From Manuscripts to Machines

Illuminated manuscripts weren’t just beautiful books. They were arguments—extravagant, materially expensive arguments—about what counted as knowledge, which ideas deserved reverence and who had the power to “enforce” this. Gold leaf, pigments, dense ornamentation: none of this was necessary for reading. It was necessary for meaning, signaling and controlling.

Illumination was a cultural stance:

Making the thing was part of knowing the thing, and signaling why knowing this thing mattered.

The Illuminated Manuscript was never passive, ornamental packaging. It was a knowledge artifact that declared the written word a privileged medium of thought, worthy of excessive care.

Computation now holds that role.

With the rise of generative AI, we increasingly think through software—through code, interfaces, simulations, models, and prototypes. But the demos that accompany technological work are usually treated as disposable and lesser-than: functional sketches on the way to a “real” product. There’s nothing inherently wrong with that.

And yet a particular kind of demo breaks this pattern by introducing markedly different qualities. These demos are highly crafted, aesthetically intentional, and—crucially—over-invested in form. Not for polish but for philosophy. Their craft is part of their claim.

These are what I call Illuminated Demos.

What makes a demo illuminated?

An Illuminated Demo is a technological demonstration that treats computational making as a serious way of knowing. Like illuminated manuscripts, it uses a saturation – an excess – of care, detail, beauty, and strangeness to signal that the artifact itself is a mode of inquiry.

Illuminated Demos:

• refuse passive consumption of novelty

• embed arguments in interaction, behavior, and form

• weave a creator’s thoughts and feelings with the technological “climate of opinion” around them (as developed by Becker)

• expose the values our tools embody

• propose alternatives primarily through working artifacts, not merely speculative words (although rare, word-only exceptions do exist)

• live inside an ongoing creative practice, not a one-off trick, alluding to a creator’s ongoing, unfinished business

Where manuscripts elevated writing, Illuminated Demos elevate computational craft as an expressive knowledge discipline.

Why illumination matters now

Our technologies—especially generative ones—shape how we think, decide, feel, and imagine. But the cultural conversation about them is dominated by abstractions: AI disruption, innovation, “the future of X.” Illuminated Demos counter this flattening by grounding ideas in crafted experiences, interfaces and prototypes.

They let us see what a technology believes. They make implicit values tangible. They reveal how tools structure thought, and how alternative designs might structure it differently.

In short:

Illuminated Demos surface and contest the values embedded in our technologies by treating the act of building itself as knowledge.

This essay is simply an attempt to name that practice, gather its examples, and chart the terrain. The definition will evolve—as it should. But for now, consider this an invitation to explore the possibilities that emerge when we treat demos not as scaffolding, but as illuminated artifacts of understanding.

Illuminated Demos, notable examples

This list is incomplete, but it should give you a good idea of what I’m looking for in an Illuminated Demo, and an Illuminated Demoer:

• ’s REAList Stack, and his and Anuraj’s miniaturized, blockchain-powered economy embodied by a robot

• Bret Victor’s Inventing on Principle

• ’s LLM daemons and barefoot developers

• Pippin Barr’s V R 3 game and The Stuff Games Are Made Of book

• Andy Matuschak’s timeful texts

• Robin Sloan Spring ‘83 Protocol, Perfect Edition and Home-Cooked Apps

• Andy Bell and Heydon Pickering’s Every Layout

• Matt Webb’s Machine Supply

• ’s Touching Computers

• ’s Spy Chrome Extension

• ’s Playground

• Arun Venkatesan’s Smart Speaker deep dive

• ’s ValTown

• Devon Zuegel’s Edge Esmeralda

• Dave Rupert’s Grid On Paper and Tale of Three Architectures

• Tl;draw’s Make Real

• The Block Protocol by HASH

And nobody does this better, or more consistently, than Ink and Switch (see Embark, for example).

AI, Fast and Slow

AI is supposed to make us write faster.
But lately, I’ve noticed something different:

AI slows down my writing. And slowing down is good.

A few days ago, I drafted a response to an important email. It was quick, efficient, “good enough.” The kind of reply you send just to clear your inbox and keep the momentum going.

My brain—ever the cognitive miser—was delighted. Check the box. Move on. No need to think harder than necessary.

But right before I hit send, something in me hesitated.
A small, inconvenient whisper: This doesn’t feel quite right.

Normally, I would have ignored that feeling. When you’re in task-positive mode, slowing down feels almost irresponsible. Everything in you wants to keep being productive, optimizing effort as much as possible.

Instead, I fired up ChatGPT to work on the draft.

And that tiny act—the friction of switching contexts—pulled me out of doing mode and into thinking mode.

The LLM didn’t replace my writing. It forced me to engage with it more deeply. I prompted. I revised. I sparred. I clarified. The process demanded more of my attention, not less.

By the end, the message I crafted was far better than the version I almost shipped.

Not because the LLM wrote it for me, but because using the LLM interrupted my speed long enough for me to write with intention.

And still, some questions lingered:

Is this just second-guessing myself?
Is using AI undermining my confidence as a writer?

Sometimes, maybe. These are for sure questions I will continue to live with.
But not in this case.

Here, slowing down was the help I needed.

We talk about AI in terms of acceleration, replacement, amplification, augmentation and automation.

But maybe its real value—especially for writing that matters—is its ability to interrupt our autopilot tendencies, add just enough friction, and return us to a more reflective way of thinking.

Sometimes the smartest thing AI can do is make you go slower.

It’s about craft, and local manufacturing, and rethinking how things get made in a world obsessed with efficiency and sameness.

(Massive thanks to Oto Vavrek and Jiří Dužár, Posedla co-founder, for answering
questions over email.)

What’s cool about Posedla?

A lot of companies claim to make “custom” products. Posedla has actually built their entire business model around it.

Their flagship saddle, the Joyseat, is more than a 3D-printed gimmick. It sits on three pillars that feel refreshing and daring when considering the future of businesses all over the globe:

• Mass customization

• Localized manufacturing

• High-quality craft

Mass Customization — A Saddle Shaped by You and Your Data

Your fingerprints are unique. So is your butt. Why, then, are most bike saddles mass-produced to ignore this? Manufacturing capability, at least, is no longer the excuse, although costs and economies of scale might still very well be valid.

While competitors like Fizik and Specialized adjust an existing shape, Posedla starts from zero—every time. Every saddle is unique because every rider is unique.

It all starts with your butt imprint. You receive a Smiling Butt Kit –yes, that’s the real name– sit on the imprint, and send your data back. Their 3S algorithm then generates a saddle that matches:

• your width

• your posture

• your riding style

• your dynamics

• your pressure distribution

They then tune stiffness by zone, using complex lattice structures that would be impossible to manufacture by any means other than 3D-printing with high-quality TPU (i.e: 3D-printing filament).

This process results in a saddle that feels nothing like a generic piece of gear with its litany of soreness and just-not-right-enough fit, and more like a conversation between your unique body shape and a manufacturing process designed to support it.

Localized Manufacturing — Craft, Proximity, Accountability

This is where Posedla gets even more interesting.

3D-printing allows them to keep production local, in North Bohemia, Czech Republic—close to the riders they serve and close to the craftspeople who understand the product.

Why does this matter?

1. Shorter supply chains means lower transportation footprint

2. Tighter quality-control loops

3. Real accountability when something goes wrong
   

Here’s an example that illustrates this beautifully:

The 3D-printed top layer (the padding) bonds to a carbon fiber shell. If that padding ever wears out or doesn’t feel right?

They can remove it, recycle it, and reprint a new one on the same shell.

The value—and the material—stays in the region. Not on a container ship.

This is the opposite of the globalized, anonymous supply chain where most cycling gear is born. It’s modern tech married to a long-standing tradition of craftsmanship and sensibility.

Subscribe now

High-Quality Products — And the Tech and People Behind Them

Posedla partners with a German company “down the street” and just on the other side of the border, to print the Joyseat’s flexible top layer. Their manufacturing process is deliciously nerdy and full of science and R&D.

Their process uses HP Multi Jet Fusion printers with TPU powder from BASF, resulting in a stronger and more durable polymer structure than UV-cured alternatives.

They can print details with 0.22 mm accuracy. This means your saddle isn’t just custom-shaped, but truly custom-crafted.

This 3D-printed top layer is attached to a “shell”. This shell is made of carbon fiber is handcrafted at Posedla’s workshop—the kind of detail that reminds you that bikes used to be things people built rather than things factories spit out.

Sustainability — Progress, Not Perfection

Let’s address the Big S for a moment…

Posedla works mostly with TPU and carbon fiber. TPU is recyclable; carbon fiber is much more difficult.

Their current approach:

• TPU padding: remove → recycle → reprint → reuse

• Carbon fiber: currently recyclable only into less demanding products (like carbon wall clocks!) until better tech emerges

Is this perfect? Of course not. No one is doing it perfectly. Not even Patagonia (this article by is an excellent read, by the way).

Here’s where I land in all this, because despair is simply not a viable strategy:

We remain skeptical because greenwashing is real.

But we should also be fair and rationally optimistic—progress matters much more than perfection, especially when a company is honest and transparent about its limitations.

Posedla is not solving climate collapse. But they are doing one thing beautifully: moving toward circularity in a space where most companies don’t even pretend to try.

In this game of inches, inches add up.

What Can We Learn From Posedla?

1. Demand better—from the companies you already support

By all means: count your blessings! Be content with what you have and be easily satisfied in your life.

But that should not apply to companies who have earned your trust and your business. Keep them accountable towards this end: doing better. For their employees, for their customers, for their communities, for the planet.

2. Get a bike fit and learn your body

Cycling pain is complicated and tends to follow this tricky maxim: “pain whispers before it screams.” Understanding your biomechanics is transformative, regardless of whether you eventually buy a Joyseat or not. This is especially true in activities like cycling and running with a ton of repetitive motions.

3. Back the builders, not the behemoths

Write a review (they still kinda matter). Share an artist’s work. Buy from a small business.

Private enterprise has never been the problem. The issues lie with concentration and abuse of power, crony capitalism, narrow-minded-and-short-sighted corporations, and myopic, lethargic and outdated government structures.

Companies like Posedla thrive when communities champion them.

The Bigger Picture?

A bike saddle is a small thing, trivial. But sometimes small things reveal deeper truths.

That we deserve products that remember us.
That manufacturing doesn’t have to be anonymous.
That sustainability can be real—even when imperfect.
That technology can bring production back home instead of sending it away.

And that a simple piece of gear can remind us what good craft feels like: something made with care, for a particular person, in a particular place by people who care.

But wait, there’s more…

For a parting gift, check out this bit of intriguing history situated in Varnsdorf, Posedla’s hometown in North Bohemia, Czech Republic.

What’s cool about Stripe Press?

Any attempt to shake loose the atrophied and calcified structures of book publishing is well worth paying attention to.

The U.S. book market is dominated by five legacy publishing houses—Penguin Random House, Hachette, HarperCollins, Macmillan, and Simon & Schuster—all founded in the 19th century.

That’s… wild. The stories that reach us by way of books are still largely filtered through institutions older than radio.

So when a modern company decides to step into that world—not to disrupt it Silicon-Valley style, but to make gorgeous, thoughtful books about science, systems, and the future—I perk up.

Enter Stripe Press, a small publishing arm inside Stripe, the internet’s financial infrastructure company. Stripe is on a mission “to increase the GDP of the internet”, and Stripe Press feels like a cultural counterpart to that ambition: a quiet, beautifully typeset bet on the value of ideas.

From their own description:

“Stripe Press produces works about technological, economic, and scientific advancement. We aim to inspire optimistic, expansive thinking about the future.”

Books as a force multiplier for optimism. I love that premise.

Beautiful books curated with a point of view

Stripe Press publishes a mix of new works and reprints—lost classics, overlooked gems, or books that didn’t get the love they deserved the first time around. At least according to their editors.

Two of my favorites:

• Working in Public by Nadia Eghbal — one of the sharpest, and most up-to-date explorations of open-source communities and digital labor.

• Stubborn Attachments by Tyler Cowen — which manages to be both morally serious and strangely uplifting.

On my to-read list:

• The Art of Doing Science and Engineering by Richard W. Hamming

• The Making of Prince of Persia by Jordan Mechner. — a diary-style time capsule about creativity and the making of a classic game. I’m thrilled this exists.

Book nerds turned fintech founders

It’s worth learning more about the Collison brothers, the founders of Stripe. They are, for instance, big book nerds: growing up in a small, remote village in Ireland, their scientist parents filled the house with books and encouraged curiosity without ever pushing a set path.

Reading became a daily ritual. Physics, biology, history, biographies, novels, they read it all. You can read more about how books shaped the brothers’ path towards founding an incredibly impactful company on this wonderful post by Thomas Yeddou.

This tracks with Stripe’s culture today, where writing well is an explicitly required competency. Clear prose is considered a vector for clarity, time-saving, and better decisions.

Patrick Collison’s personal website even includes a sprawling, fascinating, regularly updated bookshelf—part intellectual diary, part invitation.

Stripe Press feels like a natural extension of that internal culture: “what if we published the kind of books we wish existed?”

Books as strategy (or: when Disney meets fintech)

It’s tempting to look at Stripe Press as a nice-to-have brand flourish.
But I don’t think that’s right.

Two great posts—by Tyler Lasicki and Coleman McCormick—explore this more deeply, asking why a company like Stripe would build a publishing house when most tech–branded media experiments have failed, like a16z’s Future.

Coleman’s post includes a doodle I love: a sketch of Stripe’s flywheel, where Stripe’s products, including Stripe Press, appear in blue. Stripe Press is small, but at a crucial point in the loop:

It reminded me of the ole’ Disney Corporate Strategy Chart—a tangle of arrows showing how movies feed merch, which feed parks, which feed media which end up feeding all the way back to: movies. A closed loop of cultural momentum.

Disney’s former CFO Jay Rasulo put it starkly:

“Everything we do is about brands and franchises.”

Stripe isn’t building a franchise or a network of them, but I suspect something similar is happening on a quieter, more intellectual level: a belief that ideas, when circulated with care, create their own kind of flywheel.

A personal takeaway

Thinking about Stripe Press this way makes me wonder: What are the flywheels in our own lives? What systems are we unintentionally reinforcing—or could we intentionally design?

Not everything in life connects as cleanly as a Disney strategy diagram, or Stripe’s flywheel. But maybe the act of drawing our own charts, even roughly, can reveal where the energy is, where meaning accumulates, and where a small intervention –a project, a book, a side-quest– might change the trajectory of the whole system.

I see Stripe Press as one such “intervention”: a bet on the possibility that beautiful, ambitious books still matter in a world that seems to be moving further and further away from them, and that the future belongs to those willing to think expansively—and publish accordingly.

What’s Cool About NounsDAO?

There’s a lot to go through here, since NounsDAO is many things at once — an NFT project, an open-source brand, an online community, a treasury, and a funding body. It’s precisely the interaction among these parts that generates what their community has dubbed The Nouns Virtuous Cycle:

Each daily NFT auction brings new art, new members, and new capital into the DAO’s treasury.

That treasury, in turn, funds creative projects — films, clothing lines, educational initiatives, even public-goods experiments — that expand public awareness of the Nouns imagery and ethos.

As the brand spreads, more people discover and desire Nouns, fueling the next auction, the next vote, the next idea.

Check out the active proposals hoping to get funded – some are ridiculous, some lofty, yet others are incredibly practical, and together they all amount to a real representation of humanity’s varied interests.

By the way, all pixel art generated by Nouns belongs to the public domain. You can use it without copyrights! Yes, you! And I can to, and so can any other human being with some degree of computer literacy.

You can print it on a t-shirt, or make a movie with one of its characters, no problem, no royalties, no unlawful infringements.

This notion of an open source brand is powerful and quite refreshing: it represents a clean and monumental break away from the notions of intellectual property that ALL previous brands have been built upon. It’s truly something to behold and learn from.

With this new open and swarm-from-all-sides “model”, Nouns becomes a self-reinforcing ecosystem where culture begets capital, and capital begets culture — a kind of economic engine built from art and attention, sustained by the infrastructure of blockchains and the imagination of its participants.

We’ll explore it in more depth in the next sections…

How It Works (roughly)…

Every day at precisely 2 p.m. UTC, a single Noun is born: a 32×32-pixel character, minted forever onto the Ethereum blockchain. One a day gets auctioned, every day, without end.

You’re probably thinking: “bidding real money for some crypto pixel art crap is absurd”, and I get it. But consider that most of humanity works incredibly hard for what we’ve come to call real money, which is just a piece of paper and/or a cypher on your mobile app with no intrinsic value. You can’t do much with a dollar bill’s physical properties, but our collective imagination makes this lousy object quite magical.

Meaning this lousiness doesn’t stop dollars from being the currency that powers millions of transactions every single hour of every single day. It’s how you buy groceries, how you fund the arts, how you pay your bills and buy gifts for your loved ones.

This is the real absurdity, the paradox of capitalism: that, at its core, there’s just a Hungry Ghost powered by a collective delusion.

Back to Nouns. Each Noun is auctioned and its proceeds flow directly into a communal treasury governed by its holders.

That’s it. There’s no roadmap to build against, no pre-mint frenzy, no celebrity drops fueling hype cycles. Just an infinite, open-source organism that marries art and governance, pixels with culture, and ownership with experimentation in public.

…and How It Compares to Other DAOs and NFT Drops

To get at the core of how unusual NounsDAO is, consider its peers. Most NFT collections launch thousands of tokens at once, inflating hype cycles that spike and crash. Many DAOs are short-lived treasuries or private clubs with vague missions, and no real engine behind them.

Nouns avoids both traps through deliberate pace and permanence. By issuing only one NFT per day, it engineers scarcity through rhythm rather than volume. And by storing everything on-chain (i.e.: recorded on an immutable blockchain), it ensures resilience beyond any platform’s lifespan.

Once more, I want to stress the open-source aspect to it all: by releasing its imagery under CC0, NounsDAO invites competitors and collaborators alike to propagate the brand. The result is a kind of cultural flywheel: anyone can make a derivative — Lil Nouns, Nouns Esports, Nouns Coffee — and each success feeds attention back to the parent ecosystem. No trademark lawyers required; just memes defying gravity.

The DAO’s decision-making process also defies typical tokenomics. One Noun equals one vote, regardless of market price. No token weighting, no plutocracy. It’s a small but meaningful stand against the usual “whale rule” dynamics that dominate decentralized governance.

While the world continues to be obsessed with speed and speculation, Nouns’ slow, cumulative rhythm feels quietly, profoundly revolutionary.

Origins and Structure of a Living Artwork

Nouns began in 2021 with a deceptively simple proposition: what if you minted one NFT per day, forever?

The founders — a collective of pseudonymous creators like @punk4156 and @dom — deliberately stripped away the speculative tricks common in NFT launches. No “limited supply.” No exclusive whitelist. Instead, they introduced time as the scarce resource.

Each day’s Noun (referring to the pixel art here) is a unique composition of traits — heads, accessories, backgrounds — generated from open-source code. Neither the code nor the art it produces is locked down behind a license: the art is released under CC0, meaning anyone can use, remix, or commercialize the imagery.

This decision was radical in 2021, when most NFT projects guarded their IP like medieval guild secrets. Nouns, by contrast, made its art a public domain gift, betting that shared creativity would outlast private control.

The second innovation was on-chain governance. Every auction’s proceeds — now totaling hundreds of millions in ETH — flow into a transparent treasury controlled by the community itself. Each Noun grants one vote. Proposals can range from whimsical (funding an animated short film) to strategic (grants to public-goods developers, artists, or nonprofits).

It’s a kind of distributed cultural endowment, where the same community that produces the iconography also steers the deployment of its funds.

Art, Culture, and Governance Intertwined and Funded

Nouns diverges from the speculative mania that defined early NFT culture in important ways. Where many drops treated art as a vehicle for profit, Nouns treats profit as a by-product of collective authorship.

The project’s website calls it “a new kind of on-chain organization that attempts to improve the formation of on-chain avatar communities.”

But that undersells its potential. Nouns is an ongoing study in how culture governs itself when money, memes, and meaning are inseparable.

Traditional art worlds relied on patrons and institutions to sustain creativity — the Medici family funding frescoes, or 20th-century museums collecting avant-garde works.

The DAO, by contrast, turns the patronage model inside-out.

Here, the artwork itself produces the funds to sustain future art. The community decides what’s worth supporting — a documentary, a skateboard brand, an NFT museum, a coffee company — not through curatorial committees, but through votes encoded in smart contracts.

The real artworks is the infrastructure. A self-funding, self-governing organism that blurs the line between medium and movement.

The Cultural Lineage: From Guilds to DAOs

Every era invents new institutions to sustain culture. Medieval guilds maintained craft standards and communal wealth. Renaissance patrons channeled fortunes into public beauty. 20th-century co-ops and artist collectives experimented with shared ownership. Nouns DAO, improbable as it seems, belongs in that lineage — an experiment in digital guild-craft for the 21st century.

But where the guild protected trade secrets, Nouns exposes them. Where the patron imposed taste, Nouns distributes it. Its open-source brand becomes a commons of style, maintained not by hierarchy but by consent. It’s as if Bauhaus had a blockchain and a treasury.

The deeper question Nouns poses is not just what can decentralized governance fund?, but what kinds of meaning can it keep alive? Can a meme, born of pixels and protocols, nurture the same durability as a mural or a folk song? Can culture, once decentralized, sustain continuity without the anchor of institutions?

What’s cool about Campfire Cycling?

It starts with one person – someone who fell so madly in love with bicycles and camping that they decide to build a life around both.

They move to Tucson, Arizona, start pedaling into the desert with a few friends, and slowly, something takes shape: weekend rides turn into gatherings, gatherings into a community, and the community into a shop and advocacy hub.

That’s the oversimplified (and slightly romanticized) genesis story of Campfire Cycling. A spark of joy that grew into a constellation of play in the Sonoran desert and beyond!

Today, Campfire Cycling a bikepacking-dedicated shop, a touring company, and a gathering space for people who believe that adventure can – and should – begin at your doorstep and end with dirt on your legs (and your face).

Why it matters?

Bikepacking may seem intimidating from the outside: gear-heavy, planning-intensive, route-specific, and hard on the body to the point of mild masochism.

The right saddle, the right tires, the right bags, the GPS device to guide you along the correct route to your campsite and water re-supply spots… it can get heady and overwhelming.

That’s why it’s such a relief to have organizations like Campfire Cycling. They make this modality of outdoors recreation approachable, helping beginners wade through the technical clutter and focus on what matters: the joy of moving through landscapes powered by your own legs.

I can’t overstate how awesome it is to have friendly shops – and people! – narrow down the choices for beginners who get to benefit immensely – and hopefully save money and headaches along the way – from their expertise and experience. Especially when navigating particular terrains like the desert, where local knowledge becomes essential and potentially life-saving.

When I went on my first bikepacking trip earlier this year, I probably wouldn’t have made it past the gear list without the guidance of another wonderful community, Color the Valley, whose friendliness mirrored exactly what Campfire Cycling represents.

These organizations, and the spaces they create and foster, are the connective tissue of the outdoor world. They’re not just shops or businesses; they’re learning hubs, storytelling circles, and repair stations for both bikes and self-confidence.

A broader reflection: slowness, land, and the social life of movement

There’s something quietly radical about bicycles right now.

In an age obsessed with speed and efficiency, pedaling through the desert feels like an act of escapism and rebellion.

Bikepacking slows time down to a more human tempo. You start to notice temperature gradients in the wind, the hum of rubber on gravel, the smell of creosote after a storm, the Gila woodpecker foraging and nesting in a luminous saguaro...

Moving in this way, you begin to saturate your life with presence and attention, and subtle, magical wonders slowly unravel before you. It’s not just transportation; it’s a delightful way of being in the world.

Campfire Cycling needs to sell you on this, of course, but they go above and beyond mercantile pettiness: they embody this philosophy.

By promoting self-propelled exploration, they bridge recreation and responsibility. Every trip they organize, every rider they outfit, adds to a growing culture of sustainable adventure- – one that sees conservation not as a policy agenda but as an everyday practice.

As their website puts it, *”We believe in adventure that deepens our connection to the world, not our footprint on it.”*

That’s a model worth amplifying: an outdoor business that sells gear, yes, but also shows the way of restraint; that organizes trips, but also nurtures local stewardship.

Because adventure doesn’t require escape; it requires attention. Every pedal stroke is a vote for a slower, saner way of inhabiting the world.

So here’s to more campfires, more conversations, and more people rediscovering the magic of getting somewhere the long way.

And if you haven’t tried bikepacking yet, I highly recommend it.

Learn More

Campfire Cycling didn’t invent bikepacking. You can learn more about that here if you’re curious.

There’s also this wonderful resource from Bikepacking Roots with routes in Arizona. You might still want to give that a read for a little bit of history, even if you’re not considering bikepacking, or doing so specifically in Arizona.

I really do think Campfire Cycling is awesome, and that we need more orgs and people like them, people like the aforementioned Color The Valley. It also serves as a powerful reminder of how cool organizations stem from cool people doing cool things, and gathering with others to evolve.

You can book a bikepacking tour/ride with them.

You can get a custom bike built, and get your bike repaired.

You can buy wonderfully-curated gear from them.

You can find a ton of information, and inspiration, on their blog.

And while you’re at it, check out small, kindred communities like Color the Valley – proof that the best adventures often start with friends helping friends find the courage to go.

Writing along the machines

I’ve used AI to write this newsletter, but not how you might think.

AI doesn’t write for me. Few – scant!– words generated by AI tokens have made it into a published post here. But AI does help me in my writing process.

This is not a sanctimonious take on the One True Way of using AI for creative endeavors, or about not using AI at all.

We’re in an extremely experimental phase, and things will only continue to change. AI-generated content is now on par with what humans are capable of when they’re motivated by selling, advertising and being lazy, dull and “efficient”. That stages in AI’s development is well underway.

That’s why AI doesn’t write this newsletter: because asking AI to write for me is a not-even-wrong way of framing what I’m trying to accomplish here.

Having AI write for me is much like driving my car when what I want is to go for a bike ride.

My goal isn’t production or efficiency. My goal is friction! To go to the trouble of thinking! The detective work in figuring out what the hell I’m trying to say! The quest!

I want to ride my bicycle, I want to ride it where and how I want.

So I don’t ask AI to write for me. Instead, we write alongside one another!

I ask an LLM to tell me what it knows about and organization and what the LLMs would write about it. I ask it to give me three to five takes on each subject, and then to ensemble the threads I’m most interested in.

Without fail, LLMs tell me what most people would say about that organization, the industry they’re in, the problems they strive to solve and the value they aim to create. They summarize what you can glean from doing cursory research on an organization’s website plus a few Wikipedia pages.

And so far, that’s about it. Because they can’t draw a line on the sand and stand their ground! These spineless creatures of artificial intelligence! Large Language Moderates!

They cannot help it: they love to hug that middle line, and they rehash one narrative arch ad nauseam, failing to dig up any valuable nuance – much like most media outlets, I’d say.

For example, they insist on using the Open Source Software Is The Commons metaphor to understand this particular realm of the modern world I’ve been exploring here.

Mind you, OSS as Commons is not an incorrect metaphor; it simply represents a generic and partial understanding of what open source is, and what open source can be.

AI isn’t bold, it doesn’t lean forward to peek into the unknown and the possibilities beyond. Which is terrible limiting…

…and yet: incredibly helpful! LLMs help me write by telling me what not to.

I’m not an open source software expert. But thanks to my LLM collaborations, I’ve learned to sniff out for clues, truths and boo’s. LLMs are great at pointing me in the direction of the mediocre view, the midwit take, the lazy route and the sloppy, unsavory dullness of “don’t make me think” content.

LLMs will improve and my prompting, of course, can change. This is all in flux. I’m here for it.

I can tweak my prompts so that AI starts producing new thoughts and content for me, but I just don’t want to.

And if all I wanted was to “grow” this newsletter and profit, then I bet it’d be most helpful: most people would probably want to read that more than they’d want to read what I have to say. And that’s…fine! Still, it’s not what I want.

I want to write and think my way through the mud and, hopefully, come out on the other side with a new understanding and perspective, and something witty to say that’s a pleasure for you, dear reader, to chew on.

That is the wonderful practice I’m engaged in here, the one I’m inviting you to join and come along.

In the age of agents, don’t waste your own agency. Don’t give it up so easily.

I, for one, will not. And I will continue to experiment. In the meantime, why would I want to hand off the best part to the eager machines writing alongside me?

What’s cool about Putumayo World Music?

Putumayo World Music is a record label that publishes compilation albums based on the musical creations of incredibly talented musicians from around the world.

Curiously, it traces back its origins to a handicrafts-from-around-the-world shop in New York City in 1975, and only venturing into music with their first album compilations in 1993. Things…become other things!

Organizations evolve, and often transform into something quite different, like how the band Vulfpeck started as a YouTube channel, which I’ve written about here. Hunch: I’ll keep on pulling on this thread in future dispatches…

Many of us are incredibly sheltered when it comes to music, usually listening to only a few genres in a handful of languages… at most!

Words in a song matter, of course; to suggest that they don’t is nonsense. And yet…

This is the beauty and magic of Putumayo. For the most part, each of their albums is based on the music of a specific region of the world. Music from all over the globe in a myriad of foreign-to-me languages.

To date, they’ve has churned out 213 of these albums. Two hundred and thirteen! That’s an impressive number of hours of music, and plenty for you to expand your musical horizons.

Now that you know about Putumayo – and Cercle, which I’ve written about previously – it would be a sad thing to deprive yourself of the vast range of musical possibilities. Knowledge is power. And now you know.

The time so short, the music list so vast – enjoy!

Some favorites

Check out African Yoga on Bandcamp. And before we continue: a quick plug to support your favorite artists and musicians! Buy merch. Attend concerts. Listen to their music and share it widely.

Not exactly a Putumayo album, but both musicians have been featured in at least one of their albums. And what makes this particular album stand out? It was recorded in a single improvised session, and published by another wonderful label: Cumbancha.

Closing remarks

I first stumbled upon Putumayo by tuning into the Putumayo World Music Hour radio show while living abroad and driving my car on the highway. I was instantly hooked by the Bossa Nova and Afro-Cuban tunes that Rosalie Howarth played that day.

The contrast between my surroundings and the longings, memories and emotions evoked by music from far away places was thrilling, and weirdly mind-expanding. Here was not only new music but new constellations of human experience and emotion. Listening to Putumayo was a means of being transported to new lands and cultures without having to travel, a kind of cultural exchange shortcut.

But the words? I couldn’t make out what the words meant…

Since then, I’ve been able to “receive”, and enjoy, music in Arabic and French, Bambara, Songhai and Portuguese, none of which I speak myself. Why is that?

I’m not sure I can explain it, but I’ll give it a go. Paraphrasing the great John Berger, a song is a form of expression unfixed in time and place, narrating the past while filling the present with hopes of reaching future listeners. Songs lean forward, and hope to come alive by inhabiting the bodies of musicians and audience. Listen to some of the tunes I shared above and experience this for yourself.

Along with exposing people to new music (much appreciated!), I like to think Putumayo has helped musicians from around the world get opportunities that they wouldn’t otherwise have had.

And even more wonderful to consider: that thanks to organizations like Putumayo, more music has been made, produced and shared. That the cultural impact is so extraordinary significant that it is impossible to measure. Maybe we don’t need to, as long as we acknowledge, we listen, and we contribute.

Wishful thinking? Perhaps. In the meantime, enjoy the music.

What’s cool about Ink & Switch?

Ink and Switch is an “independent research lab exploring the future of tools for thought”, with the goal of “helping scientists, journalists, and creative thinkers do important work with powerful tools that respect their humanity”.

They remind us that software ≠ evil. Not necessarily, not by default. That software can be all about wonderful ideas and worthy ideals, like end-user programming, malleability and agency, unbundling siloed apps and you own your data.

What’s especially cool is that Ink & Switch puts these ideas to work by producing not only essays and research, but prototypes and working software. Real software with real potential.

Software that asks questions like “why isn’t planning a trip easier, in spite all of these tools and information?”; or “why there so many to-do list apps that don’t quite fit my wants and needs, and thus fail to use any of them?”"; or even “why can’t I get my partner and/or family on the same page about grocery shopping and making dinner?”

Much like the Xerox PARC for the new wave of tools for thought. I hope that Ink and Switch is the kind of organization that can breathe a new, healthier and loftier soul to the new machines (agents!).

I highly encourage you to check out their research projects. Like Embark, a wonderful dynamic document to organize trips and travel plans:

And PushPin, a collaborative corkboard with minimal dependence on the cloud and servers:

Read up, if you fancy, their rationale and motivations for building each of these software prototypes (products?). Product roadmaps written in prose poetry, each one an ode to the powerful feedback loop of elevating humans that make software to elevate humanity.

I’m convinced there should be more of this.

Choices maketh the app

Most – not all! – software we use is made by startups and big corporations. You know their motivations. But/and from reading Season 1 dispatches, you know that a lot of the software that those startups and corporations use – and therefore you and me and all of us – relies on open source software. And now you know the motivations of open source contributors, too.

For better or for worse, this lopsided production model is what we have, the one responsible for the magic and the surveillance, the convenience and the doom scrolling. And, importantly, it is also what we have to work with.

Ink and Switch is a hybrid in terms of that spectrum of motivations. They are steadfast in their belief that software can be much better than what it currently is, and that it can deliver on the wonderful potential for a computer: to be a bicycle for the connected mind.

Forget about patterns, libraries, or even code. At it’s core, software is made up of ideas, desires, motivations and processes that run in a computational environment.

Condensing that thought even further: software is made of choices!

All software that’s being made, sold, bought, designed and developed is encoded with those ideas, understandings and values, and the choices made by those who developed the software, what they – humans of flesh and bone, like you and me – think the world should be like, and working from their circumstances and constraints to make that world a reality.

Which – again – means that software isn’t inherently evil or good, life-changing or unnatural. The takeaway isn’t that but, instead, this: software cannot be neutral. Bias is smeared everywhere on every app you’ve ever used because choices maketh the app , and when these biases and tradeoffs are not made known explicitly, you are justified in thinking the worst.

There’s plenty of great software out there. P l e n t y. Made by open source contributors and curious, tinkering humans. Software made to spark joy, and software made like its a home cooked meal. None of the apps we use have to be blackboxes that hijack our best intentions with dark UI/UX patterns.

Ink and Switch is an organization that makes software, and the choices that went into making it, explicit and legible. You can go look at their code, and you can go ahead and read about their motivations, hold them accountable to them. You can choose to use their software or not. And with this inspiring knowledge, you can demand more of the makers of the apps that you do use.

Better yet, you can choose to build tools on top of, and elaborate on, their ideas. Build your own shovelware, but also build your own practices, even start new businesses, companies, and successful startups, based on their research and prototypes.

We need more experiments like the ones Ink and Switch is actively running. In the realms of science, democracy and technology, we need to let a thousand flowers bloom!

What’s cool about Cercle?

Cercle is an incredibly prolific organization pushing the boundaries – and excellence! and possibilities!– of music performance and experience.

In their own words, Cercle operates “at the intersection of music, aesthetics, heritage, and video”, and they aim to “create and produce innovative experiences on a global scale, highlighting unique artists on unique stages, offering a new perspective on the cultural and natural legacy of our world.”

I came to know about them while exploring new music on YouTube; a reminder that intentional interactions with algorithms can yield in something other, something quite magical.

Cercle is that: pure magic.

Particularly in post-COVID pandemic years, and compounded by the convenience of streaming services, live music has taking a hit. Some genres more than others, of course, especially those that rely more on live performances (hint: jam bands). Anecdotally, I can tell you that a few decades-old, established venues – home bases to bands like the inimitable Phish – are now closed.

What Cercle is doing is reinvigorating live music experiences by exploding the possibilities of what that might even mean. That’s the beauty, coolness and power behind this org.

I’m sure there’s plenty more that I can dig out for you that makes Cercle cool. Instead, I’ll share some of my favorite performances below, along with a little suggestion. And I encourage you to explore the Cercle Universe, which is vast and full of light and sound!

The first video on the list below, by the Icelandic mult-instrumentalist Ólafur Arnalds, is a music performance of indescribably beauty. Its beauty is simply ineffable, and I can’t tell you how many times I’ve enjoyed it since I first experienced it.

Which leads me to that suggestion: clear some time to listen to one of these Cercle performances (again, highly recommend the Ólafur Arnalds one). If you can, try to avoid sandwiching it in between busy online multitasking. Then, don’t stop there: watch them, too, attend to them as if you were there. Enjoy that delicious intimacy that only online and digital media is capable of sometimes.

Oh, the irony!

Some favorites

Thanks for reading. Catch you next time!

What’s cool about OSSPledge?

Here’s a riddle: how do you fund a global “public” good like open source software?

The Open Source Pledge answers this question with a single number: $2,000 per developer, per year.

By introducing $2,000 / dev as an anchor, CFOs and managers now know the figure. Engineering leads know the expectation. Maintainers know the scale of support they could ask for if this OSSPledge sticks.

What is fascinating is that this now has the potential to change norms and opinions. Once you start saying “two grand per developer,” suddenly there’s a benchmark. If your company has 25 developers, you’re on the hook for $50,000 annually in supporting open source software.

That reframing is the key here: it’s no longer optional, feel-good charity. It’s now part of the cost of doing business – that is, and this is the big caveat: if you care.

This entire proposal hinges on that: whether people care enough to push their companies towards taking the pledge, or elevating the pledge to the level of policy.

The pledge is important because this all can be much like paying office rent or dealing in AWS credits in terms of convenience, only this rent pays the people patching your software dependencies.

And because companies publish annual reports on where the money goes, this new expense doesn’t just collect dust in some company policy documents drawer or the bottom of the stack of some marketing blog. It now turns into public accountability.

For decades, support for open source has lived in abstractions and platitudes: “be a good citizen,” “contribute back”…similar to how some countries treat their veterans (some form of social acumen and pride, with much, much less in the way of support for services to care for those who’ve provided military services).

The OSSPledge aims to rewrite these cultural norms, with precise dollar quantities attached and a convenient mechanism to make it happen.

Since launching, OSSPledge signatories (starting with Sentry and joined by other Scary Big Companies) have reported millions flowing to maintainers and foundations. The PHP Foundation, Ruby Central, OpenSSL — substantial open source projects now directly benefiting from this “movement”.

$2,000 is not the “perfect” number. The point is that there is a number, and with that more hope in shifting support for open source maintainers from ad-hoc tips to a more systematic funding.

A maintainer once described the support they received through the Pledge as “the first time I could think in years, not months.” There’s huge potential in that, given the size of features and the length backlogs and bug lists, and the time pressures beset on developers.

Will this be a net-positive contribution to the world of open source?

On the mixing of open source and money

We must start this meditation with an expansion in the range (or scope) of what we consider economics to be responsible for studying and explaining.

We often think economics is limited to supply and demand, markets, entrepreneurship and corporations – concepts, institutions, norms and activities that stem from the logic of quid pro quo transactions. While this logic and its implications are valuable for us to study, they are only a narrow subset of the vastness of human activity.

For our purposes, it means that this logic is too limiting to be useful in explaining curious behaviors like people collaborating in their spare time and no financial reward to create open source software. The weird world of open source, too, is economics.

We can now consider economics a much broader practice – a framework, if you will – that we use to explain human behavior in its blooming, ever-changing strangeness. Certainly why we buy and sell – and how and when and under which circumstances – but also why we gather, cooperate, lie, trust and love. Why we pay to watch other people play video-games, why we have hobbies and go to concerts, and why we stand in lines to buy crap we don’t need, or even want.

Economics is weird precisely because humans are weird, because we are not rational simpletons that act the same way every time we shop, call a friend, go for a walk or decide who to marry, what career to pursue, what good music sounds like or whether we should go out for dinner or cook at home.

We are passionate and irrational, but we are also cooperative and intrinsically motivated. When people stress that humans are more than rational, they usually overemphasize the first two – passion and irrationality. But its the last two possibilities of human behavior – cooperativeness and intrinsic motivation – are the stuff of which open source software is primarily made of.

Neither the theory of the commons – how common pool resources are owned managed and used nor the Coase Theorem of the firm – why production is best organized within institutions, companies and organizations – are enough to explain open source software production.

Because of the particular dynamics between contributors, maintainers and users for each project, not all open source software can be understood as mere commons. Nor can we say that all of them should be developed under market-based terms. It’s a mix of both approaches and of many other variables at play, including the most fundamental one: the personal motivations of the people who develop them.

Software has the weird quality of approaching zero in terms of marginal cost. This little quirk is responsible for trillions of dollars in todays economy: people buy and sell software, and companies invest seriously frightening sums of of capital to build the infrastructure that allows them to profit from it.

But/and it’s also, surprisingly and ironically, the same core principle that lets open source maintainers reject the idea of charging or getting paid for their open source contributions. Open source is a legitimate and proven mechanism for producing high-quality software; software that can be as good, and as bad, as software made by companies. Software that is, in fact, a critical dependency for most of the software products and systems in existence today.

And thus, we now come back to the Open Source Pledge. It doesn’t solve every sustainability problem in software, and it doesn’t aspire to. But it does pose an intriguing challenge: what happens if we set a price tag (per developer) on the existence and, dare I say, survival of open source?

We know from projects like the web app framework Ruby on Rails that open source can not only survive but thrive without subscribing to market-based dynamics. But what if there’s a sudden influx of cash that is itself rooted in peer pressure and appeals to lofty ideals like the commons, or social responsibility? Cash thrown at open source that is slightly outside of market-based dynamics.

Now imagine a critical mass of companies adopting it. That’s billions of dollars a year flowing as infrastructure investment. Real-world investments, in real-world software, that are not pegged directly to the whims, cycles, droughts and bubbles of venture capital funding.

Do not lose sight that at the core of Open Source Pledge there is a completely arbitrary – manufactured? –  belief: that you should pay for open source! Like we’ve said before: that is simply not an absolute truth but only one possibility of many. It is similar to the belief, the social-constructed illusion, that a piece of paper with some fancy stamps and the words “In God We Trust” has any value.

It’s a cry for social responsibility at an unprecedented scale, and a cool experiment. We’ll see how open source software is impacted by it – if at all – and how open source evolves in the bumpy years to come.

Essential questions

1. What happens when we assign a fixed price to something previously treated as “free”?

Assigning a price to “free” software is less about monetizing it and more about reframing its value. For decades, companies built billion-dollar businesses on dependencies maintained by volunteers. Gratitude was expressed through conference talks or swag, but the balance sheet showed zero. By introducing a concrete number, Open Source Pledge forces companies to treat open source as infrastructure — as something that costs money to sustain, like office space or electricity.

This shift has ripple effects. A fixed price means maintainers can make future plans, not just hope for one-off donations. It means CFOs can evaluate whether they are over- or under-investing relative to peers. And it means the broader ecosystem can finally measure the gap between “usage” and “support.” The number might not be perfect, but the act of naming it creates a floor — and once a floor exists, companies are less likely to race to the bottom.

2. How does a single number ($2,000/dev/year) shift the culture of responsibility in open source?

Culture is sticky, and vague norms don’t change it. For years, “contribute back” was the mantra — but that phrase could mean anything from filing a bug report to funding a foundation. The brilliance of the $2,000 figure is that it turns a fuzzy ideal into a measurable responsibility. It offers both a ceiling and a floor: enough to be meaningful, but simple enough to calculate without endless debates.

This clarity rebalances the power dynamic between companies and maintainers. Companies can’t hide behind excuses of uncertainty or complexity — they know exactly what responsible support looks like. Maintainers, in turn, gain leverage to say: “Here’s the norm. If you’re benefiting from our work, where’s your contribution?” One number becomes a cultural anchor, setting expectations the way emissions targets or minimum wages do in other domains.

3. What’s the difference between gratitude and obligation in open source?

Gratitude is what you feel after unwrapping a gift; obligation is what you owe when the roads you drive on need fixing. Open source has long lived in the gratitude zone — companies thanking maintainers for their hard work, sometimes sending stickers or a thank-you tweet. But that dynamic left maintainers underfunded, insecure, and exhausted. Gratitude doesn’t pay rent.

Obligation changes the social contract. By framing support as a duty — a recurring financial responsibility tied to headcount — the Open Source Pledge moves open source out of the gift economy and into the infrastructure economy. Obligation is less romantic, but it’s more durable. And it’s precisely this shift, from “thanks” to “line item,” that can transform open source from a fragile commons into a sustainable foundation for everything built on top.

What’s cool about Open Collective?

If open source can be considered public infrastructure, how we fund would be represented by the potholes and cracks of its crumbling parts. We all drive on the road; almost nobody funds building said roads, at least not directly (or too willingly at that).

Open Collective (OC) aims to fill that gap with an incredibly simple-yet-profund promise: make it easy (and transparent) for communities to raise, hold, and spend money together, thus eliminating barriers and the need for each project to become its own mini-nonprofit requiring the same back-office services.

Their platform and network of fiscal hosts (more on that soon) now support over 15,000 collectives (orgs) moving about $35M/year. The overarching premise behind these impressive numbers? “Let’s show everyone the receipts!”

OC is based on the following values/pillars:

• Legibility is power. Money matters are usually opaque; OC makes it transparent, accesible and browse-able. This flips the trust equation for both donors and maintainers.
  

• Speed to legitimacy. Fiscal hosting lets a weekend project accept serious money (in the way of grants, for example) without needing to become a full-on legal entity.

• Network effects for the commons. Integrations (e.g., GitHub Sponsors) and curated initiatives (like Ecosystem Funds) make it easier to support not just the shiny top-level projects but the dependencies “hidden” underneath, way below the limelight.

• Open source all the way down. Open-sourced platform code, public budgets, and transparent community governance norms reinforce each other.

This all amounts to Open Collective being much more than a donation button (nothing wrong with donation buttons; shoutouts to buymeacoffee.com!). It’s a new framework (?) for the 21st-century commons: make money flows visible, reduce legal friction, and distribute power.

Do that consistently, and do it well, and the roads we all drive on stop cracking under our feet (or thumbs). That’s the hope at least.

What they actually do (and why it matters)

Open Collective blends software and fiscal hosting.

Software-wise, every Collective gets a public page, paid-in/paid-out ledger, and expense workflow. On the legal/finance side, fiscal hosts hold funds on behalf of projects, handle banking, compliance, and payouts.

Put together, this lets a project start taking money as early as today and pay contributors by no later than next week—without negotiating bank accounts or tax letters.

On sustainability: Open Collective’s business model is transparent too. Fiscal hosts can set their own host fee (often 0–10%). The platform then takes a 15% share of the host’s fee (not of donations), plus standard payment-processor fees—aligning incentives with hosts actually supporting projects at scale.

With some perspective, you can see that this is an incredible proposition with incredible potential: cutting down significant amounts of time and red tape for the tiny, resource-constrained organizations when they need it most.

Our world is choke-full of obscure and obfuscated financial and technological systems. Meaning OpenCollective’s transparency has to go beyond marketing fluff for it to be credible. Luckily, it does: budgets are public by default—contributors, companies, and auditors can see where money comes from and where it goes (with personal details protected).

That social pressure loop is the real magic flywheel: openness builds trust, and trust unlocks funding.

A platform that’s actually…open and impactful

The platform itself is open source, its source code available on GitHub. That matters symbolically (practice what you preach!) and practically (community-visible roadmap, issues, and contributions).

On the open-source side, the Open Source Collective (OSC)—the fiscal host that serves thousands of software projects—reported $12.5M in contributions in 2024, $9.7M paid out to maintainers (up 20% YoY), and $50M+ lifetime contributions since 2017.

Those are real, substantial numbers. And if you “humanize” these numbers, you realize we’re talking rent, childcare, and maintenance time converted into shipped software patches and healthier ecosystems.

Why this matters right now

As alluded to before, a great deal of the software we rely on relies itself on open source software, which in turn is made possible by the contributions of volunteer developers with real needs, wants, desires and pressures.

Moreover, a few recent software vulnerabilities have had widespread effects for a lot of private software sold for money, highlighting the incredible reach and pervasiveness of open source projects, as well as its criticality.

I personally – albeit somewhat indirectly – had do deal with the Log4Shell vulnerability (which was deemed as a "true cyber-pandemic") at work in 2022.

More recently, the xz utils backdoor of 2024 was another reality-check: critical infrastructure maintained by overworked humans is, potentially, a security problem at the level of an entire nation. This is not appealing to the reptilian parts of your brain that get high levels of dopamine-inducing anxiety from alarmist tabloid headlines. It’s the reality of the world that we live in, a world which is significantly enabled by software.

My hope is that highlighting this stuff, and the organizations doing something about it, makes you more attuned to this reality. And if I may, I will offer my own proposal: let’s figure out better ways to pay people to maintain the commons, or the commons will surprise you in the worst ways.

Open Collective seems to be one of the few platforms that turns this notion from an abstract pontification into a concrete, repeatable practice: move money, in the open, to the people doing the work.

Of course, the state of our economy – the macro picture, if you will – hasn’t helped. Budgets tightened in 2023, and many community-funded efforts felt the stress. Open Collective’s own note to investors acknowledged the slowdown—and then the rebound in 2024/25, with data from the Open Source Collective fiscal host showing why resiliency tools matter. It sounds obvious, but it’s anything but.

So here it is again: make it easy, transparent, and compliant, and the money seems to flow to where it’s needed.

A little aside: maybe funding of open source projects tracks well with the cycles of the art market, which would be an intriguing data spelunking project beyond the scope of this little essay.

Challenges worth keeping an eye out for

• Naming confusion (OC vs OCF vs OSC vs OCE) is a real onboarding tax. Their comms have improved, but the ecosystem could use a single “What lives where?” explainer tattooed on every page.

• Regulatory and operational load on fiscal hosts is non-trivial; the OCF wind-down is a sober reminder that community infrastructure needs responsible operations and sustainable fee models.

• Cultural shift for funders. Transparent budgets mean donors can—and will—ask hard questions. That’s healthy, but it changes grant-making behavior and expectations.

How to plug in (today)

1. Maintainers: If your project isn’t fundable in one click, fix that. Apply to Open Source Collective (if you’re OSS) or a relevant fiscal host; wire up GitHub Sponsors; publish a basic budget + wishlist.

2. Companies: Move from “we sponsor a logo” to “we fund our dependencies.” Try an Ecosystem Fund and publish what you support (and why).

3. Foundations & community orgs: Pilot a Fund on Open Collective to run mini-grants with public ledgers and lightweight ops.

If you liked this, you’ll love the rest of Season 1 of One Cool Org—subscribe, share, and tell me which collectives deserve a deeper dive next…

Essential Questions

I’m trying out a new “section” for each post that I’ll title “Essential Questions” after the work of Jay McTighe and Grant Wiggins.

Here I will aim to pose “inexhaustible” questions (my own interpretation of EQs) and give some preliminary answers. Take this as gentle embodiment of the “strong opinions, loosely held” notion in the tech world.

How does financial transparency build trust within communities?

Transparency shifts the power dynamic from “trust us” to “see for yourself.” When contributors, funders, and community members can openly view where money comes from and where it flows, suspicion is replaced with accountability. Trust no longer depends on personal relationships or opaque promises; it is earned through visible, verifiable records. In Open Collective’s case, this allows maintainers to demonstrate legitimacy quickly and funders to feel confident that their resources directly support work that matters.

More broadly, transparent financial systems can make communities more resilient. In an age of misinformation and institutional distrust, showing the receipts is a simple but profound act: it proves that fairness is not just a slogan but a commitment and a verifiable practice, one transaction at a time.

What responsibilities do companies and institutions have to fund the public digital infrastructure they rely on?

This is a tricky one, and it’s up to policymakers in office as well as Big Tech to figure this out.

What is true, however, is that every company today runs on a partially-hidden stack of open-source software and community-maintained tools. These are the digital equivalents of roads, bridges, and clean water—but unlike public works, they’re rarely funded at scale. Companies benefit disproportionately from this infrastructure, yet often treat it as free and inexhaustible. Once again, the xz utils backdoor reminded us that neglecting to pay maintainers is no longer just unfair—it’s a security risk.

Institutions have both a pragmatic and an ethical responsibility: to ensure the commons is funded so it doesn’t collapse under its own weight. This requires shifting from symbolic sponsorships (a logo on a page) to systemic investment (supporting dependencies, funding maintainers, and resourcing boring-but-critical maintenance). The cost of not doing so is far greater than the cost of proactive support.

Could Open Collective’s model of shared infrastructure redefine how we think about institutions and organizations in the 21st century?

Traditional institutions—nonprofits, foundations, governments—tend to centralize money and decision-making. Open Collective flips this by offering modular, transparent infrastructure that communities can adopt without themselves becoming formal entities. This lowers barriers to entry, democratizes access to funding, and distributes power across networks rather than concentrating it in a single organization.

In the long run, this approach could point to a new kind of organization: one that is less about buildings, charters, and bureaucracy, and more about protocols, shared infrastructure, and collective governance. It suggests that 21st-century organizations may look less like monoliths and more like ecosystems—lightweight, interoperable, and resilient precisely because they are decentralized.

Sources

• Open Collective official site

• GitHub: opencollective/opencollective

• OSC 2024 Impact Report

• GitHub Sponsors

What’s cool about OpenWater.health?

Imagine you’re at a gala in a ball room full of people where an opera singer starts singing at a particular pitch, register and tonal quality. Her singing shatters a wine glass while everyone – and everything else– in the room remains untroubled.

Now imagine that same principle applied not to glass, but to the cells in your body—selectively targeting only the ones that make you sick.

That’s the metaphor OpenWater.health uses to describe its work: harnessing light, sound, and electromagnetism with high levels of precision to diagnose and treat disease at the cellular level.

Founded by Mary Lou Jepsen, a tech world pioneer with the most impressive resume you’ll find, she and her team are bringing the logic of mass consumer electronics into medicine—using the ubiquity of smartphone-class chips and global supply chains to replace the need for developing billion-dollar, custom-built machines.

The result: functional medical prototypes, like wearable caps that can perform MRI-like imaging or deliver tumor-shrinking therapy, at a fraction of the cost.

And here’s the kicker, and why I’m highlighting them here: they’re open-sourcing it all. If they are for real about their mission and approach, OpenWater isn’t just publishing code and patents—it’s trying to open the entire medical-device stack.

How, exactly, are they going about this? Here are a few examples: software is licensed under the AGPL license, hardware under Creative Commons BY-SA (remember, licenses are important!), and patents pledged for open re-use.

Even safety data and clinical protocols are shared in public wikis, with the aim of cutting the typical medical device development cycle from 13 years and $658M to something closer to 3 years and $10M.

Imagine many of these medical R&D projects happening consistently within these new and reduced budgets and timelines over the next decade, and the world suddenly looks much different. It makes medicine start looking a lot more like Linux (the open-source and predominant OS for servers and the world's 500 fastest supercomputers!) than Medtronic (the conflict-ridden manufacturer of pioneering cardiac devices).

OpenWater is betting that openness and mass-audience appeal can do for healthcare what it did for software—make the impossibly capital-intensive a whole lot more accessible, and the lifesaving more democratic.

What’s more exciting than all this promise? Some early results that prove that it’s not just hype (more on this later).

Singing their medical praises + a quick (and important!) regulatory reality check

To put it yet another, slightly more technical way, OpenWater is placing the following bet: by using mass-market smartphone-class chips and sharing safety/efficacy data across applications, device development can compress from ~13 years and hundreds of millions of dollars to something closer to a few years and an order of magnitude less money (JAMA analysis of PMA device costs/timelines; HHS overview cited in interview).

The org is unusually transparent about regulation, too: for example, you can read its FDA Breakthrough Device request and correspondence for its stroke device directly on GitHub (Breakthrough request PDF; deficiency response with clinical data).

Yet – there’s always something, right? – it’s important to remember that OpenWater’s model still has to live within existing device rules and regulators. The FDA’s new Quality Management System Regulation (QMSR) fully aligns U.S. device QA with ISO 13485 by Feb 2, 2026 (FDA QMSR FAQ; Federal Register notice). For connected devices, the 2025 FDA cybersecurity guidance also raises the bar on SBOMs, vulnerability management, and secure design (FDA Cybersecurity Guidance).

OpenWater’s “open” approach doesn’t bypass any of this. Instead, it is a different way to satisfy it, by sharing evidence and components so others don’t keep reinventing the same, closed wheel.

Geeking out on some early signals

What follows are some examples of clinical trails run by OpenWater.health:

• Stroke triage, without a scanner. A multi-center study in Journal of NeuroInterventional Surgery evaluated OpenWater’s portable optical blood-flow monitor for detecting large-vessel occlusion (LVO) in suspected stroke patients. Among 135 ED patients across two comprehensive stroke centers, the headset achieved 79% sensitivity and 84% specificity, outperforming prehospital scales like RACE and LAMS—promising, but still needing validation in the prehospital setting it ultimately targets (PubMed abstract, medRxiv preprint).

• Depression via wearable focused ultrasound. On the neuromodulation side, the team reported an open-label feasibility study stimulating nodes of the default mode network with low-intensity focused ultrasound (LIFU). In an initial 20-patient cohort with severe major depression, they observed significant symptom improvements; the authors explicitly call for randomized, controlled trials to confirm efficacy and durability (Frontiers in Psychiatry: study overview + caveats). Underlying this is a technical advance: a wearable, steerable tFUS headset, described in the Journal of Ultrasound in Medicine, designed to hit targets through the skull with a matrix array and neuronavigation (J Ultrasound Med paper listing on OpenWater clinical page — article: “A Wearable, Steerable, Transcranial Low-Intensity Focused Ultrasound System”).

• Why this feels different: Compared with MRI-guided focused ultrasound (e.g., Insightec’s Exablate Neuro, FDA-approved for essential tremor and expanding indications), OpenWater’s tFUS aims for portable, non-MR targeting—closer to a headset than a suite-anchored procedure (FDA PMA record for Exablate; Insightec overview: company page). On the stroke side, alternatives include imaging (e.g., Hyperfine’s portable low-field MRI) or ultrasound on a chip for POCUS (e.g., Butterfly)—each democratizes access in different ways and with different tradeoffs (Hyperfine; device overview + cost context; Butterfly background). In depression, clinician-adopted comparators exist today: rTMS/Theta-Burst (non-invasive, modest-to-moderate effect sizes) and emerging, invasive DBS protocols still in trials (Abbott/TRANSCEND).

The novel angle isn’t just what they’re testing; it’s how: making the designs and safety paperwork broadly re-usable so multiple centers can iterate faster on where ultrasound and optics work best. If that reduces duplicated engineering and expands the dataset regulators see, the pathway could look very different than one company grinding through one indication at a time (MedCity News op-ed by Jepsen).

And now, let’s allow ourselves some speculations…

What could change for the better if OpenWater.health succeeds?

• Faster translation, broader access. If an open, componentized platform proves out, labs and startups could stand on shared hardware/software and shared safety/effectiveness evidence—directly targeting a status quo where novel PMA-class devices can take a decade+ and large capital outlays to reach patients (JAMA PMA cost/time study / PMC).

• Prehospital stroke triage that routes the right patient to the right hospital. Even incremental gains over clinical scales can translate into minutes saved and better outcomes if validated in the field. The JNIS results are an encouraging step, but the ambulance is the proving ground (JNIS study; broader LVO prehospital review: systematic review/meta-analysis).

• A new, non-invasive neuromodulation option. If randomized trials confirm efficacy, wearable tFUS could complement rTMS and medication for depression with better focality/depth than magnets allow, without DBS surgery (Frontiers depression feasibility; general reviews on ultrasound for depression: 2025 review).

• A public-goods rhythm to medical devices. Open designs + patent pledge + shared regulatory artifacts set a precedent: competitive businesses on top of commons infrastructure. The upside isn’t ideological; it’s practical—more shots on goal per dollar (Openwater open-source + patents; Inside Precision Medicine).

What could go wrong, plus some open questions…

• Clinical evidence still early. The stroke headset outperformed scales in-hospital; field performance, workflow fit, and impact on outcomes are unproven. The depression data are open-label; sham-controlled RCTs with durable follow-up are essential before broad claims (JNIS details; Frontiers caveats).

• Safety, governance, and “cowboy” risk. Open hardware lowers barriers for good actors and risky ones. Strong quality systems (QMSR/ISO 13485), SBOMs, secure update pipelines, and post-market vigilance are table stakes for any connected device (FDA QMSR; FDA Cybersecurity guidance; HHS OSS risk note: PDF).

• Market and reimbursement realities. Incumbents already sell cleared alternatives: MR-guided FUS (Insightec) for tremor/Parkinson’s, portable MRI (Hyperfine) for bedside imaging, widespread POCUS (Butterfly) for rapid exams—each with known CPT/reimbursement paths and clinical familiarity (FDA PMA Exablate; Hyperfine resources; Butterfly background: Jefferson paper). OpenWater will need not only data but business models that make procurement, training, and service simple.

• IP clarity over time. A patent pledge is powerful, but it must be legally durable and operationally clear as new filings and contributors join (watch how the pledge and licenses evolve: Openwater patents page).

Selected Sources

Open model & patents: Openwater Open Source · Patent Pledge

Regulatory transparency: Openwater clinical page · JNIS PubMed

What’s cool about MapLibre?

If you’ve ever panned and zoomed around in a silky-smooth web map, you’ve felt the magic of modern cartography.

Behind that experience – more often than not – sits MapLibre: an open-source library that developers around the world call the “gold standard” for map software. A friend of mine who has worked on articles relying heavily on maps and cartographic data for The New York Times put it plainly—when you want interactive maps that are powerful, reliable, and free of lock-in, MapLibre is the tool.

A brilliant achievement: from fork to foundation

”Venture Capital is a hell of a drug!”

In late 2020, Mapbox changed the license of its popular GL JS renderer, making it proprietary technology and leaving a void in the open mapping world.

The most likely outcome would’ve been fragmentation: dozens of forks, each slowly withering. Instead, the “community” made a different choice. They rallied around one project, MapLibre, to steward the future of open-source rendering. That single decision—unity instead of scatter—preserved the ecosystem.

This is a surprisingly brilliant and unexpected achievement, and marks a highlight in organizational excellence that rarely is associated with open-source projects.

Why it matters

Interactive maps aren’t just decoration—they’re infrastructure. Cities use them to visualize zoning changes. Newsrooms use them to tell stories. Researchers use them to map climate risks. And thanks to MapLibre, anyone can build these tools without surrendering autonomy to a vendor.

That’s the deeper lesson: open-source isn’t just about licenses, it’s about institutions. MapLibre shows how community governance, sponsorships, and clear social contracts can make open infrastructure sustainable.

MapLibre didn’t just keep an ecosystem alive—it elevated it. With future-proof rendering, a robust tile-server story, transparent governance, and a credible sponsor base, it offers developers something rare: maps that feel as polished as proprietary products, but remain fully independent. If you want to build maps that last, MapLibre is where you start.

I want to stress here a critical takeaway: open-source isn’t just about writing free code—it’s about building organizations (institutions?) of trust and collaboration that let technology outlast any single company or license change.

MapLibre shows how a community can preserve and even improve critical infrastructure (in this case, map rendering) when corporate priorities shift.

I am not saying corporations are evil – of which they are certainly capable of. What I am saying is that corporations can be short-sighted, and the general public short-changed because of it. And that although private enterprise is one of the main drivers of progress, it doesn’t do work independently. Embedded in every technological development is a critical open-source pillar.

In the wider context of open-source software and hardware, that lesson is profound: the real power of open-source lies not only in transparency and adaptability, but in the collective capacity to govern, fund, and maintain critical technology as shared public goods. It’s proof that communities can create resilient, future-proof alternatives to proprietary systems—giving developers, organizations, and even whole industries independence, creative freedom, and long-term stability.

How the community sustains it

MapLibre isn’t just code, it’s a coalition. The project runs on transparent governance: a Governing Board, an Advisory Council, and a fiscal host to manage sponsorships and funds.

Big players—Amazon, Meta, Microsoft, TomTom—provide financial and technical support, but the roadmap is set by the community. Quarterly check-ins keep sponsors engaged without letting them dictate direction.

This structure matters. Maintaining a renderer isn’t glamorous work: it’s bug triage, documentation, performance regression fixes, and release engineering. Volunteers make incredible contributions, but MapLibre is candid about the need for sustained funding so that maintainers can do the unglamorous, essential jobs. Few open-source projects are this honest about what it really takes.

And if you’re really itching to contribute, check out their bounty hunter developer job posting:

Responsibilities

Choose tasks from the list of published Bounty tickets in MapLibre GL JS or Native according to your skills and preferences.

Compensation

As a Developer, you work on Bounty tickets and you are not required to track time. Once you complete a Bounty, you can submit an invoice to Open Source Collective, MapLibre’s fiscal host. No further contract is required. Read more in the step-by-step bounties guide.

Bonus: What makes MapLibre technically cool

On the web, MapLibre GL JS is a TypeScript library that uses WebGL to render vector tiles in real time. The result: crisp labels, smooth zooming, and styling you control down to the layer.

For mobile and embedded apps, MapLibre Native delivers the same engine in C++ with modern backends—Metal on iOS/macOS, Vulkan for cross-platform performance. These moves result in a future-proofing of the stack as OpenGL fades.

The ecosystem even extends to the server. Martin, a Rust-based vector-tile server, lets you pipe data directly from PostGIS or MBTiles into MapLibre. That means full ownership: your data, your tiles, your rendering. Together, it’s a coherent, modern toolkit that empowers developers instead of locking them in.

What’s cool about Arduino?

If you’ve ever played with electronics, there’s a mighty-high chance you’ve crossed paths with Arduino. (And if not, you’re definitely doing it wrong).

What started in 2005 as a scrappy master’s thesis project at the now-defunct Interaction Design Institute Ivrea (IDII) in Italy has since grown into one of the most influential open-source hardware organizations (movements? shall we call it a movement?) in the world (Wikipedia, Arduino).

The founders—Massimo Banzi, David Mellis, David Cuartielles (and others; this is a contentious subject, more on that later)—wanted a simple and affordable “system” for design students to prototype interactive ideas.

The result was a microcontroller board anyone could program with via an easy-to-use IDE that abstracted away the often complicated details of electronics so that people could focus on their own objectives.

From blinking LEDs – the “hello world” of electronics – to robots and satellites, Arduino became the launchpad for millions of experiments.

Arduino is the kind of organization – and product, platform and community – that simply wouldn’t exist without open source, and it embodies its essence like almost no other open-source org out there.

It lowers barriers, builds community, and empowers people everywhere to turn imagination into reality. It has survived internal conflicts, embraced new directions, and kept its identity rooted in making electronics easy and accessible. Whether you’re a kid wiring your first sensor or an engineer designing industrial IoT, Arduino makes building, playing and learning with electronics possible, and more human.

Attribution is messy

Before we continue, a quick word on attribution. Attribution is a tricky business in areas like marketing, the arts, archeology, evolutionary biology, science and academia – but it can get especially sanguinary (h/t to DC for leading me to the right word here) in the worlds of open source hardware and software.

I’ll let you come up with your own conclusions but, at the very least, Hernando Barragán deserves to get credit for starting Wiring as his master’s thesis project at IDII. His work served as necessary groundwork for Arduino to exist. Putting it bluntly: Wiring is Arduino’s amniotic fluid.

On that note, The Untold History of Arduino is a great read. In it, Barragán himself shares his side of the story, and that dives deep into the particulars.

I’d like to highlight Wiring’s original objective, which I’ve alluded to before:

to make it easy for artists and designers to work with electronics, by abstracting away the often complicated details of electronics so they can focus on their own objectives.

Beautiful. Whatever economic system we come up with should encourage and reward this kind of project.

From a tiny Italian classroom to a Global Movement

I guess we are calling it a movement after all.

By 2020, Arduino counted 30 million active users worldwide, with 300,000 official Arduinos commercially produced, and 700,000 official boards in users' hands. (Wikipedia). Its boards are now staples in classrooms, labs, and makerspaces the world over.

And in case you care about rankings, Time Magazine sets Arduino at #9 in the 2025 World’s Top EdTech Companies, underscoring its role in transforming how STEAM is taught (Arduino Blog). Nobody should pay too much attention to rankings like these, but transforming STEAM education is well worth celebrating.

I’m glad to highlight Arduino here. The org is proof you don’t need to make learning (about electronics, in this case, but maybe anything) something that’s elitist or expensive in order for it to be wonderful.

How to stay open-source through hell and high water?

At its core, Arduino is about more than boards and kits—it’s about openness, democratization if you’re more keen on that term. Both their hardware and software are released under permissive licenses, inviting remix, rebuild, and re-imagining (Arduino). The community—millions strong, truly—shares code, schematics, tutorials, and inspiration freely.

But openness always has its costs and tensions. And we should always be thinking in degrees of openness (as alluded to before with Barragán’s Wiring project lacking the proper recognition). But there’s more: Arduino weathered a trademark dispute and organizational split in the mid-2010s before reuniting as Arduino AG.

This saga highlights a persistent challenge: how does a mission-driven open-source organization protect its identity, sustain itself financially, and still honor community contributions so that contributors remain engaged? (Hackaday)

These questions aren’t unique to Arduino, but its willingness to evolve governance while staying committed to open access is part of what keeps it “cool”. They provide fertile ground for many a “lessons learned” for makers, engineers, entrepreneurs and curious learners in years to come.

I’ll say one more thing on this topic…We can treat this as evidence that open-source can be a strategic and/or necessary phase in the lifecycle of an organization. Organizations open up or close down behind shut doors out of necessity but/and also due to strategic pivots and updates to their goals and objectives. Even a change in staffing members can be the deciding factor in moving towards openness or closed-ness.

Looking Ahead

Arduino is now a mature, well-heeled organization, one pushing boundaries in education, IoT, and industry. In classrooms, kits like Arduino Alvik provide hands-on ways to spark creativity and problem-solving in STEM (Elektor). In industry, Arduino unveiled Rileva ME, a wireless sensor for predictive maintenance, alongside professional learning paths at the 2025 Automate conference (Automate.org).

With this launch as an example, we can also see amateurization and professionalization going hand in hand. The narrative that open sourcing a project lowers the value of each contribution by increasing the pool of contributors is only one side of the story. More people, too, have the opportunity to professionalize, if the project is “guided” to that end.

Arduino as an org also champions an IoT Manifesto built on fairness, openness, and sustainability—an antidote to the corporate “black box” approach to connected devices (Wired). This transparency is essential to the world, and our fights for the right to repair need to be backed up by our will to repair – to tinker, learn and pry those sealed electronic boxes open.

In research, Arduino is finding its way into surprising places—from soft robotics to air-quality monitoring (arXiv, PMC).

My hope is that learning a bit more about Arduino gets people thinking about do’s and don’t’s when considering joining, starting, backing or contributing to an open source project and community.

What’s cool about RepRap?

If you own a 3D printer, or ever considered buying one, you can thank RepRap.

RepRap is short for: self-REPlicating RAPid prototyper 3D printer, and also how most people know the open source community behind the project. What I love most is that RepRap is one of those rare idealistic organizations with strong values that has unleashed a wonderful innovation into the world by making it affordable by design.

3D printing is a form of additive manufacturing (AM) that uses fused filament fabrication (FFF) technology, where a single layer of polymer – materials made up of many simpler molecules like cellulose (made up of simple sugars), natural rubber and plastics – is deposited after another.

But before the explosion of 3D printers small enough to fit atop your desk, a 3D printer was a massive machine that only highly-specialized manufacturing companies could justify owning and operating, since they’d run you north of USD $20,000.

The goal of the RepRap project and community has always been to create an cheap-enough 3D printer: a self-replicating machine that was capable of creating 50% to 70% of its own parts, with the remaining components easily and affordably sourced.

And they succeeded immensely. RepRap’s open source hardware approach resulted in the creation of the first desktop 3D printer, a dramatic decline in costs and an explosive early adoption by hobbyists and makers, primarily using these machines for rapid prototyping.

They allowed people like Josef Prusa, founder of Prusa Research, to start tinkering with these machines and their capabilities, and by doing so, in turn, expanding their possibilities.

Thanks to RepRap, and with a little bit of tinkering, you can now own a 3D printer that runs on your home desk for USD $500 (or even less).

If you own one, I’d be curious to know what might’ve convinced you to get one. And if you don’t, I’d be curious to know if you’d consider buying one, and why. After all, there are a number studies (like this one and this other one) showing that owning a 3D printer and printing common household goods, has an ROI of more than 100% in five years. And hobbyists who own 3D printers absolutely love them.

The strangeness of open source 3D printing economics

Economics is most often thought of too abstractly, too narrowly. Inflation, taxes, interest and exchange rates. Sure, those are important. And supply and demand, opportunity costs etc…, those are important too.

But just like history is, fundamentally, all about “what happens”, economics is all about “how we behave”. Often “how we behave” is puzzling enough that it turns out to be much more fertile ground for analysis and understanding than trying to get at the root of “why” we might behave the way we do.

This is all to say that economics is complex, multi-faceted and unequivocally bereft of simple, straightforward answers. That there’s so much more to the story than dictums like “capitalism is a terrible system run by profit-starving pigs who can’t get enough of making more money” or “socialism is misguided and always leads to communism” or “tax the rich” (for example, in the name of social justice) or “increasing the number of dollars circulating always results in higher inflation“ (most of the time, probably, but not axiomatically).

If I learned one thing as an Economics student in college, the best answer is always: “it depends”.

There are obviously observable and predictable phenomena that economics can, and does, explain accurately. What I am advocating for is that we think more deeply and more critically when we hear these seemingly conclusive statements about our economic systems.

If we base our reasoning about our economic systems on premises like “people are primarily motivated by money” and “more money makes people bad”, then we stand to be wrong in our understanding of a lot that happens in the world. We can’t believe in these one-liners and also glue bumper stickers urging others to “shop local”.

If capitalism were such a terrible system as to be utterly immoral, then we wouldn’t have people like Adrian Bowyer pouring their hearts and souls into an open source project like RepRap that stands to make him zero dollars.

Capitalism has many flaws. Of course, if we could improve upon this system, perhaps there would be more people like Adrian. Or perhaps we would need people like him, and projects like Rep Rap, a bit less.

But I’m still grappling with the economics of open source, it’s existence at the height of “late-stage” capitalism. It is clear to me that the relevance of open source is increasing. That it is both enabled by capitalism, and set in direct opposition to most of the capitalistic goals. I strive to continue deepening my understanding of it, and sharing what I learn along the way.

We love or hate stories of rags to riches, but we have a much harder time assessing and celebrating the economic impact of contributions from organizations like RepRap. I’m certainly at fault here, and trying to remedy that.

My sense is that although 3D printing seems like a thing of the near past for most people, on the decline after it peaked in 2018 and certainly a much less exciting technology than AI, we still haven’t seen the economic effects of its “full potential” (what that means, I’m not sure, but think distributed manufacturing and distributed peer production).

There’s another “lesson” I learned as an Econ student that is best summarized in this quote from Henry Hazlitt:

The bad economist sees only what immediately strikes the eye; the good economist also looks beyond. The bad economist sees only the direct consequences of a proposed course; the good economist looks also at the longer and indirect consequences. The bad economist sees only what the effect of a given policy has been or will be on one particular group; the good economist inquires also what the effect of the policy will be on all groups.

But 3D printing is maturing before our eyes.

It has impacted the toys and games industry, for example: MyMiniFactory, just a single 3D printing repository among dozens, is saving consumers well over $60 million a year in offset purchases.

It has empowered scientists to create new experimental laboratory equipment and tools.

And it is going mass market, with a bunch of companies, like Bambu Labs – which you might love or hate, depending on where you stand on the morality of not playing by the rules of open source – driving costs down and making printers even more user-friendly and plug-and-play, more like kitchen appliances. 3D printers are the new microwave, the new air fryer.

This revolution is happening in great part thanks to RepRap, an open source hardware project. How strange indeed.

What’s cool about this org?

Prusa Research (aka Prusa) is an incredibly prolific 3D printer manufacturing company that has recently taken definitive steps away from open source.

Founded in Prague, Czechia in 2012 by Josef Prusa, the company employs over 1,000 people who are – mostly – working out of a single, incredible HQ and manufacturing facility, developing software, filament and making 3D printers.

(Prusa is also the company behind PrusaSlicer, PrusaMent and Printables)

Prusa is built “on top” of a wonderful open source project: RepRap. Josef and his brother were heavily involved in, benefited from and contributed to the RepRap open source projects and community prior to starting their own company.

But things changed around 2023, with the company partially abandoning their open-source ethos due to competition from others, and releasing their first two 3D printers that were not 100% open source.

Echoing Josef’s words in his state of open source hardware discussion of 2023:

The open-source movement relies on the fact that everyone involved plays by the same rules.

These rules are difficult to enforce, and largely unattractive to competitors who “poison the well” of goodwill built by open source communities.

But/and “open source” can also be an excuse.

In the words of Armin Ronacher:

For the price of one Prusa MK4 with shipping I could buy three Bambu A1 printers or two Bambu A2 printers plus the AMS addon which adds multi-color printing.

There is absolutely no reason to buy a Prusa MK4 today unless you want to support Open Source or a European company (Prusa is from Czechia).

Wether Bambu Lab and other manufacturers are playing nicely or not remains an open question. Of course Prusa wouldn’t be happy about it, especially if Bambu Lab and others are filing patents in their own countries, backed by hungry, potentially unfriendly governments.

But things change. Open source tends to evolve in a cycle of "amateurization”: companies and organizations make distribution better, wider and easier, lowering the bar for more and more people to participate, thus increasing the numbers of amateurs and late-adopters.

Who is to say that we – consumers, makers, prosumers – don’t benefit from simpler, easier-to-use powerful 3D printers like Bambu’s flooding the market, enabling more people to get into 3D printing?

It remains to be seen.

A few insights

• “Open source” is a dynamic, evolving concept

• “Open source” is also loosely and vaguely interpreted by different companies and jurisdictions around the world

• The amateurization cycle is embraced by those who usually stand to gain from it, and decried by incumbents and purists

• Licenses are critical components dictating outcomes in the shades-of-grey spectrum of open–to close–source

• Open source business models are challenging to sustain, while open source remains an attractive component for companies, particularly at their earlier stages

• Convenience usually triumphs over the principles of purists

Manufacturing matters

We’ve been able to run a printing press from our home offices since the advent of home printers. With modern 3D printers like Prusa’s, we can essentially run a small, nimble factory.

Along with a PC or laptop, that’s three revolutions in history cozying up on your desk, accessible as packaged machines requiring little more than a few clicks to run in our homes, reliably, powered by affordable, reliable electricity. With the entirety of internet to learn from…(and without even getting into AI).

A study from 2017, for example, made the following claim about American households and 3D printers:

Buying a 3-D printer represents a return of investment of over 100% in five years.

My sense is that the real revolution with these technologies is yet to come. That the constraints to our thinking, imagining and creating are slowly…vanishing.

I hope we can skillfully navigate the noise, the confusion and the fear-mongering, and come up with protocols and standards that foster experimentation, learning and a better relationship with the environment.

What’s cool about this org?

Media Burn collects, preserves, and distributes documentary and experimental media produced by artists, activists, and community groups.

Our mission is to create positive social change by amplifying underheard voices, both in contemporary dialogue and the historical record.

Media Burn gets us thinking about what is worth remembering, and the default forgetfulness of our media worlds.

(You’ll be pleased to know that Tom Weinberg, founder of the Media Burn archive, was a collaborator of Ant Farm).

If we stay in the lanes of mainstream media, we only get to experience a narrowly-defined slice of what’s out there. We risk never being exposed to a fuller picture of life and its overlooked histories. That strikes me as a huge loss.

I’m not anti-YouTube; I think YouTube is great. In fact, in a previous post I described how Vulfpeck started as a YouTube channel, for example. There are hundreds of great content creators out there who might not be doing what they do if it weren’t for YouTube.

But, after all, it is a commercial platform of the contemporary, walled-garden internet. As such, it is prone to enshittification, a termed coined by Cory Doctorow to explain the lifecycle of platforms like YouTube (and Substack and X and TikTok etc…):

Here is how platforms die: first, they are good to their users; then they abuse their users to make things better for their business customers; finally, they abuse those business customers to claw back all the value for themselves. Then, they die.

What Media Burn offers, then, is not a replacement but an extension pack to the internet of video content. You get more out of the internet than YouTube’s abundance of hard-to-escape sameness. (Hint: Media Burn has a YT channel!)

Through Media Burn’s work and footage archives, you can dive into the quirky, weird, hobbyist, politically charged realms of video production, from documentaries to low-budget films and advertisements for some of the strangest projects out there, like the afore-mentioned Bluestar.

Or Electronic Masks. Or this cozy documentary about a double dutch rope jumping competition in NYC: Pick Up Your Feet: The Double Dutch Show.

I tell you: these archives are full of gold. Checkout their Watch Videos page and see for yourself. And for curated guides to topics in their collection, visit their Digital Exhibitions page.

Media Burn also enables creators in making new projects. They help people use their footage and archives to create new works, encouraging projects of all scales, from studio-funded feature films to student projects shot on smartphones. You can read more on their Licensing page to learn more.

Beyond preserving their own collections, they also work with clients of all sizes to digitize their obsolete videotape and audiotape media. According to their website:

Media Burn is one of only a handful of facilities in the United States with the capacity to play and digitize the many formats that were used during the 50-year-history of commercially available videotape.

That’s really cool, and more of a necessity than we might think.

The entropy of history and media

In data transmission and information theory, entropy is a measure of the information lost in a transmitted signal or message.

For instance, in an oral culture, only those thoughts that can be formulated into sayings, proverbs, and other dicta are likely to survive the entropic effects of oral transmission.

Video has its own entropic effects going on, which Media Burn tries to address by doubling-down on preservation efforts.

History, too, is subject to entropic forces and effects, primarily in the shape of political forgetfulness, our collective inability to tell the story of what really happened and, instead, record the story only of those with power during a given era.

Here I’m using the term political in its broadest sense: how we make decisions, as a collective, about what is valuable, just and desirable. Because forgetfulness is something that happens, but/and what we choose to do about it still requires a great deal of consideration.

Ways of remembering

For at least 200 years, more information and knowledge was better. It was the beacon of hope on which the internet was predicated upon. Until somewhat recently, sometime in the twentieth century, information became too vast to the point where it became detrimental to our well-being and functioning.

I’m not offering a solution, but I do encourage more experimentation! We will continue to struggle with this overwhelming abundance of information, and we’ll need to continue to come up with strategies to live and thrive in it.

It’s critical to highlight that forgetfulness happens both intentionally – think conquerors burning the villages, art, artifacts and literature of the conquered – and by default.

We forget when we fail to convince others of the the value of what we’re trying to continue to remember. We forget when those who cared about remembering retire, lose their cognitive abilities or die. We forget by sheer volume of content produced.

What can be remembered? Af first glance, this These we could call objects of history: events, stories, bodies of knowledge, songs, literature, rituals, oral traditions and many more.

What’s key to keep in mind is that what can be remembered also changes, as do our means to remember, encode, store and transmit. Media Burn acknowledges this, admitting that video killed the radio star, and that something else is already killing video; they’ll at least rescue some of the remains.

History’s quirks

You’ll have to fact-check these, but there’s a thread on Reddit about unlikely simultaneous historical events, shared by Jason Kottke here.

Here’s a sublist from that post:

• Spain was still a fascist dictatorship when Microsoft was founded.

• There were no classes in calculus in Harvard’s curriculum for the first few years because calculus hadn’t been discovered yet.

• Two empires [Roman & Ottoman] spanned the entire gap from Jesus to Babe Ruth.

• When the pyramids were being built, there were still woolly mammoths.

• The last use of the guillotine was in France the same year Star Wars came out.

• Oxford University was over 300 years old when the Aztec Empire was founded.

The past ain’t fixed

Here I mean fixed as in unchanging. The past is constantly changing, through the expansion of our curiosity and interest, technological advancements and the need to look at our recorded history through the lens of current events and cultural impact.

Of course, this process can be politicized and weaponized. The term revisionist has become an excuse to squeeze out supporting evidence for whatever public policy, cult ethos or fear-mongering agenda out of history. History, and the past, are infinitely malleable.

The goal, then, is not to fix the past and preserve it without change. We should, instead, continue to keep it alive by asking questions like:

• Who’s story is this?

• What are the recurrent motifs of history and in what ways have they changed or remained the same?

• Can technology help us remember? And how does it impact our forgetfulness?

• How can we know what really happened in the past?

• How do we remember/forget as individuals, and as a culture/society?

Forgetfulness is history’s entropic force. But we can choose to fight back against this form of entropy, and how we choose to fight matters a great deal.

What’s cool about this org?

Geomys is set up as an "organization of professional open source maintainers". They provide maintenance and support for critical packages in the Go programming language ecosystem.

How do they make money? A relevant question, given that most open source maintainers are not getting paid (or paid enough) for the software they publish and maintain. Geomys is backed by clients in retainer relationships.

Presumably, their clients have identified the need for specific software libraries, frameworks and packages that are critical to their businesses.

Given the importance of open source – which we dive into in the next section – this is an encouraging and optimistic model for financially sustaining key open source projects, and it appears to be working well.

Most recently, Geomys have started acting as a "maintainer of last resort" for security-related Go projects in need of new maintainers.

This is a promising model, and it will hopefully get replicated and expanded into the wider communities of open source projects beyond the Go language ecoysstem.

(Thanks to Simon Willison for putting Geomys on my radar)

Why is open source software so critical?

Quoting Nadia Eghbal’s report on the state of open source software:

Our modern society—everything from hospitals to stock markets to newspapers to social media—runs on software. But take a closer look, and you’ll find that the tools we use to build software are buckling under demand. Nearly all software today relies on free, public code (called “open source” code), written and maintained by communities of developers and other talent.

Much like roads or bridges, which anyone can walk or drive on, open source code can be used by anyone—from companies to individuals—to build software.

I get it, diving into this is getting into the weeds of Nerdom, and you’ve got bigger fish to fry.

But/and a lot of what we value in the world today would not exist without open source code being maintained by people who volunteer their time and talents to make code…and to share it with the public, making it available for everyone – and anyone – to use.

What’s the value of open source?

• It has dramatically lowered the cost of creating software and launching companies (e.g., startups like Imgur began with almost no capital thanks to free libraries).

• It has democratized access to learning: anyone can download a web development framework like Ruby on Rails for free, enabling self-taught programmers and diversifying who can participate in tech.

• It gives developers mobility, bargaining power, and fosters innovation through rapid iteration and modification (forking).

However, the hidden labor required to maintain this infrastructure is undervalued and often invisible. The people, their labor and the open source code they publish and maintain form the “invisible” backbone on which nearly all modern software gets built.

What’s more: everyone uses and relies on it—from small startups to Fortune 500 firms to governments. Why? Because it is cheaper, easier, and more efficient than building proprietary software.

There are risks involved, of course, since there is no free lunch. When things breakdown (which happens more often than you think), they often do so spectacularly. The Heartbleed bug in OpenSSL, for example, revealed how fragile this reliance on open source software can be.

The open source thread

If you read the last post published here about Precious Plastic, open source is a common thread between this post and that one.

The economics of open source software are incredibly lopsided to the point that it feels like open source shouldn’t exist. Even a young Bill Gates could not fathom a world where open source became critical infrastructure:

Who can afford to do professional work for nothing? What hobbyist can put 3-man years into programming, finding all bugs, documenting his product and distribute for free?

It turns out, a lot of people could back then, and have done so for many years to this day.

Which makes the “ ‘why' they do it?” question not as intriguing…why do any of us do anything?

What’s fascinating is that our world relies on them this much, and that, perhaps, this isn’t sustainable in the long run and we might want to rethink how this works. How it could work.

Don’t get me wrong: I’m sure open source maintainers don’t do it just out of the goodness of their heart. It’s some sort of effective altruism ploy. But our reliance on them is simply of too big a scale to ignore the pressures on them and the changing nature of software development and publishing, particularly in the AI-bubble we’re in.

Also, this should get us all thinking about the worlds that exist – an the orgs inhabiting those worlds – within the spectrum of “tooth and claw” capitalism on one end, and utopian socialism on the other.

Because, at the very least, there are entire sections of our global economy that rely on the gifts, talents and volunteered time of a few hundred human beings writing, and sharing, code.

A few recommendations

Working in Public: The Making and Maintenance of Open Source Software, by Nadia Eghbal

Roads and Bridges Report, also authored by Nadia Eghbal and published by the Ford Foundation

The Mozilla foundation on open source and AI: For the sake of our digital future, open source must win.

What’s cool about this org?

Precious Plastic is a wonderful open-source, community-driven organization that helps people build businesses from plastic waste.

They do other really cool things, too, like share R&D findings about how to turn organic materials (like food waste) into biodegradable products like plates, bowls or cups.

Plus, it’s also cool to see how organizations form and evolve relative to the challenges they’re trying to tackle.

Per the organization’s website:

Our solutions see people as the key element to fix the plastic mess. Precious Plastic approaches count on people to bring about the necessary change. Small steps, multiplied by millions. That's where we can win our battle.

We don't believe in techno-utopian, fix-it-all, dream technology.

Precious Plastic is a combination of people, machines, platforms and knowledge to create an alternative global recycling system.

It’s an impressive grassroots organization leveraging the power of the internet and open source ethos to “share all the outcomes of our research & development online”.

But, also…

Precious Plastic is In Trouble

That’s right. For all the amazing work they do, they are also at a pivotal moment. In their own words, the org is at “an important breaking point where we either give it a big boost or it will die”.

You can read more about it here.

I highly recommend giving that a read. Open-source, community-driven organizations are vital, now more than ever, and Precious Plastic has been running one for a while.

They’ve succeeded in their mission of building a global community of recyclers, and now they are at a crossroads and have some lessons learned to share.

They have set September 1st, 2025, as the date in which they will announce the future plan for Precious Plastic.

Plastic (and plastics reciclying) is contentious

Oh and by the way…plastic production and waste are intimately related with climate change, which is of course a highly politicized issue, but this post has no political agenda.

To put it bluntly, I don’t care about politics and other people do politics better than I do.

My position on this is best summarized by Tim Urban from Wait But Why:

Let’s ignore all the politicians and professors and CEOs and filmmakers and look at three facts:

Fact 1) Burning Fossil Fuels Makes Atmospheric CO2 Levels Rise

Fact 2) Where Atmospheric CO2 Levels Go, Temperatures Follow

Fact 3) The Temperature Doesn’t Need to Change Very Much to Make Everything Shitty

Now that that’s out of the way, let’s move on. But/and before we dive more deeply into Precious Plastic, let’s set the stage for the severity and scale of the problem.

It is so sticky a problem, in fact, that it serves as a perfect-storm example of the kinds of problems we will be forced to face as a global human civilization.

Some facts about plastics (and we won’t even get into microplastics):

• 400 million tons of plastic waste are produced worldwide each year, but only 5-6 percent is recycled.

• Plastic usage and waste have been on the rise for the last 30 years

• Significant stocks of plastics have already accumulated in aquatic environments, with 109 Metric tons (Mt) of plastics accumulated in rivers, and 30 Mt in the ocean.

• Plastics have a significant carbon footprint, contributing 3.4% of global greenhouse gas emissions throughout their lifecycle. In 2019, plastics generated 1.8 billion tonnes of greenhouse gas emissions, with 90% coming from their production and conversion from fossil fuels.

– via OECD, Global Plastics Outlook

And here’s a lawsuit filed by California’s Attorney General against ExxonMobil for allegedly engaging in a decades-long campaign of deception that caused and exacerbated the global plastic pollution, and sheds a light on the petrochemical industry’s role in the plastic pollution crisis.

The jury is out on whether plastic recycling moves the needle or not. In the meantime, orgs like Precious Plastic are taking matters into their own hands instead of twiddling their thumbs.

On the emergence of structure relative to challenges-to-be-tackled

A quick sidebar…I’d like to think that there’s a latent contribution to the theory of organizations here… My modest start towards that is that an organization’s structure – from it’s hierarchy (or lack thereof), to its cadence and rhythms, incentives, metrics etc… – is in constant reciprocity to the challenges that the organization is tackling.

In short, that Precious Plastic’s structure could be plastic-y. That its structure mirrors, say, plastic waste’s ubiquity – 400 Metric tonnes of plastics are produced every year. Or plastic’s stickiness – it takes anywhere from 20 to 500 years for plastics to decompose, depending on the material and structure.

I’m sure this happens implicitly with organizations of all kinds. Data centers and their anonymous architecture and secretive, inconspicuous pervasiveness is one example that comes to mind. Bit what if this mirroring was made explicitly, as a conscious choice?

Glorified waste, precious solutions

It’s brilliant. To treat plastic as a precious material, “to turn waste into wander”.

Through Precious Plastic, you can:

• Learn to start a business from plastic waste and/or your own recycling center

• Learn to build DIY machines to recycle plastic waste

• Buy and sell your own machines

• Get the latest from the biodegradable materials R&D

• Learn how to make furniture, construction materials, modular structures and high-precision moulding

I highly recommend perusing their website and online community

Lessons from running an open-source, community-driven organization

I’ll mention this again here because I think it’s worth a read:

Precious Plastic is In Trouble. Don’t want to click a link? That’s fine. Here’s an excerpt:

#5 Open Source community

[…] We went all-in on giving. And believe(d?) that sharing Open Source will bring contributions back one way or another. Contributions to support the Precious Plastic Community can be made in various ways. And many members do contribute something back which is great. However we’ve also been observing quite some established organizations that take more than they give, building business around Precious Plastic but not contributing anything back. It’s allowed, since it’s all Open-Source. But the mentality of only taking things and not contributing will eventually kill a community-driven project like this.

#6 It is bad designed

We don’t blame those for not contributing enough back. We see this as a fault on us. The project wasn’t designed to have a healthy financial model and relation with the community. We were always fully focussed on giving to the community, not us being a financially sustainable organization. Funny example of this is the recent PPOSF (Precious Plastic Open Source Fund). We received a €100K donation. Which was amazing, but we decided to give it all to the community so they can continue developing their projects. Not to sustain the organisation itself. You could see this as a humble move from us, give it to the community. But it isn’t, it’s ineffective, because now we have to bother you with our problems. Ideally we don’t have to do this and you don’t have to worry about us.

#7 No long term team

As you can imagine all of the above problems make it hard for a team member to have a long term perspective in Precious Plastic. Even though the team has been very small, effective and works with many volunteers it has continued to be a struggle to pay everyone every month without worrying. Over time this brings lots of stress and uncertainty, especially if the team members by themselves grow up and need more stability in life.

Some cool projects affiliated to Precious Plastic


ROBRIES

Opo Lab