I was gonna make this a blog post but realized I already wrote something like this already (https://blog.webb.page/WM-078). I've been thinking more about the overall vision lately and I needed to write so I can focus on coding. It's actually annoying how blocked I feel when that happens. Anyways.
Something's gotta give.
What was supposed to be the great liberator, killer of gatekeepers, democratizer of information, has become an attention enslaver, embracer of greed, fear monger of the easily deluded, nay a highly efficient extraction machine that cannot be sated. I'm talking about the internet, of course.
You know it's bad when the retired execs of some of the worst behaving of these corpo behemoths forbid their offspring from engaging with the product they gleefully sicc'd on millions. Late onset morality is wild, especially when it comes with a price. Prices? Personal wealth and the cost to mankind. Anyways...
I won't pretend that the internet has always been altruistic but it did seem for a time that there was more of us (decent folks with morals who just wanted to share cool stuff with other cool people) than there were of them (sickos). Advertising existed (banner ads) and they did get terrible (popups, pop-unders, fake close dialogs and download buttons) but man...with enough experience you could figure things out. Okay, so things were terrible back then too but the terrible things weren't invisible! If you can see it, you can react to and flat out refuse it. The insidiousness of dark patterns today is that you don't even realize you're being manipulated half the time.
All this to say, a lot of the internet back then was built based on the assumption of good intentions (https://blackqueer.life/@FinalGirl/116324559586041902). So we got the rant out the way, how do I think we could fix the many ills of the internet today?
Rebirth
High-level overview of my thinking is going scorched-earth and...rebuilding on the assumption of good intentions. Gotta be honest, that's pretty weak but if we can do so while keeping in mind that those with perverse incentives will want to colonize this new space, maybe we can introduce safe guards.
The issues with decentralization are that 1) consolidation will always be easier and 2) decentralization is often a barrier to entry, be it financial or time or effort. Nerds don't want to admit it but most people are struggling with two, if not all three of those points.
Another issue with decentralization is more of a people problem; some projects use the term as an ideological symbol of purity and reject any sort of consolidation, even if it makes things easier for passersby to jump in and start interacting with said project. Expecting everyone to run their own infrastructure is laughable when the goal is widespread usage (otherwise, carry on). Decentralization for decentralization's sake is self-congratulatory with no purpose. If there's no purpose, what's the point?
Centralized infrastructure subsidized by paying customers is more sustainable than begging for donations. That being said, revenue does not have to compromise our privacy, manipulate us, or steal from us. Business models that align with user interests do exist. Such as...providing a service and charging for it. That's it. Imagine respecting users! Wow, what a novel concept!
We've moved fast and broke things; trust, cooperation, and comradery. For shame.
The Stack
The Neue Internet is a suite of interconnected products and protocols, each useful as standalone things (perfect for sustainability) while being infinitely more powerful and capable together. Every stack builds upon the one below it. This is why the foundational layer has to be strong af.
Identity Layer
The DNS is commonly described as "the phone book of the internet." Dap then, is just a different kind of phone book. Think Gemini or Gopher in that there's a new protocol that comes with it, just more secure (boring/unsexy stuff like DNSSEC and complicated maths). With Dap, there's no gatekeeper dictating anything or imposing exorbitant entry fees like ICANN.
Dap democratizes the issuance of TLDs (top-level domains) via Vickrey-style auctions. This is the fairest method I've seen but it doesn't stop there! Every naming system has an issue with squatters; it's unfortunate that squatting is lucrative. Currently, owning a large portfolio of TLDs within Dap comes with usage requirements enforced by the blockchain but validated off-chain. This needs a rework to make gaming this non-trivial...there's also the use case of personal TLDs; you wouldn't expect someone to create hundreds (or even dozens) of domains like you would a .com, .neue, &c.
Owning your name is the first step. Proving it is the next. Porting your identity to a new platform you want to try should be easy. This would work like "Sign In with Apple/Google" buttons.
Payments Layer
Advertising took over the internet because HTTP 402 Payment Required was never fleshed out. Commerce is such an integral aspect of society it seems like quite an oversight of RFC authors to leave a placeholder in an official spec for literal decades.
Fortunately, Cloudflare and Coinbase (look, I know) got together and just did it themselves. Here's your regular reminder that you don't need permission to try and make the 'net better. Anyway, that got me thinking about creating the Stripe of crypto for my personal projects (dogfooding) before sharing it with others.
I call my solution, NeueCash, and the first product that will use it is my video platform, Nickel. I wanted a way for people to tip creators and dealing with the Stripe API kinda pissed me off. Payment providers these days are super opionated with their design and impose their Fisher-Price design sensibilities on their customers; I am not a fan. Like ISPs, payment providers should just be dumb pipes and get out of my way.
Further down the line it'd make sense to create a bank-like product for spending, yield (savings), and social payments (splitting bills with friends, that sort of thing). If you advertise yourself as a bank though, regulators swarm on you. That's the reason why all these neobanks say in their site footer something like, "we're not a bank, we're a financial services company but look, here's our partner bank and they're FDIC-insured!"
The world has not moved beyond needing nation currencies yet so this not-a-bank product would likely be what political dissidents and people in oppressive regimes use, aside from crypto folks and early adopters.
Communication Layer
Email is the de facto communication standard, period. Maybe in Asia can one claim that an app is more important but at least for the Western world, email reigns supreme. When you attempt self-hosting your own email server you gotta deal with so many damn acronyms and acrhaic protocls that haven't been updated in decades. JMAP is cool but it's lipstick on a pig; cute, but unpleasant to look at. When's the last time you saw a cool email client only to realize that it requires Gmail? That's not an email client, that's a Gmail client. Email sucks so bad that people flock to Gmail and think of that as the best we can do...knowing full well that Google reads your emails. SIGH
E2EE (end-to-end encryption) is a baseline feature for communication apps today. PGP is such a complicated slog to deal with and thus, most people don't (not even the tech-inclined). I think the plus trick with creating email aliases is well-known, but Fastmail takes this a step further and empowers you to create aliases via their UI and dictate where mail goes.
Email is open by default which is what you want for communication. One of the ills of today is that your email has been leaked or sold and if it hasn't, it will; just wait. This is one of the reasons why spam is prevalent. The other reason is that spreading it is cost-effective (practically free).
I think there should be a financial barrier to email. Even a one cent paywall would discourage most spammers. You'd want an allowlist for people you actually want to hear from though, and even for transactional emails like receipts. Imagine if you will:
- yo@paul.webb: allowlist, close-friends and family only; anyone else would get a rejection message
- tx@paul.webb: allowlist, transactional emails, possibly even a read-only inbox
- office@paul.webb: my email server requests $5 from the rando trying to email me
- junk@paul.webb: wild west, free reign, insane unread counts
My first pass at this specification is the Inbox Protocol (https://useinbox.org) and it builds upon Dap and NeueCash. As it stands though, it could work on clearnet (the regular internet) today if someone was inspired to build it.
Content Layer
So we've got the foundation, money, and communication. Now what to do on this Neue Internet? My short-form video platform I mentioned before, Nickel, will have first-class support for Dap but it should by no means be the only video platform in town. It wouldn't work for long-form content like documentaries or Bobbin Threadbare's excellent Let's Play of Deus Ex, for example.
At a bare minimum, we need blogging and code forges and some kind of social media. I tried my hand at that last thing but decided Mastodon was good enough. It shouldn't be too difficult to make a Dap-aware instance.
I have ideas for opinionated blog and code platforms that I'll keep to myself for now. When everything looks the same, I get inspired to do something different and I think that every Github alternative looking like a Temu variant is a missed opportunity.
Discovery Layer
Several years ago I had this concept of a P2P search engine where people could host their own "index" and fine-tune access; I call it, "Queree" (query). There would be a primary index of course, but I'd surface independent indexes whenever possible. The main use case I optimized for was scientists and researchers sharing experiment progress, journals, &c. It should work as an intranet search engine as well.
When I get around to building Queree it'll be atop Kagi. No one is doing search better than them.
The great thing about Dap-backed identity is being able to do things like trust-weighted reputation. In a world of AI and scams, it's increasingly difficult to gauge what's a real review. I want human-curated discovery!!
The Roadmap
NeueCash > Dap > Inbox > EOL
Why Now?
To be clear, we could have this now but people are disenchanted and feel there's no point. Y'know how people feel optimistic on January 1st? I hope people will feel the same way about the Neue Internet, but actually stick around.
The tech exists, or it will soon. It's just that no one's combined everything into a cohesive, compelling package. "Ecosystem" is an ugly word in the tech world because it always refers to a walled garden that demands your allegiance.
At the end of the day, the Internet is a vehicle for text and images. Let's assume we somehow convinced 8 billion people to join some aspect of the Neue Internet. That'd generate between 2-8 TB of data, which any small business could afford. Realistically speaking, I'd probably get a few thousand people interested in what I'm doing enough to cost me anything in terms of storage.
I've bootstrapped and provisioned countless servers and APIs over the past decade and only made maybe $40? Who knows how much I've spent and I don't care.
Every self-initiated project, in hindsight, has touched on the broader vision I now have of what I call The Neue Internet. Where I was floundering before, I have guidelines and roadmaps...at least 10+ years worth.