TL;DR
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A former engineer who helped build Meta’s Mononoke says Anthropic’s new version control system, Scion, closely resembles Mononoke’s design, including its use of derived data to speed up operations such as blame. The criticism concerns attribution, not a claim that Anthropic was wrong to reuse open-source ideas; Anthropic’s full response and the extent of the similarities are not established in the supplied material.
A former engineer who helped build Mononoke, Meta’s source-control server, says Anthropic’s newly described version control system Scion closely follows ideas from the earlier system, including a technique called derived data for speeding up operations such as blame. In a post published October 15, the engineer argued that Anthropic’s account gave too little credit to Mononoke, raising questions about attribution as AI-assisted development draws on established software designs.
The author compared Anthropic’s description of Scion, published the day before, with their own 2022 account of Mononoke’s derived-data approach. Both descriptions discuss using derived data to make operations such as blame queries more efficient. The author says the similarities go beyond independent teams arriving at a shared solution, but the source material does not provide a technical review establishing how much of Scion’s architecture matches Mononoke’s.
The criticism is specifically about public credit. The author says Anthropic credited Sapling for “hard-won lessons” but did not mention Mononoke, and argues that this framing risks making it sound as though Claude designed the system’s ideas. The author also says the Mononoke team developed many of those ideas years earlier. These are the author’s assessments; the supplied material does not include a response from Anthropic or a detailed statement of the company’s design process.
The author says they do not object to Anthropic reusing Mononoke’s design, describing the spread of open-source ideas as a benefit. Their concern is that the system’s earlier contributors should be credited. They point to Mononoke’s open-source availability and argue that AI models can readily encounter existing designs, making acknowledgment more important when those ideas influence new systems.
Credit for AI-Assisted Design
The dispute highlights a challenge for software teams using AI tools to create or explain technical systems: similarity does not by itself establish copying, but a model’s ability to draw on published work can complicate how teams identify where design ideas came from. In this case, the author links Scion’s reported use of derived data to Mononoke’s earlier design and says the public account should have named that precedent.
For open-source developers, the question is practical as well as reputational. Projects often build on prior systems, and credit can make technical history easier to trace and help readers understand which ideas were reused or adapted. At the same time, the supplied source does not establish whether Scion’s implementation reproduces Mononoke’s code, whether the systems share additional architectural details, or what obligations apply. The story is therefore a dispute over resemblance and recognition, not a confirmed finding of plagiarism.
As an affiliate, we earn on qualifying purchases.
From Mononoke to Scion
Mononoke is a source-control server built by a team at Meta. The author says the team developed its design in 2017–18 and later described its derived-data approach in candidate materials written in 2022. The author also notes that the Mononoke and Sapling teams studied other prior work, including Google’s SrcFS and database write-ahead logs. That acknowledgment situates Mononoke itself within a broader history of software design rather than presenting it as having no antecedents.
Anthropic described Scion’s derived-data concept on October 14, according to the author’s timeline. The public comparison followed the next day. The author says they also wrote a follow-up post to address questions and concerns, and references a separate blog post by Vicent Marti for Cursor as an example of a system with related insights but important design differences. The supplied material does not include that follow-up or the Cursor post’s technical details.
“I think the level of imitation goes well beyond having similar insights.”
— The former Mononoke engineer who wrote the comparison
As an affiliate, we earn on qualifying purchases.
What the Comparison Establishes
The supplied material does not include Anthropic’s response, a full technical comparison of Scion and Mononoke, or Scion’s source code. It therefore remains unclear how extensive the architectural overlap is, whether the systems share implementation details beyond the derived-data concept, and what information Anthropic considered when developing or describing Scion. The author’s use of “plagiarism” is an allegation about attribution, not a confirmed legal or technical finding.
It is also unclear whether Anthropic will revise its description or add acknowledgment of Mononoke. The source says a follow-up post was published on October 15, but does not provide its contents, so any additional explanation or response there cannot be assessed from the material available here.
open-source code collaboration tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Possible Clarification From Anthropic
The next useful developments would be an Anthropic response and a more detailed, side-by-side account of the two systems’ designs. Such information could clarify whether Scion’s derived-data approach is a close adaptation, a broader shared pattern, or one component of a substantially different architecture. The author’s October 15 follow-up may also address questions raised by readers, but its claims are not included in the supplied source.
Until those details are available, the confirmed development is the public criticism and the specific resemblance the author identifies. No conclusion about copying, intent, or legal responsibility is established by the material provided.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the allegation about Anthropic’s Scion?
A former engineer who helped build Mononoke says Scion’s design closely resembles the earlier system, citing their shared use of derived data to make operations such as blame more efficient. The claim has not been independently established in the supplied material.
What does the author say Anthropic failed to credit?
The author says Anthropic credited Sapling for “hard-won lessons” but did not mention Mononoke, whose team the author says developed many of the relevant ideas years earlier.
Is this a confirmed plagiarism finding?
No. The source presents the author’s view that the similarities go beyond shared insights and criticizes the omission of credit. It does not provide a legal ruling, a complete technical comparison, or an Anthropic response.
The author says they do not object to reuse of Mononoke’s design and sees open-source sharing as beneficial. Their stated concern is whether contributors to earlier work receive appropriate acknowledgment.
Source: hn
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
