At Avail, we want to accelerate the unification of web3. Glad to have the fantastic set of backers to help us achieve it.
Read more about the Avail Thesis:
The day is here! We’re unfolding the cards!
Avail raises $27m, led by
@foundersfund
and
@dragonfly_xyz
to accelerate the unification of web3!
With this, we’ll build the Avail trinity - a three-phase roadmap aimed at unifying multiple ecosystems in web3.
Consisting:
• Avail…
The unification drop was all about gratitude. Avail has always been built on the shoulders of giants. It was built by, for and supported by the community. And this was our chance to recognise them. We did not want to take it lightly and wanted to be as thorough as possible. I…
I'm happy to see 35K+ unique LCs running in last hour. But please don't DDoS our RPC with tens of millions of requests.
Expect degraded performance while we fight this wave off.
We have i̵n̵t̵e̵r̵n̵a̵l̵l̵y community aligned 100,000+ active light clients.
A one-of-its-kind data availability sampling peer-to-peer overlay network of 100K stable LCs now operating on top of our testnet.
Thank you everyone who accepted this challenge.
- Use Cosmos SDK to build a rollup on Avail.
- IBC between two rollups on Avail.
- IBC between an Avail rollup and another Cosmos chain.
Great work by the
@vitwit_
team and ofc
@RollkitDev
.
PS: IBC works here but still needs a lot of work to be secure.
Hello World 🌍,
We're excited to share educational videos on setting up a Cosmos chain as a Rollup with
@RollkitDev
and
@AvailProject
for DA 🛠️. Check out how we demonstrate
#IBC
connections between rollup chains too.
The
@AvailProject
team is in London this week. Let's catch up.
13 (Wed): Oxford Blockchain Conference & StarkCity London
14 (Thur): HER Con & Pragma London
15-17 (Fri-Sun): EthGlobal London
Attending a few other side events as well. See you around!
PS: Some good hackathon…
With EIP4844, if you want to submit 10 KB data as a blob, you still need to consume 128 KB blob. In fact, for anything <128 KB, you pay for the entire blob.
If you have more than 128 KB, you pay for multiple blobs. But there is no guarantee of them being in the same block.
On…
The Miden VM and rollup design are one of the best IMO, and I keep referring to them in various discussions. Have a watch if you want to know why it is special.
Discussions with
@bobbinth
are always a learning experience. This was no different.
Wake up anon, a new video from our whiteboard series just dropped. And this time, it is with
@0xPolygonMiden
💁
If you can't spare the time for the full video, here's a quick thread on how Miden is building a new VM built for the ZK future. 🧵
The
@AvailProject
light client challenge starts next week. Exact dates TBD.
Why we want you to run it
- The LC is a fruit of months of development. This is a crucial stress test.
- We are expecting tens of thousands of LCs, all with diff network configs, archs, etc - something…
Today we open-source the prototype of OpEVM, an EVM-compatible sovereign optimistic rollup SDK, with
- working fraud proofs
- decentralized sequencer set
- permissionless watchtowers
- slashing
On top of
@AvailProject
, i.e., w/o need for a smart contract on the base layer 🧵
1/3 We’re launching a new prototype of the OpEVM SDK.
Built as a collaborative project between Avail,
@eiger_co
and
@equilibrium_co
from the Equilibrium Group, this open-sourced tech is a promising alternative to traditional smart contract-based rollups.
The Dencun upgrade hits Ethereum, enabling data blobs as part of EIP4844 (protodanksharding). We were waiting for this for a long time and really happy it finally hits mainnet.
Now that rollups can post blobs on Ethereum, a new era ushers upon us!
Happy Dencun Day!
The Dencun upgrade going live is an awesome feat of engineering and systems coordination on a running network that secures billions of dollars of economic activity.
Huge congratulations to everyone on the dev teams and the community. It is a step that will go down in history as…
Proof of valid execution can take many forms. ZK & fraud proofs are leading the way, but always felt TEE is another solid tech to explore (all with diff trust assumptions).
@Ternoa_
is showing the way.
Blockchains should drop the charade and stop hiding behind social slashing whenever they cannot enforce a rule within the protocol. DA layers are no different.
Validiums using attestation bridges from external DA don't benefit from using a specialised DA layer. Period. 🧵
Can someone explain (or point me to the actual code/doc) to me how can light clients slash a full node in Celestia? It seems impossible.
What if all the full nodes just not respond to the LC requests? Instead, they just sign the header file and post it to Ethereum through…
New pod episode out with
@AvailProject
co-founder
@prabalbanerjee
!
In this ep we unpack:
- What Avail is building
- Validity proofs vs fraud proofs
- Decentralized block building
- Avail's DA solution design decisions
- Safety vs liveness tradeoffs
Links below ↓
@AvailProject
team will be in Bangalore this week. If you are around and want to talk, let us know.
Roughly, we will be at Interfaces, Protocols Conf, Founder's House, Polygon Connect, Polkadot Pulse, EthIndia and few other places.
See you there.
Clash of Nodes mark our final lap towards the mainnet. We will slowly reveal the stages and challenges as the weeks pass by, to allow all types of participants to interact with the network, find bugs, help us test stability and climb the leaderboard ... but more on that later 🤐
It's here - Avail's Clash of Nodes incentivized testnet is officially up and running! 🌟
This is a major step towards Mainnet, with a chance for validators, light clients, and everyone to test our network, earn points, and more. Ready for a journey?
Here's what's ahead...…
Don't mistake our silence for non-readiness. We believe in heads-down building and shipping. We like to walk the talk and our partners already know that.
Also, this list is far from exhaustive. Many many other things lined up in the next few weeks and months.
I got the impression while at Token 2049 that not many people know how far
@AvailProject
is in terms of production readiness and implementation complexity
Some stats:
- started within
@0xPolygon
in late 2020 and spun out by me and
@prabalbanerjee
in Mar 23
- mainnet in Q1 2024…
As I'm leaving Istanbul, the last week around
@EFDevconnect
has been awesome. Teşekkürler Türkiye. A quick recap:
Sunday was full of anticipation and nervousness. We at
@AvailProject
were hosting our maiden event, the first of the Hot Take Series. The day was full of catching...
If you're still having doubts about performance of data availability sampling (DAS) and KZG opening checks, just try out our web light client:
But here is what it took to build it, why it is WIP and what lies ahead 🧵
How easy is it to run an Avail light client?
Like
@thunder
mentioned on stage today, it's as simple as clicking a link.
Run yours now at and tell us how it works!
There are largely two approaches to DAS
1. DAS by LCs participating in consensus
2. DAS by LCs outside to consensus
The design/arch/economic challenges for the two are diff. 🧵
DAS is the critical dependency in Ethereum's long-run scaling competitiveness and (worryingly) the least specified part of the roadmap.
We know how to do DAS in the dPoS BFT setting (cf Celestia), but not afaik in the 100k validator setting.
Well, if you have been remotely following the L2 space and attending
@EFDevconnect
, you should not miss this panel.
Six gigabrains are going to be on-stage to discuss the L2 landscape.
And check out the many other talks and spicy panels on the Battle of the L2s.
Cya on 13th 👋
📢Attention Devconnect participants
Here is a panel you do not want to miss.
Come learn about the future of L2 and L3 scaling initiatives being built for developers
The L2 Landscape panel will be hosted by
@bkiepuszewski
from
@l2beat
with:
@ben_chain
from
@optimismFND
…
🎙️Last week,
@annarrose
chatted with
@prabalbanerjee
, co-founder of Avail. They covered Prabal’s career from working at
@0xPolygonLabs
to kicking off
@AvailProject
, they revisit DA + then get into the weeds about edge-cases in a DA-based stack. Enjoy!
Build on an L2 of your choice, Avail DA has your data availability needs covered.
We’re excited to announce that developers building rollups on
@0xPolygon
,
@arbitrum
,
@Optimism
,
@StarkWare
and
@zksync
can now utilise Avail DA.
Technical deep dives coming soon! Meanwhile, you…
After the debates about "Ethereum alignment" are over..
After the fights about who copied whose open-source code is over..
Let's hope to build stuff which gets used by average people. Let's align to use each others work freely in order to make the community successful.
At
@AvailProject
we are strong believers of the free and open-source software thesis, and as such, most of the work we push out has permissive and/or copyleft licensing.
We recently launched our version of the OP Stack with Avail as the DA layer, and we highly encourage…
Some will keep on doubting the possibility of appchains and we will keep on showing their potential.
If you're surprised by these numbers, wait till we release the total amount of transactions we handled.
Let thousands of rollups bloom.
🚀 Exciting news! Karnot x Avail hits a new high: 3100 forks & 2000+ app chains deployed!
We're overwhelmed by the support & participation, giving us great insight for future growth. Read more 👇
DA deep dive w avail cofounder: professor
@prabalbanerjee
on
@cryptotownpod
🧑🏻🏫
- a cryptology phd who loves fifa
- the art of being a researcher
- how avail cofounder writes code
- state bloat: the biggest prob evm chains will face 🚨
- analysing diff rollup stacks
- everything…
I think
@tkstanczak
mentioned during his
@ETHGlobal
London keynote, that it doesn't matter what you build during the hackathon, but it matters if you continue building it after the hackathon...
And I cannot agree more.
Hackathons are just a great way to:
- switch from being a dev to being a user/consumer of your chain
- zoom out to see how protocol is getting used vs how it was intended to be
- know the gaps on your docs/version incompatibilities/testnet vs latest
- gain insights on tooling
We are thrilled to announce the official launch of Avail's Kate Testnet! This major milestone brings us one step closer to the mainnet release. Check out our blog to learn about it!
Off late, there has been a lot of discussion about rollup sequencer design [or rollup orientations].
The 3 popular ones:
1. Centralized sequencer
2. Based rollup
3. Shared sequencer (not discussing today)
I feel people confuse 1 & 2 when they talk of properties of a rollup. 🧵
Allow us to reintroduce ourselves! 📢
We’re Avail, the essential base layer for modern blockchains.
Check out our new website (with our new look 👀) and join us as we reshape the future of blockchain development, making it accessible to all developers.
Tomorrow, 21st July,
@anuragarjun
is giving a talk on
@AvailProject
's architecture and use cases at the
@modular_summit
.
If you're in Paris, please attend it and give us feedback. Quite a few interesting talks and panels happening at the Modular Summit.
The
@AvailProject
team is in Paris this week for
@EthCC
, Modular Summit and other events. Say hello if you encounter someone from the team.
I'm unable to visit due to rejected visa, but see you at the next conference.
It's pretty simple. We wanted to connect
@AvailProject
with Ethereum in the most secure manner. And nothing better than ZKP based consensus verifying bridge.
For ZKP, there aren't many better than
@SuccinctLabs
. Hence this partnership.
Why bridge needed? What makes it unique?
🧵
Creating the essential base layer for rollups is something our team is very passionate about. That’s why we are excited to share our progress of a two-way bridge connecting Avail with Ethereum, after collaborating for months with
@SuccinctLabs
, one of the best in the ZK space.
The Kate testnet was put to rest today. It's soul still hovers in those 357+ nodes. Until they all stop, the soul cannot rest in peace.
Please migrate over to our Goldberg testnet.
It was fun to talk about bridges with some awesome speakers. Among many other things, we discussed about the validity proof secured proof of consensus bridge from Substrate to Ethereum. Hope to share more soon, but in the meantime...
Judging the awesome projects at
@ETHIndiaco
was a privilege. 2000+ hackers submitting 360+ projects is just 🤯
Congratulations to every participant!
Huge kudos to
@devfolio
@ETHGlobal
teams for organising this. Can't imagine the effort behind the scenes to make this success
🧵
Just to clarify that Polygon PoS has deterministic finality today. It has had it since it launched, almost 3 years back.
This would only accelerate the time to deterministic finality by making 2/3+ sigs accessible to users earlier than checkpointing.
When apps are on the base layer, it sometimes creates weird problems. If the app becomes very valuable, then any app level bug might result in fork of the chain. The app users dominate the chain dynamics.
This is one of the reasons I'm bullish on appchains. App users (and any…
Get prepared for another set of intellectually stimulating conversations around some of the most relevant topics by some of the best minds in the space.
If you're around Denver at that time, you wouldn't want to miss it.
HOT TAKE SERIES IS BACK! 📢
This time it's the Execution Race at
@EthereumDenver
, where we're looking to find out which execution layer will reign supreme.
RSVP here -
We are excited to introduce the first building block of the Nim AI framework, which provides economic ownership and verified access to AI models.
Ligetron 1.0 by
@ligero_inc
, is a memory-efficient ZK system powered by Ligero Proofs. We created the first proof in the world for…
Shout out loud to our speaker on the topic of
#data
availability 🥳
🎙️ The Co-Founder
@prabalbanerjee
from
@AvailProject
to a groundbreaking approach of enabling greater decentralization and reliability for
#rollup
users
RSVP ➡️
Well, I do not agree with some of the opinions here, but that's for another time.
Wanted to point out that
@AvailProject
is not a Data Availability Committee. It's a DA focused PoS L1 chain, with DAS enabled LCs & upto 1K validators.
@LoganJastremski
@smyyguy
@swmartin19
Your Twitter timeline says modular...
But there's a high throughput, single-shard blockchain thesis growing in the shadows...
@LoganJastremski
reframes the scaling conversation in today's episode!
→
→
→
I'm glad someone asked this question
Ofc,
@ri5hitripathi
may have done it for fun, just to show that it can be done. But let's talk about the potential of this design.
Today, watches are already used for payments. In a zk-appchain world, this can be taken to the next level. 🧵
Let me write the same thing as it's power cannot be overstated.
User (your wallet) has a proof (that it can independently verify) that all your rollup transactions that were ever submitted to the chain were correctly executed.
No reliance on smart contract. Just blobs and…
I want to talk about the unique “Sovereign” rollup architecture that comes out of the box with Sovereign SDK because, although it’s not an aspect of our SDK that we’ve deliberately marketed so far, it enabled a solution to a costly user problem that was not possible otherwise.…
Good thread. But disagree with the mental model.
If you use contracts today, how do you verify they are battle tested? You see ERCs followed, OpenZeppelin contracts imported and then the delta over it? Maybe audits reports?
Rollups will be similar. SDKs will be battle tested...
The easier it gets to launch a rollup, the more we will see rollups of very poor quality or with outright malicious intents. Small systems will get less scrutiny than big ones, it's unavoidable. Be aware aping into them, dyor, ask questions.
Had a great conversation with
@LoganJastremski
🤝
Discussed modular designs, where it stands out and some of its limitations.
And of course talked about
@AvailProject
and some of its design choices.
Dropping a fantastic discussion with
@prabalbanerjee
- The Co-Founder of
@AvailProject
There are many small nuances between architectures, and many are pretty misunderstood.
Prabal explains:
- increasing throughput via block size and block times
- Light clients and data…
The modular era has arrived and NEAR is set to take centre stage.
Introducing the NEAR Data Availability Layer.
Hot off the press at NEARCON23, NEAR DA is set to launch to offer secure, cost-effective data availability for ETH rollups. 🌐
NEAR DA marks NEAR's entry into the…
Some very interesting EIP-4844 numbers: 🧮
* As of now, Ethereum already has ~11.7k blobs on mainnet
* Relatively speaking, blobspace is currently more competitive than it was at launch, as
@Optimism
,
@arbitrum
,
@ourZORA
and
@Base
have joined in after
@Starknet
and
@zksync
Era.
IMO, the power of probability amplification is similar to compounding.
You have a randomized algo with prob of correct (boolean) output = 51%. Suddenly you have the power to make it (100-δ)%, for any δ.
We use the same principles in LCs to get very high confidence of DA.
Great thread (as always) by
@bobbinth
🙌
My $0.02: It is great that we have very cheap zkEVM proofs. If you want to build on status quo w/ all existing tooling and well understood limitations - go for it.
But if you want to build innovative apps, you may need to look beyond.
A quick take on zkEVM vs. zk-optimized VM performance (specifically in the context of rollups).
First, a bit of historical context: even as recently as a couple of years ago, many people believed that zkEVMs would not be practical in the near or even medium term future.
The execution layer landscape has become quite crowded. The rise in rollup frameworks has exacerbated this trend.
We did our best to map out the landscape and came to some interesting (possibly controversial) conclusions.
Read the full piece here:
TLDR↓
I'm sure relay bug will be fixed, but p2p propagation of blobs are hard. With 6 blobs, no erasure coding, is still hard.
But this also shows why you need LC verifiability. To know whether blobs are missing, today you need an Ethereum full node. Ethereum LCs cannot detect this.
We have observed blocks containing kzg commitments, but w/o the release of blobs over p2p the last few days, it happened 10-20 times per day.
The same issue has happened over 20 times in the last hour(!), presumably due to a spike. Are relayers or builders looking into this?
Congrats
@sovereign_labs
!
This is a big step towards spinning up your own validity proof secured sovereign rollup. Looking forward to seeing what new doors this opens up.
ZK rollups are the future of blockchain applications. And now, anyone can build one.
Introducing the alpha release of Sovereign SDK, a toolkit for developers to easily launch their applications as zk-rollups.
This is a great effort by the PoS team to bring down deterministic finality time on the chain. 👏
Let's talk about the availability-finality dilemma in blockchains and how constructions avoid the dilemma by using hybrid ledgers 🧵
[image credit:
@praetorian
🙏]
Rollups remain pretty elusive to average devs today. If you have a business logic and you want to spin up a rollup (with even broken security guarantees), it is hard.
But there are very capable teams working on it. In the very near future, we are going to have
- Any Rust code in…
data availability explained by
@AvailProject
cofounder, prof
@prabalbanerjee
📔
0:41 - blockchains: origin to rn
3:47 - what's da, what does it solve?
8:11 - types of da
10:08 - dac & das
12:56 - tx process with da layer
13:52 - scenarios: with & w/o da layer
on
@cryptotownpod
…
Time is indeed money.
Timing attacks are underrated. While this targeted PoS protocols, we had earlier worked on a paper titled similarly looking at griefing attacks in Bitcoin lightning network.
Full paper:
h/t
@MazumdarSubhra
1/ Time is Money: Timing Games in PoS Protocols
tldr: a block proposer may strategically delay their block proposal to maximize MEV capture, while making sure the block is still early enough to become part of the canonical chain.
I'm sure that finality will be restored. It might have been already. But this is a gentle reminder of the power of a hybrid ledger. Block production continues even if finality gets delayed. And instant finality chains can't give you that.
The beacon chain stopped finalizing about thirty minutes ago. I don't know why yet, but in general the chain is designed to be resilient against this, transactions will continue as usual and finalization will kick in when the problem is resolved.
Just got to know that intern doesn't have any days off even on the weekends because apparently crypto doesn't sleep?
Who's building a solution to this?
Judging new technology solely by old metrics is like assessing an elephant's capabilities based on its ability to climb a tree.
Here are some common misconceptions about modular chains. From complexity, cost, performance, congestion pricing to experimentation and hard forks.
0/ There has been a lot of debate around modular vs monolithic chains. And we have seen new nomenclatures being invented. To address some of the most common misconceptions, here is a post that tries to debunk these myths. Summary in 🧵👇
Interesting paper: "Snapping Snap Sync: Practical Attacks on Go Ethereum Synchronising Nodes"
- Applies *only* to PoW geth nodes performing snap sync
- Needs very small amount of mining power
I feel bootstrapping is Achilles heel of blockchains. Need more research in this space
1/ We are excited to announce Succinct Processor 1 (SP1), our first generation, 100% open-source zkVM that proves arbitrary Rust programs.
SP1 targets an order of magnitude performance improvement vs. existing zkVMs, and is already up to 28x faster for certain programs.
Wonderful essay on peer-review.
While it talks about how peer-review couldn't prevent error in papers, it doesn't quite talk about how hard it gets to publish new ideas w/o being institutionalised. PhD becomes not about learning how to do research, but how to publish papers.
And as always, thanks a lot for participating on the testnet. All the feedback we are getting is helping us improve. A very cool lineup of challenges are being set. So, stay tuned.
Keep following and keep engaging on X/Discord.
Avail ofc has VectorX, but that is just bcoz it's the status quo and most immediate need. We are heavily committed to make Avail decentralised. But we are also pushing for Validiums where users themselves verify DA using DAS and L2 execution proofs, from inside their wallets.
It's pretty simple. We wanted to connect
@AvailProject
with Ethereum in the most secure manner. And nothing better than ZKP based consensus verifying bridge.
For ZKP, there aren't many better than
@SuccinctLabs
. Hence this partnership.
Why bridge needed? What makes it unique?
🧵
4/ We have not started the LC related challenges yet. But as we prepare, we wanted to stress test internally before letting others use it. With 1000s of LCs running, we are fine tuning last bits to make sure UX remains good when we test on the wild. LC challenges coming soon™
I was eagerly waiting for
@luketchang
and team to release this. No need for me to speak much about the dev team or the utility of the product. Looking forward to seeing this in production. Sure they will come up with more awesome stuff.
1/ For crypto to reach its next stage of growth, it must be used less for speculation and become more like money. We believe privacy is a necessary condition for this to occur.
@nocturne_xyz
we’re excited to introduce a new primitive to the Ethereum ecosystem—private accounts 🧵
... ~300 validators on this testnet. We are humbled by the huge interest in running validators. But don't get disheartened if you don't get to run a validator. There will be other ways to contribute to the security and stresstesting of the network.
In classical monolithic chains,
- you need full node to verify block correctness
- you check by re-executing blocks
- light clients only verify header consensus (sig checks for PoS)
Hence, full node requirements are high and LCs have inherent trust assumptions
@teemupai
DA layers are optimized to ensure that app full nodes had access to txn data of specific blocks. This is super important for ORUs (security) and ZKRs (usability).
When
@realDonaldKnuth
says so, you got to admire the algo.
And when it is by Sourav Chakraborty who taught you approx algo & advanced algo courses during postgrad, you wish you had paid more attention to the lectures and learnt some more.
Link:
3/ With large number of validators comes big monitoring responsibility. We are working on letting validators know when they are not producing blocks/offline so that the network remains healthy + they don't get slashed. This is easier now that most vals have identity setup.
Last, but not the least, I've heard few finalists used
@AvailProject
internally without us giving any bounty or booth. It is great to know that people are using us bcoz they find it useful.
If you were one of the projects, let us know. Something cool might be waiting for you 😉
Friday was about
@alt_layer
's Rollup Frontier Day. Also wrapping up meets with people within the team & many others from outside
Many other things happened through the week. But that's for another time. Back to stress testing our prod-ready codebase on incentivized testnet 🧑💻
It's here - Avail's Clash of Nodes incentivized testnet is officially up and running! 🌟
This is a major step towards Mainnet, with a chance for validators, light clients, and everyone to test our network, earn points, and more. Ready for a journey?
Here's what's ahead...…
1/ The first cohort of validators have been fully onboarded and second cohort is being onboarded. Currently, 185 validators actively secure the network, with ~40 waiting. The proposal to make val count 220 is in motion and should take effect soon. To reiterate, we will have ...
For L2 txns, this should offer the highest form of security. For Ethereum, which acts as a bridge really, it either needs to
- check ext DA attestation
- dec operator set does DAS and agree on DA
- forget everything and just "trust me bro"
The last one is most hyped rn somehow.
At Avail, we want to offer a robust and secure DA API, so that app devs can focus on building the next generation of trust-minimized applications without worrying about bootstrapping security or inheriting complexities of shared monolithic chains.
5/ The next challenge will be balance transfer. If you are just starting off, make sure you acquire some balance from the faucet. If you already have non-zero balance, should be enough to participate for now, so don't fret. No need to run any node to participate in this challenge