write jellyfish book

[zhenfeizhang]

something similar to https://github.com/zcash/halo2/tree/main/book would be really helpful

[chancharles92]

Hi @zhenfeizhang @alxiong , how about assigning this issue to @mrain so that he can be more familiar with the codebase along the process?

[mrain]

Is it a better way to first enhance the rust doc? The codebase may need better commentations.

[chancharles92]

Is it a better way to first enhance the rust doc? The codebase may need better commentations.

Sounds good

[alxiong]

Sounds good, the only concern would be, we can improve the documentation more strategically given that our constraint system API (i.e. jf_plonk::circuit::*) might be updated significantly. (contingent on our investigation into Halo2)

I'd say start with jf_plonk::proof_system::*which is likely something we will keep. WDYT? @chancharles92 @mrain

(and, until we have a clear comparison with halo2, plonky2 and zk-garage's API design, our code base would be subject to major changes, and I personally would prefer to write a rustbook after we have a more stable plan.)

[chancharles92]

Sounds good, the only concern would be, we can improve the documentation more strategically given that our constraint system API (i.e. jf_plonk::circuit::*) might be updated significantly. (contingent on our investigation into Halo2)

I'd say start with jf_plonk::proof_system::*which is likely something we will keep. WDYT? @chancharles92 @mrain

(and, until we have a clear comparison with halo2, plonky2 and zk-garage's API design, our code base would be subject to major changes, and I personally would prefer to write a rustbook after we have a more stable plan.)

Sounds good, or @mrain can work on other small issues first (e.g. one of #89 #62 #81), and continue on this after finalizing the circuit front-end.

[zhenfeizhang]

Sounds good. It is better to improve the code APIs first before working on this ticket.