Binance Research just published a paper called "Web3 Social: Road to Mass Adoption".
Many players are discussed including DeSo, Lens, and Farcaster. On DeSo, it says:
"[...] building custom L1s is a double-edged sword. While it allows fast transactions and low content storage costs, it loses valuable connection with the existing dApp and identity ecosystem on chains like Ethereum."
Their opinion, like mine, is that modularity and composability will be key factors in the success of Web3 Social.
Guess I'm not the only one who believes that Lens checks all the boxes.
i've spent enough time on deso's documentation and code to conclude the following:
1) @nader is exaggerating by omission, and your initial analysis of 30 tps is fair. 2) what deso has to offer is far beyond code optimizations that can be addressed when the load justifies it
just to be clear, the current processing power of deso is more than enough to power things through. we're really talking about hypotheticals of incredible widespread global adoption here.
Nader says that DeSo can handle 45k posts per second.
⚠️ Don't trust, verify.
We already know that the actual blockchain can only confirm around 30 transactions per second. So maybe Nader is talking about the mempool?
Fine, let's bowl with bumpers on 🎳:
Each node has a mempool that can store up to 250 MB worth of transactions.¹ New transactions are rejected once it is full.²
Using 218 bytes as the average transaction size (from DeSo docs)³, at 45k TPS roughly 9.81 MB of transactions are added per second:
45,000 * 218 bytes = 9,810,000 bytes
So we divide the max mempool size by 9.81:
250 MB / 9.81 MB/s = 25.48 seconds
That means that at a rate of 45k posts per second, DeSo would start rejecting new transactions after only 26 seconds.
@nader Where is my misunderstanding?
¹ github.com/deso-protocol/core/blob/d6458b608d1bf1ca9be7dea2f72c0ed244421049/lib/mempool.go#L36
² github.com/deso-protocol/core/blob/d6458b608d1bf1ca9be7dea2f72c0ed244421049/lib/mempool.go#L829
³ docs.deso.org/about-deso-chain/readme#a-simplified-scaling-roadmap
--
What is a mempool:
node.deso.org/posts/f26fd17550bc933b3938eea8ac15d109a2426b457d9564d72480edb711765f3e
"We already know that the actual blockchain can only confirm around 30 transactions per second". Says who?
node.deso.org/posts/74587eeade495b12f4c928da0e3cf854122ca382a130bc16234ec6d484ed9a28
Binance Research just published a paper called "Web3 Social: Road to Mass Adoption".
Many players are discussed including DeSo, Lens, and Farcaster. On DeSo, it says:
"[...] building custom L1s is a double-edged sword. While it allows fast transactions and low content storage costs, it loses valuable connection with the existing dApp and identity ecosystem on chains like Ethereum."
Their opinion, like mine, is that modularity and composability will be key factors in the success of Web3 Social.
Guess I'm not the only one who believes that Lens checks all the boxes.
research.binance.com/static/pdf/Web3_Social_ Road_to_Mass_Adoption.pdf
and old mentor of mine used to say: trust, but verify :)
Is there a way to get this to scale better? Or is this a permanent feature of the blockchain?
Is this where the missing transactions have gone?
such a bummer... i was led to believe that all media is stored on block.
now i see @nader suggests it will be delivered before end of year
.... but what then?
what about one video 251 MB stored on chain ?
i've spent enough time on deso's documentation and code to conclude the following:
1) @nader is exaggerating by omission, and your initial analysis of 30 tps is fair.
2) what deso has to offer is far beyond code optimizations that can be addressed when the load justifies it
just to be clear, the current processing power of deso is more than enough to power things through. we're really talking about hypotheticals of incredible widespread global adoption here.
we have time to figure those things out
#Nader
member.cash/t/Nader