Dom stop fucking me in the ass (please)
@dominic_w
#icp
internet-computer:native
16
Icy Pee retweeted
It's time to ditch PHP. Here's how your website can move to Motoko, in under 20 seconds.
Hand the job to your coding agent. Give it the migration skill, ask it to move your site, and it looks everything over and plans the move. Nothing changes until you say yes. Then it rebuilds the site in Motoko, the Internet Computer's programming language, and puts it live.
Your pages and links stay the same, and there's no PHP runtime, web server or OS left to patch.
Icy Pee retweeted
Tokenized equities recently reached a new peak: $4.8B in circulating value. And this is only the beginning.
Fingerprint is building more ways to invest in, swap and earn from tokenized equities, onchain.
Join the waitlist 👇
fingerprint.family/
Icy Pee retweeted
Replying to @PlayRematch
(1) I know it’s not the current vibe or direction but there’s a huge gap for a stripped back look in a stadium environment for 7 v 7 + pro clubs style mode. Normal goal nets. Stadiums. Possible offsides or only offsides from balls played within own half. Goal hanging…..
Icy Pee retweeted
Cant use coingeko anymore! They will push a hard Solana first narrative to ensure their coin reins supreme; even though #ICP consistently crushes it in every aspect.
Icy Pee retweeted
If banks all zero what’s the case with coinbase… what shall I do with my internet-computer:native ??
If banks all zero what’s the case with coinbase… what shall I do with my internet-computer:native ??
Icy Pee retweeted
🚨HUGE: $ICP has just recorded its LOWEST average monthly transaction fee EVER! 🔥
According to @dfinity and @chainspect_app, the average transaction now costs just $0.0000518!
To put that into perspective, the price of a $5 coffee could cover approximately 96,500 transactions on ICP!
Icy Pee retweeted
A lot happens before the ticker. A whole market develops around it.
Invest in private companies
Earn from stock trading
Swap onchain stocks
More of the market. One Fingerprint. Soon.
Dfinity Founder wants Exchanges to use ICP...
@Dfinity Founder Dominic Williams (@dominic_w) is unhappy about the current attacks on exchanges.
$ICP is trying to provide a tamper-proof custody service for institutions using its threshold cryptography technology built on the Internet Computer.
According to Williams, Dfinity's Chain-Key design has no single points of failure.
Instead of a private key system, it uses a decentralized signing mechanism via autonomous network canisters triggered only by approved approval chains.
This message is clearly directed at exchanges.
Cannot believe I’ve been seeing people slagging off Billy Mitchell.
Makes you wonder if these people have other main hobbies or are just Wednesday “supporters” and not football fans. Absolutely fine but I wouldn’t personally be so opinionated if I wasn’t in to the sport. #swfc
Icy Pee retweeted
The Code Is Public. So Is the Patent.
NEAR's own public repository implements a signing flow that maps closely onto a DFINITY patent, explicitly cites the underlying Groth–Shoup work, and NEAR is now replacing that flow with an architecture that removes the patent-sensitive steps.
That is the conclusion I reached after comparing the patent claims directly against NEAR’s public code and development history.
Not community rumors.
Not tribal narratives.
Not price speculation.
The patent, the source code and the migration history are all public.
This is not a court ruling. It does not prove intent, theft or legal liability. But the technical overlap is substantial enough that this can no longer be dismissed as ordinary crypto drama.
The patent
The central document is U.S. Patent 12,695,626 B2, titled:
“Digital signatures with re-randomized presignatures.”
It lists Jens Groth and Victor Shoup as inventors, DFINITY Stiftung as assignee, claims priority from September 30, 2021, and was granted on July 28, 2026. Google Patents
The independent Claim 1 describes a signing method with five essential steps:
A presignature is computed.
A signing request for a message is received.
A non-predictable public re-randomization value is computed after that request.
That value is used to re-randomize the presignature.
The message is signed using the re-randomized presignature. Google Patents
That sequence matters because it is not merely a broad description of threshold cryptography.
It describes a particular order of operations.
And NEAR’s stored-presignature signing flow maps closely onto it.
The NEAR flow
In NEAR’s public MPC repository, the relevant legacy path can be reduced to this:
Presignature computed and stored
→ signing request arrives
→ public re-randomization value δ is derived
→ presignature is re-randomized
→ signature is produced
The implementation contains types and functions named:
PresignOutput
RerandomizedPresignOutput
RerandomizationArguments
derive_randomness
rerandomize_presign
The re-randomization value is derived using HKDF-SHA3-256 from a combination of the public key, key-derivation tweak, message hash, presignature commitment, participants and public entropy originating from the network.
The resulting value is then used to transform the stored presignature before the ECDSA signing protocol is executed.
That is not merely conceptual similarity.
It is a recognizable claim-by-claim technical mapping:
Compute a presignature: present.
Receive a signing request: present.
Generate a public, unpredictable re-randomization value for that request: apparently present.
Use it to re-randomize the presignature: present.
Sign with the resulting presignature: present.
The most legally sensitive question would be how a court interprets the requirement that the re-randomization value be “generated after receiving the signing request.” But NEAR’s execution path appears to calculate the final delta from request-specific data after the request enters the signing pipeline.
That is a real issue for patent counsel to examine—not a superficial resemblance.
The dependent claims make the overlap more interesting
The patent’s dependent claims narrow the method into several specific implementations:
Threshold signatures.
Pseudo-random generation of the re-randomization value.
Applying a hash function to the message, public key or other inputs.
Deriving public randomness from a random beacon.
Discrete-logarithm signatures.
ECDSA over elliptic curves. Google Patents
NEAR’s system uses threshold ECDSA, derives delta pseudorandomly through HKDF-SHA3-256, includes the message hash and public-key-related data in the derivation, and uses public block entropy as an input.
Whether NEAR’s block entropy legally qualifies as the patent’s “random beacon” would require formal claim construction. But the patent itself describes a random beacon broadly as publicly available randomness supplied by the distributed network. Google Patents
That makes Claims 5 through 10 particularly relevant to any serious infringement analysis.
The repository identifies the underlying work
The historical record is also difficult to ignore.
In November 2024, NEAR merged PR #50:
“Simon/rerandomize key derivation.”
The commit history explicitly described the change as:
“Rerandomize using the GS21 variant.”
And the implementation itself identified the re-randomization as a variant of GS21.
GS21 refers to the Groth–Shoup work underlying DFINITY’s research on ECDSA with additive key derivation and re-randomized presignatures.
This proves something important—but not everything.
It proves that NEAR’s developers were aware of and intentionally implemented ideas derived from the Groth–Shoup work.
It does not, by itself, prove that they knowingly infringed a patent, misappropriated confidential information or intended to harm DFINITY.
Scientific work can be publicly readable while implementations of it remain subject to patent rights. A paper being public does not automatically create a patent licence.
That distinction matters.
There is also an additive key-derivation overlap
NEAR’s Tweak implementation derives keys using the algebraic form:
x′ = x + tweak
X′ = X + tweak·G
The same DFINITY patent contains dependent claims covering additive shifting of a secret and public master key:
d′ = d + e
D′ = D + eG. Google Patents
The symbols differ.
The algebraic construction is essentially the same.
That does not automatically establish infringement of Claims 16 and 17, because dependent claims incorporate all limitations of the claims from which they depend. But it adds another technically significant correspondence that cannot simply be waved away.
Not every claim maps
There is also an important difference, and it should be stated clearly.
Claim 11 describes a specific additive re-randomization formula:
r = r′ + δ
R = R′ + δG
NEAR’s stored-presignature implementation uses a multiplicative transformation in which the nonce commitment is scaled by delta, while related secret shares are adjusted by its inverse.
Therefore, Claim 11 is not an obvious case of literal infringement.
That matters because a serious analysis should identify both the matches and the differences.
But Claim 1 is broader. It does not require the precise formula found in Claim 11. It requires that a public, non-predictable value generated after the request be used to re-randomize a presignature before signing.
The broader Claim 1 is where the strongest potential mapping remains.
Then NEAR changed the architecture
On September 14, 2026, NEAR merged PR #4386, introducing a new presign_and_sign protocol.
Instead of calculating and storing a presignature before the message is known, the new path consumes fresh cryptographic triples only after the message hash is available. Presigning and signing occur within a single online protocol, and no presignature is stored. GitHub
The new path also removes the old request-specific re-randomization step.
The pull request describes the key-derivation adjustment as effectively using δ = 1, rather than deriving a fresh delta and using it to re-randomize a stored presignature. The repository’s own review discussion explicitly contrasts this with the classic path and notes that delta no longer binds R to the message hash. GitHub
Technically, the difference is striking:
Old path
Stored presignature
→ request-specific δ
→ re-randomization
→ signing
New path
Request arrives
→ fresh presign computation
→ immediate signing
→ nothing stored to re-randomize
The new construction pays for that simplification with two additional online communication round trips.
Then, on September 25, PR #4503 integrated the new path into NEAR’s node software. Its author stated that the team planned to remove the existing Cait-Sith implementation later, while temporarily retaining it as a fallback during the upgrade. GitHub
The migration has continued.
On October 21, 2026, NEAR merged PR #4792, extending the online-presign architecture to foreign-chain transaction signing. The repository also references a separate effort, PR #4751, to remove the stored-presignature signing path from the node altogether. GitHub
Is this a patent design-around?
Technically, it looks like one.
The older architecture contains the sequence most relevant to DFINITY’s Claim 1:
stored presignature + post-request public δ + re-randomization + signature.
The replacement architecture removes:
the stored presignature
and
the post-request re-randomization of that stored presignature.
Those are precisely the patent-sensitive steps.
But GitHub does not establish why NEAR made the change.
A design-around is not an admission of infringement. It can be good engineering, a security simplification, an operational trade-off or prudent intellectual-property risk management.
Therefore, the accurate conclusion is not:
“GitHub proves NEAR changed the system because it feared DFINITY’s patent.”
The accurate conclusion is:
“GitHub proves that NEAR is replacing the relevant flow with one that removes the elements most likely to trigger the patent claim.”
The motive remains an inference.
The architectural change is a fact.
What the public evidence establishes
The public record establishes that DFINITY holds a granted U.S. patent covering re-randomized presignature signing flows.
It establishes that NEAR implemented a stored-presignature threshold-ECDSA flow with request-specific public re-randomization.
It establishes that NEAR’s repository explicitly referenced the Groth–Shoup work.
It establishes that multiple elements of DFINITY’s independent and dependent claims appear in NEAR’s implementation.
And it establishes that NEAR is now migrating toward an architecture that eliminates the stored-presignature and re-randomization sequence.
What the public record does not establish is a judicial finding of infringement, intentional theft, criminal conduct, willfulness, damages or liability.
Those conclusions would require patent claim construction, deployment evidence, territorial analysis, validity review, prosecution-history analysis and a formal legal process.
The real conclusion
Dominic Williams’ entire public account should not automatically be treated as proven.
But the central technical allegation can no longer be dismissed as unsupported.
NEAR’s own repository provides enough evidence to justify a professional claim chart comparing its stored-presignature implementation against US 12,695,626 B2.
It also provides enough evidence to justify serious discussions about attribution, licensing and potential remedies.
DFINITY invested years in foundational cryptographic research.
NEAR publicly implemented a closely related signing flow.
The underlying work was explicitly recognized in its repository.
And the relevant architecture is now being replaced.
This is not about tribal loyalty to ICP or hostility toward NEAR.
It is about whether organizations that fund difficult, first-principles research can protect the resulting inventions—and whether other projects can commercialize those inventions without permission, attribution or compensation.
NEAR deserves the opportunity to answer.
But that answer should be technical and claim-by-claim.
Not slogans.
Not social-media deflection.
Not another round of crypto tribalism.
The code is public.
The patent is public.
The migration is public.
Now the explanation should be public too. ∞
#WorldComputer #InternetComputer #ICP
🤖 Made with AI
Icy Pee retweeted
ICP'S NETWORK ACTIVITY IS SURGING.
New addresses on Internet Computer are up 45% since summer, and daily active users are back to early-2026 levels, according to Santiment.
However, $ICP is still about 64% below last November's Caffeine AI peak.
Is ICP seriously undervalued?
Icy Pee retweeted
A certain Mr Carlos Carvalhal is currently out of work ... 👀
The man who had a dream...
Thoughts?
#swfc
Icy Pee retweeted
Stood up the challenge in the clubs darkest time , thank you for your efforts henrick 🦉#swfc
Sheffield Wednesday Football Club confirms that Henrik Pedersen has left the club with immediate effect.
Henrik led the first team through an exceptional period in the club's history, providing stability and leadership during a time of significant uncertainty and change.
The club would like to place on record its sincere thanks to Henrik for his contribution and wishes him every success in the future.
Sporting Director Simon Wilson said: “First and foremost, I want to thank Henrik for his professionalism, commitment and contribution during an exceptionally challenging period for Sheffield Wednesday.
“We knew the start of this season would be difficult due to the scale of change across the club, but our expectation was always that we would develop as the season progressed, establishing a recognisable style of play and greater consistency in our performances.
“Although we have accumulated points, we haven’t seen enough progress in those areas. We believe making a change now gives us the best opportunity to establish those foundations and build towards the second half of the season and beyond.
“We remain ambitious about what we can achieve this campaign, and our focus remains on building a Sheffield Wednesday team with a clear identity, capable of delivering sustained improvement and long-term success.”
Matt Hamshaw will take interim charge of the first team while the club undertakes the process of appointing its next Head Coach.
The club will provide further updates when appropriate.
The World needs to wake up.
$ICP
Is THE answer.
Buts it’s not me that needs to make the institutions aware and adopt
2
127
If people realise how big this is you need to look into what DFINITY is doing and Internet computer.
It’s becoming all too common.
internet-computer:native
@dfinity @dominic_w
#RevolutDown #AWS #INTERNETCOMPUTER
How can Revolut allow for a SPOF so big ????
App is down, Payments don't work, cards ( even physical ) don't work and you can't transfer money out.
Insane fumble...
1
200























