block withholding attacks (BWA):kind of attack when validators refuse to publish new blocks.

#Done
#research

rollup data availibility

اکی ما تراکنش ها رو به صورت خلاصه شده میزاریم توی لایه ۱ و سر در دسترس پذیر بودن دیتا ها میگیم هر وقت کسی بخواد میتونه بیاد با استفاده از این تراکنش ها و تراکنش های وارد شده به کانترکتش به اون استیت نهایی برسه

سوالی که برام پیش اومده اینه :

یه ادرسی دریافتی داشته تو لایه ۲ و این ادرس تو لایه ۱ اصلا فعال نبوده
خب اون ادرس هم به صورت خلاصه شده توی لایه ۱ قرار میگیره یا نه ؟

اگه بگیره چجوری اون داستان دسترس پذیری دیتا رو هندل میکنن و …
چک کن ببین خلاصه کردن ادرس اصلا روش رایجی هستش یانه و ...

#do
#research

Parallel computing in consensus
(Julia)

#research
#Do

a blockchain that drop its history in every 2 years and provide zk proof for validity of the last state instead of keeping history

* search about attack vectors and weakness
* pros and cons ?

Efficient Lossless Compression of Trees and Graphs

https://users.cs.duke.edu/~reif/paper/chen/graph/graph.pdf

#Do

#research
#Do

EVMX (384 bit)

https://vitalik.ca/general/2021/06/18/verkle.html



So what is this little extra that we need as a proof? To understand that, we first need to circle back to one key detail: the hash function used to compute an inner node from its children is not a regular hash. Instead, it's a vector commitment.

A vector commitment scheme is a special type of hash function, hashing a list . But vector commitments have the special property that for a commitment and a value , it's possible to make a short proof that is the commitment to some list where the value at the i'th position is . In a Verkle proof, this short proof replaces the function of the sister nodes in a Merkle Patricia proof, giving the verifier confidence that a child node really is the child at the given position of its parent node.

ZK-SNARKs are hard because the verifier needs to somehow check millions of steps in a computation, without doing a piece of work to check each individual step directly (as that would take too long).

We get around this by encoding the computation into polynomials.

A single polynomial can contain an unboundedly large amount of information, and a single polynomial expression (eg. ) can "stand in" for an unboundedly large number of equations between numbers.

If you can verify the equation with polynomials, you are implicitly verifying all of the number equations (replace with any actual x-coordinate) simultaneously.

We use a special type of "hash" of a polynomial, called a polynomial commitment, to allow us to actually verify the equation between polynomials in a very short amount of time, even if the underlying polynomials are very large

there are many (50+) variants of SNARKs

#Do
#research