Federico Carrone has a team that is working on this: https://twitter.com/federicocarrone/status/1585304643344560128

Trey Del Bonis (who wrote this: https://tr3y.io/articles/crypto/bitcoin-zk-rollups.html) is involved as well.

I'm not aware of anyone else working on it, but I hope to see more join. There are several different ways to build rollups on bitcoin, would be great to see different approaches tried or at least fully researched so we have some hard data about what approach would work best technically and also be most politically feasible.

For the validity rollups report, I was time-boxed, I had four months to produce the report. I already had a lot of background knowledge on the subject, so I wasn't starting from zero, which helped. The time was split about 2.5 months on research and 1.5 months on writing/editing. I had a rough idea ahead of time how much time I wanted to devote to writing/editing and made sure that my research schedule did not cut into that. That's usually how my process goes when I am time-boxed: have a deadline, estimate time needed for writing/editing, and the balance is time for research.

Generally speaking what I do is I research and think until I think I understand something well enough to write about it and then I write about it. Sometimes the writing takes several iterations as I work through my thoughts. During the research process I also do a lot of learning/thinking "in public", mainly on Twitter, which helps me refine my thoughts as other people give feedback or share more info.

thanks! I am still doing research on rollups and related protocols like validia chains. I also try to stay up to date with bitcoin privacy and crosschain research broadly speaking.

I am also a full-time Sovryn contributor, where my main focus is Zero: https://github.com/distributedcollective/zero but I help with a bunch of other projects in that ecosystem, too.

so far the response has been mainly positive or neutral. a couple of negative comments but nothing substantial, stuff like "this will never happen, bitcoin doesn't change" (which is obviously wrong, since bitcoin changed to adopt taproot/schnorr just over a year ago). given that validity rollups align so well with bitcoin values like trust minimization, decentralized validation, open and permissionless access, etc, I was not really surprised by the overwhelmingly positive response.

this wasn't a direct response to my report but ruben somsen has an interesting theory that if bitcoin full nodes switch to what I call "proof-sync" clients like zerosync then rollups would not be safe to use due to data unavailability concerns: https://twitter.com/SomsenRuben/status/1566709837395894272

I have some thoughts about this but I'm waiting for a longer-form exposition of this theory before I form a solid opinion and respond to it.

it's hard to pick "most surprising", so I'll mention a few things that stand out as having surprised me:

My main focus is Zero: https://github.com/distributedcollective/zero but I help with a bunch of other projects in the ecosystem, too. like right now we are working on integrating Zero with Mynt, a bitcoin-backed stablecoin aggregator.

the actual work I do touches on a number of different skills/activities. mostly a mix of product and project management, testing, ui/ux writing, and documentation.

this past year a lot of learning came from two efforts: research for my validity rollups report, and online resources I've used to learn more about product management.

in the case of the rollups report, I will defer to my acknowledgements: https://bitcoinrollups.org/#acknowledgements

it's hard to pick someone I learned the most from there. they are all sharp thinkers who have a lot of deep knowledge to share in their respective areas of expertise. I also learned a lot going through the work cited in the references section: https://bitcoinrollups.org/#references

in the case of product management, most of what I've learned has come from taking reforge classes. they create high-quality educational content relevant to a number of different professions, probably most relevant to people working at tech companies but maybe other industries too. I recommend anyone looking to level-up their skills to check out the courses and free content they have to offer: https://www.reforge.com/programs

And as a secondary question, how does a validity rollup compare to mimblewimble?

The main difference is certainty over data availability, which ultimately boils down to certainty of ownership security (i.e. that if you own coins, no one else can move them without your private key and you can transfer the coins at-will as long as you pay a market-rate fee).

The merkle root is not enough for a holder of coins on the sidechain to unilterally produce a proof and withdraw their coins, they need data about the entire state of the sidechain to be able to prove that they own some coins in the current canonical state. And there's a problem that can occur called a "data withholding fault" where sidechain block producers create a block, commit the hash to the parent chain, but don't actually broadcast the block. Then users are stuck, because they don't have enough data to produce a proof that convinces parent chain full nodes they own some coins in the current canonical sidechain state. Maybe they have data to show that they owned coins in the past, but they could have spent the coins in the block that wasn't published -- it's uncertain without the proof, so parent chain full nodes won't let those users withdraw their coins back to the parent chain. And since the block producers aren't publishing those blocks, even if the user could get a transfer tx confirmed in a sidechain block, no one would know if the tx was confirmed since the block data isn't available. So then the sidechain coins are worthless. If this was actually an attack by sidechain block producers and not just a technical failure, at this point they can extort users ("we will give you the block data if you transfer us x% of your coins" or something like that).

Rollups solve the data availability problem by publishing rollup block data (or a compressed form of it that still has enough information to reconstruct the rollup state) inside of the parent chain blocks. This, combined with the use of validity proofs to ensure the correctness of rollup state transitions, makes the rollup ownership security equal to the parent chain ownership security, since the data availability guarantees are equivalent and coins cannot be transferred on the rollup without cryptographic proof of correctness.

scrambled eggs, hash browns, and sausages with a dab of butter and bourbon maple syrup.

I am one of a small handful of people who has been writing about crosschain bitcoin protocols for about ten years, and it's a recurring theme in a lot of my work. Even today this topic is pretty niche so that is probably a distinctive trait.

can you give an example?