IOHK | Dr. Peter Gaži, presenting Ouroboros at MIT.

[@IohkIo]

New and improved with all slides by popular request!

[@Rogzy2]

Thanks god !! Great job

[@davidnachman6398]

Thank you for all the new video updates. If I make a suggestion, the background music/sound in the video is distracting. It would be more pleasant to listen to the lectures without this added layer of music/sound.

[@jakubstec3706]

Hi guys :) On question for you. Cardano has slot-leaders; how resistant network will be to DDoS attacks?

[@claudiusarnellius2465]

I was there for this presentation. Excelsior!

[@ChanelCoco555]

At 33:24: let's say a malicious node, when it's slot leader, feeds a block to half the network, and nothing to the other half. Then let's say the next slot leader is an honest node that builds off the malicious block. Doesn't this mean that half the network will see this new block as coming from a block that doesn't exist? I know it'll contain the hash of the previous block, but one can imagine this happening beyond one "missing" block. In general, how would they catch up? On a related note, at 21:54, it says that adversarial parties can send messages to "arbitrary subsets" of the network -- but how can they guarantee this unless they know the network topology? If the network is a greater-than-2-edge connected graph, shouldn't it be impossible for a single node to, say, split the nodes in half and send only messages to one half?

[@Oskimaa]

I've been wondering for a while why not use network ping as source for randomness... I.e. one node pings another node and uses the hash of that ping when selecting the next block generator. Network traffic changes all the time and thus ping is effectively random.

[@kingofgimp]

PROBABLY because given a set of physical parameters, one could narrow down the "randomness" of one's ping across a fairly common normal distribution. True random number generators are a hard problem because there are nearly no physical phenomena which are NOT deterministic or stochasitically deterministic... That said, neat idea!

[@larposan]

Because it would be impossible for other nodes to verify what the result of the ping was, and thus adversarial node would just cheat and say the ping was such that it gave him another slot as a leader.

[@cool-wl1ex]

Local randomness is not verifiable across the network.

[@davidealy6356]

Appreciate this presentation. More info please!

[@BjornWiese]

what happens when adversary has the majority of stake?? And isn't that going to happen at some point when staking rewards are being proportionally paid??

[@wims5889]

This is the beginning. We have not heard of how quantum computing will break this in milliseconds and alternative solutions💃

[@roadto23bitcoin68]

Awesome presentation! I feel like I've learned alot and I'm ecstatic to start staking my ADA!

[@amalnathsathyan6029]

You might be a millionaire by now !😊👍

[@0xD201d]

I love IOHK's style.

[@elindauer]

Crypto is a technical marvel. It's amazing how much human ingenuity goes into creating these assets of all humanity.

[@THEHADIDA]

Wow, 52:00 pretty much sums it all up

[@HPRaceDevelopment]

This isn't an easy watch for a laymen like me. Could this be improved?? I've watched every youtube video on "probably safe" and find them a challengin watch

[@slomnim]

so underviewed

[@zacheryabraham6083]

will ouroboro rise? why does it seem died? is ouroboro a good investment?

[@peny79133]

카르다노 가즈아~~

[@paudelamarianaoancea3761]

boooooooooooooooooooooooom