The 100 oldest .com domains on the internet | Hacker News Reader

The 100 oldest .com domains on the internet

theforrester.wordpress.com

59 pointsamerf148 comments

Interesting if you haven't seen it before, but that blog post is kind of terrible. (Calling them "WWW" domains? In 1985?)

Wikipedia has the list too, along with the corresponding domains in .org, .net, etc.:

https://en.wikipedia.org/wiki/List_of_the_oldest_currently_r...

I wondered about the accuracy of these dates which are pulled from whois, but they seem to agree with the timeline outlined in RFC 921, which lays out the transition to DNS from the hosts.txt file previously FTPd around.

http://tools.ietf.org/html/rfc921

With the dominance of the WWW now, most people don't realize the Internet (and the ".com" nomenclature) predates the WWW.

I worked at BBN (#2) for 2.5 Years right when I graduated from college. Genius level colleagues; The types of guys that would download computer viruses and decompile them "Just to look at the code."

A funny story is when they called me for an interview, my Mom said "It's BB&T on the line." and my response was "I bank with Bank Of America." Then she said "Just take the call!"

Not to be confused with BB&N, the private school just down the street from it...

That did confuse me for years. I thought there were a bunch of child prodigies working for BBN.

This list makes the rounds every year or so, but I always like seeing it; one of the domains is a special one to me, although I was not there at the time it was registered. UniPress Software (unipress.com, #80 on the list) was my first job while in college in the late 90s, and gave me the ~14 year experience of creating a product that went through all of the stages from "figuring out if people want a web-based application that does ___" to being a grown up, enterprise software that was acquired by a very large company. The two founders had real software-startup instinct before startup/entrepreneur was something written about, a love for what it is that we do, and are still close friends. It was a great place to be, a great time, and awesome people to be around. The two founders reached the age of retirement and sold the company years ago, but have been involved in two startups since, just because it's what they like. It makes sense that they would have bought a domain when it was a "pioneer" thing to do.

Most recently, seeing that Twitter was becoming very popular, Mark had been pumping out PHP code using CodeIgniter and made www.mediaroost.com in his spare time, a Twitter Management platform. It didn't take off, he retired it, but had a good time. Going way back, this is the same guy who wrote C compilers and sold Whitesmiths Ltd which released the first commercial C compiler.

(and going back before my time again, but interesting history - UniPress had also purchased Gosling Emacs which then became UniPress Emacs, controversially asking Richard Stallman to stop distributing GNU emacs source code.)

YOU EVIL SOFTWARE HOARDER!!!!

Ha ha, just joking. ;) I worked for UniPress too, on Emacs. There was a certain professional animosity between Stallman and UniPress. Stallman had a justifiable axe to grind against UniPress, which he expressed by calling UniPress "Evil Software Hoarders".

But between the individual people, it was more like the relationship between the cartoon sheepdog and the coyote, that was friendly and respectful after everyone had punched out of work.

Just after RMS's house had burned down, Mike "Emacs Hacker Boss" and I were wandering around a science fiction convention and ran into RMS (who is a fixture in the east coast SF con scene). Mike sincerely asked him, "Richard, I heard a terrible rumor that your house had burnt down. Is that true?" To which RMS replied without missing a beat, "Yes, but where you work, you must have heard about it in advance."

We all had a great laugh from that -- RMS is a very funny guy who's quick on his feet, with a sharp sense of humor! Just don't let him get under your skin, like when you post a baby announcement to a mailing list for arranging dinner get-togethers on the other side of the country than he lives: http://www.art.net/~hopkins/Don/text/rms-vs-doctor.html

Here's an old picture of JSOL, RMS, Liz and Mike, where RMS is holding a gerbil wrapped in duct tape, about to ask, "I don't know, why do you wrap gerbils in duct tape?" (Now you can ask google if you really want to know...) http://www.art.net/~hopkins/Don/images/jsol-rms-gerbil-liz-m...

Sat with a fellow student at 3am in a UMIST lab in early 90s. Trying to resolve 'levi.com', 'cocacola.com' etc. on some early rev Mosaic ... rarely find a registered brand ... "someone needs to make page with a list of ones that work" ... go back to COBOL project.

There was no web in 1985.

But there was an Internet. Companies registered domains for email addresses, FTP servers, etc.

There's gotta be a way to create a site where you can catch up on stuff that is often resubmitted to reddit and HN (and even digg!). I don't know how one can do that without human moderators or expensive and difficult AI.

People who have never seen this list before find it interesting. People who have been on reddit and HN see it every other month. Some way to please both groups so everyone is happy would be nice.

Reddit has related links that no one uses. There is also a comment bot that finds reports. AI looks for Similar titles and site links. It's more complicated when trying to tailor it to a persons history. Digg and reddit have tried AI frontpage customization by history, but idk if it was ever to bring new people to catch up on old posts.

Facebook sort of did it with old posts randomly coming up on the side to trigger nostalgia I guess.

Taking into account vote history, you can tumble through old popular posts not similar to anything voted on yet. Maybe push it out to groups of people who haven't seen it to keep the comments and discussions to thrive.

Can use the API to gather the needed info

> Taking into account vote history, you can tumble through old popular posts not similar to anything voted on yet. Maybe push it out to groups of people who haven't seen it to keep the comments and discussions to thrive.

I like this. Also instead of a related link, just show the related things. And maybe stack overflow style user moderation.

This is a TED talk by the creator of think.com: http://www.ted.com/talks/danny_hillis_the_internet_could_cra...

He touches on his early realization that he could have taken a ton of domain names & made some money, but that would have gone against the spirit of the internet at that time (i.e. don't take more than you need.)

Sad to see that "think.com" is so underused. Oracle would do well by me to donate it to one of the many educational initiatives for use today (think.com resolves to this http://www.thinkquest.org/en/)

Recently I went looking at the registration for love.com, but that was a relative newbie, being registered in 1998. How did it take so long for someone to snatch that up? (sadly, that too is being wasted by a corporation: AOL)

Totally agree. It's a shame there isn't a rule in place whereby domains need to be used or handed back.

Problem with such rule is how would you define "used domain". For example I own one domain whose almost sole purpose is that it is part of package name in few Java systems written 10 years ago (and few hostnames belonging to that domain are in configuration of who knows how many things).

Perhaps not a rule, but a general onus on any large corp to give up domains that aren't being used for something "useful", to donate them or at least lease them out.

With real estate property taxes encourage an efficient distribution of resources. The domain fees could increase, however it would be difficult to assess the value of a domain for a property that is so nonfungible that it is only valuable to the company that has a made up word as its name.

Ha! Oracle and donate - this is an oxymoron. Just like Steve Jobs, Larry doesn't believe in giving away stuff for free. He's one of the greediest people I know.

This reminds me of a book published, back in the early 90's that contained a list of every public email address on the internet. It was smaller than the average phone book.

I imagine an updated edition would be a little larger. Or have really tiny type...

So why was Symbolics first?

It was an MIT AI Lab spinoff which helps explain why it was early. As to why it was earlier than, say, BBN though I have no idea.

BBN has an entire /8 for itself at 4.0.0.0. I'm sure whoever made that decision saw there were maybe a dozen addresses and figured, "whats the point? people can reach us at 4.3.2.1."

BBN built the Interface Message Processor that routed ye olde arpanet.

And why is it a pure ads page now? Is this the future for every site out there?

Because the domain keeps getting listed in various articles as "the oldest .com domain", and keeps getting people visiting it, and so it's passive income for whoever owns it.

Just an assumption: they had relatively large business case for DNS and user-friendly global networking without site-specific configuration in general as they were selling what essentially amounted to free-standing workstations that did not really need centralized infrastructure. As opposed to many other workstation vendors at that time who either didn't care (e.g. Xerox) or assumed that their product would be used en-masses as part of some existing network and centrally managed (e.g. Sun).

My guess is that they just happened to be incorporating at the right time, tech companies already in existence when .com registrations were introduced that needed to do networking probably had Arpanet addresses and waited a while to switch over.

Interesting to compare this list to the current Fortune 500 list to get a list of companies that were first on the net and also managed to stay very profitable over the last couple of decades.

Just from a glance, looks like IBM and AT&T made both lists. I'm sure there are others.

Northrop.

Previous occurrences/discussion of same thing here:

https://www.hnsearch.com/search#request/all&q=oldest+domains...

Anybody got older snapshot?

http://web.archive.org/web/19981207002851/http://stony-brook...

http://web.archive.org/web/20031206082643/http://stony-brook...

this one is the latest worthy snapshot

The Wayback Machine let me down again. Until I waybacked the Wayback Machine.

Interesting that some companies really advertise on the oldest domain, http://symbolics.com. (Or it's fake so others want to advertise, too.)

An aged domain typically has higher page rank. I imagine it has a ton of backlinks from just existing in that time. Companies pay for its page rank to be funneled into them by association.

Looks like it's someone pretty excited to have it: http://symbolics.com/about-symbolics/

#62, cts.com

About 4 years ago the company I worked for had this in their domain portfolio and at one time offered shared hosting from this domain in the mid-late 90's to Present. There are likely a lot of legacy accounts still utilizing the subdomain hosting accounts but we decided to take advantage of domains that were no longer used. Most of these accounts were tied to a dial-up internet service (Concentric).

There were a ton of old broken links for subdomains like mycompany.cts.com where the client had closed their account. I set up 301 redirects in the htaccess file for any subdomain which was no longer in use and the actual cts.com domain to redirect to one of our services websites. The PageRank for that website went from 3 to 7 for about a year then it dropped back down to around 4-5.

Fun times!

As someone who was being born when those domains were being registered, can anyone shed some light onto what the internet was like back then? (my understanding is that there wasn't a www back then)

If you've never read it before, you're in for a fun read - https://en.wikipedia.org/wiki/History_of_the_Internet

Back then email was still the primary civilian use of the internet, although by the mid-80s BBS's and USENET were firmly entrenched. https://en.wikipedia.org/wiki/Bulletin_board_system https://en.wikipedia.org/wiki/Usenet

The early walled gateways (AOL, Compuserve, etc) and then the wider web as you know it didn't really take off until the early 90s.

I finished my studies in 1993 and I have used internet a lot during the 3 last years (mid 1990 until mid 1993). ncsa mosaic has really arrived in my school only at the beginning of 1994, after I left.

During my studies, the main internet tools were telnet and ftp. There was no firewall and it was really easy to connect to any computer anywhere. It was also really easy to find password (ypcat passwd then dictionary based attacks). My girlfriend was in another school (500km away) and I was using a small utility named xhtalk to monitor when she was connected in order to chat with her. With xarchie, you could search a software or game by its name and it returned an anonymous ftp server.

I played go on internet using igs in 1992. There was something similar for chess.

The main information source was the newsgroups. There was already hyperlink navigation with gopher.

The version of linux on my PC was 0.99 (version of the kernel). I was using it mainly as an X server to access the school computers when the rooms were closed.

Has anyone gotten in early on namecoin domain names? There are still a lot available when I checked. Unfortunately the process seemed really complex so I kind of gave up for now. Any thoughts?

Ha - #69 = my first name - kai.com - registered by Intel - I guess they are looking after it for me : )

I recommend not visiting symbolic.com. Tacky advertising and nothing else.

IBM, HP, Apple, Intel are right there. I didn't see Microsoft though.

The oldest item on that list is one day older then I am...

dig @new.toad.com toad.com -t AXFR

wutttttt

Reply on news.ycombinator.com