Happy Birthday to the First Dot-Com

​Today, the domain name symbolics.com is the property of XF.com Investments, a Dallas-based "virtual real estate investment firm." From its meager website, it seems reasonable to conclude that the crux of XF's real estate is that single domain name. As the first ever .com website, symbolics.com is something of an internet historic site. Currently, it's mostly inhabited by advertisements for advertising.

2015 visitors to said domain are greeted with a weird illustration that appears to be hyperlinked in random places to modal windows supplying random facts about domain names and internet history. It takes a bit of hunting to discover this functionality, and I kind of suspect you'd find something very similar in searching for jQuery example projects.

It's pretty clear that symbolics.com in 2015 exists as a click sink. And, indeed, every year around this time, the March 15 anniversary of the registration of symbolics.com, there will be a new wave of clicks accompanying a new wave of blog posts about symbolics.com. There's something kind of perfect or curiously meta about the first ever .com—the top-level domain originally intended to signify a commercial enterprise—now existing as an attempt to cash in on the clicks of curious online sightseers.

The first round of six top-level domains (also: .org, .mil,.gov,.edu, and a bit later, .net) was defined in October of 1984 via a document known as RFC 920 (RFC: "request for comment"). The domains were to be administered by the Defense Data Network Network Information Center, via SRI International, née the Stanford Research Institute.

RFC 920 defined domains as such:

Domains are administrative entities. The purpose and expected use of domains is to divide the name management required of a central administration and assign it to sub-administrations. There are no geographical, topological, or technological constraints on a domain. The hosts in a domain need not have common hardware or software, no reven common protocols. Most of the requirements and limitations on domains are designed to ensure responsible administration.

The problem in 1984 was that more and more people were using the budding internet for more and more things; no longer was it the domain of researchers and military types. Manually routing all of this traffic around was getting messy and it got to the point that users would occasionally be asked to disconnect. The web needed order.

A key piece of that order were to be top-level domains. We can imagine TLDs as sort of switchboard operators or even phonebooks. While the early internet would've represented one giant disorderly phonebook, the newly partitioned internet split that phonebook into smaller volumes. To find a .com address, one only needs to search the .com "namespace." As a bonus, the internet could even now host identical lower-level domain names so long as they existed in distinct TLDs. The internet was becoming the internet.

That internet begins at a somewhat hidden domain called the root. From here, the web grows as a tree, with the very next nodes being the top-level domains. When you visit Motherboard, your brower's first stop is (theoretically) this root directory, which would then direct you to the next more specific directory, e.g. .com or .org or whatever. There are 13 root domains in total and, for the most part, they're distributed (via anycast), though a few still exist in single locations: Aberdeen Proving Ground in Maryland, the University of Southern California.

The root zone file is tiny by today's standards. It contains the names and numerical IP addresses of all of the top level domains. While the root is in practice usually skipped over thanks to various caching schemes, in theory it's purpose is to point toward the next subdomain (the TLD), which itself contains the next sub domain (if it exists) and on down the line, until you eventually get to the desired destination. It helps to imagine the entire internet as just a big old file system: folders and documents, from top to bottom.

Today, the internet is home to 810 top-level domains, most of them being country-specific, while several more are currently being considered, including .med, .mail, .geo, and .shop. Mostly, it's a wonder that the original domain hierarchy was able to survive and scale up like it has at all. It wasn't really meant to.

"I don't recall anybody ever thinking we were creating an organizational structure to encompass hundreds of millions of entities covering the entire planet in support of all human activities," recalled e-mail pioneer Jack Haverty, one of the original internet architects for Oracle, in a Verisign short-history. (Verisign administers the .com and .net domains, as well of two of its root servers.) "And it certainly wasn't supposed to last for 30+ years, even as an experiment. It just happened to turn out that way."

Soon enough, symbolic.com and the whole fuss will seem brutally anachronistic. Most news from the domain name marketplace is that it's in decline and "internet real estate" isn't what it once was. Which makes sense: web navigation since Chrome has tilted more and more toward search rather than punching in actual addresses, and appification isn't helping the URL case. (Client-side scripting and the uncrawlability of dynamic websites will presumably throw a wrench into this trend, but I'm not sure to what extent.) The domain hierarchy will persist and grow, but eventually we won't have to see it.