22. Alternative Consensus Mechanisms

MIT MAS.S62 Cryptocurrency Engineering and Design, Spring 2018
Instructor: Tadge Dryja
View the complete course: ocw.mit.edu/MAS-S62S18
KZbin Playlist: • MIT MAS.S62 Cryptocurr...
Alternate consensus including unique node lists, proof of stake and space, variants, directed acyclic graphs, and proof of idle.
License: Creative Commons BY-NC-SA
More information at ocw.mit.edu/terms
More courses at ocw.mit.edu

[@mitocw]

Please note that lectures 9, 19, 20, and 21 are not available.

[@IrakliSafareli]

:(((

[@kristiantrujillo5443]

why

[@andso7068]

@@kristiantrujillo5443 No permission from speakers.

[@oldeucryptoboi]

A real bummer: They bought guest speakers for these 4 videos and couldn't get them to sign a release form? What a shame. Sharon Goldberg contributed to Bitcoin and Ethereum. I was looking forward seeing her talking about peer to peer network.

[@leeroyduncan4610]

What in the world freeze funds ....what's up with this misinformation

[@erichanintokyo]

What in the world? Freeze funds? Modified Code? Fire this guy.

[@KryptoPionier]

Hello Eri, same thought during my research for a ripple consensus mechanism video. Not sure if this guy is serious here?! Is it maybe outdated? Best Nicolas

[@newbraveworld9733]

You're surprised that Ripple is a centralized scam project?

[@lyonrockdigital6165]

You can't freeze XRP. An issuer of an asset on the XRPL can freeze their IOU, but not XRP

[@FF-bf5oq]

So basically Tadge snitched on Ripple to Gary Gensler here. LOL

[@deepthoughts8228]

😂😂

[@shymaaarafat1342]

I think one should get back to anay Eth-2 in 2021 with new insight after learning about these attacks/threats and what did they change/add in 3yrs to avoid/mitigate them before they start applying it

[@LOGICZOMBIE]

GREAT WORK

[@RiccardoVincelli]

Great stake!😂

[@cauebraga]

1:14:40 very interesting view

[@davidreynolds9649]

Is he just confusing mining with full node?

[@DavidVaughan00]

Trying to build a proof-of-work or proof-of-work-like system in an attempt to achieve "better economy of scale" seems totally absurd to me. Isn't the act of wasting resources inherent to the functioning of these systems? I don't see how the concept of "economy of scale" even applies here. Take proof-of-work (POW) vs proof-of-space (POS) for example. Say you're mining in a POW-based currency and you have infrastructure such that, on average right now, it will cost you $X in electricity and cpu-purchases in order to make $(X + Y) worth of the currency when you mine a coin, making $Y in profit. Well, if instead you're mining in a POS-based currency, despite the fact that now you're purchasing hard drives and disk-IO (or whatever) instead of CPUs and compute, you could just as well spend $X on *those* resources in order to achieve $(X + Y) worth of money in *that* currency. Maybe it takes tons more hard drives that it would've in CPUs, but that's irrelevant; in the end you're spending the same amount of resources for the same amount of profit. You might say that different currencies might allow miners to take different amounts of profits for the same input amount, but I think that would be due to the the perception of value of the currency itself, probably not so much to do with the proof system. If you think POS-type work is "less costly" than POW-type work, then really that just means the miners will put in more and more resources until it *is* just as costly. Regardless, if you wanted a proof system that was less costly, why wouldn't you just take the existing POW system found in bitcoin, but lower the difficulty so that coins get mined faster? I think it would truly be equivalent. And it's easy to see, in that scenario miners would just mine more and harder until the production efficiency slows to exactly where it started. And if personal resources aren't the bottleneck and instead it's something else (for example maybe it's the case that the hard drive manufacturing industry just cannot build them fast enough to keep up with demand from miners while the cpu manufacturing industry can [I have no idea whether this might actually be true; just an example]), then we're in a situation where miners will increase their resources until all of the top miners are equally powerful, and so have equal chances at the next block. If we wanted this, why not just make it an equal-chances lottery instead of work-based? Tl;dr: The entire mechanism of these proof systems is that whoever can spend the most resources has the highest chances to get the next reward. Changing the unit of the spent resource does nothing; in the end the equivalent amount of each resource will be spent, and so it'll be equally wasteful. Correct me if I'm wrong pls but that's how it seems to me.

[@DavidVaughan00]

Immediately after writing all this I think of two answers to the question: -- Maybe once a miner hits their resource limit, this would produce fewer negative externalities if spent on one resource than another (ie maybe buying $1M worth of hard drive work contributes less to global warming than $1M worth of CPU work). -- Maybe the work achieved in one proof system can actually be considered more useful than in another, outside of crypto mining (ie maybe POS-mining helps drive production of better hard drives which can then be used by consumers, while maybe POW-mining drives production of useless chips that are only good for crypto mining). Is this the whole idea?

[@georgionic]

I think it’s more about how much pollution PoW is generating. Paying $10,000 for hard drives releases less CO2 than paying $10,000 for electricity does.

[@cicero1930]

Georgio Nicolas we need to come up with ways for bitcoin to pollute more not less

[@consistencyye7178]

@@cicero1930 innovative solar energy

[@consistencyye7178]

@@cicero1930 to the point to power bitcoin farms instead of coal

[@djlenton200804]

Could do with getting your facts right!

[@zes3813]

no such thing as nerx about it, usex etc techx etc s ok, not nerx, idts, also no such thing as madx or disagreex or stupix or not, just u r stupix, cepux etc, think, say , do any nmw an dany s perfect