Welcome to the Old Internet, Again. – Richard Bettridge – Medium

Welcome to the Old Internet, Again.


This year I started collecting vintage computers. It all started when I was gifted a Macintosh LC 575 in fantastic shape. I ordered a network card from ebay and fitted it into the Mac.

Once I had network connectivity I realized I really couldn’t do a whole lot with it.

I was hoping to check out archived websites from the 90’s using the Internet Archive: Wayback Machine.

However the Wayback Machine injects modern javascript into these archived sites. This will break things on old browsers that have no concept of some of the words and syntax used in modern JS.

So, with my web development background I decided to read up on the Wayback Machine’s API and devise a way to turn their content into something that can be consumed on old systems.

Some all nighters over a weekend later I had a prototype of what would now be called The Old Net.

The Old Net
Welcome to the Old Internet Again!theoldnet.com

I would have preferred The Old Web as I feel that would more accurately describe what it is, but that domain was being sat on for a cool $2500.

Along the way my collection grew from just the Macintosh to two original iMacs (the fruity coloured kind), an IBM PS/1 486, and a not-so-vintage Pentium 4 system.

Having this fun filled collection at my disposal I was very determined to keep investing time into The Old Net. As I started to add features, and as I started to use the system, I realized that it was evolving from something that simply acted as the Wayback Machine for old browsers. It was really starting to feel like I was surfing the old web.

Additionally it was starting to become something that was just as good and fun to use on modern systems.

I built up the homepage and manually curated links of great sites. I brought memorable graphics to the forefront. Built a working guest book (please sign it). Invested into better tooling to increase the chance of a site rendering properly or at the least being debuggable. I improved the performance of the site by standing up an image and document cache.

At this point I decided it was time to release it into the wild and let people in the Vintage PC communities know about it. It was well received and much appreciated.

From there I started to record search terms so that I could show most popular and most recent sites. Hey it’s almost like I’m building a primitive search engine at this point.

It was then when it started to come together. I was discovering things that I would have never thought to type into the address bar. I was able to get a sense of what other people longing for nostalgia thought were their go-to sites.

This is where it deviated greatly from the experience on the Wayback machine. On the Wayback machine you have to know what you are looking for, and there is effort to search, get a response, find a particular snapshot, view it, repeat. What I created significantly removed the obstacles and time taken to complete that cycle.

It stopped feeling like you were loading up archived documents and started feeling like you were clicking around the internet you fondly remember.

At this point the site was in good shape, you could go to a domain, search for sites by a fuzzy query, get a list of popular sites, etc.

But much of the content from my childhood that made the old internet great wasn’t .com’s. It was homepages.

Homepages were a wonderful part of the old web. An endless variety of sites made by all sorts of people. Many people who really didn’t know how to use a computer, and in-spite of that, figured out how to use one well enough to slap some HTML down and upload to an FTP.

Home pages aren’t so easy to find in the Archives. They’re there, but all of the search tools of the past simply do not work anymore. Do you remember webrings? They’re broken. Anything /cgi-bin/? Forget about it. There’s no longer any servers to handle your request.

I realized I could write a parser that would determine if a page that a user was going to was a geocities homepage, and if so it would be tagged and displayed in a specific section of http://theoldnet.com.

Now, all of the lost geocities sites have the potential to be rediscovered and never lost again. This made me very happy and I hope people take advantage of it.

The original goal of the project was to support old browsers and it was brought to my attention by someone on Reddit that I did not support the very oldest browsers.

These browsers have no concept of the “Host” header and they break entirely thanks to the Content-Type header (text/html) being hijacked to also include the character set (text/html; charset=UTF8). I had to put in some work but I got the ;charset=UTF8 removed and I got my server to handle the missing Host header.

The project now worked in Netscape 0.9 and Mosaic 2.5

The entire project has been fun and novel. But I think it’s not as superficial as that. I think I’ve started to discover something that we’ve been deeply missing in the new internet. I talk about this briefly on my homepage http://theoldnet.com/~rich/. It is a semi-serious/semi-being-silly rant. I think there’s at least one good point buried in there.

The timing of this project has been fortunate because on March 12th it is the 30th anniversary of the web. I would love everyone to be able to use my project to relive some of those early days. Though the Wayback Machine doesn’t have archives of the very first days of the internet, it does have the mid 90’s which is when I believe most people of that era first became aquatinted with what the internet was. #Web30 #ForTheWeb #90sInternet