Read on Twitter
Next EIP-1559 Implementers' call has been scheduled 
My goals for the call are to (hopefully
) get to a decision about how we bundle/don't bundle 1559 & escalator fees, and get agreement on a plan for testing/analyzing the change.
The call will be streamed
! https://twitter.com/TimBeiko/status/1266353721975373825
My goals for the call are to (hopefully
) get to a decision about how we bundle/don't bundle 1559 & escalator fees, and get agreement on a plan for testing/analyzing the change. The call will be streamed
! https://twitter.com/TimBeiko/status/1266353721975373825
Agenda 
: https://github.com/ethereum/pm/issues/184
Notice the time is *19*:00 UTC (vs. the usual 14:00)
See you all on June 25th
: https://github.com/ethereum/pm/issues/184Notice the time is *19*:00 UTC (vs. the usual 14:00)
See you all on June 25th
Fee Market Implementers' Call #3 in 20 minutes!
Packed agenda today, with the goal of hopefully coming to agreement on a combination of EIP-1559, escalator fees and possibly EIP-2718.
Agenda: https://github.com/ethereum/pm/issues/184
Stream:
Packed agenda today, with the goal of hopefully coming to agreement on a combination of EIP-1559, escalator fees and possibly EIP-2718.
Agenda: https://github.com/ethereum/pm/issues/184
Stream:
I won't be live-tweeting as I'm facilitating the call, but I'll post the highlights of the call here once it's over 
Apologies, we're having technical issues with the stream. We won't live-stream the call but will upload the recording to the YT channel later today. My bad!
Call over, here are the takeaways for the main agenda items 
To Bundle or Not To Bundle - There was a lot of discussion of the impact bundling EIP-1559 and Escalator Fees would have on implementation complexity, user privacy, further analyses of the change, and, most importantly, the timeline at which the change could be deployed.
Everyone on the call seemed to have a weak to strong preference for ideally bundling things, but not at the expense of introducing large delays in the work.
Also, there are open questions, raised by this piece ( https://twitter.com/hasufl/status/1275809861692002304) about whether Escalator fees are better done outside the protocol.
Next step: I'll follow up with champions and implementers of the EIP to work through the technical details of what bundling the EIPs would entail, and we'll share our findings with the community 
Testing, Testing, TestingEIP-1559/Escalator fees are a pretty sweeping change and we want to be sure we test it accordingly. Again, there was a lot of discussion on the call about how to best test things, ideally incrementally.
The first thing we'd like to figure out is whether 20M gas blocks (25, now?) are even viable! To do that, the simplest way seems to be to create another Ephemeral Testnet (like the Berlin YOLO one) running Besu & Geth's current implementations of 1559.
Once the Ephemeral testnet is up, it should be possible to write scripts that send a ton of transactions through it to see how it reacts. Assuming it survives, we can then try more sophisticated transaction-sending patterns (i.e. spikes of usage).
Next step here is for @JHancock, the Besu and the Vulcanize teams to get this up and running.
Once that is done, we'll likely want more rigorous analyses of the EIP specification, but at least we will have shown that 1559 _can_ work on a live network. Obviously, the EIP will then go through the normal AllCoreDevs process, including being deployed on testnets.
There was some conversation about potentially using existing testnets to do this (and perhaps lowering the block gas limit), but @dannyryan mentioned we definitely don't want to break whatever testnet Reddit is using, so we'll err on the safe side 
EIP-2718 ( https://eips.ethereum.org/EIPS/eip-2718 )This EIP was proposed as a way to make it easier to add new transaction types to Ethereum. If we were to combine 1559 & Escalator fees, it would make implementation easier, and IIRC the folks working on Account Abstraction would also want it
Again, the main concern with this EIP is whether using it will slow things down. To gauge other core devs' feelings about it @lightclients will bring it up on this week's AllCoreDevs call. Based on feedback there, we'll see if we want to have it as a dependency for 1559.
Last but not least, funding While a few of us working on 1559, myself included, are able to do this as part of our day job (thanks @PegaSysEng!), it's not true of everyone working on it, and we don't have funds to pay for, say, a formal audit of the EIP.
. @JHancock proposed putting together a @gitcoin grant to help compensate some of the work for the implementation. It would not only provide funds, but could provide a strong signal to core devs if many people are contributing funds to the grant (even small amounts).
Rough rules we'd use to allocate the funds are that people who *don't* want to receive funds _and_ are familiar with the work could act as administrators of a multisig, which could pay out things like non-Besu implementations, an audit, etc. Feedback welcome!
Next step here is for me to put together a draft grant and share it in the 1559 discord and Cat Herders channels to get feedback. Hopefully we have something up by the end of the week 
... and that was it! Again, apologies we couldn't livestream. I've just uploaded the recording to Youtube for those interested in watching:
Thanks everyone who attended
Thanks everyone who attended
