Retrosurfing

It’s 2 AM. I’m in my bedroom. The window is open, and I can hear the crickets. I’m the only one up in the house; everyone else is asleep. “Maybe I should go to bed,” I tell myself. “Or maybe I’ll just continue to surf the Web for the just another hour…”

I’m sitting in front of my slow Packard Bell computer with its Pentium processor, 16 MB of RAM, a 2 GB hard drive, and one… big… blocky… cathode-ray tube display… with bulky speakers on its sides. It was my first computer at home, but not the first one I used—for a few years before, I was frequently using my uncle’s PC running Windows 3.1.

On my computer running Windows 95, I’m on the Internet, connected with my 56 kb/s dial-up modem. (Although, I know the maximum speed is actually 53.3 kb/s.) It’s late, very late, but this is the best time to be connected and keep the phone line for myself. There was no risk for anyone to pick up the phone and disconnect me by mistake. Also, it’s free to connect at night!

I’m browsing the Web using Internet Explorer 4, because I find it more stable than Netscape 4. I’m chatting with friends on ICQ, IRC, or some unheard Perl script in a cgi-bin directory running on an Apache Web server which refreshes its page everytime you post a message. I like some anime, notably Sailor Moon and Evangelion, so I browse the Anime Web Turnpike (Anipike, for short) then try navigating through a Japanese Web site even if the font couldn’t be displayed. I download hi-res images of something I like, but I know I’ll have to patiently wait a minute or two for each progressive JPEG file to download and go from a blocky mess to a clear picture. Good thing my graphics card can handle more than 256 colours, or else the dithering would make the image look horrendous! But, the download can wait. I normally queue URLs of those in a download manager and put some of them off for later—I don’t want the transfers to hog the speed for whatever else I am doing at the moment.

I could always use more free hours of dial-up, but I’m too sleepy. I queue one MP3 or two for download in Napster. Downloading a single track takes half an hour in the best conditions. I schedule my computer to shutdown in an hour or two, giving some times for the downloads to finish, and I go to bed. It’s now Sunday morning—I’ll probably sleep in until the afternoon.

That was the way I was spending many of my weekend nights when I was in high school in the late 1990s. I didn’t have a car, and I had a friends but they were not living nearby. The world outside, for me, seemed much larger, but online, or in “cyberspace” as we called it back then, it seemed much, much smaller, and it kept shrinking.

The Web has been a companion for me, and for many others, during the past few decades. It began as a side project by TimBL, a scientist working on a particle accelerator to share documents. Who knew it would spread like a wildfire and affect so many lives for years to come?

It made me sociable, it made me knowledgeable, it gave me friends, entertainment, jobs, and a voice. What attracted me to it at first was its social value. I could easily make friends online, find people who loved the same things I did. Like now, I enjoyed writing, so I learned HTML to be able to write about myself and the topics I enjoyed, then share my thoughts with the world with a site I published on whatever free hosting provider I found in the time: GeoCities, Angelfire, Tripod, Xoom… For me, that was the most interesting and rewarding thing. I’m a Web developer today thanks to that. It’s not what I expected, not really what I wanted—I was hoping to be an author or actor—but it often worked in my favour.

Now, 22 years later, I longed for those times.

It’s a little bit strange. I have a smartphone with me at all times now with a fast Internet connection, even outside. I don’t have to be in the “computer room” to be on the computer. I don’t even need a desktop computer. Even my notebook is at times obsolete! No matter where I am, I’m on the Web, and I can talk to anyone and do most of the things I need. What we dreamed of being able to do back then, we have it today! So why would I miss those times?

I yearn for a few things from back then. First, not everyone knew how to be online or cared to be on the Internet. I didn’t have everyone from work or from home watching my every move online 24/7. I had my circle of friends online, and my other circles of people in my life offline—they rarely mixed. Well, that’s until I found the omnipresent omnipresence invasive and decided to block dozens and dozens of people out of my online presence recently. Still, the fact that I even had to do that… I also didn’t feel my every move online was tracked or analysed for commercial purposes or research… even though they probably were, just like in the present.

Not everything was conglomerated into a handful of popular sites. If we talk about the Internet today, we think YouTube, Facebook, Twitter, Amazon… To most, the Web is synonymous to the Internet, and to many, the Web is just those sites. But back then, the Internet, the “Information Superhighway” was a trail in digital outer space waiting to be explored, while the World Wide Web was its World Wild West and yet the most handsome new kid on the block. There were even magazines—yes, actual publications printed on paper—reviewing every unknown corner of the Web, any little site that could have an obscure topic or a flashy design. (Heh, Flash…) Every Web site had personality: Web design was new—not always pretty, but often imaginative—and it was something with which I had fun experimenting. Many surfers enjoyed it too, and about half the people of my age online had their own site, their own space online telling their stories… They were webmasters. We’d all exchange our URLs to each other on our sites’ guestbooks, join chains or link exchange networks… The Web was nascent and we all enjoyed learning to grow up with it.

What if there was a way to take a trip back into those times?

I may not have a phone booth in which I can jump into and pick a place and time, but I can maybe turn my computer into my very own TARDIS.

During my hunger for nostalgia last weekend, I headed to the one place I know with all the sites of the past… The Internet Archive. They’ve been crawling the Web site 1996 and they’ve been storing historical copies of anything they can find and store over the years. With the Wayback Machine, you can not only input a URL to visit, but also ask to see it from a specific time! You can visit Apple’s front page when they introduced their iBook—the MacBook of the 1999—or check out a budding YouTube thinking about being a dating video site in 2005… There’s so much that massive digital archive can tell us.

But, browsing a retro site on a modern browser doesn’t feel right. Fonts are anti-aliased, all displays have 16 million colours, resolutions are large, pixels are dense, browser UIs are clean, neat, and minimalist… I mean, it works, but a bit like how watching VHS tapes looks somehow better or at least feels better on an old CRT with two large channel dials on the side, I believe those sites would be best viewed in Netscape than Firefox.

Thus, I explored today’s Web to find anything that will let me run outdated operating systems on my computer.

Turns out, finding all the software I needed was quite easy. I struggled a bit with setting up everything in virtual machines, notably networking, but I managed to make everything work. For a few hours, I had many failures: setting up TCP/IP in Windows 3.1 for me was impossible, Windows 95 in VirtualBox on macOS just wouldn’t boot… Eventually I set up Windows for Workgroups 3.11 and Windows 95C in VirtualBox on Windows 10. Those were the closest to what I’ve used back in the day. I also installed Mac OS 8.5 in SheepShaver on macOS Catalina. I never had a mac before Mac OS X, but hey, why not?

Okay, that was all fun for that retro feeling, but… All I get out of all this is toying around with historical OSes and browsers. There’s a big part missing… Where’s the Web?

Sure, I can try to browse the Web. But the Web of 2020 isn’t made for a W3 browser from 1998! Start accessing sites in those apps and quickly you find that practically nothing is working. In Netscape 2, for example, SSL was practically new, and so was JavaScript and CSS. Everything looks terrible, and most sites won’t even work because they all redirect to their more secure HTTPS version now, with encryption way too advanced for that navigator! Although, really, what else was I expecting? It’d be crazy to make any Web site backwards-compatible for a browser that is more than 20 years old.

What about old sites from the Internet Archive… Well, turns out it’s not so great. Sure, the pages were archived from decades ago, but their custom code embedded in them for navigating through their content was made for browsers today. In the old browsers, JavaScript errors pop up left and right, and it messes up the rendering of the pages. It’s as if those dusty apps travelled back to the future, our present, and had to deal with something new to them but old to us. Crap…

However, maybe I can bring the restore the retro Web thanks to that archive. Maybe I can find a way to bring that authentic browsing experience from the 1990s.

And yes, fair enough, there is a way.

A few hours later, I wrote timeprox, a Node.js ES6 HTTP server which you can use as a proxy to browse sites from the Wayback Machine. (JavaScript. There’s another thing I didn’t expect to evolve and spread far and wide!) Start the server, set it up as a proxy in your old browser, try to visit an address, and voilà! My server will request the page you want archived in 1998 (or whatever it can find from later), strip the Internet Archive’s custom code, and show it to you, along with all the images and other assets. You can click the hypertext links all you want and surf the Web of the past! Or retrosurf, as I like to call it.

(I found out later that a similar proxy already exists, written five years ago, called peabody—you know, in reference to Mr. Peabody the dog entering the WABAC machine, or “Wayback Machine.” The name of mine is simply short for “time proxy.” So, no, the “prox” here isn’t short for “proximo” or some old term from Rhode Island, US to describe a list of candidates at the elections. Nothing fancy.)

After some tweaking and trial and error, I have to say, the proxy works rather well! I was quite proud of making this work. The sudden surge of nostalgia as I browsed the old homepage of Netscape was exhilarating!

I visited all the sites I could recall, all the places I used to frequently visit when I was in my bedroom during my days in high school or in class during college. Yahoo!, GeoCities (which used to be independent), Infoseek, Anipike, that old Perl chatroom I mentioned, an early Japanese video-on-demand site with anime shows, and even my old site, called E-Rémi or Érémi in 2001. (A title to which I cringe every time I see it. The way I wrote it in Japanese, eレミ, thus making it unreadable by the average visitor, is making me shiver further. I like to think I learned from those mistakes. In any case, I enjoyed experimenting with its Web design and mimicking Apple’s Aqua look from the day.)

For the rest of the evening, I was excited to retrosurf through Web 1.0. The small animated GIF banners with dithered colours, the jagged fonts, the image maps as menus, the repeating motifs set as backgrounds, blinking or scrolling lines of text, design made for small monitors with a resolution of 800×600 pixels… The experience brought back so much memories. Not only memories of what I did online, but what I also did offline, how I was back then, in which environment I was while being connected. Where I went to meet people, what I used to read and watch for hours on end, the activities I used to do… Not only my online life began then, it essentially became my life.

That said, after browsing the Web of yesterday, the nostalgia melted into a desire to relive those days. But, I didn’t actually fall into a portal back in time. What I was seeing was only a snapshot, an image of the time back then, a faded photograph of what I used to enjoy. I could access Hotmail, but of course I couldn’t log in to check my mail from 22 years ago and send messages from then as if I was a time traveller. I could visit Yahoo!, but I can’t type a query and search like I can on Google today. I checked out the chatroom where I socialised, but I can’t send messages and talk with people as they were in 1998.

Those were only memories. The craving to be there and then again eventually faded away… into the realisation that we cannot live in memories and those days are long gone.

I can’t do anything but to look forward… Like going to bed for work tomorrow instead of staying up until 2 AM on the Internet to write this article.

If you wish to retrosurf too and relive the memories from the Web of then, here are some links to which you can refer:

timeprox

The Wayback Machine HTTP proxy I wrote.

WinWorld

Archive of old software and operating systems.

Windows 3.1 Networking with VirtualBox

Video tutorial on how to make networking, well, work in Windows for Workgroups 3.11 (despite the title mentioning Windows 3.1).

Socket 3: Install & Configure MS-DOS 6.22 & Windows 3.1 using Oracle VirtualBox

Although I ended up with WfW 3.11, this tutorial did offer some good pointers on how to set up the VM, notably with display drivers.

Guide to DOS on Virtual Box

The networking setup of this guide didn’t work for me, but maybe it’ll work for you.

E-Maculation: Setting up SheepShaver for OSX/macOS on Intel Macs

Detailed steps on how to have old Mac OS versions work in SheepShaver on macOS.

Redundant Robot: SheepShaver

A companion tutorial to the one above, with essential downloads.

ΩMEGALOGIC: Retrocomputing

Article with more resources on retrocomputing.

Feel free to ask me questions about setting up VMs or simply to reminisce.

P.S. Ironically, while restoring this article, instead of digging through old database backups and code, it was easier to pull the content and images from The Internet Archive. Thank you, Wayback Machine!