Unbundling Threads

Published Originally on Synthetic Futures in Janurary 2023.

When you unbundle an industry you get diverse new types of products and services at more intuitive layers of the stack. From Google PageRank’s algorithmic unraveling of Yahoo’s directory system, and social networks reformatting paperbound address books, to MP3s liberating tracks from CDs. Unbundling gives innovative startups, creators, and other independent entrants the edge they need to convert the Innovator’s Dilemma into breakout opportunity. It’s a direct path to challenge legacy players, as making established markets open up through modularization hot swaps contrarian perspectives into them.

You can build now, off these new verticles, compelling and ever new products and services while the old incumbents continue to stumble. Crushed by the weight of their legacy choices, and costs of internal teams who owe them no loyalty, they will continue to compromise the quality of service and performance of individual offerings in ways no amount of reengineering can overcome.

What the results of the better part of the past two and a half decades have continued to surface, is how even once innovative startups, primed to disrupt everything and ‘eat the world’, are not immune themselves from statically bundling back up into faceless enterprises of the kind they once set out to undo. This time with more ping pong tables and beer on tap, until the layoffs ran them dry. Facebook, AirBnB, Uber, Tesla, AOL and Twitter, to name a few of the countless many, are ripe for overhaul. Or, have already begun their steep cliff dive into nostalgia. Fracturing along the fault lines. The Old King is Dead, Long Live the King. Unbundle, bundle… the cycle continues.

Or does it?

Joseph Marie Jacquard first debuted his upscaled modification to the hand loom at the 1801 Paris exhibition. This many geared and levered contraption hung from the ceiling, weighed around 4000 pounds and took up 72 square feet of floor space at 16 feet high. The most complicated machine invented at the time. Thousands of punch cards engaged and moved threads in a syncopated fashion to automate textile production with shockingly high resolution and precision.

With his device, productivity in weaving increased ten-fold. It rose from requiring multiple skilled weavers engaged in back-breaking labor to produce just 2 inches of fabric a day, to needing only one operator now with capacity to generate up to 2 feet of fabric daily. And it didn’t simply or directly improve the quality of life for those making their living from textile manufacture, but the increased demand from faster production times now made it possible for anyone to purchase and wear fabrics that had previously been reserved and available to only the most wealthy elite.

This profound step towards mechanized weaving democratized a controlled market previously dominated by oligopolies. It was an unbundling of the established textile industry through increased open production, where smaller firms and independent weavers could enter the market to compete head-on with the larger players. The fragmentation of the industry allowed these independent workers to focus on specific niches and verticals, from raw materials, to dyes, to spinning, weaving and tailoring. In turn, this also increased competition and innovation — so much so, the programmable loom became one of the most important and immediate precursors to modern computing.

Bundling, unbundling and rebundling. These simple seaming conceptual structures describe the process of breaking apart complex systems, platforms, industries and market segments into their base components, offering opportunities to reason fresh from first principles, and then recombine them in novel ways as they take to market in more nimble units, with further consequence.

Unbundling as a process has been accelerated across nearly all industries with the rise of the internet and market saturation in the digital age. Previous monoliths have become dismantled piecemeal across verticals and horizontals, with greater choice for consumers, competitive advantages for savvy teams, and economic dynamism for society at large.

We’ve reached the end of history, or economics, then. With formulas for how new opportunities rise from the study and ashes of old legacies.

Not quite.

There is something uncanny working away, teasing out and unwinding loose threads at the edges of the fabric, where industries that thought they had it all figured out once and for all are now becoming undone again.

With a surreal promise of content-product-message hypersaturation in every industry, and counterintuitive increase in self-determinative independence in both voice and earnings, open source AI models offer something else, delivered through (and field welded to) the decentralized web.

It teases at the sudden inversion of who gets to create, and who we choose to consume from.

What else does this mean? And for fashion?

Despite its disruptive patterns, from the mid 20th to early 21st centuries the fashion industry has made itself into one of the least dynamic industries despite all outward appearances. Manufactured glamor and shallow glitz cover up the truth of how the machinery works under the surface. Only a handful of mega-conglomerates and supplier cartels control the supply chain, from materials sourcing to manufacturing and distribution. Dogmatism makes it feel impossible for indie designers to enter the market. And that’s exactly how incumbent brands like it.

From prehistory to the nearly present ahead, textiles have made the world. Precisely. They’ve spread alphabets, funded the Renaissance and set the inkwork to parchment for the development of double entry bookkeeping, borders crossing commerce, and computing.

Are we expected to give up all hope that innovation in fashion — and throughout creative industries thanks to it — will continue?

Fabrics are the original story of human ingenuity, covering (and even pre-dating) our entire lifespan as a species. It was the use of natural fibers from plants and animal skins that ensured the survival of our ancestors against harsh environmental conditions as they compensated for evolutionary hair loss. Homo Erectus are believed to have first made and worn clothes around the onset of the Ice Age and as they migrated into glacial Europe, but it’s likely far older. It was fabrics and the fabrication process that allowed us to wrap and store food, boil water, create and carry fire, seeds and even make new tools. Twisted fibers and tanned animal hides offered a more durable material for creating ropes, baskets, nets, shelters and weapons for hunting. And beyond utilitarian functionality, especially for those prehistoric hominids that stayed in the warmer tropical environments of Africa and south x southwestern Eurasia, animal-based apparel match the first appearance of personal adornments for style.

The first rudimentary dyes paleo archaeologists know of date back to the Neolithic period, around 10,200 BC, with primary sources consisting of roots, berries, bar, leaves, wood and other organic substances like ever friendly neighborhood fungi. When European colonists spread to the Americas, natural dyes came with them, dominating the global textile market until the invention of the first synthetic dyes marching onwards from the mid 19th century. In 1856 William Perkins serendipitously synthesized the chemical concoction for mauveine whilst experimenting on formulas for quinine, for which future inhabitants of malarial climates would be forever grateful. And, it was through the process of home-dying silks and conversing with the Scottish Pullar factory that the first set of synthetic dyes were pioneered. Perkins’ purple, and the adjacently developed mix of high quality and consistently cheap and vibrant colorants, changed history.

These synthetics seeped into other industries, explosively expanding the dissemination of knowledge as the printing industry grew. They spurred innovation in photography and film with new primary sources of color. Pigments were used to enhance the visual appeal of meats, jams, baked goods and even hair dyes. Most notably in the fields of biology and medicine, the dyes were used to stain cells in research leading to breakthroughs in cell anatomy, the understanding of the makeup of our DNA and chromosomes and even cures or treatments for pathogens and diseases like Cholera and Cancer. The increased demand for textiles and its constituents continued to feed the Industrial Revolution and unlock new forms of automation and autonomation.

By 1930, textiles had shaped more efficient forms of production and manufacturing with the introduction of the concept of “Jidoka”, a Japanese term developed by Toyota Motor Corporation. The practice took hold when Sakichi Toyoda invented a textile loom that stopped automatically when thread broke. Previously when thread broke, the machine would churn out massive amounts of defected fabric, meaning that each machine had to be constantly monitored by an operator. With Jidoka, the combination of automation and human intelligence improved production quality and reduced waste and errors, enabling Toyota to transition into becoming one of the leading global vehicle manufacturing companies of the 20th and early 21st centuries — where its “lean manufacturing” or “Just-in-Time” system is now one of the most well known and studied modern forms of factory powered craftsmanship.

Advances in textile technology have further enabled new forms of information records, storage and transmission. In China, the practice of painting and calligraphy on silk scrolls became a highly regarded art form, accelerating the development of writing systems, along with the invention of paper in the 2nd Century BCE. One of the earliest studied examples of textiles used for accounting dates back to the Quipu method developed by the Inca civilization. Strings were tied in various knotted and colored patterns for numerical representation and record-keeping. In the Renaissance Mediterranean and spreading to the merchant city states of the Hanseatic League, doubly-entry notation soon replaced single entry as essential clusters for textile production and trade in Florence and Venice used the wealth they generated from this industry to fund the patronage of artists, architects, and intellectuals who were at the forefront of the Renaissance. If Leonardo DaVinci rings a bell, you have the ancestors of the clothes you are wearing to thank.

In more modern times, especially during the past century, the impact of textiles on information technology has been even more revolutionary. The Jacquard loom, one of the earliest computing devices, was later used by Charles Babbage and Ada Lovelace in the development of the analytical engine and set the precursors to modern day programming from binary. It is no coincidence that modern hardware and software engineers commonly use the term “multi-threaded” in computer architecture. All the more, the invention of fiber optic cables, used to reliably transmit data at lightspeed across global distances, have enabled the internet, video conferencing, telemedicine and almost every device we all use 24/7, non-stop.

Electronic functionality has also extended to surfaces beyond conventional recognition, with advances in conductive fabrics. These threads, made of silver and copper, are machine woven and knit into fabrics to create electronic circuits and sensors that can be integrated into wearable technology, smart textiles and alternative products and applications.

This list could be endless.

So many have refined the function of fashion for civilization to a point where its impact is so profound that it seems to disappear — indistinguishable from the threadstuff of everyday life.

The next step is digital, decentralized, diffuse.

What will be obvious in retrospect, is not so often obvious in the thick of battle. The laws of disruption carve unique opportunities for indie creators.

Designers can build loyal followings by targeting niche markets. Offering distinct styles differentiate from the hypersaturated competition, and overtake established players.

Just as the iPhone transformed how we see and interact with content, practically yesterday in the historical view, image and text synthesis now transforms how we use and create all forms of design and media. As much of a new interface as a new canvas on every kind of surface, anyone with an internet connection and enough effort to watch a few videos or read some Reddit threads to catch up through the learning curve can tell compelling stories, iterate through a relentless number of branding ideas, and co-author strategies for the spread of personal and / or commercial information.

For web3 fashion, synth machines accelerate the unbundling of the fashion industry, and of adjacent creative sectors, across all verticals. From design and prototyping, to manufacturing, distribution, retail, marketing, and vast stores of personal record keeping in the zk-encryption secured public memory.

AI and web3 have long hovered too often as buzzwords poorly understood by the mass market, at the fitful edge of legitimacy. Yet, their creative power is obvious, whether you analyze them in good faith or fail to do so. Unlocked by text-to-image and image-to-image diffusion, combined for persistent distribution through NFTs and decentralized marketplaces, they obliterate barriers to entry while drastically reducing the usual overhead of traditional production and sales channels. When you consider their use in open source 2D/3D digital modeling tools like Blender, enhanced by fashion synth AOP editors, a new canvas for prototyping, development and publication of information, media, design and creativity fails entirely to capture the full extent and power of the transformation underway.

With the help of ChatGPT, and some practical insights, we can break open each of these key use cases to gain a better understanding of what’s involved in the refounding of everything.

Design and Prototyping

Proprietary closed source digital fashion editors, like CLO3D, promised to provide designers — particularly those with empty pockets — an efficient tool for digital design and prototyping before big commitments to physical production. Although valuable to a growing community given a real taste of what sophisticated 3D software can do, the lack of file interoperability and locked down codebase of these editors limit accessibility and prevent more novel advancement of fashion manufacturing processes and methods. Some go as far as issuing T&Cs that claim the company owns all models made within the app under their closed licenses. Indie creators are still hopelessly stuck in a self-disinvestment loop while the few mega-conglomerate players who have the time and resources to invest more heavily in these editors continue to flail forward far ahead.

Blender, as the pre-eminent open source all-purpose design tool, offers a free and interoperable alternative. Unfortunately, it lacks comparable and sufficiently powerful digital textile modeling tools, for now.

In both cases, even when prototyping in 3D, the process of converting a prototype into a usable physical pattern for garment design, fit, and construction is almost impossible for most designers. Not simply because of a lack of functionality in these applications. Pattern blueprints and templates in the fashion industry are also inherently secretive and proprietary, by tradition, where even general sizing systems are notoriously difficult to tailor to without industry exclusive know-how.

It all operates as an esoteric mass capital accumulation priesthood. An admittedly stylish one, squirreling away secrets in hushed tones amid sparkling runways.

This restrictive in-group culture where insiders hoard expertise down to the buttons and thread is not new. It has been de rigueur for many centuries — which is why what Marx Zeigler did in the 17th century was so radical. The publishing of Weber Kunst und Bild Buch, the very first practical ‘how-to’ weaving pattern book, was made open and available to all. It contained detailed instructions, motifs and weaving drafts for the weaving of decorative wool & linen textiles and sparked an ethos of knowledge sharing which continues to advance science, engineering, and industry today.

In a similar way, All-Over-Print fashion synth editors carry on Zeigler’s legacy to provide the foundation for rapid prototyping and dissemination of unique garment collections made from open source fashion templates.

With the ability to generate endless varieties, designers can respond to growing environmental concerns directly, by testing in digital, and porting only their selected outputs to IRL production. The open source nature of these editors also provides the components of innovation essential for adoption of zero-waste pattern templates, making sustainable fashion design and prototyping directly available to independent and aspiring designers.

Further, as these AI open editors save time and labor, they feed a secondary abundance by freeing people to choose more personally satisfying work. This creative process shortens the distance for indie creators to reach commercial markets without sacrificing your integrity.

Manufacturing

The corporate chokehold on fashion manufacturing has also resulted in major overproduction of unsold lines, and unfathomable amounts of apparel waste. Bad bets on massive production runs go directly to the landfill. Recently the Atacama Desert in Chile was the latest victim. Used as a dumping ground for over 39,000 tonnes of non-biodegradable synthetic clothing.

In many cases items skip the storefronts, with tags still on, as they head straight to the trash mountains from the factory floor.

Mass produced fast fashion is responsible for unconscionable environmental damage, with fashion as a whole known as the second most polluting industry globally. Demand for perpetually new items at lower prices feeds cycles of ever lesser transparency, flaunting their lack of accountability, while making a show of their exploitation of workers in countries with no respect for personal liberties, inherent human rights, or the kinds of simple decency most consumers of these products would take for granted.

It’s not all Mad Max, yet. The unbundling of the industry taking place in design and prototyping, through fashion synth editors, web3, and decentralized social media, are also clearing the way for the rise of local micro-manufacturing co-ops.

Here, with less reliance on intermediaries, small batches of designers can directly manufacture their own collections, and develop global fulfillment network relationships.

In more macro-economic terms, decentralized manufacturing means developing nations can incubate new sources of employment that stimulate broad based economic growth, with practices that are more ethical, accountable, and sustainable at every step of the supply chain.

Distribution

Direct relationships between creators and consumers have long been the dream pitched by disruptive technologies. As quiet quitting by wage workers, long treated as expendable, takes hold throughout more of the traditional retail distribution pipeline… what happens when any retail clerk can moonlight as an indie hybrid fashion model, chatslinger, micro-influencer, and synth designer?

High markups and slow response times to consumer demand fade when faced by relentless competition for attention.

Decentralized marketplaces, decentralized social media, and NFT-stamped cultural assets are just plain better distribution channels than all of algorithmic and brick-and-mortar-verse obstacles in traditional commerce. Direct-to-consumer channels put profit and sustainable success in the hands of designers. Jokes about how mangled those hands might be by the first few months of Stable Diffusion models aside, as those are already becoming obsolete thanks to newer ControlNet libraries.

AI has long promised, and is finally delivering, the ability for creators to optimize inventory management, improve supply chain visibility and enhance customer experience. Personalized online and IRL shopping experiences which more accurately provide valuable insights into consumer behavior, preferences, and trends are real now. This greater fluency, with endlessly generated information, can make informed decisions about critical components of a successful distribution strategy, from pricing and shipping logistics to geographical sales targeting, more of a skill to be honed than a daydream deferred.

Retail

It’s no secret that traditional strip malls and shopping centers are struggling to adapt. Living on borrowed time since the beginning of online shopping’s dominance, low effort and big box storefronts aren’t looking so good after the pandemic. Virtual try-on and augmented reality are hyped as the vanguard disruptive technologies to traditional fashion retail. However, if these new outlets are made in the same hub and spoke, top down control styles as current retail spaces, immersivity will be inconsequential.

Decentralized interfaces, in comparison, are not limited to the screens and surfaces that we are accustomed to today. As AI continues to feed user generated content, VR and AR tech, as well as more immersive and interoperable game-like experiences, continue to advance. The metaverse economy is delayed by big tech layoffs, but still set to open vast new territories for fashion content creation and sales. New designer stores will bleed across virtual, latent, and IRL spaces to stake their claim to highly personalized and interactive engagement funnels. Transactions are no longer limited to traditional checkout counters or mobile phone UX flows. Instead, they are more likely to morph into collectible quests, and loyalty driven exclusive content unlocks, as isekai-like QR portals are spread throughout cities worldwide.

The infinite canvas that synth models give to creators, along with improved local manufacturing facilities, also means that the very nature of what our apparel represents — what it does for us — evolves. Beyond protection, status and style, our worn fabric surfaces will become embedded with smart wearables and invisible electronics over time. These new mobile storefronts, social media carriers, and sales channels will communicate with us through what we wear.

Marketing

Marketing for an indie designer today is significantly different than it was a decade ago. In a number of ways, the overwhelm of social media algorithms driving overnight virality promised a more equitable jumpstart for creators able to leverage these platforms quickly to build a following, and grow interest in products. If your content can feed the algorithms what they want, your products can potentially reach tens of millions of would be customers, on a non-linear timescale, sometimes even while you sleep.

Of course, it’s still the traditional content and fashion industry cartels who have made the most of algorithmic social apps, particularly with their ability to dominate brand advertising, and to pay for whatever it takes to outrank you, as the networks become oversaturated with precisely calibrated junk.

What the recent AI advances have shown is that the way we respond to, and take in, content is changing. Moving from cheap slogans to narrative driven art, creators who make works that are driven by strong narratives, whether political, societal, instructional, or niche cultural, reinforce the meme that designers are storytellers. It’s David vs. Goliath. And we all know who we’d rather be, or bet on.

Serialized content storyboards printed on apparel connects buyers to products and each other, through messages, subversive instructions, and memes as they don’t follow in the footsteps of webtoons, but better yet, seek what they sought.

With unbounded limits to production capabilities, the use of AI is also facilitating the shift towards video content as the primary way for brands to connect with their audience — with interactive audiovisual content requiring a higher level of creativity, storytelling and technical skills, acting as the new litmus test for distinguishing between genuine organic creator led content and truly effortless one-armed-bandit gambling addiction crap.

This counter cultural approach emphasizes the advantage of public domain content, where the spread of ideas is a radical incentive. Open source design and CC0 creative content promote hyper spreadability, remix culture and guerilla street marketing tactics as high quality and captivating posters, stickers and other street art works make public spaces more vibrant, enjoyable, useful, and diverse.

And as if the positive potential of decentralized social media in the unbundling of marketing isn’t obvious enough, it is also worth mentioning again that with open source base layer protocols like Lens, it’s now more within reach than ever for future-tech savvy designers to spin up their own social media platforms to open-up algorithms, uprank your own influence, and grow your own communities free from the edgelord snark endemic to the platform lords of old.

Whatever your fashion, gear up. The unbundling of every industry awaits.