Domains as "Internet Handles"
A little while ago I cam across a post by Dan Abramov , a name that until then didn’t ring a bell, but who appears to be a former Meta employee and member of the React core team. The post links to a website made by Abramov , that addresses the issues of how, quote, every time you sign up for a new social app, you have to rush to claim your username , how, quote, if someone else got there first, too bad and how, quote, that username only works on that one app anyway . The website goes on: This is silly. The internet has already solved this problem. There already exists a kind of handle that works anywhere on the internet—it’s called a domain . A domain is a name you can own on the internet, like or . Most creators on the internet today don’t own a domain. Why not? Until recently, you could only use a domain for a website or custom email. But personal websites have mostly fallen out of fashion, and each social app sports its own kind of handles. However, open social apps are starting to change that. These apps let you use any internet domain you own as a handle Abramov highlights a familiar pain point: On every new platform, users must scramble to secure their preferred username, often discovering it was taken years ago. Domains, he suggests, solve this by offering a globally unique namespace. However, this solution introduces an even greater scarcity problem, amongst other more important issues. Short, meaningful domain names have been scarce for decades. Most desirable combinations of common words, short names, or initials were claimed long before modern social platforms even existed. For example, just like our author, I, too, would have loved to use or as my handle on e.g. Bluesky . Sadly, however, I’m more than two decades late for that, as the former seemingly belongs to a Russian company, and the latter to a namesake somewhere in Bavaria, Germany. Domain marketplaces and registries still list alternatives , but these often come with premium or recurring fees far exceeding what the average user is willing to pay. When platforms require domains as identity tokens, a user whose preferred domain is unavailable loses access to that identity everywhere , not just on a single platform. Unlike usernames, which can often be adapted with simple variations (e.g. adding punctuation), domains offer no such flexibility. TLD constraints mean that once a desirable domain is taken, there may be no practical semantic alternative. Domain scarcity does not solve the “handle availability” problem, it instead exacerbates it by moving contention from individual platforms to the internet’s global naming infrastructure. Usernames exist within individual platforms and their loss, while inconvenient, usually has contained consequences. Losing a username typically means losing access to a single isolated data silo (platform). Domains, by contrast, are subject to a multilayered hierarchy of control involving domain registrars, TLD operators, ICANN-affiliated registries and the DNS root zone. By using a domain as a cross-platform handle , users tie their entire online identity to this centralized, multi-stakeholder governance structure. Misconduct, even just alleged, on one platform could result in escalations to a registrar or registry, potentially leading to domain suspension. A suspended domain invalidates not just a handle on one platform, but an entire online identity across all services using that identifier. The risks extend beyond platform moderation. A compromised mailbox, a malware incident on a web server, or an automated threat-intelligence flag from entities such as the internet’s favorite bully Spamhaus can lead to domain suspension. In such scenarios, users may face lengthy appeals processes involving opaque third-party entities that wield far more power than a typical platform operator. Domains were designed for hosting services, not for acting as the cornerstone of individual identity. Using them as universal handles places disproportionate power in the hands of infrastructure operators who were never intended to serve as arbiters of personal identity. If you’re a long-time reader of this website you probably already knew that privacy must come up at some point. Well, here it is: Traditional username-based systems allow users to separate their personal identity from their public persona. After all, not everyone might want others to know about their activity in the Taylor Swift forum of FanForum.com , and that’s fine. Domains, however, increasingly erode this layer of privacy. While privacy-respecting domain registrars still exist, the mainstream domain ecosystem overwhelmingly encourages or requires KYC, traceable payment methods and paid WHOIS privacy services to maintain the illusion of privacy. Most users will register domains using a credit card or similar traceable payment method through large commercial registrars. Even if WHOIS privacy is enabled, metadata leakage and billing records remain. In the context of social identities, this creates an environment where domain-based handles can be correlated with real-world identities far more easily than pseudonymous usernames. A user posting under a domain such as time-to-get-swifty.com could find their identity exposed not through any platform breach, but simply through the structural nature of domain registration. Usernames are free. Domains are not. Even the cheapest domains incur recurring costs. More desirable, short, memorable, or branded names often command high premiums or elevated renewal fees. While this financial burden may appear negligible to, let’s say, former well-paid Meta employees who consider their online presence a professional asset, the majority of internet users do not attach the same value to domain ownership. For many, especially outside tech-centric circles, the ROI of maintaining a personal domain is negligible or non-existent. A farmer participating in an agricultural forum is unlikely to find value in purchasing and renewing a domain like solely to participate in an online community. Any identity system that introduces ongoing financial requirements creates unfair barriers to participation and risks entrenching socioeconomic inequality in digital spaces. Abramov ’s argument positions domains as a universal, user-controlled solution to fragmented identity systems. While his vision aligns with broader goals of data portability and user autonomy, domains introduce significant drawbacks that usernames do not suffer from: Greater scarcity and reduced availability, centralized infrastructure vulnerabilities and governance risks, reduced privacy and increased traceability, and recurring financial burdens for users. With statements like “You don’t have to squat handles anymore. Own a domain, and you can log into any open social app” the author makes it sound like domain names are less exclusive than simple usernames, when it’s clearly the other way around, and they fail to recognize that squatting is far worse of an issue for domains than it is for simple usernames. Moreover, the reliance of on conventional DNS infrastructure undermines the self-sovereignty that decentralized identifier systems aspire to. Without a complementary decentralized naming layer (e.g. Handshake ) domain-based identities merely exchange one set of constraints and issues for another (vastly more dangerous and impactful) one. For these reasons, users and platform developers should think carefully before adopting domains as universal “internet handles” . Usernames, for all their imperfections, remain simpler, safer, more private, and more equitable for everyday identity on the web, at least until the truly decentralized future is here. While one might say that the handle is merely a representation of the underlying decentralized ID , a loss of the domain will nevertheless come with functional implications across every service that uses it. Luckily, platforms that implement domain handles continue to offer accounts under their own domains for the time being, so that at least for uninformed users nothing really changes (on the surface). Note: I have an account on a platform that supports domain handles and I am using the feature in order to be able to make informed statements. The account is, however, nothing that is crucial to my existence on the internet. If my domain should spontaneously combust that account would be the least of my worries. Instead, I’d be more troubled about this site and its related services, which is why I have a fallback domain . While I’m sure the author of internethandle.org didn’t intend to, some statements on the website “sound” somewhat out of touch, or at the very least tone-deaf , e.g.: Most creators on the internet today don’t own a domain. Why not? Until recently, you could only use a domain for a website or custom email. But personal websites have mostly fallen out of fashion […] Dan , personal websites haven’t fallen out of fashion , but have suffered under the World Wide Web altered (dare I say destroyed ?) by the very companies you supported building as part of your previous roles and, to some extent, as part of the technologies you’re working with. Just because you, and the people you surround yourself with, seemingly don’t care about the small web it doesn’t mean it has fallen out of fashion ; If anything, personal websites are gaining popularity and are the weapon of choice against the enshittification of the web by companies like Meta and others.
Swift
Web Development
CSS
Frontend
Open Source
Design
Web
Tutorial
JavaScript
HTML
Backend
Ruby
API
JSON
Culture
Music
Security
DevOps
Programming
Business
C++
Rust