantine fault tolerance 00:21:33.750 --> 00:21:36.980 align:start position:0% synchronous Byzantine fault tolerance safety<00:21:34.260> and<00:21:35.250> notice<00:21:35.940> the<00:21:36.030> word<00:21:36.150> they<00:21:36.600> fork<00:21:36.900> at 00:21:36.980 --> 00:21:36.990 align:start position:0% safety and notice the word they fork at 00:21:36.990 --> 00:21:38.060 align:start position:0% safety and notice the word they fork at the<00:21:37.050> top<00:21:37.230> and<00:21:37.410> they<00:21:37.530> have<00:21:37.620> safety<00:21:37.920> at<00:21:38.010> the 00:21:38.060 --> 00:21:38.070 align:start position:0% the top and they have safety at the 00:21:38.070 --> 00:21:39.620 align:start position:0% the top and they have safety at the bottom<00:21:38.370> and<00:21:38.490> the<00:21:38.790> way<00:21:38.880> we<00:21:39.000> calculate<00:21:39.120> safety 00:21:39.620 --> 00:21:39.630 align:start position:0% bottom and the way we calculate safety 00:21:39.630 --> 00:21:42.050 align:start position:0% bottom and the way we calculate safety here<00:21:39.810> you<00:21:40.260> know<00:21:40.380> is<00:21:40.650> using<00:21:41.490> the<00:21:41.550> safety<00:21:41.730> Oracle 00:21:42.050 --> 00:21:42.060 align:start position:0% here you know is using the safety Oracle 00:21:42.060 --> 00:21:43.070 align:start position:0% here you know is using the safety Oracle that<00:21:42.180> I<00:21:42.210> described<00:21:42.510> earlier<00:21:42.630> where<00:21:42.870> you<00:21:42.960> look 00:21:43.070 --> 00:21:43.080 align:start position:0% that I described earlier where you look 00:21:43.080 --> 00:21:43.650 align:start position:0% that I described earlier where you look at<00:21:43.170> these<00:21:43.260> cleat 00:21:43.650 --> 00:21:43.660 align:start position:0% at these cleat 00:21:43.660 --> 00:21:45.330 align:start position:0% at these cleat which<00:21:43.780> see<00:21:44.440> each<00:21:44.500> other<00:21:44.650> agreeing<00:21:44.950> can't<00:21:45.220> see 00:21:45.330 --> 00:21:45.340 align:start position:0% which see each other agreeing can't see 00:21:45.340 --> 00:21:48.000 align:start position:0% which see each other agreeing can't see each<00:21:45.430> other<00:21:45.580> disagree<00:21:46.000> now<00:21:46.650> we<00:21:47.650> can<00:21:47.860> watch 00:21:48.000 --> 00:21:48.010 align:start position:0% each other disagree now we can watch 00:21:48.010 --> 00:21:51.240 align:start position:0% each other disagree now we can watch more<00:21:48.880> nodes<00:21:49.240> go<00:21:49.780> faster<00:21:50.140> and<00:21:50.470> here's<00:21:50.980> the<00:21:51.160> you 00:21:51.240 --> 00:21:51.250 align:start position:0% more nodes go faster and here's the you 00:21:51.250 --> 00:21:53.070 align:start position:0% more nodes go faster and here's the you see<00:21:51.460> some<00:21:51.970> interesting<00:21:52.090> stuff<00:21:52.480> were<00:21:52.600> like<00:21:52.870> a 00:21:53.070 --> 00:21:53.080 align:start position:0% see some interesting stuff were like a 00:21:53.080 --> 00:21:54.660 align:start position:0% see some interesting stuff were like a bunch<00:21:53.560> of<00:21:53.650> a<00:21:53.770> bunch<00:21:54.160> of<00:21:54.250> blocks<00:21:54.370> will<00:21:54.550> get 00:21:54.660 --> 00:21:54.670 align:start position:0% bunch of a bunch of blocks will get 00:21:54.670 --> 00:21:55.830 align:start position:0% bunch of a bunch of blocks will get finalized<00:21:55.030> at<00:21:55.120> the<00:21:55.210> same<00:21:55.390> time<00:21:55.600> sometimes 00:21:55.830 --> 00:21:55.840 align:start position:0% finalized at the same time sometimes 00:21:55.840 --> 00:21:58.020 align:start position:0% finalized at the same time sometimes right<00:21:56.170> because<00:21:56.770> every<00:21:57.220> the<00:21:57.490> thing<00:21:57.760> that<00:21:57.880> kind 00:21:58.020 --> 00:21:58.030 align:start position:0% right because every the thing that kind 00:21:58.030 --> 00:21:59.730 align:start position:0% right because every the thing that kind of<00:21:58.060> realize<00:21:58.300> like<00:21:58.450> blocks<00:21:58.960> can<00:21:59.230> contribute<00:21:59.530> to 00:21:59.730 --> 00:21:59.740 align:start position:0% of realize like blocks can contribute to 00:21:59.740 --> 00:22:03.330 align:start position:0% of realize like blocks can contribute to the<00:21:59.860> finality<00:22:00.220> of<00:22:00.340> many<00:22:00.520> many<00:22:00.700> blocks<00:22:02.040> and<00:22:03.040> and 00:22:03.330 --> 00:22:03.340 align:start position:0% the finality of many many blocks and and 00:22:03.340 --> 00:22:05.610 align:start position:0% the finality of many many blocks and and basically<00:22:04.300> because<00:22:04.570> every<00:22:05.200> block<00:22:05.410> you<00:22:05.590> have 00:22:05.610 --> 00:22:05.620 align:start position:0% basically because every block you have 00:22:05.620 --> 00:22:07.800 align:start position:0% basically because every block you have will<00:22:06.460> like<00:22:06.640> agree<00:22:07.000> and<00:22:07.150> disagree<00:22:07.270> with<00:22:07.660> other 00:22:07.800 --> 00:22:07.810 align:start position:0% will like agree and disagree with other 00:22:07.810 --> 00:22:10.860 align:start position:0% will like agree and disagree with other blocks<00:22:08.400> large<00:22:09.400> number<00:22:09.730> of<00:22:09.760> them<00:22:09.910> and<00:22:10.120> which 00:22:10.860 --> 00:22:10.870 align:start position:0% blocks large number of them and which 00:22:10.870 --> 00:22:12.210 align:start position:0% blocks large number of them and which blocks<00:22:11.140> you<00:22:11.260> disagree<00:22:11.620> with<00:22:11.770> and<00:22:11.860> agree<00:22:12.070> with 00:22:12.210 --> 00:22:12.220 align:start position:0% blocks you disagree with and agree with 00:22:12.220 --> 00:22:13.530 align:start position:0% blocks you disagree with and agree with actually<00:22:12.520> will<00:22:12.610> determine<00:22:12.910> how<00:22:13.180> the<00:22:13.240> safety 00:22:13.530 --> 00:22:13.540 align:start position:0% actually will determine how the safety 00:22:13.540 --> 00:22:16.230 align:start position:0% actually will determine how the safety Oracle<00:22:13.900> turns<00:22:14.110> out<00:22:14.430> and<00:22:15.430> so<00:22:15.550> remember<00:22:15.940> the 00:22:16.230 --> 00:22:16.240 align:start position:0% Oracle turns out and so remember the 00:22:16.240 --> 00:22:17.910 align:start position:0% Oracle turns out and so remember the colors<00:22:16.540> not<00:22:16.780> forking<00:22:17.200> up<00:22:17.350> here<00:22:17.620> at<00:22:17.680> the<00:22:17.740> top 00:22:17.910 --> 00:22:17.920 align:start position:0% colors not forking up here at the top 00:22:17.920 --> 00:22:19.320 align:start position:0% colors not forking up here at the top lots<00:22:18.160> of<00:22:18.250> 4k<00:22:18.490> like<00:22:18.580> look<00:22:18.730> at<00:22:18.820> the<00:22:18.880> red<00:22:19.000> line<00:22:19.180> it 00:22:19.320 --> 00:22:19.330 align:start position:0% lots of 4k like look at the red line it 00:22:19.330 --> 00:22:22.980 align:start position:0% lots of 4k like look at the red line it kind<00:22:19.690> of<00:22:19.720> goes<00:22:20.230> and<00:22:20.530> changes<00:22:21.040> right<00:22:21.880> like<00:22:22.150> you 00:22:22.980 --> 00:22:22.990 align:start position:0% kind of goes and changes right like you 00:22:22.990 --> 00:22:24.300 align:start position:0% kind of goes and changes right like you don't<00:22:23.140> really<00:22:23.290> know<00:22:23.530> if<00:22:23.740> these<00:22:23.860> blocks<00:22:24.130> are 00:22:24.300 --> 00:22:24.310 align:start position:0% don't really know if these blocks are 00:22:24.310 --> 00:22:25.620 align:start position:0% don't really know if these blocks are confirmed<00:22:24.670> or<00:22:24.880> they're<00:22:25.060> gonna<00:22:25.120> be<00:22:25.240> consensus 00:22:25.620 --> 00:22:25.630 align:start position:0% confirmed or they're gonna be consensus 00:22:25.630 --> 00:22:27.750 align:start position:0% confirmed or they're gonna be consensus but<00:22:25.990> the<00:22:26.110> ones<00:22:26.260> down<00:22:26.410> here<00:22:26.440> are<00:22:26.650> right<00:22:27.070> so<00:22:27.280> it's 00:22:27.750 --> 00:22:27.760 align:start position:0% but the ones down here are right so it's 00:22:27.760 --> 00:22:29.790 align:start position:0% but the ones down here are right so it's very<00:22:27.910> much<00:22:28.030> like<00:22:28.420> a<00:22:28.960> four<00:22:29.260> key<00:22:29.410> thing<00:22:29.590> it's<00:22:29.770> a 00:22:29.790 --> 00:22:29.800 align:start position:0% very much like a four key thing it's a 00:22:29.800 --> 00:22:31.800 align:start position:0% very much like a four key thing it's a kind<00:22:29.980> of<00:22:30.010> a<00:22:30.070> blockchain<00:22:30.250> like<00:22:30.610> thing<00:22:30.790> but<00:22:31.690> this 00:22:31.800 --> 00:22:31.810 align:start position:0% kind of a blockchain like thing but this 00:22:31.810 --> 00:22:33.390 align:start position:0% kind of a blockchain like thing but this is<00:22:32.020> really<00:22:32.380> this<00:22:32.860> is<00:22:32.920> really<00:22:33.220> what<00:22:33.340> I'm 00:22:33.390 --> 00:22:33.400 align:start position:0% is really this is really what I'm 00:22:33.400 --> 00:22:33.810 align:start position:0% is really this is really what I'm talking<00:22:33.640> about 00:22:33.810 --> 00:22:33.820 align:start position:0% talking about 00:22:33.820 --> 00:22:36.810 align:start position:0% talking about this<00:22:34.480> is<00:22:34.810> kind<00:22:35.350> of<00:22:35.500> what<00:22:36.250> I<00:22:36.280> was<00:22:36.430> saying<00:22:36.580> when<00:22:36.760> I 00:22:36.810 --> 00:22:36.820 align:start position:0% this is kind of what I was saying when I 00:22:36.820 --> 00:22:38.550 align:start position:0% this is kind of what I was saying when I say<00:22:36.910> that<00:22:37.210> I<00:22:37.420> can<00:22:37.750> achieve<00:22:37.870> asynchronous 00:22:38.550 --> 00:22:38.560 align:start position:0% say that I can achieve asynchronous 00:22:38.560 --> 00:22:40.650 align:start position:0% say that I can achieve asynchronous Byzantine<00:22:39.430> fault<00:22:39.640> tolerant<00:22:39.970> consensus<00:22:40.390> with 00:22:40.650 --> 00:22:40.660 align:start position:0% Byzantine fault tolerant consensus with 00:22:40.660 --> 00:22:42.840 align:start position:0% Byzantine fault tolerant consensus with the<00:22:40.690> overhead<00:22:41.320> of<00:22:41.350> Nakamoto<00:22:41.950> consensus<00:22:42.400> if 00:22:42.840 --> 00:22:42.850 align:start position:0% the overhead of Nakamoto consensus if 00:22:42.850 --> 00:22:46.740 align:start position:0% the overhead of Nakamoto consensus if you<00:22:43.420> if<00:22:43.600> they<00:22:43.750> do<00:22:45.120> manage<00:22:46.120> to<00:22:46.480> send<00:22:46.630> the 00:22:46.740 --> 00:22:46.750 align:start position:0% you if they do manage to send the 00:22:46.750 --> 00:22:49.170 align:start position:0% you if they do manage to send the messages<00:22:47.050> around<00:22:47.170> a<00:22:47.290> round-robin<00:22:48.180> then 00:22:49.170 --> 00:22:49.180 align:start position:0% messages around a round-robin then 00:22:49.180 --> 00:22:50.700 align:start position:0% messages around a round-robin then eventually<00:22:49.690> you<00:22:49.930> get<00:22:50.020> to<00:22:50.140> a<00:22:50.170> point<00:22:50.380> where 00:22:50.700 --> 00:22:50.710 align:start position:0% eventually you get to a point where 00:22:50.710 --> 00:22:54.690 align:start position:0% eventually you get to a point where every<00:22:51.340> additional<00:22:51.850> message<00:22:53.550> finalized<00:22:54.550> an 00:22:54.690 --> 00:22:54.700 align:start position:0% every additional message finalized an 00:22:54.700 --> 00:22:57.690 align:start position:0% every additional message finalized an additional<00:22:55.090> block<00:22:55.240> and<00:22:55.710> this<00:22:56.710> kind<00:22:57.520> of 00:22:57.690 --> 00:22:57.700 align:start position:0% additional block and this kind of 00:22:57.700 --> 00:22:59.760 align:start position:0% additional block and this kind of overhead<00:22:58.120> for<00:22:58.360> finality<00:22:59.080> of<00:22:59.320> consensus<00:22:59.680> at 00:22:59.760 --> 00:22:59.770 align:start position:0% overhead for finality of consensus at 00:22:59.770 --> 00:23:01.860 align:start position:0% overhead for finality of consensus at every<00:22:59.920> block<00:23:00.130> like<00:23:00.640> one<00:23:00.940> message<00:23:01.360> per<00:23:01.600> block 00:23:01.860 --> 00:23:01.870 align:start position:0% every block like one message per block 00:23:01.870 --> 00:23:03.540 align:start position:0% every block like one message per block for<00:23:02.200> finale<00:23:02.530> at<00:23:02.710> any<00:23:02.860> level<00:23:03.130> of<00:23:03.190> thought<00:23:03.310> on<00:23:03.520> a 00:23:03.540 --> 00:23:03.550 align:start position:0% for finale at any level of thought on a 00:23:03.550 --> 00:23:05.520 align:start position:0% for finale at any level of thought on a threshold<00:23:03.850> is<00:23:04.150> like<00:23:04.720> theoretically<00:23:05.170> optimal 00:23:05.520 --> 00:23:05.530 align:start position:0% threshold is like theoretically optimal 00:23:05.530 --> 00:23:07.230 align:start position:0% threshold is like theoretically optimal because<00:23:05.710> you<00:23:06.580> know<00:23:06.640> you<00:23:06.730> can't<00:23:06.940> really<00:23:07.090> get 00:23:07.230 --> 00:23:07.240 align:start position:0% because you know you can't really get 00:23:07.240 --> 00:23:09.660 align:start position:0% because you know you can't really get lower<00:23:07.450> than<00:23:07.630> one<00:23:07.750> message<00:23:08.050> for<00:23:08.170> block<00:23:08.670> and 00:23:09.660 --> 00:23:09.670 align:start position:0% lower than one message for block and 00:23:09.670 --> 00:23:10.770 align:start position:0% lower than one message for block and this<00:23:09.760> kind<00:23:09.970> of<00:23:10.000> little<00:23:10.210> interesting<00:23:10.510> here<00:23:10.630> you 00:23:10.770 --> 00:23:10.780 align:start position:0% this kind of little interesting here you 00:23:10.780 --> 00:23:14.070 align:start position:0% this kind of little interesting here you see<00:23:10.930> that<00:23:11.760> this<00:23:12.760> validator<00:23:13.360> has<00:23:13.510> more<00:23:13.810> weight 00:23:14.070 --> 00:23:14.080 align:start position:0% see that this validator has more weight 00:23:14.080 --> 00:23:15.870 align:start position:0% see that this validator has more weight than<00:23:14.380> these<00:23:14.590> two<00:23:14.890> there's<00:23:15.340> why<00:23:15.520> the<00:23:15.580> fourth 00:23:15.870 --> 00:23:15.880 align:start position:0% than these two there's why the fourth 00:23:15.880 --> 00:23:18.660 align:start position:0% than these two there's why the fourth choice<00:23:16.060> rule<00:23:16.680> this<00:23:17.680> validator<00:23:18.130> even<00:23:18.610> though 00:23:18.660 --> 00:23:18.670 align:start position:0% choice rule this validator even though 00:23:18.670 --> 00:23:20.700 align:start position:0% choice rule this validator even though you<00:23:18.730> saw<00:23:18.850> that<00:23:19.030> block<00:23:19.270> whatever<00:23:20.140> preferred<00:23:20.590> to 00:23:20.700 --> 00:23:20.710 align:start position:0% you saw that block whatever preferred to 00:23:20.710 --> 00:23:22.410 align:start position:0% you saw that block whatever preferred to built<00:23:20.860> on<00:23:20.950> his<00:23:21.100> like<00:23:21.250> initial<00:23:21.640> block<00:23:21.850> than 00:23:22.410 --> 00:23:22.420 align:start position:0% built on his like initial block than 00:23:22.420 --> 00:23:24.000 align:start position:0% built on his like initial block than these<00:23:22.600> ones<00:23:22.810> because<00:23:22.960> these<00:23:23.200> two<00:23:23.440> guys<00:23:23.470> have 00:23:24.000 --> 00:23:24.010 align:start position:0% these ones because these two guys have 00:23:24.010 --> 00:23:27.690 align:start position:0% these ones because these two guys have less<00:23:24.190> weight<00:23:24.400> than<00:23:24.430> this<00:23:24.670> one<00:23:25.500> and<00:23:26.500> so<00:23:27.250> this<00:23:27.580> is 00:23:27.690 --> 00:23:27.700 align:start position:0% less weight than this one and so this is 00:23:27.700 --> 00:23:29.550 align:start position:0% less weight than this one and so this is kind<00:23:27.880> of<00:23:27.940> what<00:23:28.120> I<00:23:28.150> mean<00:23:28.330> when<00:23:29.020> I<00:23:29.050> say<00:23:29.290> that<00:23:29.440> like 00:23:29.550 --> 00:23:29.560 align:start position:0% kind of what I mean when I say that like 00:23:29.560 --> 00:23:30.960 align:start position:0% kind of what I mean when I say that like you<00:23:29.800> know<00:23:29.920> we<00:23:30.010> can<00:23:30.130> achieve<00:23:30.310> really<00:23:30.670> really 00:23:30.960 --> 00:23:30.970 align:start position:0% you know we can achieve really really 00:23:30.970 --> 00:23:33.890 align:start position:0% you know we can achieve really really efficient<00:23:32.130> asynchronous<00:23:33.130> consensus<00:23:33.490> a<00:23:33.580> few 00:23:33.890 --> 00:23:33.900 align:start position:0% efficient asynchronous consensus a few 00:23:33.900 --> 00:23:37.920 align:start position:0% efficient asynchronous consensus a few so<00:23:35.010> somehow<00:23:36.060> you<00:23:37.060> know<00:23:37.150> I<00:23:37.180> know<00:23:37.330> it's<00:23:37.750> like<00:23:37.900> a 00:23:37.920 --> 00:23:37.930 align:start position:0% so somehow you know I know it's like a 00:23:37.930 --> 00:23:40.050 align:start position:0% so somehow you know I know it's like a bit<00:23:38.110> much<00:23:38.230> but<00:23:38.560> I<00:23:38.830> shared 00:23:40.050 --> 00:23:40.060 align:start position:0% bit much but I shared 00:23:40.060 --> 00:23:43.250 align:start position:0% bit much but I shared consensus<00:23:40.780> safety<00:23:41.020> proof<00:23:41.260> that<00:23:42.010> talks<00:23:42.460> about 00:23:43.250 --> 00:23:43.260 align:start position:0% consensus safety proof that talks about 00:23:43.260 --> 00:23:46.170 align:start position:0% consensus safety proof that talks about you<00:23:44.260> know<00:23:44.350> these<00:23:44.560> local<00:23:45.060> determinations<00:23:46.060> of 00:23:46.170 --> 00:23:46.180 align:start position:0% you know these local determinations of 00:23:46.180 --> 00:23:48.120 align:start position:0% you know these local determinations of estimate<00:23:46.570> safety<00:23:46.870> and<00:23:47.110> how<00:23:47.410> they're<00:23:47.740> related 00:23:48.120 --> 00:23:48.130 align:start position:0% estimate safety and how they're related 00:23:48.130 --> 00:23:51.210 align:start position:0% estimate safety and how they're related to<00:23:48.310> consensus<00:23:49.240> safety<00:23:49.710> we're<00:23:50.710> able<00:23:50.800> to<00:23:50.950> define 00:23:51.210 --> 00:23:51.220 align:start position:0% to consensus safety we're able to define 00:23:51.220 --> 00:23:54.090 align:start position:0% to consensus safety we're able to define of<00:23:51.370> watching<00:23:51.640> consensus<00:23:51.940> protocol<00:23:53.100> through 00:23:54.090 --> 00:23:54.100 align:start position:0% of watching consensus protocol through 00:23:54.100 --> 00:23:56.190 align:start position:0% of watching consensus protocol through like<00:23:54.310> basically<00:23:54.790> by<00:23:55.030> first<00:23:55.750> defining<00:23:55.900> the 00:23:56.190 --> 00:23:56.200 align:start position:0% like basically by first defining the 00:23:56.200 --> 00:23:57.270 align:start position:0% like basically by first defining the binary<00:23:56.530> consensus<00:23:56.800> protocol 00:23:57.270 --> 00:23:57.280 align:start position:0% binary consensus protocol 00:23:57.280 --> 00:23:58.950 align:start position:0% binary consensus protocol by<00:23:57.460> the<00:23:57.520> safety<00:23:57.700> proof<00:23:57.970> and<00:23:58.150> making<00:23:58.630> a<00:23:58.720> really 00:23:58.950 --> 00:23:58.960 align:start position:0% by the safety proof and making a really 00:23:58.960 --> 00:24:00.990 align:start position:0% by the safety proof and making a really like<00:23:59.200> minor<00:23:59.620> change<00:24:00.010> that<00:24:00.580> doesn't<00:24:00.850> affect 00:24:00.990 --> 00:24:01.000 align:start position:0% like minor change that doesn't affect 00:24:01.000 --> 00:24:03.260 align:start position:0% like minor change that doesn't affect the<00:24:01.120> safety<00:24:01.270> proof<00:24:01.540> at<00:24:01.660> all 00:24:03.260 --> 00:24:03.270 align:start position:0% the safety proof at all 00:24:03.270 --> 00:24:05.670 align:start position:0% the safety proof at all and<00:24:04.270> then<00:24:04.750> we<00:24:04.870> were<00:24:04.960> able<00:24:04.990> to<00:24:05.170> actually<00:24:05.590> add 00:24:05.670 --> 00:24:05.680 align:start position:0% and then we were able to actually add 00:24:05.680 --> 00:24:08.340 align:start position:0% and then we were able to actually add validate<00:24:06.100> a<00:24:06.130> rotation<00:24:06.610> and<00:24:06.850> remove<00:24:07.350> finality 00:24:08.340 --> 00:24:08.350 align:start position:0% validate a rotation and remove finality 00:24:08.350 --> 00:24:09.900 align:start position:0% validate a rotation and remove finality thresholds<00:24:08.860> also<00:24:09.190> with<00:24:09.400> very<00:24:09.700> minimal 00:24:09.900 --> 00:24:09.910 align:start position:0% thresholds also with very minimal 00:24:09.910 --> 00:24:11.850 align:start position:0% thresholds also with very minimal changes<00:24:10.360> to<00:24:10.450> the<00:24:10.540> protocol<00:24:10.870> and<00:24:11.080> the<00:24:11.620> reason 00:24:11.850 --> 00:24:11.860 align:start position:0% changes to the protocol and the reason 00:24:11.860 --> 00:24:13.770 align:start position:0% changes to the protocol and the reason why<00:24:11.950> we<00:24:12.070> can<00:24:12.220> do<00:24:12.340> this<00:24:12.490> and<00:24:12.760> why<00:24:13.030> it's<00:24:13.210> like<00:24:13.360> so 00:24:13.770 --> 00:24:13.780 align:start position:0% why we can do this and why it's like so 00:24:13.780 --> 00:24:15.570 align:start position:0% why we can do this and why it's like so little<00:24:13.990> work<00:24:14.140> is<00:24:14.380> because<00:24:14.890> of<00:24:15.160> the<00:24:15.340> fact<00:24:15.520> that 00:24:15.570 --> 00:24:15.580 align:start position:0% little work is because of the fact that 00:24:15.580 --> 00:24:17.580 align:start position:0% little work is because of the fact that the<00:24:15.730> proofs<00:24:16.000> that<00:24:16.750> were<00:24:16.840> lying<00:24:17.080> on<00:24:17.230> are<00:24:17.380> not 00:24:17.580 --> 00:24:17.590 align:start position:0% the proofs that were lying on are not 00:24:17.590 --> 00:24:19.110 align:start position:0% the proofs that were lying on are not changing<00:24:18.160> while<00:24:18.580> we're<00:24:18.790> making<00:24:19.030> these 00:24:19.110 --> 00:24:19.120 align:start position:0% changing while we're making these 00:24:19.120 --> 00:24:20.460 align:start position:0% changing while we're making these changes<00:24:19.330> right<00:24:19.600> there<00:24:19.780> we<00:24:20.230> have<00:24:20.260> like 00:24:20.460 --> 00:24:20.470 align:start position:0% changes right there we have like 00:24:20.470 --> 00:24:22.950 align:start position:0% changes right there we have like abstract<00:24:20.890> safety<00:24:21.340> proofs<00:24:21.960> concrete 00:24:22.950 --> 00:24:22.960 align:start position:0% abstract safety proofs concrete 00:24:22.960 --> 00:24:25.410 align:start position:0% abstract safety proofs concrete protocols<00:24:23.470> which<00:24:24.010> we<00:24:24.220> can<00:24:24.370> tweak<00:24:24.610> without 00:24:25.410 --> 00:24:25.420 align:start position:0% protocols which we can tweak without 00:24:25.420 --> 00:24:29.220 align:start position:0% protocols which we can tweak without changing<00:24:25.990> the<00:24:26.080> safety<00:24:26.320> proof<00:24:27.390> and<00:24:28.390> you<00:24:29.110> know 00:24:29.220 --> 00:24:29.230 align:start position:0% changing the safety proof and you know 00:24:29.230 --> 00:24:30.570 align:start position:0% changing the safety proof and you know we<00:24:29.380> were<00:24:29.470> able<00:24:29.560> to<00:24:29.710> achieve<00:24:29.830> this<00:24:30.130> really<00:24:30.400> nice 00:24:30.570 --> 00:24:30.580 align:start position:0% we were able to achieve this really nice 00:24:30.580 --> 00:24:35.220 align:start position:0% we were able to achieve this really nice kind<00:24:31.030> of<00:24:31.270> latency<00:24:31.990> numbers<00:24:33.750> so<00:24:34.750> that<00:24:34.960> you<00:24:35.110> know 00:24:35.220 --> 00:24:35.230 align:start position:0% kind of latency numbers so that you know 00:24:35.230 --> 00:24:39.000 align:start position:0% kind of latency numbers so that you know I<00:24:35.490> think<00:24:36.490> that<00:24:37.450> you<00:24:38.110> know<00:24:38.170> this<00:24:38.290> is<00:24:38.440> hopefully 00:24:39.000 --> 00:24:39.010 align:start position:0% I think that you know this is hopefully 00:24:39.010 --> 00:24:41.670 align:start position:0% I think that you know this is hopefully gonna<00:24:39.220> be<00:24:39.340> educational<00:24:40.140> hopefully<00:24:41.140> gonna 00:24:41.670 --> 00:24:41.680 align:start position:0% gonna be educational hopefully gonna 00:24:41.680 --> 00:24:43.260 align:start position:0% gonna be educational hopefully gonna lead<00:24:41.890> to<00:24:41.980> new<00:24:42.070> consensus<00:24:42.460> protocols<00:24:42.820> you<00:24:43.180> know 00:24:43.260 --> 00:24:43.270 align:start position:0% lead to new consensus protocols you know 00:24:43.270 --> 00:24:44.400 align:start position:0% lead to new consensus protocols you know we<00:24:43.390> actually<00:24:43.480> have<00:24:43.780> a<00:24:43.810> bunch<00:24:44.080> of<00:24:44.140> Kuznets 00:24:44.400 --> 00:24:44.410 align:start position:0% we actually have a bunch of Kuznets 00:24:44.410 --> 00:24:45.900 align:start position:0% we actually have a bunch of Kuznets protocols<00:24:44.740> that<00:24:44.800> were<00:24:44.950> mentioned<00:24:45.160> here<00:24:45.430> that 00:24:45.900 --> 00:24:45.910 align:start position:0% protocols that were mentioned here that 00:24:45.910 --> 00:24:46.950 align:start position:0% protocols that were mentioned here that were<00:24:46.000> derived<00:24:46.210> according<00:24:46.570> this<00:24:46.660> process 00:24:46.950 --> 00:24:46.960 align:start position:0% were derived according this process 00:24:46.960 --> 00:24:50.330 align:start position:0% were derived according this process replicating<00:24:47.410> things<00:24:47.590> like<00:24:48.000> concurrent 00:24:50.330 --> 00:24:50.340 align:start position:0% replicating things like concurrent 00:24:50.340 --> 00:24:52.530 align:start position:0% replicating things like concurrent instead<00:24:51.340> of<00:24:51.430> sequential<00:24:51.760> ones<00:24:52.030> or<00:24:52.240> like<00:24:52.390> an 00:24:52.530 --> 00:24:52.540 align:start position:0% instead of sequential ones or like an 00:24:52.540 --> 00:24:55.470 align:start position:0% instead of sequential ones or like an integer<00:24:53.820> through<00:24:54.820> these<00:24:54.910> domain<00:24:55.210> for<00:24:55.360> time 00:24:55.470 --> 00:24:55.480 align:start position:0% integer through these domain for time 00:24:55.480 --> 00:24:59.550 align:start position:0% integer through these domain for time stamping<00:24:55.810> so<00:24:56.350> here's<00:24:57.040> a<00:24:57.310> little<00:24:57.700> secret<00:24:58.470> mmm 00:24:59.550 --> 00:24:59.560 align:start position:0% stamping so here's a little secret mmm 00:24:59.560 --> 00:25:03.210 align:start position:0% stamping so here's a little secret mmm just<00:24:59.920> now<00:25:00.100> I<00:25:00.870> published<00:25:01.870> the<00:25:02.260> prototype<00:25:02.680> for 00:25:03.210 --> 00:25:03.220 align:start position:0% just now I published the prototype for 00:25:03.220 --> 00:25:08.610 align:start position:0% just now I published the prototype for the<00:25:03.490> for<00:25:04.150> these<00:25:04.890> fourteen<00:25:05.890> consensus<00:25:07.620> and 00:25:08.610 --> 00:25:08.620 align:start position:0% the for these fourteen consensus and 00:25:08.620 --> 00:25:11.940 align:start position:0% the for these fourteen consensus and also<00:25:09.430> this<00:25:09.880> white<00:25:10.690> like<00:25:10.960> a<00:25:10.990> specification<00:25:11.470> in 00:25:11.940 --> 00:25:11.950 align:start position:0% also this white like a specification in 00:25:11.950 --> 00:25:14.850 align:start position:0% also this white like a specification in very<00:25:12.640> much<00:25:12.790> detail<00:25:12.850> of<00:25:13.390> all<00:25:14.110> of<00:25:14.560> this<00:25:14.650> stuff 00:25:14.850 --> 00:25:14.860 align:start position:0% very much detail of all of this stuff 00:25:14.860 --> 00:25:19.230 align:start position:0% very much detail of all of this stuff and<00:25:15.990> more<00:25:16.990> incomplete<00:25:17.620> description<00:25:18.160> of<00:25:18.670> the 00:25:19.230 --> 00:25:19.240 align:start position:0% and more incomplete description of the 00:25:19.240 --> 00:25:20.940 align:start position:0% and more incomplete description of the correct<00:25:19.450> my<00:25:19.540> construction<00:25:19.840> process<00:25:20.290> given<00:25:20.860> at 00:25:20.940 --> 00:25:20.950 align:start position:0% correct my construction process given at 00:25:20.950 --> 00:25:22.110 align:start position:0% correct my construction process given at a<00:25:21.040> very<00:25:21.280> very<00:25:21.490> high<00:25:21.670> level<00:25:21.700> of<00:25:22.030> abstraction 00:25:22.110 --> 00:25:22.120 align:start position:0% a very very high level of abstraction 00:25:22.120 --> 00:25:25.670 align:start position:0% a very very high level of abstraction much<00:25:23.080> higher<00:25:23.320> than<00:25:23.590> I<00:25:23.650> even<00:25:23.920> hinted<00:25:24.310> at<00:25:24.400> today 00:25:25.670 --> 00:25:25.680 align:start position:0% much higher than I even hinted at today 00:25:25.680 --> 00:25:28.920 align:start position:0% much higher than I even hinted at today so<00:25:26.680> you<00:25:27.220> know<00:25:27.310> this<00:25:27.910> kind<00:25:28.240> of<00:25:28.270> represents<00:25:28.900> the 00:25:28.920 --> 00:25:28.930 align:start position:0% so you know this kind of represents the 00:25:28.930 --> 00:25:33.330 align:start position:0% so you know this kind of represents the state<00:25:29.260> of<00:25:29.410> the<00:25:29.500> work<00:25:29.680> today<00:25:30.450> and<00:25:32.040> you<00:25:33.040> know<00:25:33.130> I 00:25:33.330 --> 00:25:33.340 align:start position:0% state of the work today and you know I 00:25:33.340 --> 00:25:35.160 align:start position:0% state of the work today and you know I gear<00:25:33.640> I'm<00:25:33.970> excited<00:25:34.720> for<00:25:34.840> you<00:25:34.900> guys<00:25:34.930> to<00:25:35.110> check 00:25:35.160 --> 00:25:35.170 align:start position:0% gear I'm excited for you guys to check 00:25:35.170 --> 00:25:37.410 align:start position:0% gear I'm excited for you guys to check it<00:25:35.320> out<00:25:35.530> and<00:25:35.710> to<00:25:36.220> take<00:25:36.430> a<00:25:36.460> look<00:25:36.550> at<00:25:36.910> the<00:25:37.180> correct 00:25:37.410 --> 00:25:37.420 align:start position:0% it out and to take a look at the correct 00:25:37.420 --> 00:25:38.910 align:start position:0% it out and to take a look at the correct my<00:25:37.510> construction<00:25:37.870> consensus<00:25:38.350> protocols<00:25:38.710> and 00:25:38.910 --> 00:25:38.920 align:start position:0% my construction consensus protocols and 00:25:38.920 --> 00:25:41.670 align:start position:0% my construction consensus protocols and see<00:25:39.460> if<00:25:39.640> you<00:25:40.210> know<00:25:40.330> you<00:25:40.960> find<00:25:41.200> them<00:25:41.320> as<00:25:41.410> easy<00:25:41.560> to 00:25:41.670 --> 00:25:41.680 align:start position:0% see if you know you find them as easy to 00:25:41.680 --> 00:25:45.450 align:start position:0% see if you know you find them as easy to reason<00:25:41.830> about<00:25:41.980> as<00:25:42.250> I<00:25:42.340> do<00:25:44.070> and<00:25:45.070> then<00:25:45.190> we're<00:25:45.280> done 00:25:45.450 --> 00:25:45.460 align:start position:0% reason about as I do and then we're done 00:25:45.460 --> 00:25:48.780 align:start position:0% reason about as I do and then we're done right<00:25:45.840> so<00:25:46.840> so<00:25:47.110> Nate<00:25:47.350> Danny<00:25:47.700> worked<00:25:48.700> really 00:25:48.780 --> 00:25:48.790 align:start position:0% right so so Nate Danny worked really 00:25:48.790 --> 00:25:50.640 align:start position:0% right so so Nate Danny worked really hard<00:25:48.910> to<00:25:49.390> help<00:25:49.960> me 00:25:50.640 --> 00:25:50.650 align:start position:0% hard to help me 00:25:50.650 --> 00:25:52.260 align:start position:0% hard to help me get<00:25:51.190> ready<00:25:51.340> for<00:25:51.400> DEFCON<00:25:51.730> and<00:25:51.760> get<00:25:52.000> the<00:25:52.090> code 00:25:52.260 --> 00:25:52.270 align:start position:0% get ready for DEFCON and get the code 00:25:52.270 --> 00:25:54.210 align:start position:0% get ready for DEFCON and get the code published<00:25:52.690> get<00:25:52.840> the<00:25:52.900> papers<00:25:53.200> ready 00:25:54.210 --> 00:25:54.220 align:start position:0% published get the papers ready 00:25:54.220 --> 00:25:56.490 align:start position:0% published get the papers ready you<00:25:54.550> know<00:25:54.640> million<00:25:55.060> thanks<00:25:55.240> to<00:25:55.360> them<00:25:55.510> Carl<00:25:55.900> who 00:25:56.490 --> 00:25:56.500 align:start position:0% you know million thanks to them Carl who 00:25:56.500 --> 00:25:59.550 align:start position:0% you know million thanks to them Carl who just<00:25:56.710> spoke<00:25:56.940> last<00:25:57.940> was<00:25:58.720> the<00:25:58.780> first<00:25:59.020> person<00:25:59.470> to 00:25:59.550 --> 00:25:59.560 align:start position:0% just spoke last was the first person to 00:25:59.560 --> 00:26:01.410 align:start position:0% just spoke last was the first person to really<00:25:59.830> actually<00:26:00.490> the<00:26:00.580> first<00:26:00.670> developer<00:26:01.120> on 00:26:01.410 --> 00:26:01.420 align:start position:0% really actually the first developer on 00:26:01.420 --> 00:26:04.350 align:start position:0% really actually the first developer on the<00:26:01.710> correct<00:26:02.710> reconstruction<00:26:03.220> Casper<00:26:03.550> Greg 00:26:04.350 --> 00:26:04.360 align:start position:0% the correct reconstruction Casper Greg 00:26:04.360 --> 00:26:05.970 align:start position:0% the correct reconstruction Casper Greg Meredith<00:26:04.780> who<00:26:05.080> is<00:26:05.380> here<00:26:05.620> somewhere<00:26:05.800> but<00:26:05.950> I 00:26:05.970 --> 00:26:05.980 align:start position:0% Meredith who is here somewhere but I 00:26:05.980 --> 00:26:08.760 align:start position:0% Meredith who is here somewhere but I haven't<00:26:06.160> seen<00:26:06.280> him<00:26:06.370> yet<00:26:06.490> has<00:26:07.480> you<00:26:08.470> know<00:26:08.590> really 00:26:08.760 --> 00:26:08.770 align:start position:0% haven't seen him yet has you know really 00:26:08.770 --> 00:26:09.540 align:start position:0% haven't seen him yet has you know really introduced<00:26:09.130> me<00:26:09.190> to<00:26:09.220> the<00:26:09.340> correct<00:26:09.490> my 00:26:09.540 --> 00:26:09.550 align:start position:0% introduced me to the correct my 00:26:09.550 --> 00:26:10.440 align:start position:0% introduced me to the correct my construction<00:26:09.850> process 00:26:10.440 --> 00:26:10.450 align:start position:0% construction process 00:26:10.450 --> 00:26:12.420 align:start position:0% construction process and<00:26:10.540> help<00:26:10.960> me<00:26:11.050> really<00:26:11.260> like<00:26:11.410> formalize<00:26:12.100> and 00:26:12.420 --> 00:26:12.430 align:start position:0% and help me really like formalize and 00:26:12.430 --> 00:26:13.830 align:start position:0% and help me really like formalize and like<00:26:12.730> kind<00:26:13.030> of<00:26:13.060> take<00:26:13.270> this<00:26:13.360> mathematical 00:26:13.830 --> 00:26:13.840 align:start position:0% like kind of take this mathematical 00:26:13.840 --> 00:26:16.200 align:start position:0% like kind of take this mathematical approach<00:26:14.020> finally<00:26:14.830> metallic<00:26:15.370> you<00:26:15.730> know<00:26:15.850> has 00:26:16.200 --> 00:26:16.210 align:start position:0% approach finally metallic you know has 00:26:16.210 --> 00:26:18.450 align:start position:0% approach finally metallic you know has been<00:26:16.330> working<00:26:16.390> with<00:26:16.990> me<00:26:17.080> on<00:26:17.320> this<00:26:18.040> stuff<00:26:18.250> for 00:26:18.450 --> 00:26:18.460 align:start position:0% been working with me on this stuff for 00:26:18.460 --> 00:26:21.660 align:start position:0% been working with me on this stuff for years<00:26:18.700> like<00:26:18.910> three<00:26:19.150> years<00:26:19.180> more<00:26:20.130> you<00:26:21.130> know<00:26:21.250> so 00:26:21.660 --> 00:26:21.670 align:start position:0% years like three years more you know so 00:26:21.670 --> 00:26:22.860 align:start position:0% years like three years more you know so thanks<00:26:21.940> for<00:26:22.030> think<00:26:22.330> thanks<00:26:22.600> a<00:26:22.630> lot<00:26:22.750> for 00:26:22.860 --> 00:26:22.870 align:start position:0% thanks for think thanks a lot for 00:26:22.870 --> 00:26:29.930 align:start position:0% thanks for think thanks a lot for everyone<00:26:23.020> and<00:26:23.290> thanks<00:26:24.070> you<00:26:24.100> for<00:26:24.250> listening 00:26:29.930 --> 00:26:29.940 align:start position:0% 00:26:29.940 --> 00:26:33.140 align:start position:0% [Applause] 00:26:33.140 --> 00:26:33.150 align:start position:0% [Applause] 00:26:33.150 --> 00:26:40.729 align:start position:0% [Applause] [Music]