Decentralized Digital Distribution

Three major technologies underpin the next generation of the internet: transaction-enabled APIs, digital currencies, and BitTorrent style file distribution.

Modern software is built on the API, or application programming interface. It’s a method that enables one piece of software to talk to an independent piece of software. APIs depend on “calls”, which is where one piece of software communicates to another piece of software via the API. It’s like a phone call between two pieces of software to trade data.

Traditional APIs don’t have any transaction method attached to the API calls. When you’re using software on your own computer for your own use, there’s no real justification for an API call to be priced. But when software is exposed to the internet and used by millions of people, the infrastructure costs of hosting and bandwidth for supporting all of those interactions gets very expensive.

Recently major websites have begun charging for access to their API, such as Reddit and Twitter. Those APIs aren’t built with an inherent currency transaction method. The only way for these websites to commercialize their API access is through setting up a separate account for each user, independent of the API, then including an identity in the API call that relates back to that user account. The websites then sell a large number of calls to the user account, typically in the millions or billions of calls per month range. This is for a number of reasons, including that API calls themselves are usually very low-cost and only have a significant cost when there’s a lot of them. And because one user interaction can result in dozens or hundreds of calls.

But mostly it’s because the companies involved want to take larger payments from fewer parties due to the costs of processing transactions. Transactions have a relatively fixed processing cost, if we take a VISA/Mastercard situation as an example, the recipient may pay $0.35 per payment plus 3% of the transaction. This means that any payment less than around $0.36 loses money. And at that cost profile, for a $1 payment, more than 1/3 of the payment goes to the transaction processing. There’s also the administration and overhead of accepting payments. Someone has to monitor the system, fix errors, resolve disputes, and other needs. This further increases the desire for fewer but larger transactions.

Then comes the scale of the business – a billion-dollar business does not want one billion one-dollar transactions. “But a can of coke…” you might say, but Coke doesn’t work that way. Coke licenses their products to bottlers, who then sell in bulk to distributors, who then sell in bulk (but smaller bulk) to retailers, who then sell to individuals. This multi-tier approach helps to ensure that the transaction size at each level of interaction relates appropriately to the transaction size at the next level down. Coke isn’t selling a billion cans at $1 each, they’re selling a million cans to a thousand buyers, and that thousand buyers sell a thousand cans to a thousand buyers, who then sell one can to each retail customer. This is an example, don’t take it for gospel. The whole point is that the transaction size needs to scale to the business scope to avoid a massive overhead cost for dealing with a huge number of tiny transactions.

But digital systems transform this entire ecosystem by enabling a standards-based automation of services for compensation, without significant human interaction, through automated computer processes. That’s how the internet works, there’s no person strapping your data to their back and carrying it across the internet to you, it happens automatically.

This is where digital currencies come in. By attaching a digital currency transaction method to the API, the API call can incorporate a payment method automatically, so that the API only responds to a service call that includes the correct payment. This method is actually built into HTML and has existed from the start, it’s called “402 Payment Required”. We’re familiar with 404 Not Found, or 501 Service Unavailable, but most people have never seen a 402 Payment Required notice in their browser because it was anticipated by the original developers, it’s never really been implemented. Bitcoin, which most people think of as the “first” digital currency, was never the first. There were many digital currencies developed in the original .com boom of the late 90s, but they were never widely adopted. Bitcoin was actually inspired by these first attempts at digital currencies, and attempted to improve on them.

By using a modern digital currency method, we can incorporate microtransactions into API calls, and then use the 402 Payment Required service (and other methods like it) to include a transaction method directly into the API. This would require a wallet for the web servers and end users to be able to pay with, but wallet methods are built into most modern browsers – when was the last time a Chrome user had to manually type in their debit card number?

Adopting a browser or server extension that holds a transaction method for 402 Payments is a fairly trivial exercise at this point, albeit one that hasn’t previously been popular. That’s largely because there’s not been a reason for it yet, as the lack of a reliable microtransaction method in the early internet led to the monetization in another way – through Google Adwords. Google realized that they could subsidize internet activities without user-based payments through data mining those user actions and selling access to “eyeballs” through Adwords. This is how Facebook monetizes their service too. As well as most other websites, including news sites, and content sites like Youtube.

This realization made Google one of the most “valuable” and profitable companies in history. The spin from user-pays to intermediary-pays financed an explosion of “free” internet services over the last few decades. This was a major evolutionary transition for the internet, to shift the focus away from the person visiting the website, and refocusing the purpose onto maximizing the advertising potential of the content of the website. The internet isn’t user-focused, it’s advertiser-focused, and that shift in purpose, and the change in values the shift in purpose represents, has led to a massive deterioration in the quality of the internet and the way that users interact with content.

Let’s be frank for a moment: Are you tired of ads? Are you sick of that crap? Always someone shoving some worthless garbage in your face and begging for money? Websites that have no useful content and are SEO’d to hell and back, just to “get the clicks” to monetize emptiness through advertising? Aren’t we all collectively exhausted by constantly being data mined and advertised?

This can all be repaired by stripping out the advertising-orientation of the internet and replacing it with a user orientation. Currently the user isn’t the customer, they’re the product, and the advertiser is the customer. The only way the user becomes the customer is if the user is paying, and right now, there’s no good way for the user to pay. Even when the user can pay, for example with subscription based services, the user ends up getting locked in, data mined, and monetized, because the user-paid services are “platforms”, not methods.

The advertising approach shifted attention away from core internet systems – computer software methods – and into monolithic, corporate-owned “platforms” that lock users in, lock data in, and monetize the user twice, first from paying for a service, and second from any data that the user contributes to the service that the company can sell. This “platform” approach doesn’t benefit users or creators, it benefits the platform, the advertisers, and the platform’s investors. Look at the newest developments of Google Bard, that is currently datamining all of the content stored in Gmail and Google for Business in order to improve something else that Google will sell. This platform approach to AI, “metaverse”, and other major technical developments (like crypto “platforms” such as FTX) have neutered and stymied the value of new technologies by forcing those new technologies to be constrained under the paradigms of the limitations (and cost/benefit profiles) of the previous paradigm. This doesn’t benefit users, or content creators, it benefits the platform owner and the advertisers.

The same can be said of creative platforms like Youtube, Instagram, TikTok, Soundcloud, Spotify, and more. These aren’t software methods where the content is owned by the creators, and the creators are paid for access to the content. These are platforms where the content is made by the creators, hosted by the platform, the platform is paid by advertisers (and some subscribers) and then the platform pays a pittance to the people whose content contribute the value that justifies the payment. In this platform-based approach, the people contributing the value by producing content get essentially nothing, while nearly all of the money is kept by the platform intermediary (and their financial backers).

The justification for this is manifold:

  • The platform costs a lot of money to develop to make everything possible. We can call this “Coding Costs Money”. This is true, but addressable in another way.
  • The hosting and transmission of content is expensive. We can call this “Hosting is Expensive”. This is true, but addressable in another way.
  • The people who finance the development and operation of the platform deserve a return on their investment. “Finance Requires Return on Investment”. This, like the rest, is true, but can be resolved in another way.

There are more reasons given but these are the main ones we’ll worry about for now. Let’s discuss them briefly in turn.

Coding Costs Money

A platform is simply a codebase, like any other software. Yes, software can be expensive to develop, but that expense is one-time. New features and capabilities are also expensive to develop to extend the platform (or other software) but again, that’s one-time. Digital systems costs of replication are essentially free – what does it cost you to copy-paste text from one place to another? A tiny bit of time, a tiny bit of electricity. It’s not like transporting a house from one spot to another, it’s not like copying the design of a house and constructing a second one. The essentially-free cost of copying digital products is very well established and understood.

Companies will hire a vast host of developers to build their codebase, and most of those developers are using existing frameworks and applications to build that private codebase. It’s a universal joke among developers that they’ll use Google, StackOverflow, Git projects, and other publicly accessible or open source tools to solve coding challenges. And then that code that was developed by someone else, who receives no compensation, and copy-pasted around the world, is incorporated into a private codebase that is privately monetized for the private gain of the private company.

Coding costs money, but most of the money made from coding doesn’t go back to the actual original developer, or even the developer who copies and implements the code – the costs of hiring the developers are borne by a private organization using private money, but a big portion of that code development is stolen. These massive development teams will regularly pirate code snippets that someone else already wrote, from some resource they find for free online. The developer then stitches them together, or uses existing algorithms, math structures, and open-source toolkits, plugins, and libraries to implement what is later used for private profits.

This means that, while major platforms may hire thousands upon thousands of developers, a significant portion of that code base is absolutely dependent on either the volunteering of labor from a non-employee, or the theft of that non-employee’s previous labor. Private codebases are filled with code that is not legally owned or possessed by the company monetizing it. And most of that private code that really is written in-house is simply the glue that’s holding other pieces of code together that someone else wrote a long time ago. The maintenance of closed-source private codebases represents an enormous and almost completely unnecessary duplication of human effort, and replication of functions and code segments whose development costs were already paid for by someone else previously.

The privately paid developers of these major platforms are in many cases simply hunting out the code they need and pirating that content for private benefit – the same objection that content platforms make to complain about people using their platform without paying! How absurd is the hypocrisy of a private enterprise complaining that people are watching their content without paying, when much of the code that their platform relies on was copied from someone else without paying?

Hosting is Expensive

Running a cluster of webservers, or entire data centers, to serve content is incredibly expensive. Facebook is currently investing $800m into a data center in Kansas City. As of 2021, Facebook had 18 data centers that cost it $20 billion. Major content organizations like Google, Netflix, Hulu, Disney, and others have several to dozens of data centers around the world to host their content. One of the reasons given for Youtube being so expensive is the sheer cost of hosting that data, and transmitting it to whoever wants a copy.

Nearly all data centers will have some form of local hosting of a fraction of the content on these platforms, often called a “shard”. Getting a “box” that hosts a “shard” can cost that data center hundreds of thousands of dollars, but provide them a good return on that investment. These costs are further allayed by “content distribution networks” (CDN) that act as load balancing and rehosting facilities for content on-demand. Some CDNs will pay to be located into the data center, and get paid from the content providers they’re hosting. Other CDNs will get paid to be located in the DC, to extend and improve the content distribution of other services and content providers hosted inside the DC, so that the DC can improve the efficiency of their own operations.

On top of the hosting expenses (having computers that store the files and send them), there’s also the internet service costs to transmit those files to other places, like CDNs or users. Laying and operating fiber networks have huge up-front costs, but the long-term operation costs are relatively low. For example, Time Warner Cable was running about a 95% IRR before it was sold off as Spectrum in the late ‘10s. This is because Time Warner absorbed billions of dollars of investment to build out the cable (and later, fiber) network. Time Warner charged high prices to users to repay the costs of implementation of the network, but that network’s expected life is much longer than its amortization life.

Once the network was in the ground, the company only had to keep it running. They kept charging the pre-amortization prices to users, with some tweaks like increased speeds over time as the technology at the ends of the network improved. Modern lasers on older fibers can pump many times the amount of data as the original lasers that powered the network when it was built. The fiber doesn’t need to be replaced, just the lasers at either end. So once the original costs of implementation were amortized, leaving behind only the operating costs, but the charges to the users stayed about the same, the enormously expensive network was enormously profitable.

All that to say, running a major content platform (no matter what the content) incorporates huge costs for hosting and data transmission.

Finance Requires Return on Investment  

Almost nobody intentionally loses money. Anyone paying for something does so because they perceive the benefit of the purchase to outweigh the cost. This is true for everything, all the way down to a cup of coffee. A coffee shop sells you a cup of coffee because your money is more valuable to them than keeping the cup of coffee to themselves. And it’s asymmetrical in every single exchange: The purchaser of that cup of coffee considers the coffee to be more valuable than the money they’re giving for it. This asymmetry exists in every transaction. A purchaser of a home finds the home more valuable than the price of the mortgage, while the mortgage provider considers the payments more valuable than the home. A seller of corporate equity considers the cash more valuable than the equity, while the purchaser of equity considers the equity more valuable than the cash. When this asymmetry does not exist, in nearly all cases, a transaction is not made. Even charity experiences this – when someone donates to charity, they consider the benefits (tax write offs, feeling good about ones’ self, helping others, whatever) is more valuable than the money they give up. Everyone wants an ROI.

The problem is that building a software platform requires an enormous amount of up-front investment. And building the hosting and distribution network for that software platform and its content requires an enormous amount of up-front investment. These investments incur risk, and the higher the risk, the higher the expected return profile to justify the investor taking on that risk. A VC firm, for example, has a target return of 75% IRR from its investments in order to generate a 100x multiple on invested capital (MOIC) from a small number of “wins” (and a much larger number of duds). The VC firm needs a 100x MOIC from a few investments so that, after the overhead of running the VC firm, and the subtraction of the carried interest that the VC firm keeps, it can return a target 30x MOIC to its own investors. This math is absolutely brutal, but it’s required to make investments at the scale that is necessary to build a Google or Facebook.

Reaching these return profiles requires that the invested companies behave without mercy towards their users and clients. There are countless startups that lose money hand-over-fist to gain marketshare, and once they have penetrated the market, they turn the screws on everyone and begin outrageous price and cost increases to repay the original risk capital.

Amazon was the low-cost seller for the better part of two decades, but now is often the same price or more than local stores. Facebook drove most news sites into impoverishment, but now Facebook is so filled with ads its almost impossible to use. Netflix, and eventually other streaming companies, drove video rental out of the market, and basically killed cable (an argument is to be made that cable committed suicide, but that’s neither here nor there), but now continually raise their prices. Uber drove out cabbies, but now an Uber is the same price or more than a cab. These companies use investor money to subsidize the cost of business to gain market share, establish a hold in the market, and then impose extreme price increases to generate investor returns.

Now that Doordash is far more expensive than picking up your food yourself, people are shifting back to just getting it themselves. Netflix and Disney+ are so expensive that users are returning to piracy. Youtube is so ad-slathered or expensive that it’s in a constant battle with ad-blockers and content downloaders. Twitch relies on content creators, but it’s so hard to make money on Twitch now that it’s not worth the bother. People stop driving for DoorDash because it doesn’t pay them enough, which makes it harder and more expensive to get food. People stop driving for Uber because it doesn’t pay enough, so people stop using Uber. 

This dynamic is playing out now for Reddit, Twitter, Youtube, and other platforms. Spotify has killed most direct music sales, yet artists can’t make money on Spotify. At every turn, the demand for investor ROI ends up destroying the value for users and the content creators that the platform depends on to generate the value that justifies people paying for the service.

This platform-oriented Web 2.0 approach of massive investment against massive ROI demands from the investors, creates a destructive spiral where old businesses are driven out of the market, but the new businesses that replace them can’t price themselves at a sustainable price point, so once they start raising prices to repay investors, both the content creators (or in the case of DoorDash and Uber, the drivers) and the users both exit the platform because the value to both sides of the two-sided market is destroyed by investor demands.

Paying for Coding

With modern approaches to code development, including open source and public code repositories, anyone can publish a codebase that can be used or improved upon by others. But as established earlier, copy-paste doesn’t cost money, but it also doesn’t get someone paid. However, if the payment methods are inherent to the codebase, including the codebase of the repo, the downloading or copying of a codebase can be designed to pay the original authors (coders) of the codebase through a transaction-enabled API – the same thing that we’re discussing now.

Right now this monetization of code is established through licensing of the code bases, or transaction methods that are inherently separated from the actual code being used. You aren’t, in a general sense, able to pay to download an open source codebase from Github, implement that codebase in a way that generates revenue, and then steer revenue from the use of the open source code back into the development team that originally wrote it (or copy-pasted it together from somewhere else).

With transaction enabled APIs, the use of a codebase can be organized to pay its creators as the code is used, in the same way that a sales tax for the purchase of a product pays into the organization that built the infrastructure that the store is using. There’s no reason why these API payment methods wouldn’t be able to be designed to pay the original coders whenever the software is used. Maybe only small amounts of money, but as we know from Superman 3, or Office Space, a fraction of a cent multiplied by a million or billion times quickly turns into significant money.

“But if it’s open source someone can just remove the payment method!” That’s where the code license comes in that ties the software back into a legal system. Sure, someone can take that codebase and strip out the payment systems, or redirect where the payment flows go, but we have a legal solutions for that already, even if the technical solutions fail. Corporations may not be able to stop us from keeping a copy of a streaming movie (it’s called a WebRip), but they can sue under DRM and copyright laws if people try to sell that copy.

There are other methods too, but the point is that making it illegal to take out these payment methods is already an established part of law. Sure, enforcement could be an issue, but that’s an issue for everything, isn’t it? Rape and murder happen daily despite them being illegal. But for large companies, the financial risks of committing crimes will, in theory, overrule the potential for benefits.

The point is that having a corporate owner of a codebase who pays coders to write code, then charging someone for an explicit purchase or subscription price to the user, or intermediating advertising into the product aren’t the only ways that code can be monetized. Thanks to the ongoing march of progress and the increasing maturity of APIs and digital currencies, we can also monetize code with or without a corporate intermediary by building that codebase to include transaction-based APIs that automatically pay the coders a fee each time their code is used, copied, downloaded, implemented, ran, or whatever method makes sense for the code in question.

And that code, developed to underpin a method to pay for sending a webpage (or some other static content, really any kind of file, it doesn’t matter what’s in the file as long as it’s static), will pay both the webpage and the coder who wrote the software. As more software extensions and improvements are developed, those contributors will get paid from the user of the software too. This is simply digital rights management from a coder-first approach, instead of a “corporate copyright owner”-first approach that DRM usually embodies.

By registering who wrote what code through embedding an identity into the codebase, and by having a payment mechanism in the code, a developer can get paid when their code is used without having an employer that intermediates their income stream. Using this method, the coder gets paid when their code is reused, regardless of who reuses it (assuming the user follows the license and doesn’t modify the code). And if someone improves or updates the code, they can add themselves to the license and get paid as well. This produces a method where the more code is used and reused, the more the original developer and their contributors get paid. It replaces a free-for-all copy-paste bonanza where coders are dependent on employers, with a direct-payment meritocracy where better coders whose code is more in demand make more passive income from the user of their code. It also relies on trust and verification, and existing legal systems for licenses and compliance, but literally every aspect of the economy relies on those elements to some extent, so pretending like those are defeating factors is disingenuous and counterproductive – merely excuses for maintaining the employer-as-intermediary approach to coding.

Paying for Hosting

Yet let’s consider the “dark” side of the internet. BitTorrent has been in use since 2001, and in the mid-00s, before content companies (music, TV, and movies) shifted to streaming, BT represented around 1/3 of all internet traffic. Following the adoption of streaming services, in the late ‘10s BT had been reduced to only about 2.5% of internet traffic (or, less than 10% of its prior use) showing how successful streaming was at combatting piracy. However, traffic has two “directions”, downstream and upstream. Downstream traffic comes from organized internet services, but upstream traffic comes from individual users. Even in the late ‘10s, BT was still 1/3 of upstream traffic, showing that it continues to be very broadly used.

The reason BT is so attractive from a technical standpoint is the same reason why BT is attractive to pirates – there’s no centralized hosting, which means no hosting costs, and no transmission costs. Sure, BT commands internet traffic, but the hosting expenses are incorporated into the personal internet service costs of the users, instead of being borne independently by content providers.

BT creates a “hash”, or mathematical proof, of a specific file, and the hash is much smaller than the file itself. Anyone with the file itself can “seed” the network by advertising that they have a copy of the file, and prove that the file is the correct file. Anyone who wants the file can download the hash, and the hash can then call a listing of seeds from the network. Everyone who has a copy “raises their hand”, and the BT client/server application then begins pulling pieces of the file from anyone who has a copy. This is a “distributed” file host, where the clients that have copies of the file for their own use are also the “hosts” or “servers” of the file for anyone else who wants a copy.

Even with its reputation for piracy, BT is often (or occasionally) used on the back-end of major content provider services to spread the costs of hosting across multiple shards. Around 2016, I learned in a discussion with a person at Sprint that Sprint had BT built into its app store to reduce its own hosting costs. Whenever a Sprint customer would call for an application from the store, the store hosting would act as a seed for the file, but any other Sprint customers who had the application would also quietly (and without the knowledge of the owner or user of the phone) send parts of the application package to whoever was requesting it. Currently Facebook, Twitter, Amazon, broadcasters around the world, software companies like Blizzard, and many others use the BT protocol on the back-end of their service to distribute files. 

Consider how websites work now: A browser calls for content, and the server sends a copy of that content. The files are held in cache for display. When the website is reloaded, it may draw from the cache, or it can be forced to re-call the server by using CTRL-F5. Many news websites, for instance, will not reload from cache but from calling the server again. For dynamic pages, the server calls happen on scroll or other user interactions. Right now, let’s focus on static pages and “content”, or video and audio files.

When someone visits the NY Times webpage, they call a page, and that page is held in cache. A cookie is left in the browser that identifies the user so that NYT can keep track of how many “free” page loads they get. But if the same page is reloaded, it’s not drawn from cache, it pings the NYT server again, which deducts another page load from the “free” pages. Many news sites will completely load a page, then throw a paywall over the top of it, and kill the scroll bar so that users can’t use inspect-element to kill the paywall and read the page. This is a complete waste of bandwidth! The page wants to get paid, but they send the entire content body, then crap over the top of it so that it can’t be viewed. And what happens if the user reloads the page? They send the same content again, and crap on it again! The news host is still not getting paid, but continuing to absorb the entire cost of serving the content, but making it unusable. This is as sane as serving a child in the lunch line, then if the child can’t pay, throwing away the tray of food. More intelligently designed sites will first check permissions and only deliver a stub of a page if the user doesn’t have permission to load the site. This can be seen with Financial Times, Insider, and most Business Journal sites. But despite being more intelligent, there still isn’t a built-in 402 payment system, and hosting is still centralized.

For other forms of content, like streaming video, the hosting and transport costs are wasted to an incredible amount. Netflix may cringe over how much bandwidth a 4k video uses, but Netflix also has intentionally designed their system so that it doesn’t play video from cache, forcing the video to be re-streamed each time it’s viewed. Obviously the video has to be transferred the first time somehow, but the only reason for not designing the system to cache the entire file and reuse the same file when the same content is replayed is to make it harder for people to rip copies of the content and rehost it elsewhere.

The same goes for any other video and audio content, regardless of host. The entire file is hidden in cache, but obfuscated to make it hard to use, and the content is re-sent each time it’s viewed. Content companies may complain about the costs of hosting and transmission, but how much hosting and bandwidth costs are wasted for no reason other than to make it harder for pirates to get full, high-quality copies? These content companies are creating the problem intentionally, blaming someone else, then complaining about the costs of managing the problem they intentionally created.

There are fundamentally better ways to do all of this by taking advantage of the ongoing development of software systems over the last few decades, particularly the lessons learned from CDNs.  This conversation is currently limited exclusively to static files. A static file is any website or AV file that does not change over time – livestreams and content like Reddit comments or Twitter feeds don’t count, those are dynamic. Resolving hosting and distribution for dynamic files can be addressed later, using a similar method, but static is an easier description for the time being.

Instead of obfuscating the browser cache, consider it a content archive that represents the full version of every webpage or file that is transmitted to the browser. Send all of the browser content via BitTorrent magnets, and enable all users to act as Seeds for content they have hosted. Include a user license file that’s attached to a payment method in the browser, and maintain a side-by-side one-to-one archive of content licenses that have been paid for. Now, any static content only has to be sent to the user once, and the user pays for the content at the point it’s transmitted, not as a separate function.

The user also becomes a Seeder of any content they have stored, which lets them take advantage of the unused bandwidth of the ISP they’re already paying for, to reduce the hosting and distribution costs for the primary content creator (whoever that is) and in doing so, reduce the overhead costs of operations for the content creator. Anyone who pays the API to access the content gets a permanent license file, and that license file enables the BitTorrent Leeching and Seeding of the content. This produces an inversion of the traditional hosting dynamic, where the more popular a website or piece of content is, the more expensive its hosting and distribution. By treating all internet files as a BitTorrent file, paying a license for that file at its API call, and caching the full file in an archive on the user’s machine that also acts as a Seed, the more popular a file becomes, the lower the hosting and distribution costs are for the content creator.

This approach would collapse the operational cost not only for major content hosts like Netflix or Disney, but actually reduce the overall bandwidth consumption by users because each file would have to be transmitted once, no matter how many times the user views the file. Now, rewatching the same movie on Netflix, Disney, video on YouTube or TikTok, listening to the same song on Spotify or Soundcloud, or any other static file, will only require a single file transmission, instead of a new transmission for every single repeat use of the file.

And because hosting is distributed to all users this way, and all users are paying for access to the content through a readily accessible pay-per-copy method, this ends up being far cheaper for ISPs too!

Everyone wins using this distributed content hosting – the content creators get paid directly, the content platforms have dramatically lower hosting and distribution costs, the ISPs have lower bandwidth usage, and the users get a better quality of service because they don’t have to constantly redownload the file, and their speed to receive the file depends on the total network bandwidth from all the users, not just the bandwidth provided by the main content host.

The only drawback to this distributed method is that the content host doesn’t get to control the file from end-to-end, 100% of the time, and charge ongoing subscription fees. This is an outcome of a pay-per-license approach where a user keeps a copy of the content that they call for. It wouldn’t stop content platforms from charging fees for content, or from imposing advertising on the content, but once the user had paid for receiving the file, that is the end of the platform’s control over that file.

That said, by building in the license to the content, it also inhibits the pirating of the content while disincentivizing piracy by giving users back control over their own computers and files, and freeing them from the burden of constantly being advertised to.

Generating a Return

This is the part where it gets interesting. Coders have an inherent incentive under this method to contribute to the codebase, thereby improving the product. And since coders get paid as their code is used, they don’t need an intermediary employer to justify their contributing. This is open-source writ large and turbocharged. It also legitimizes and commercializes the copy-paste nature of modern coding.

Content platforms are incredibly expensive because someone has to employ all the coders, someone has to pay for the hosting and distribution, someone has to manage the rights and licenses, and someone has to finance the creation of content.

But under the described method, the need for employers to pay coders to develop is relaxed or removed, because coders get paid automatically as their coding contributions are used. The need to pay for massive centralized data centers is relaxed or removed, because the content is distributed through a decentralized, highly efficient method that benefits from central hosting, but doesn’t rely on it, and is able to take advantage of the caching that users computers already provide, and transform that cache into a BitTorrent distributed hosting network. The need to manage rights and licenses is automated through the inclusion of a content license through the paid-API method and the proof of payment incorporated into paying for the BitTorrent hash enabled by the paid API method. This leaves only the financing of the content creation as a primary cost for the content creators.

But content creators are already bearing the burden of financing the content! Whether it’s small independent filmmakers, YouTube, Instagram, and TikTok “influencers”, independent Soundcloud musicians, or professional musicians whose music is being distributed through Spotify, Apple Music, YouTube Music, movies and TV shows created by Netflix, Hulu, Paramount, Disney, HBO, or whoever, someone is already financing the content.

This approach greatly increases the potential return for whoever is producing the content, however professional or amateurish, because it nearly eliminates the other costs – code development, hosting, transmission, etc – and disintermediates the payment of the content creator for their contributions. By separating the content creator’s payment from the subscription of a user to a platform, or the imposition of advertising into the content, and establishing a method where content creators get paid directly through the paid API system by each user obtaining a copy of the content, the content creators have the opportunity to get paid based on the demand and popularity of their content, and the value attributed to it by consumers, instead of whatever negotiated agreement they accept from a platform, or whatever share of advertising dollars the platform pays them from advertisers.

This new approach to digital decentralized content distribution, and the inherent payment methods built into the API, HTML, and browsers, enables a new realization of value for the entire world of content creation and its distribution. This is a dramatic change for existing major platforms that may take as much time for them to adapt to as online content distribution took the first time – but probably not, because they learned a lot from those experiences.

For the platforms that are their own content creators, like Disney, Hulu, Paramount, Netflix, and others, this is a reimagination of their business model that may take significant adjustment, but will end up being much cheaper and more lucrative than the current streaming model. This approach is also reasonably seen as a major boon to news websites. Instead of cajoling people to sign up for subscriptions, or slathering ads everyone hates onto the content they actually want, the news website can simply establish a reasonable price for their articles and get a payment from everyone who downloads their article-file, whether it comes from their original host-Seed or from a distributed Seed from others who’ve already downloaded it.

And for content creators that create SEO optimized AI-sourced crap pages that exist only to shovel advertising into people’s faces, this approach essentially eliminates the value of clickbait, enabling real, human-authored, intelligent content to thrive again. At least in theory, the slimeballs will figure out how to game the system eventually.

For the platforms that rely on 3rd parties to create content, such as YouTube, TikTok, Instagram, Soundcloud, Spotify, and other similar platforms that primarily act as intermediaries for payment to the content creators, this is going to be a harder pill to swallow, as they’ll lose their ability to command a subscription model, control content creator compensation, and control cash flows from advertising. And to that I say, get fucking bent! Nobody has the right to control the livelihood of others and exploit others hard work for their own benefit. If they can do so in a voluntary fashion, whether because the content creators prefer it, or in an involuntary fashion because the content creators have no other choice, then more power to them. But if the content creators can seize their own destiny and escape the clutches of these platforms to get paid directly by the people consuming their content, then why wouldn’t they?

This approach requires a few pieces to work – the payment-enabled API system, including the codebase that makes it work; the integration of the payment-enabled API into BitTorrent files, along with a license method; the implementation of HTML 402 calls into static content using the payment-enabled API; and modifications to existing browsers to transform the cache into an archive, and support a digital currency microtransaction method.

Though not discussed here, it’s suggested that the BT DRM license file method be adapted from current NFT systems and the license histories stored in the blockchain in some way, so that users can prove they own content and redownload it using their license file if their own archive gets deleted, whether because they ran out of storage space (which is becoming a rarer issue day by day) or because they switched to a new device and need to transfer files to that new device.

None of these tasks are technically or logistically daunting, they just need to be done, and haven’t been done before.