How Many .com Domain Names Are Unused? (singaporedatacompany.com)

Christopher Forno, CTO at Singapore Data Company writes: When looking for .com names, I've been frustrated by how many are already taken but appear to be unused. It can feel like people are registering every pronounceable combination of letters in every major language, and even the unpronounceable short ones. Is there rampant domain speculation, or do I just think of the same names as everyone else? Let's look at the data.

There are currently 137 million .com domain names registered. Of these, roughly 1/3 are in use (businesses, personal websites, email, etc.), another 1/3 appear to be unused, and the last 1/3 are used for a variety of speculative purposes. I started by crawling a random sample of the domains from the top-level .com DNS zone file, until reaching 100,000 valid domains. [...] For most categories I've included a random sample of screenshots from that category, excluding redundant ones: Content (31% or ~43 million), Ads (23% or ~31 million), No Web Server (11% or ~16 million), Empty (9.2% or ~13 million), For Sale (7.1% or ~9.8 million), Error (5.7% or ~7.9 million), Parked (4.8% or ~6.5 million), Gambling (3.0% or ~4 million), Mail (2.6% or ~3.5 million), Redirect (1.1% or ~1.6 million), Private (0.64% or ~0.9 million), and Porn (0.59% or ~0.8 million).

Comments

All 3-4 letter combinations taken

[xpiotr]

Probably 5 letter combinations too... https://whoapi.com/blog/we-are...

godaddy is the culprit

[magarity]

Years back I tried an experiment: put a domain name in a browser and not no response. Went to GoDaddy to register it and was told it was taken. Tried in the browser again and got a 'this address is for sale!' banner and an email to the address I had given GD offering to sell it to me within minutes. GD pretty much exists to suck up domain names people submit and then try to sell their idea back to them.

Re:godaddy is the culprit

[Anonymous Coward]

If this is truly the case then you should be able to put together a reproducible test of randomized words into a domain. At that point you could document it and possibly even automate it. Once it's that obvious you can get some attention from media, ICANN and/or some other places.

Re:godaddy is the culprit

[Anonymous Coward]

It usually takes a little more than just trying to resolve the domain name. It is well known that some domain registrars reserve domains that you look up through their web site (e.g. to see if they're available). There's even a name for it: domain name front running.

Re:godaddy is the culprit

[fermion]

Go daddy did not innovate this as it has been done since day one. Large corporations were too dumb to register thier domain early, then too greedy to pay what was a time fair market value as it was cheaper to sue. It is like having a food truck, having someone else but the parking lot you use, then suing them to prevent them from using the lot as they please because you are entitled to the space for your business.

It was legal and reasonable speculation just like any other. Some made money, most lost a lot. There is no real harm because we are not talking real estate in London or Paris.

Re: Squatters

[Dynedain]

Back in the first internet boom (when we wore onions on our belts since it was the style at the time) a startup registered the 20,000 most common US surnames as .com domains. They were selling email accounts. In the inevitable bankruptcy the whole thing was sold to Tucows.

Re:Ooh! Ooh! I know!

[Solandri]

Even if there is rampant speculation, and assuming a 26-character alphabet, there are 1.46*10^11 domain names between two and 10 characters. If 137 million of these are registered, that means only 0.0001% of all possible domain names have been taken.

About a decade ago, I was trying to think of a domain name to register for a website. I wrote a short program to put together English phonemes to generate every possible pronounceable word (up to 7 letters).. Then I had the program do a whois database lookup for each fictional word as a .com domain, outputting a list of the unregistered domains. (Yeah, I'm probably part of the reason why ICANN now makes you solve a captcha before doing a whois lookup).

All the 4-letter .com domains were gone. Most of the 5-letter domains were gone too, and the few which weren't sounded horrible. Most of the 2-syllable 6-letter domains were gone. But there were lots of 3-syllable 6-letter domains still available. There were lots of 7-letter domains still available. So many that I killed the program part-way through (at domains starting with the letter 'f' if I remember). The list of available domains it was generating was becoming so long it would've taken me too much time to look through it, trying to find one that seemed decent.

Re:How do they know it's not in use?

[dissy]

How do they know it's not in use?

He says if no web-server was running, he went by DNS records.

Just because a domain doesn't have a website doesn't mean that it isn't used for something.

From his report, he has two categories for "No A record but does have MX" and "No MX record but does have A records with no apparent web server"

Mail (2.6% or ~3.5 million)
Any domain not in any other category, but with MX DNS records (for email), I categorized as Mail. I did not attempt to see if the mail server was working or if delivery was possible. It's possible that many of these domains are not actually used for email, but I've given them the benefit of the doubt.

No Web Server (11% or ~16 million)
If I was unable to connect to, or receive a valid response from, port 80 or 443 for either the top-level domain or the www subdomain and the domain had no MX records, I placed the domain in this category. Some of these domains likely have some non-web use, such as an FTP or video game server, but I expect them to be a small fraction. Additionally, the crawling server was only configured for IPv4, so any IPv6-only websites would have been grouped here.

It would seem all the other categories were determined from data returned by a web server.