Is All Software Converging
I'm Jacob - I run Sancho Studio a software consulting group - I write about cryptography, craft, and human experience.
We are living through a great convergence of software tooling. Tools that used to be distinct: a code editor, a project management suite, a design canvas - are all melting into a single, agentic slurry.
Linear declared issue tracking dead, Figma opened its canvas to coding agents, and Notion shipped autonomous Custom Agents all within weeks of each other, all using nearly identical language. If you squint, everything is beginning to look like everything else.
This is not slop. There are a lot of bright people working on these problems. It seems that the incentive structure (n.b. investors, market) expects an adapt-or-die approach to AI. So firms respond to changing technology, selective forces don't always select for novelty.
Why is so much software starting to offer exactly the same functionality - chat boxes and agents?
At 10,000 ft, agentic coding tools like Claude Code, Codex, Conductor are all offering some variation of a model in a loop. Sometimes exactly the same model with a differing UI. The cost to copy those UIs is rapidly approaching zero.
At 5,000 ft, old tools are rebranding as new tools. Notion was a team wiki and dynamic database, now it's a team of agents on top of a wiki. Linear used to be an issue tracking tool, now it's a team of agents on top of an issue tracking tool.
At 1,000 ft small software is also converging. Every morning there is a new orchestrator or agent sandbox launched on X. Sometimes big players like Vercel will smoke a frog between the hours of 1AM and 4 AM PST and launch "the future of apps" the next morning. Rinse and repeat.
As models get better, the marginal cost of deploying "intelligence" will approaching zero. The technology and its interface get applied to everything. Yes there are costs - token costs - but they are pennies on the human-dollar.
Rhythms in convergence
Seeing this convergence reminded me of the Industrial Revolution. The past does not repeat itself, but it rhymes - Industrialization converted a massive amount of production from craft-based (i.e. human) to machine-based (i.e. mechanized factories). "Ready to Wear" clothing revolution changed humanity's relationship to clothing. Before the mid-1800s almost all clothing was bespoke. Tailors, tanners, hand stitching.
Cheap, mass produced garments are ubiquitous today. So much so that most people don't know anyone who knows anyone who can make a garment entirely from raw materials, let alone individual artisans who specialize in roles in the production process.
One of the great benefits to industrialization is abundance, but it also comes with waste and conformity as the cost to produce plummets. Software is going through an industrialization process - machines are producing computer code which is producing software. Vibe coding produces software yes, but on average we call it slop.
The software is becoming "Ready to Use". Every tool has a chat box, a helpful assistant, and maybe actions it can take. The Universal Interface as I'll call it.
I predict that the Universal Interface will be everywhere. Your grocery delivery app will have an agentic interface to curate your delivery. Your dog walking app will have an agentic interface to schedule your walker. Your dating app will have an agent interface to find you a mate.
This type of 'chatification' of every software tool is cheap, lacks character, and is a minimization of effort. Why build specialized tools when we can throw one massive omni-tool at it? Both standardized clothing and now Ready to Use software are undoubtedly innovative. We lost our craft of tools and our culture around process along the way.
The danger of this convergence is not that it's boring (it totally is). It's that it constrains how we think. Typing on a keyboard versus writing by hand changes how you think.
Tools are thought
Recall a transition in thought that none of us lived through: the move from Roman numerals to Arabic numerals. If your aspirations in life were to count sheep, LXXXVIII was just fine. Trying to divide those sheep into groups of XIX was a lot harder. The tool limits the thought. Rome had the notion of nothing - nihil - but no notion of the number zero. The number zero is a revolution in thought as much as it is in notation.
The antidote to this convergence is more character (n. the mental and moral qualities distinctive to an individual). Be outlandish, inquisitive, divergent - all to explore what it means to create tools in our own image. I'm not so sure that agentic tools are created in our image or even help us think new or different thoughts.
Tools might hold an emotional place in your life. It may have been a camera, or a piano, or a laptop (for me it was a bicycle). The things we create with tools define us and can be seen in the style of a film, or the resonance of a piece of music. Our creations define our species, and the tools used to create are equally important in both the myth and novelty of those creations.
Tools are expression
This reminds me of something Johann Wolfgang von Goethe wrote in his novel Wilhelm Meister's Apprenticeship:
"A man should hear a little music, read a little poetry, and see a fine picture every day of his life, in order that worldly cares may not obliterate the sense of the beautiful which God has implanted in the human soul."
In the story, the theater manager Serlo offers this advice to young Wilhelm Meister as he navigates the tension between his mundane commercial duties and his artistic calling, implying that unless we intentionally curate beauty, the sheer weight of "being useful" will inevitably erode our capacity for wonder.
Goethe wasn't just talking about being a "cultured" person. He was talking about survival. He was warning us that the "worldly cares" (the chores, the productivity, the "agents", the endless copying of the same clothing and condiments) will eventually scorch out our ability to feel the soul because our creations are cheap, easy, and ultimately meaningless.
Is everyone just copying each other? Mostly, yes. Because it's easy and there's intense competitive pressure to do so during times of new technology. Natural selection doesn't always select for the ideal traits - just the ones necessary to survive. Perhaps that's where we are with tool making.
Special thanks to Sophie, Greg, and Michael for reviewing and editing this writing.
Epilogue - Divergent contributions
My father sent me the Goethe quote this morning which prompted me to write this. He runs a company that sits between writing and technology. He's always been an admirer of the arts be it visual or literary. It's helpful to process the work done by thinkers before us - take what we can and leave the rest.
Also this week, a disappointing interaction meeting the CEO of a startup. I expressed concerns about the effects of AI on labor, society, and income inequality. The response I got was "we'll have all these new friends to talk to". If we create a country of geniuses in a data center hopefully we can solve critical frontier science - which would benefit all of humanity...
By definition it would also make human brains as obsolete as the cow's brain.
I expressed that the humanities are a study of the human condition and can teach us a lot about new technologies. The response I received was "the humanities don't know anything about technology".
It broke my heart to hear that. Philosophy is categorized in the humanities. Some of the best scientific thinkers were philosophers first. The humanities are the study of ourselves by ourselves.