Comment by teiferer
8 hours ago
2/3 down the article I gave up. What are you trying to tell me? What is this revolution about? How are agent things fundamentally different and how are they being solved? What is this "second coming"?
Also, why "second"? Was git the first? But then what about all the other things before it? CVS was huge before, for better or worse.
The author has a severe case of being oblivious to the fact that not everyone knows what they know, or views the world the way they do. As shown for example by their assertion that "we all stopped coding manually around December 2025", as if AI critics don't exist.
As a result you have to look at what they share about themselves to infer where they're going with things, because they never explicitly state it. All traits amplified for the worse when people use chat bots too much, in my experience.
Scrolling to the end, they write:
> In 2005 a dozen version control systems fought to replace Bitkeeper, and the winner ended up ruling software development for two decades. Nobody in that race predicted that the decisive factor would be a hosting site with a social network on top.
They had a VCM start-up around their own technology, then Git displaced all other technologies, and then Github ate everyone's lunch. So the first revolution not-so-coincidentally overlaps with the time they ran a SCM startup, and they ignore anything that happened before they entered the space. Funny that.
So the implicit context is the business case and programmer culture around VCM systems.
Meaning the second revolution is anything that challenges Git and Github.
They also keep bringing up gigantic mono-repos and how Git can't handle the scale at which AI increases the amount of code bloat or commits.
So what I read between the lines is that they believe that there are enough people who don't want to actually address the automated hyperscaled Wirth's Law in the room, and are willing to pay services and technology to adapt around it instead.
That's their new business case that will power the second "revolution".
Which sounds horrifying to me, and the worst part is that I can actually see it happening if this AI bubble doesn't pop fast enough.
I'm not an AI critic (I use it for autocompletions and asking some questions) but still write my own code.
I'm not sold on the agentic workflows that AI labs are pushing and haven't yet figured out if/when/how to integrate it into my workflow.
There are two general things I'm weary of with agentic workflows compared to the chat-based workflows:
1. the agentic workflows are much more inclined to go and do the thing rather than involve you in the loop -- e.g. if you are trying to design/plan something;
2. the agents are happy to go and run any command -- you can get them to prompt to confirm the action, but they could easily do something like wipe your home directory, install a random package, or something else.
> They also keep bringing up gigantic mono-repos and how Git can't handle the scale at which AI increases the amount of code bloat or commits.
I don't understand why everyone keeps saying "AI" can write code and somehow the rest of the world can't keep up when we haven't even solved a more basic problem that "AI" itself can't keep up with the code it writes. Claude on the web has to search and scan to find the relevant parts and for whatever reason there is a size limit to how much context I can give to claude and it is laughably low. Shouldn't we be trying to see if it can keep up with itself first? Also not all project files fit in the "context" so once you are in search mode anyway, why is the limit so arbitrarily small? Why not allow at least about 1GiB of plain text? Claude tells me 1 GiB is 250 million tokens and it is too much for current capacity but I thought the whole point of project files was not all of it is in the context window or whatever. It says it is in search mode anyway... So only search for the relevant bit and put only that in the context. You can limit your search to a particular submodule or something but still have broader context where required
>> > They also keep bringing up gigantic mono-repos and how Git can't handle the scale at which AI increases the amount of code bloat or commits.
I also doubt very much that the average monorepo is bigger than the linux kernel. While the latter is a single software as things go, it contains quasi independent subsystems. And those repos don't usually have a merge flow that is as smooth as the Linux kernel. If you have chaotic process, then the result won't be very good and it's not git's fault.
> "we all stopped coding manually around December 2025", as if AI critics don't exist.
It was obviously a generalization. The number of AI critics (which eschew AI for coding) are a minority. I would say tiny minority at this point.
> Wirth's law is an adage on computer performance which states that software is getting slower more rapidly than hardware is becoming faster.
So, datacenters will fix that.
Is this an ironic "tell me you don't learn from history without telling me you don't learn from history" joke?
I read it yesterday and had a discussion on Git Discord. The author believes people will slop Google-scale repos and Google-type SCM is needed now. Some say a USB-C cable has more computing power than the Apollo program, so maybe.
Banks, airlines, power infra, and other high-stakes customers got computerized in the 60s and 70s. IBM still holds a chunk of that market. Microsoft and Google are not going anywhere. So I guess all that slop will serve something less critical than all of the above.
USB-C charger, not cable.
My Claude says that a decent cable has an ARM Cortex-M0 running at tens of MHz, while the Apollo Guidance Computer was 1MHz. I am afraid to think what a charger packs.
Same. I had no idea if the article was for me and what I was going to get out of it, so I just gave up.
I gave up after 5 seconds.
Blog authors need a straight up TL;DR because I ain't reading all of that.
Well, I like a good story, there is big potential in this long form stuff that's more than a tweet or two. There can be a great reward when the crux is revealed after building a foundation and then looking at it from different angles. But if that crux never comes then it's really just leaving a feeling of disappointment and waste of time that I invested into reading all this. And that does a disservice to everybody else writing blog posts.
I think that's part of why tiktok and yt shorts really took off. It's so short that if you realize it's bad and move on then your wasted investment in terms of time and energy was negligible. (Times 200 that's a different story but signal-to-noise is still high enough that people accept this.)
FWIW, I flagged and downvoted both of your comments. The article is a very interesting overview of the space and recent developments, and having to scroll past multiple paragraphs about your low attention span and reading difficulties to get to some actual technical discussion was quite frustrating!
2 replies →
This is my dilemma. Do I feel like giving up because it isn’t written well or because my attention span is shot? Hard to tell.
It isn't written well.
fully agree. cant even started to read this. just skipped directly.
You should try TikTok, I bet you’ll like it.