Rendered at 11:08:26 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
BryantD 14 hours ago [-]
From the Anthropic announcement:
“Source control for huge monorepos has prior art, notably Google’s Piper and Meta’s Sapling, so we were able to build Scion with their hard-won lessons in mind.”
Mononoke is a component of Sapling, as I understand it. In fact, the repo link provided in this article links to a subdirectory of the Sapling repo. I’m not totally sure I understand the complaint.
WhitneyLand 13 hours ago [-]
If I credit my newly introduced car as inspired by Corvette, then sure, it might be ok not to mention the engine (Mononoke is the server part of Sapling)
However if my new car's engine is a near copy of the one in the Corvette, and my announcement gushes about the engine using the same terms the Corvette design team invented...
And you take the statement: "Mononoke is open source and in every LLM’s training set."
There is not much resemblance with any cleanroom implementation anymore.
tcdent 8 hours ago [-]
To me, this sounds like it's less about the technology, whether it's open source or not, and more about the people in the room that are influencing the design, direction, and implementation of it. All of which are very likely to have previous positions at major companies responsible for the products mentioned.
Perhaps if there is any ambiguity, it's intentional on the part of caution around intellectual property provenance.
sunshowers 14 hours ago [-]
The level of inspiration goes well beyond "hard-won lessons". Something like Lore (slop design document aside) can claim that level of inspiration. Here the data model is a direct copy, down to the terminology.
To be uncharitable, this is marketing for Anthropic's impending IPO.
whiplash451 14 hours ago [-]
I see correlation instead. You IPO when you grow like crazy. You build something like scion when you grow like crazy.
muglug 14 hours ago [-]
And I think Not Invented Here has turned into Not Invented By Our Agents at many places.
taneq 12 hours ago [-]
Devil’s advocate: This could turn out to be a good thing for software security. If each company’s stack is home grown, there will still be vulnerabilities (probably even more of them) but they’ll be less portable compared to today’s monoculture where an exploit for a common component will work almost anywhere.
Terr_ 11 hours ago [-]
The generation process will be the monoculture. "Oh, model X generates a ton of code with vulnerability Y."
cyanydeez 8 hours ago [-]
except, once you get real analysis on open models, you're likely going to see the same insecurities everywhere because the models have blind spots. The same blind spots. The reason distillation works so well is because of the shared nature of their training corpus.
devin 13 hours ago [-]
I remain unconvinced that they even need a new VCS to achieve what they want to. One has to wonder what the level of churn is at this number of PRs.
tmp10423288442 13 hours ago [-]
Who cares about your opinions about what number of PRs is right? They want to make more PRs (particularly to a monorepo), they need a new VCS.
Even if they weren't making a lot of PRs, Github has been melting down enough this year that they've probably encouraged customers like Anthropic to move off.
devin 12 hours ago [-]
I was casting doubt on whether or not they need a new VCS to achieve their goal in the first place, not providing an opinion about the "right" number of PRs. Like most things coming out of these labs it's hard not to look at it as marketing. Their "changes merged per day" is absolutely meaningless unless you know the nature of those changes. I could make a graph like this by making tons of agent-based empty commits. That's not a problem that actually needs a solution, though. Furthermore, they _didn't even build a new VCS_, they copied one that existed and made a bunch of noise about it.
steveklabnik 12 hours ago [-]
Google and Meta hit these sorts of limits over a decade ago. They then built new VCS stuff to handle it, Meta's part is the subject of this submission. This was even pre-agents.
Git is just not suited to truly large monorepos, and it won't ever be. That's totally fine, of course. Less of a monoculture here would be great.
charcircuit 3 hours ago [-]
Microsoft hit it too and then improved git and made it faster. Right now Anthropic seems to be under 1 commit per second. I don't think that would make the git server break a sweat.
solenoid0937 12 hours ago [-]
Did they make a bunch of noise about it, or did HN? This place loses its mind every time OpenAI or Anthropic posts a blog.
SaucyWrong 12 hours ago [-]
There is no distinction at this point. If you know your blogs always get picked up by HN, and they do know this, then you know your blog is a de facto and gratis marketing channel and will use it as such.
Analemma_ 12 hours ago [-]
fwiw I've heard accounts that Google's source control is also creaking under the weight of AI code changes, and that there's already a new frontend to it with backend revisions on the way too. When GitHub and Google and Anthropic all simultaneously have VCSs that start buckling I think it's safe to say it's a real phenomenon and not marketing.
andai 12 hours ago [-]
more = better
dpe82 13 hours ago [-]
Effectively nobody considering buying into the IPO cares about Anthropic's source control system. They could be passing around floppy disks and so long as they were still shipping nobody would care.
cyphar 13 hours ago [-]
Not about the details but the marketing story is all about how "agents" are building so fast they need to design new infrastructure for the new world, etc. Extra points if you spin it as though the "agents" have built some new never-before-seen architecture that is clearly better than the "human scale" stuff.
The intended audience is obvious from first paragraph alone:
> At Anthropic, coding is no longer the bottleneck to building systems or doing research. Claude is able to autonomously iterate on research loops, micro-optimize complex programs, and build new services from the ground up at a speed and scale we haven’t seen before. The bottleneck, now, is all of the developer infrastructure built for human scale: source control, builds, and CI.
Translation:
> We are so far at the cutting edge we need to build shiny new infrastructure to replace the smelly "human scale" cruft. How could you /not/ value us at $2T?
solenoid0937 12 hours ago [-]
Both things can be true, they're scaling fast and it looks good for IPO.
vhantz 12 hours ago [-]
Scaling what exactly? Anthropic doesn't release any new products except models. I'm sure you don't need tens of thousands of pushes every day to train models. Scaling fast is so far just their words, nothing else indicates it.
simonw 12 hours ago [-]
> Anthropic doesn't release any new products except models
Claude.ai
Claude Code CLI
Claude Code Desktop (multiple platforms)
Claude Code on the web (aka Claude Code Cloud)
Claude Design
The Claude API
Claude SDK libraries
Cowork (just retired and merged into regular Claude)
Claude Tag
Claude for Legal
They ship a lot of products.
sroussey 10 hours ago [-]
Also: bun
Even that one repo is breaking GitHub this week…
vhantz 12 hours ago [-]
And still in no way warrants tens of thousands of pushes. These all overlap pretty broadly
vhantz 12 hours ago [-]
Yeah that's all the same thing
bonsai_bar 10 hours ago [-]
How…?
henrymerrilees 12 hours ago [-]
That would arguably make it look worse.
solid_fuel 13 hours ago [-]
> Here the data model is a direct copy, down to the terminology.
That’s not surprising to me - they designed and built it using their LLMs, of course it’s duplicating prior work. That’s essentially the core of what LLMs do, after all.
simonw 13 hours ago [-]
Why is that bad?
Mononoke and Sapling are open source systems. Why shouldn't new software take inspiration from how they are designed?
sunshowers 13 hours ago [-]
From my post:
> And, to be clear, I do not have a problem with Anthropic copying Mononoke’s design. (This is part of why we put out open source — to benefit humanity through the spread of good ideas.)
> But as LLMs continue to assist in designing new systems, they are likely to index heavily on existing designs. Mononoke is open source and in every LLM’s training set. If you ask Opus 5.5 or GPT-6 Astra (thanks Matthew Fowle) about prior art, they both immediately point out the similarities to Mononoke.
> Giving credit to those that came before us is now easier — and more important — than ever. There is no longer an excuse.
simonw 13 hours ago [-]
So you'd like Anthropic to update their announcement so it credits Mononoke in addition to crediting Sapling?
Fair enough.
sunshowers 13 hours ago [-]
Yes, and I would like it to explicitly say that their data model is very similar to Mononoke's, and avoid even the appearance of the idea that Claude came up with it de novo.
simonw 13 hours ago [-]
I didn't think that their article implied Claude had come up with the design.
I thought it was likely that their team of 7 included ex-Meta and ex-Google people with knowledge of those company's internal version control solutions.
menaerus 3 hours ago [-]
> and avoid even the appearance of the idea that Claude came up with it de novo.
If we s/Claude/Few Meta engineers/, would de novo still hold?
gorgoiler 13 hours ago [-]
I saw RMS give a talk once on Free Software. He wore three quarter length cargo pants and sandals with no socks, and spent a good amount of the talk picking a scab off his shin to the point where he started bleeding. It got on his fingers. No one said anything.
There are things in this world that people can and should be stopped from doing, and there which anyone can do unhindered if they so wish. In between those two boundaries there’s a standard of decent behavior that one should strive for in order to be considered a good person.
Giving credit isn’t required, but many consider it basic decency.
13 hours ago [-]
LogicalRisk 13 hours ago [-]
"I built a thing in the style of X" (which is even more charitable than what Anthropic piece said) is different than "I incorporated X from Y to build a thing this way."
teaearlgraycold 14 hours ago [-]
> hard-won lessons in mind
Another Claude-written article
stasomatic 11 hours ago [-]
Yea, hard-won, who talks like that... oy vey
bayindirh 14 hours ago [-]
> Another Claude-written article
This is the load-bearing seam.
wiml 13 hours ago [-]
It's not just obvious, it's evident.
bayindirh 13 hours ago [-]
Why it's important: It's the honest take.
MeetingsBrowser 14 hours ago [-]
When Fable first released, I had it "one shot" a project using an initial spec and a team of agents. I was blown away with how good the results were.
And then I noticed some links to images from random small github projects and realized somewhere along the way agents had searched for existing projects and copied almost everything.
I wonder if something similar happened here on a larger scale.
bushido 13 hours ago [-]
GitHub is to agents what StackOverflow was to developers back in the day (in a meme kinda way).
More seriously, I'm curious if you checked on the licensing of said codebases, were they permissive or strong copyleft?
NewJazz 4 hours ago [-]
Lolol nobody is checking that shit. Do most people prompting AIs even understanf what copyleft is?
hetspookjee 14 hours ago [-]
I think a lot is memory really. For fun you can try pointing just about any model against the older kaggle challenges and it’ll come up with more or less the winning solution from memory. It’s here where you see a Gemma 27b perform about the same as opus 5.5
elmer2 14 hours ago [-]
It's the same for CTF challenges. Many get 'solved' because a writup was found and steps were followed.
Terr_ 11 hours ago [-]
I asked it to sketch out something from a research paper, to later find out most of the code was translated from another implementation of the paper.
peterbower 7 hours ago [-]
It is like a vacuum cleaner.
6879346626 14 hours ago [-]
Is this what is expected of how to cite sources? Because if this is the expectation nobody does it.
There are two cases and only one is an issue for citations:
- Mononoke was directly consulted and not attributed
- Concepts were reinvented or used as a derivative (Sapling cited Mononoke and implements something similar)
For one I know Sapling, Git, and Jujutsu, but up to today I have never heard of Oxide or Mononoke and I am not exactly compelled to look up the research papers of any of them, at most I read through their high level documentation.
sunshowers 14 hours ago [-]
To be clear, Mononoke is the server-side part of the Sapling system as deployed at Meta. Scion appears to be very closely inspired by Mononoke.
henrymerrilees 12 hours ago [-]
Sapling does a bit more than cite Mononoke. And Oxide was not involved in Mononoke's development. These are kind of details that would not be expected of a casual user. However, when valuations soar into the trillions, we can expect companies with dedicated writing staff to communicate attribution properly.
henrymerrilees 5 hours ago [-]
*the kind of... sorry, that typo made my answer sound a bit sarcastic which was not the intent.
evolve-maz 10 hours ago [-]
I've always worried about the level of "IP laundering" which is done by LLMs.
While that's not technically the case here (given Mononoke's license) it is still quite negative in my view. We will have fewer people wanting to share ideas openly since LLMs are so good at not only finding them, but also not attributing them. This also leads to a loss in diverse collaboration, which is probably one of the biggest drivers for good software.
Right now there's probably a lot of these "novel" ideas floating around which models can present as their own, but that's raiding our coffers with no inbound future stream.
pmbauer 13 hours ago [-]
Anthropic effectively doing a "git clone" with extra steps here.
Uncharitable?
I don't think so.
If this were about functionality, why not use Sapling/Mononoke or extend it and contribute back to the open source community?
Because that doesn't juice the IPO valuation.
thewhitetulip 9 hours ago [-]
The most horrible thing is they don't read their own code. They read the "intent"
On one hand they say AI will end the world. On other hand they say well, we just merge PR by reading the "intent" which I think is the AI summary
cmrdporcupine 10 hours ago [-]
I think it's worse than just IPO juicing. I think it's "look what our agents can do" and also pushing more on the "software engineers are becoming obsolete" line that Dario public-pronounces. Getting high on their own supply.
Projects seem to be getting approved or pushed there purely on the basis of showing off that they can vibe-build infrastructure.
Which.. yes, you can. But now maintain it. Understand it. Manage thousands of people using it. Those are part of the reasons we re-use proven solutions rather than rolling our own -- not just because the coding part is hard.
Same thing happened with the Bun rewrite. "Oh wow, look what we can do! What's that, nobody adult supervision telling us that this is a bad idea? Let's go!"
Havoc 47 minutes ago [-]
Can’t say I have much sympathy for big tech and their employees whining about stolen IP, data and insights.
They’ve been the most egregious offenders as of late
mad_vill 14 hours ago [-]
Oh wow. Being pretty new to meta and seeing the anthropic announcement I was thinking to myself “wow that sounds a lot like sapling”.
I guess that is the case.
schleck8 12 hours ago [-]
They mentioned Sapling in the post and even hyperlinked it
cmrx64 14 hours ago [-]
hi rain :) Anthropic’s post reminded me a lot of your work, glad this is getting some visibility. do you think the “time has come” for some of these ideas that previously may not have spread because they need too much learning to appreciate? I wonder what new norms will emerge in creative collectives.
sunshowers 14 hours ago [-]
Thanks! Yeah, I think overall several more organizations are going to reach the scale that Meta reached a while ago. I'm not sure if they'll go in the monorepo or polyrepo direction, though; I can see arguments both ways.
alper 2 hours ago [-]
I thought the Anthropic project did read as a tall order to be completed inside a company from scratch.
notnullorvoid 6 hours ago [-]
I don't expect Scion's design to be very good, and the implementation is likely pretty bad, unless it is a near 1 to 1 copy of existing open source.
The amount of training data for incremental/derived data systems is quite sparse in the grand scheme of things. I've seen little to no improvement over the last year when testing new models in that area, they always make some pretty bad mistakes, and what is correct is very inefficient.
sunshowers 5 hours ago [-]
Thanks. Indeed, the Scion design is quite good because Claude was pointed directly to Mononoke and asked to analyze the code base and copy it. I am very proud of what our team built by making contrarian bets at exactly the right spots.
wakamoleguy 14 hours ago [-]
Agents tend to act differently than humans here. I would have expected most humans to notice and acknowledge the connection. Passing it off as one's own insight definitely feels like plagiarism. Agents are generally not equipped to deal with this type of thing, since looking for prior art is a detour from their goal or an exercise in post-hoc rationalization of knowledge built into the weights.
Whether it is plagiarism or hacking, the buck has to stop with the operators. It's not a good look.
8 hours ago [-]
designwithjawad 2 hours ago [-]
As a product designer, what do I need to know?
arjie 13 hours ago [-]
The company is predominantly FAANG people. They’re going to just rebuild what they know. I don’t think you need any kind of conspiracy here. One year into ad tech and everyone was using HyperLogLog. Everyone was using 20 years ago LMAX-style disruptor, and single threaded event queue powered by loop called governor. It’s just what happens. People carry in their head what they know.
IP maximalism is the present zeitgeist but that’s just how knowledge flows.
sunshowers 13 hours ago [-]
Help me understand. When you write a post about designing a system, do you not do some basic diligence about what existing systems you resemble?
Open source is not IP maximalist by any means. I am merely asking that the appropriate credit be provided.
arjie 13 hours ago [-]
No. I actually don’t do that and I don’t think that all human work should be O(n) deduplication gated.
It isn’t something I care very much about, except when it’s used as a certificate of achievement. I almost never do literature searches with an attempt to dedupe. I only ever do it to see if someone has explored something better than I have. That’s often the case but other times it is not.
hyperpape 12 hours ago [-]
If they knew it from working at Meta, it wouldn't be O(n), it would be O(1).
I agree that there will be cases where you implement tricks that others have had before you, without knowing. A few years ago, I had a delightful idea, experimented and wrote it up. One commenter directed me to a post that described it as "folk knowledge". There will be other cases where you encountered an idea before, but didn't remember it.
Still, crediting what you can easily know you cited is good. Even aside from credit, comparisons to existing systems are part of how we understand what a system does.
sunshowers 13 hours ago [-]
Okay, well, I actually do think it is part of the basic duty of an engineer to do that.
gavinray 13 hours ago [-]
I do, solely on the basis of "there are a lot of smart(er) people in the world, and very few truly novel problems".
The first step in approaching anything difficult is seeing whether someone else has already solved it.
arjie 13 hours ago [-]
I see. Well, I suppose we differ on this matter of opinion. May the chips land where they may.
kccqzy 13 hours ago [-]
Absolutely not. When I create projects I only credit other existing systems if they are a direct inspiration. If I independently come up with the same idea as someone else I see no reason for me to do the due diligence and credit them.
sunshowers 13 hours ago [-]
Okay, well, in this case (even accepting your argument) I've been told there is direct overlap in contributors so there is no feasible way they couldn't have known.
mycoliza 12 hours ago [-]
Oh, is that why they only listed the initials of the contributors in their blog post? The plot thickens!
simonw 11 hours ago [-]
That's a pretty weird detail of this. Anthropic don't usually do that - other posts have named the contributors in full, for example:
The politics of AI are getting pretty heated and I suspect engineers at key companies are increasingly reticent to get doxxed and harassed.
zamalek 12 hours ago [-]
What happened is that it wasn't a FAANG engineer replicating what they know, rather one asked Claude what to do and it told them to replicate what it knew.
That's how the knowledge that it was a facsimile was lost.
otterley 10 hours ago [-]
Oh yeah? How do you know? Were you sitting next to them? Have any screen recordings you can share?
cmrdporcupine 11 hours ago [-]
Rebuilding what you know from the previous FAANG you worked at is expected on the whole.
What is douchey is presenting the whole thing as some sort of intense novelty and stroke of genius required in the age of agents when it's just doing what the other BigCo companies did to scale their monorepos already. Like a decade (or more) ago.
The reality it is when I first read the thing I was -- as a Xoogler -- like "Oh, yeah, so, like Piper & SrcFs and the stuff that Meta also does" and really didn't understand why they wouldn't have started with something off the shelf inside of vibe-conjuring a clone that they would then have to support
To me it speaks ill, again, of the engineering culture at Anthropic. "We do these things not because they are useful, but because they are novel."
Don't they have better things to do at Anthropic than building their own source control or rewriting Bun from scratch in another language? Like, y'know, basic site reliability? Or a coding harness that doesn't eat multiple GB of RAM?
simonw 11 hours ago [-]
Which off the shelf solution would you have used?
Google's isn't open source. Meta's is an option, any others you would consider?
cmrdporcupine 11 hours ago [-]
Well, sure, the article here and Meta's, etc.
Even Google started with Perforce originally and only forked/rebuilt when it stopped being able to scale. And did so in a very incremental fashion.
simonw 7 hours ago [-]
When Google built their alternative to Perforce it wasn't possible for a team of seven to build a new system from scratch and port the entire company over to it in five months.
cmrdporcupine 3 hours ago [-]
I'd wager it's still not possible.
You can write it, but can you maintain it? Can you even fully understand what was built? Can you staff the team to manage the bugs? The potential data loss?
micimize 12 hours ago [-]
Whether the ant post amounts to plagiarism, it is definitely somewhere between extremely lazy and intentionally misleading. No particular idea has its inspirations or origins traced or cited, there is no comparison to comparable tooling or discussion of prior work, and there is absolutely nothing to actually position or discuss the project in a manner that might lead to "Contributing to the wider ecosystem."
If only they had had a claude write the post, it'd be full of citations, inspiration credit etc.
flakiness 14 hours ago [-]
Can anyone explain the difference between sampling vs mononoke?
Edit: I was too lazy. From the README: "Mononoke is the server built for the Sapling Source Control System"
sunshowers 14 hours ago [-]
It's a bit complicated:
1. Sapling can refer to either the system as a whole, or just the client people use (`sl`, forked from `hg`).
2. EdenFS is the virtualized FUSE filesystem (closely modeled after SrcFS at Google).
3. Mononoke is the distributed system that implements the backend server storage + the API they describe for checkout-less writes.
All three come together to form the core of the Meta source control experience. From what I can tell, Anthropic built its own version of 1 and copied 2 and 3, plus also built out a code review tool separately.
MiroslavPokorny 4 hours ago [-]
Says a lot about Anthropic given they havent open sourced Scion.
matt3210 13 hours ago [-]
Did they know or did the agents just have that in training data?
duhhhhh1212 14 hours ago [-]
Damn. It’s hard not to see it as plagiarism. I can only imagine the surprised Pikachu face you must have made reading that announcement from Anthropic.
dboreham 14 hours ago [-]
Similar to how git was "inspired by" BitKeeper and Monotone?
sunshowers 14 hours ago [-]
I do not think Linus has ever claimed that Git wasn't following in the footsteps of those systems.
InsideOutSanta 14 hours ago [-]
Neither has Anthropic. They mention Sapling as an inspiration.
sunshowers 14 hours ago [-]
Scion is far more similar to Mononoke than Git is to BitKeeper.
kccqzy 13 hours ago [-]
GNU is extremely similar to the proprietary Unix flavors that came before it. Is that a problem?
sunshowers 13 hours ago [-]
No? Did you read my post? The problem is the lack of attribution.
kccqzy 13 hours ago [-]
I fail to see why they also need to mention Mononoke when they already attributed inspiration to Sapling, when Mononoke is a component of Sapling.
zamalek 12 hours ago [-]
Having your AI rewrite a projects is a very different level of "inspiration" than normal inspiration.
What would you classify the rust rewrite of Bun as? Because that's what we're looking at here.
nomel 14 hours ago [-]
It's totally wild seeing entities being criticized for writing software based on/very similar to other open source software. Ffs, this used to be what open source software was about. It was actually the fucking point [1].
> And, to be clear, I do not have a problem with Anthropic copying Mononoke’s design. (This is part of why we put out open source — to benefit humanity through the spread of good ideas.)
Followed by:
> But as LLMs continue to assist in designing new systems, they are likely to index heavily on existing designs. Mononoke is open source and in every LLM’s training set. If you ask Opus 5.5 or GPT-6 Astra (thanks Matthew Fowle) about prior art, they both immediately point out the similarities to Mononoke.
I think it is the duty of any responsible engineer to look for and cite prior art and related work, if only to compare notes.
rcxdude 12 hours ago [-]
> I think it is the duty of any responsible engineer to look for and cite prior art and related work, if only to compare notes.
The patent system actively discourages this by tripling damages for knowing violation, as does the copyright system by the threat of accusations of copying that are best defended against by 'I never saw it'. It's much safer, legally speaking, to explicitly avoid doing this.
nomel 11 hours ago [-]
If you've ever worked in a big engineering org, they always give very specific instructions to NOT look for patents/prior work if you are researching, or have come up with, an idea.
nomel 13 hours ago [-]
I did, but those additional requirements are outside the licenses and the legal framework, and more wanting them to match your personal vibe/moral framework around open source. There's case law around this going back to the 80's. A third party is allowed to independently produce, and be extremely inspired, by open source code, and then can license it however they want. Protecting copied functionality is NOT PROTECTED. Again, that was the fucking point. You could, for example, swap out compatible software if you wanted! You wouldn't be tied to one vendor!
Open source licensing protects direct copying. It does NOT protect re-implementing from scratch. See all the lawsuits around Java. You should not be using open source software if you don't want others to read your source!
sunshowers 13 hours ago [-]
Yes, I am aware that that credit is not required by law, nor in this case do I wish for credit to be required by law. I am appealing to norms, to basic decency.
webdood90 13 hours ago [-]
> I think it is the duty of any responsible engineer to look for and cite prior art and related work, if only to compare notes.
Maybe 5 years ago that would be the expectation but all etiquette has been thrown out the window with AI. I think you expect too much.
duhhhhh1212 14 hours ago [-]
It’s wild to see people rudely commenting without reading the article. It used to be that people read the article before opening their mouth.
nomel 11 hours ago [-]
I read it, and it appears I interpreted the situation correctly. See sibling.
14 hours ago [-]
thewhitetulip 9 hours ago [-]
Did Linus also have access to the BitKeeper source code?
ezst 7 hours ago [-]
So it's mercurial all the way down, then?
gizajob 14 hours ago [-]
This headline is very confusing for fans of Basic Channel / Chain Reaction records.
fhdkweig 14 hours ago [-]
It is also confusing to fans of Studio Ghibli.
stavros 12 hours ago [-]
And fans of Toyota.
cmrdporcupine 9 hours ago [-]
Monolake, not Mononoke?
forrestthewoods 14 hours ago [-]
I really really wish Sapling could be used externally. It is by-and-large TheWay.
Centralized, virtual file system, all commits automatically on cloud, large binary files totally fine, Mercurial CLI. It’s fantastic. People who only use Git really don’t understand how much better things can be.
Shish2k 14 hours ago [-]
Working for a few years at Facebook, and then returning to the normal world, it was like stepping back in time -- I really think that the mainstream git culture (by which I mostly mean "github") achieved a local maxima around 2012, and has been holding the industry back since :(
(Like GH finally has some support for stacks, 15 years late, with an implementation which is somehow more complicated and less powerful than the prior art...)
ezst 7 hours ago [-]
As someone who came to terms with dying on the mercurial hill, someone who forgot more gitisms than there are to learn in the first place because git cli's inconsistencies and poor taste hurts and neuters my memory, yeah, I relate.
Most people aren't VCS nerds, most are of a generation that only ever knew git. Learning git was a prerequisite they engaged with out of necessity, not by choice, not after exploring the problem space and pondering about different approaches. Learning git was difficult, and the more they learned the more they discovered they had more to learn, including the little things, obscure details and a plethora of paper-cuts. For them (again, I suspect they are a majority), the perspective of learning something else comes with the expectation of a similar daunting effort and steep cognitive reluctance.
That's the paradox of git as I see it: had it been a simple and easy system to pick-up, it would probably have been replaced, and/or the VCS space would be much more competitive.
forrestthewoods 14 hours ago [-]
> [Git/GitHub] achieved a local maxima around 2012, and has been holding the industry back since :(
I agree so hard. It’s inertia is too strong and we’ve stalled progress for a decade.
esafak 12 hours ago [-]
It really isn't too strong; people here are craving for a Github killer.
steveklabnik 12 hours ago [-]
The startup I work at is working on exactly this. So I agree :)
(Okay not literally exactly, for example, jj/git not hg. But the vibe is the same.)
Qwuke 12 hours ago [-]
Hey Steve, I read your recent blogpost on the ERSC site - have you or David (or anyone else at ERSC) felt like there's a Rusty-enough topic around what your startup is doing to give a talk at Rust NYC/Rust conference soon?
There's not really a VCS meetup in NYC, but I got to hear David and others at Rust NYC wax eloquent about Sapling vs Piper and Buck vs Bazel a couple of times, so I think the audience is there!
steveklabnik 12 hours ago [-]
Rust NYC is certainly one of the things I have missed since moving to Austin :)
Maybe, we'll see. I think one thing that's tricky these days with startups is that being super transparent technically about what you're up to is much harder to justify. Look what happened with Origin: tobi vibed up a copy over the weekend after it was announced. Obviously there's balance to be found here, but that's partially why we play our cards a little closer to our chest.
forrestthewoods 12 hours ago [-]
I am delighted to hear this. Please do not forget support for Windows.
Let me share with you my previous screeds on this topic:
They’re semi-adjacent in that they are about build systems that don’t suck. (buck2 is also mostly TheWay but also not usable outside of Meta and has room for improvement).
Save us from Git and CMake Steve! Please I beg you!!
steveklabnik 11 hours ago [-]
To be clear, we are not building a build system, just a VCS. :) we do use buck2 internally though.
I do care about Windows, having used it for years at this point. The client is git or jj, so Windows support is there for sure. For some of the other stuff we’re doing, we’ll see what customer demand is like. Obviously there are tons of devs on Windows, but we also have to make calls about priorities too.
bill10 5 hours ago [-]
and then what?
j45 10 hours ago [-]
Appreciate the perspective, more to learn from.
exe34 13 hours ago [-]
Once it's been through the slopotron, the provenance is laundered away.
Another effort I couldn't have done without Astra. Even more so. All I did, the only reason I got cited, was because I'd asked Astra.
sunshowers 14 hours ago [-]
Thanks!
Mawr 6 hours ago [-]
> I think the level of imitation goes well beyond having similar insights.
Uhuh:
> To solve this, we moved the diff-friendly representation to being derived data (i.e. an index).
Coming up with the idea of an index is not a novel thought. Nothing Anthropic described in their post was anything unexpected that one working from first principles that their problem presented would not have come up with.
sunshowers 5 hours ago [-]
Except the resemblances extend to the entire server, not just the idea of derived data. The idea of derived data becomes apparent only once you have bought into the core Mononoke insight about Bonsai changesets, later copied wholesale and presented as the core Scion insight about Scion changesets.
As I suspected at the time of writing the post (but could not verify) and as was later confirmed in private correspondence to me, Anthropic did not come up with this from first principles. Claude was pointed directly at Mononoke and asked to re-engineer it.
verdverm 9 hours ago [-]
Not to be confused with Google's Scion (kubernetes based openclaw platform)
“Source control for huge monorepos has prior art, notably Google’s Piper and Meta’s Sapling, so we were able to build Scion with their hard-won lessons in mind.”
Mononoke is a component of Sapling, as I understand it. In fact, the repo link provided in this article links to a subdirectory of the Sapling repo. I’m not totally sure I understand the complaint.
However if my new car's engine is a near copy of the one in the Corvette, and my announcement gushes about the engine using the same terms the Corvette design team invented...
Then can you see how a mention might be nice?
If you take the statement: "Pretty much 100% of our code is written by Claude Code" (https://x.com/bcherny/status/2015979257038831967)
And you take the statement: "Mononoke is open source and in every LLM’s training set."
There is not much resemblance with any cleanroom implementation anymore.
Perhaps if there is any ambiguity, it's intentional on the part of caution around intellectual property provenance.
To be uncharitable, this is marketing for Anthropic's impending IPO.
Even if they weren't making a lot of PRs, Github has been melting down enough this year that they've probably encouraged customers like Anthropic to move off.
Git is just not suited to truly large monorepos, and it won't ever be. That's totally fine, of course. Less of a monoculture here would be great.
The intended audience is obvious from first paragraph alone:
> At Anthropic, coding is no longer the bottleneck to building systems or doing research. Claude is able to autonomously iterate on research loops, micro-optimize complex programs, and build new services from the ground up at a speed and scale we haven’t seen before. The bottleneck, now, is all of the developer infrastructure built for human scale: source control, builds, and CI.
Translation:
> We are so far at the cutting edge we need to build shiny new infrastructure to replace the smelly "human scale" cruft. How could you /not/ value us at $2T?
Claude.ai
Claude Code CLI
Claude Code Desktop (multiple platforms)
Claude Code on the web (aka Claude Code Cloud)
Claude Design
The Claude API
Claude SDK libraries
Cowork (just retired and merged into regular Claude)
Claude Tag
Claude for Legal
They ship a lot of products.
Even that one repo is breaking GitHub this week…
That’s not surprising to me - they designed and built it using their LLMs, of course it’s duplicating prior work. That’s essentially the core of what LLMs do, after all.
Mononoke and Sapling are open source systems. Why shouldn't new software take inspiration from how they are designed?
> And, to be clear, I do not have a problem with Anthropic copying Mononoke’s design. (This is part of why we put out open source — to benefit humanity through the spread of good ideas.)
> But as LLMs continue to assist in designing new systems, they are likely to index heavily on existing designs. Mononoke is open source and in every LLM’s training set. If you ask Opus 5.5 or GPT-6 Astra (thanks Matthew Fowle) about prior art, they both immediately point out the similarities to Mononoke.
> Giving credit to those that came before us is now easier — and more important — than ever. There is no longer an excuse.
Fair enough.
I thought it was likely that their team of 7 included ex-Meta and ex-Google people with knowledge of those company's internal version control solutions.
If we s/Claude/Few Meta engineers/, would de novo still hold?
There are things in this world that people can and should be stopped from doing, and there which anyone can do unhindered if they so wish. In between those two boundaries there’s a standard of decent behavior that one should strive for in order to be considered a good person.
Giving credit isn’t required, but many consider it basic decency.
Another Claude-written article
This is the load-bearing seam.
And then I noticed some links to images from random small github projects and realized somewhere along the way agents had searched for existing projects and copied almost everything.
I wonder if something similar happened here on a larger scale.
More seriously, I'm curious if you checked on the licensing of said codebases, were they permissive or strong copyleft?
There are two cases and only one is an issue for citations: - Mononoke was directly consulted and not attributed - Concepts were reinvented or used as a derivative (Sapling cited Mononoke and implements something similar)
For one I know Sapling, Git, and Jujutsu, but up to today I have never heard of Oxide or Mononoke and I am not exactly compelled to look up the research papers of any of them, at most I read through their high level documentation.
While that's not technically the case here (given Mononoke's license) it is still quite negative in my view. We will have fewer people wanting to share ideas openly since LLMs are so good at not only finding them, but also not attributing them. This also leads to a loss in diverse collaboration, which is probably one of the biggest drivers for good software.
Right now there's probably a lot of these "novel" ideas floating around which models can present as their own, but that's raiding our coffers with no inbound future stream.
On one hand they say AI will end the world. On other hand they say well, we just merge PR by reading the "intent" which I think is the AI summary
Projects seem to be getting approved or pushed there purely on the basis of showing off that they can vibe-build infrastructure.
Which.. yes, you can. But now maintain it. Understand it. Manage thousands of people using it. Those are part of the reasons we re-use proven solutions rather than rolling our own -- not just because the coding part is hard.
Same thing happened with the Bun rewrite. "Oh wow, look what we can do! What's that, nobody adult supervision telling us that this is a bad idea? Let's go!"
They’ve been the most egregious offenders as of late
I guess that is the case.
The amount of training data for incremental/derived data systems is quite sparse in the grand scheme of things. I've seen little to no improvement over the last year when testing new models in that area, they always make some pretty bad mistakes, and what is correct is very inefficient.
Whether it is plagiarism or hacking, the buck has to stop with the operators. It's not a good look.
IP maximalism is the present zeitgeist but that’s just how knowledge flows.
Open source is not IP maximalist by any means. I am merely asking that the appropriate credit be provided.
Written about it here before: https://wiki.roshangeorge.dev/w/Blog/2026-09-08/Intellectual...
It isn’t something I care very much about, except when it’s used as a certificate of achievement. I almost never do literature searches with an attempt to dedupe. I only ever do it to see if someone has explored something better than I have. That’s often the case but other times it is not.
I agree that there will be cases where you implement tricks that others have had before you, without knowing. A few years ago, I had a delightful idea, experimented and wrote it up. One commenter directed me to a post that described it as "folk knowledge". There will be other cases where you encountered an idea before, but didn't remember it.
Still, crediting what you can easily know you cited is good. Even aside from credit, comparisons to existing systems are part of how we understand what a system does.
The first step in approaching anything difficult is seeing whether someone else has already solved it.
- https://www.anthropic.com/engineering/a-postmortem-of-three-...
- https://www.anthropic.com/engineering/how-we-contain-claude
That's how the knowledge that it was a facsimile was lost.
What is douchey is presenting the whole thing as some sort of intense novelty and stroke of genius required in the age of agents when it's just doing what the other BigCo companies did to scale their monorepos already. Like a decade (or more) ago.
The reality it is when I first read the thing I was -- as a Xoogler -- like "Oh, yeah, so, like Piper & SrcFs and the stuff that Meta also does" and really didn't understand why they wouldn't have started with something off the shelf inside of vibe-conjuring a clone that they would then have to support
To me it speaks ill, again, of the engineering culture at Anthropic. "We do these things not because they are useful, but because they are novel."
Don't they have better things to do at Anthropic than building their own source control or rewriting Bun from scratch in another language? Like, y'know, basic site reliability? Or a coding harness that doesn't eat multiple GB of RAM?
Google's isn't open source. Meta's is an option, any others you would consider?
Even Google started with Perforce originally and only forked/rebuilt when it stopped being able to scale. And did so in a very incremental fashion.
You can write it, but can you maintain it? Can you even fully understand what was built? Can you staff the team to manage the bugs? The potential data loss?
If only they had had a claude write the post, it'd be full of citations, inspiration credit etc.
Edit: I was too lazy. From the README: "Mononoke is the server built for the Sapling Source Control System"
1. Sapling can refer to either the system as a whole, or just the client people use (`sl`, forked from `hg`).
2. EdenFS is the virtualized FUSE filesystem (closely modeled after SrcFS at Google).
3. Mononoke is the distributed system that implements the backend server storage + the API they describe for checkout-less writes.
All three come together to form the core of the Meta source control experience. From what I can tell, Anthropic built its own version of 1 and copied 2 and 3, plus also built out a code review tool separately.
What would you classify the rust rewrite of Bun as? Because that's what we're looking at here.
[1] https://en.wikipedia.org/wiki/Open-source_software_movement
> And, to be clear, I do not have a problem with Anthropic copying Mononoke’s design. (This is part of why we put out open source — to benefit humanity through the spread of good ideas.)
Followed by:
> But as LLMs continue to assist in designing new systems, they are likely to index heavily on existing designs. Mononoke is open source and in every LLM’s training set. If you ask Opus 5.5 or GPT-6 Astra (thanks Matthew Fowle) about prior art, they both immediately point out the similarities to Mononoke.
I think it is the duty of any responsible engineer to look for and cite prior art and related work, if only to compare notes.
The patent system actively discourages this by tripling damages for knowing violation, as does the copyright system by the threat of accusations of copying that are best defended against by 'I never saw it'. It's much safer, legally speaking, to explicitly avoid doing this.
Open source licensing protects direct copying. It does NOT protect re-implementing from scratch. See all the lawsuits around Java. You should not be using open source software if you don't want others to read your source!
Maybe 5 years ago that would be the expectation but all etiquette has been thrown out the window with AI. I think you expect too much.
Centralized, virtual file system, all commits automatically on cloud, large binary files totally fine, Mercurial CLI. It’s fantastic. People who only use Git really don’t understand how much better things can be.
(Like GH finally has some support for stacks, 15 years late, with an implementation which is somehow more complicated and less powerful than the prior art...)
Most people aren't VCS nerds, most are of a generation that only ever knew git. Learning git was a prerequisite they engaged with out of necessity, not by choice, not after exploring the problem space and pondering about different approaches. Learning git was difficult, and the more they learned the more they discovered they had more to learn, including the little things, obscure details and a plethora of paper-cuts. For them (again, I suspect they are a majority), the perspective of learning something else comes with the expectation of a similar daunting effort and steep cognitive reluctance.
That's the paradox of git as I see it: had it been a simple and easy system to pick-up, it would probably have been replaced, and/or the VCS space would be much more competitive.
I agree so hard. It’s inertia is too strong and we’ve stalled progress for a decade.
(Okay not literally exactly, for example, jj/git not hg. But the vibe is the same.)
There's not really a VCS meetup in NYC, but I got to hear David and others at Rust NYC wax eloquent about Sapling vs Piper and Buck vs Bazel a couple of times, so I think the audience is there!
Maybe, we'll see. I think one thing that's tricky these days with startups is that being super transparent technically about what you're up to is much harder to justify. Look what happened with Origin: tobi vibed up a copy over the weekend after it was announced. Obviously there's balance to be found here, but that's partially why we play our cards a little closer to our chest.
Let me share with you my previous screeds on this topic:
* https://www.forrestthewoods.com/blog/dependencies-belong-in-... * https://www.forrestthewoods.com/blog/using-zig-to-commit-too...
They’re semi-adjacent in that they are about build systems that don’t suck. (buck2 is also mostly TheWay but also not usable outside of Meta and has room for improvement).
Save us from Git and CMake Steve! Please I beg you!!
I do care about Windows, having used it for years at this point. The client is git or jj, so Windows support is there for sure. For some of the other stuff we’re doing, we’ll see what customer demand is like. Obviously there are tons of devs on Windows, but we also have to make calls about priorities too.
Another effort I couldn't have done without Astra. Even more so. All I did, the only reason I got cited, was because I'd asked Astra.
Uhuh:
> To solve this, we moved the diff-friendly representation to being derived data (i.e. an index).
Coming up with the idea of an index is not a novel thought. Nothing Anthropic described in their post was anything unexpected that one working from first principles that their problem presented would not have come up with.
As I suspected at the time of writing the post (but could not verify) and as was later confirmed in private correspondence to me, Anthropic did not come up with this from first principles. Claude was pointed directly at Mononoke and asked to re-engineer it.
https://googlecloudplatform.github.io/scion/overview/