Pop Chunhapanya

> I miss the method which translates custody group into subnet, I think we need it or I don't understand an idea a bit. So instead of this The idea...

I think this is ready to be merged?

@jimmygchen Sorry for the late reply. I would like to compare this PR to attestation subnets and attestation committees. In the phase0 spec, we decoupled subnets from committees as well....

@jimmygchen Thank you for the feedback. > I'm not sure we need to design this with increasing to over 128 subnets in mind at this stage, or if there's any...

It seems that everyone in this PR is already convinced that this PR is useful. So it means it's ready to be merged?

Since `verify_and_notify_new_payload` is an EL function rather than the CL, I don't think what is specified in the consensus-specs has to be exactly the same as what is done in...

Why not just completely remove `activation_eligibility_epoch` instead of leaving it unset? It doesn't seem to beak any backward-compatibility, right? If it breaks, it shouldn't break more than just unsetting it,...

I'm not sure if `WithIgnoreIneed` should be added to the upstream repo, but I will implement it in my local repo anyway because I need it for my simulations in...

> As part of the discussions surrounding EIP-7594 (peerdas), it was highlighted that during sampling and/or data requests, the sampler does not have timing information for when a samplee will...

Is there any rationale of having the timeouts in the first place? Having two constants, `TTFB_TIMEOUT` and `RESP_TIMEOUT`, rather than just a single constant like `TIMEOUT` quite surprised me.