“Geocities is to start charging from April 2, and it encourages users who want to continue using FTP for upload to upgrade to a premium package. Premium users are unaffected by the move,”The Register, 2 March 2002.

“It seems that everything Yahoo! touches ends up dead and burning in hell with Undertaker.” — World of Wonders, 24 April 2002, http://www.geocities.com/Area51/Labyrinth/2003/

As you may remember, Yahoo! not only killed GeoCities. First and foremost Yahoo! has bought it.

It happened in January 1999, but became clear to GeoCities users half a year later, when Yahoo! changed their terms of service. In In Protest of Section 8, published in 2014, I collected reactions and evidence of that user boycott campaign, which some days later forced Yahoo! to take a small step back.

After, things calmed down. Not that users were happy with annoying adds, the malfunctioning PageBuilder editor, traffic restrictions, security holes, etc, but as a frustrated webmaster of TimesSquare/Arena/1256/ nailed it: “all the other online services were either just as bad or go belly-up six months later.”

Full quote from this page, abandoned February 2002:

“Now that I have no need for Geoshities anymore, I have one message for those folks at Yahoo: SUCK IT. SUCK IT HARD. You bastards have caused me and my fellow browsers nothing but trouble for years now, and not just with your webpage services. You sick sons of bitches bought out and ruined Webring.com, one of the greatest services to fandoms all across the net, and for that I will never forgive you. The only reason I haven’t moved earlier was because all the other online services were either just as bad or go belly-up six months later. But now I have my own domain, and I don’t have to put up with you bullshit anymore. So again I say, SUCK IT.”

My collection of “angry” farewell messages was growing steadily but slowly, until I reached April 2002, when the next big move of Yahoo! made users furious and pushed them to look for other hosting services: the shutdown of FTP access for the free home page service.

Here is an official announcement from Yahoo!, Maria Technosux left on her page:

Summary by Denise:



In practice it meant that users could not make their own HTML files any more for free. Non-paying users could only use PageBuilder to create, edit, and publish, effectively becoming tied to a content management system. In 2018 it sounds like the most normal way to publish on the web, but not in 2002. Additionally, PageBuilder was not a particularly sophisticated or reliable tool, which should be a subject of another investigation. (Listen to wise old man in the meantime.)

Users were neither eager to pay, nor to being trapped inside PageBuilder. Below are some expressive reactions which replaced the complete initial content previously published on these pages, and remained until Yahoo!’s final move in 2009:




Researching for this post I stumbled upon reviews of GeoCities at Webhosting ratings.com. They all are precious documents, but the one I want to quote is also a perfect summary of Yahoo’s achievements in their role as owners of GeoCities:

“At one point they changed the TOS agreement to try to grab rights to web content, and refused to let you log into your account, even to cancel and erase everything, without agreeing to the new TOS. THAT was stopped by massive protest. Had I not already left, the final straw was when they discontinued FTP access and email forwarding to anyone but paying customers. By that point I wasn’t about to pay them a cent. I switched to Earthlink and never looked back. My old files are still just sitting there collecting dust…..it just kills me that they took a wonderful, friendly community like Geocities and just smashed it to bits. Yahoo, you STINK!”
Review by Lesley A., submitted on May 17, 2003

The story with Yahoo! taking away FTP is unfolding in front of my eyes simultaneously to another crime happening today, which is Google killing the URL in the version 69 of the Chrome browser, taking away web user’s right to call the site by its address, to give a command to the browser. And though the rhetoric of the two companies is very different—Yahoo! wanted users to pay for the option to control, Google is selling the absence of control as a blessing—I can’t stop myself from drawing a parallel, as both are about denying essential web users’ freedoms.

What was it that made the web so powerful? What made it grow? I would say 3 things:

– Open (and readable) source code.
– A transparent addressing system that allowed to locate and link.
And, of course,
– FTP, i.e. control over your files and independence from particular host service.

These are technical and not really spectacular features, but maybe we should talk more about them than about Animated GIFs, Comic Sans, and other effects defining the epoch  visually but not conceptually.

I’m spending the last hours of 2017 organizing and categorizing pages last updated in the end of 2001. It’s the time when Harry Potter fanfic starts to get illustrated with stills from the film, not pictures from the book; when N’Sync fandom gets more vibrant than Backstreet Boys fandom; when you see a bit more of cat web sites than one year before, but still more dog lovers are out there; when GeoCities users call Yahoo! names for suspending their sites for too much traffic.

However, these are just side notes. The most striking content from 2001 is websites that were made or modified in reaction to September 11. Up until today I looked at 97 of them, and there will be more sad, angry, devastated, patriotic, conspiracy pages appearing in the coming months.

Below is a compilation of 50 screenshots that catch the first days and weeks after the tragedy, and reveal verbal and visual narratives emerging around it.

(Click images to show full screenshots.)

tl;dr: 1TB of KB Age switched to Internet Explorer 6 and Windows XP, but we don’t go for 1024×768 … yet.

Our Tumblr blog One Terabyte of Kilobyte Age Photo Op started in February 2013. It became popular, people were exited about seeing a new snapshot of the past every 20 minutes. Web pages last updated before the advent of e-commerce and social networks, framed by the sweetest browsers with the cutest logos provoked nostalgia even among the followers who were born after these pages went live on the web.

Fifteen months later we made an unpopular but necessary change. Screenshots switched from being created with Netscape to Internet Explorer 5. To say the truth we were really postponing that decision as long as possible, but the 19 March 1999 release of IE5 was the border after which we couldn’t represent the web inside of a browser that no one used any more.

It is almost 2018, almost 5 years from the launch, and we are on the eve of the next big sad change. This time it is not connected to any particular moment in the history. It just happened that some weeks ago the repository of snapped sites got empty, and it was time to restart Dragan’s screenshot factory and discuss the appearance of pages that were last updated in October 2001.

One thing was clear, time has come to change to Explorer 6  and stay with it till the last screenshot we have.

Another thing we seriously considered these last days was switching the screenshots’ resolution to 1024×768 (XGA).

I wanted to use this break in the Tumblr blog’s flow to address the issue, which was eating me from the inside for quite some time already: the eligibility of 800×600 (SVGA) to represent web pages of the 3rd Millennium.


There appeared more and more pages stating that they are best viewed in 1024×768; more and more often I stumble upon sites that are made for this resolution.


Add to it my personal memories about having 1024 monitor already in 1999. Plus the knowledge that Yahoo!’s Page Builder templates were designed for a 1024 pixels wide window, already 2 years prior to the time we are looking at now…


…and you know how I feel. Namely, I feel like I am a very mean researcher who makes old web pages look older than they are.

On the other hand, what I just described as “more and more” is only a perceived increase. Pages that are made for wider screens are still a minority among GeoCities pages last updated around the turn of the century: 3 of 100, as a quick test of 100 web pages last updated in July 2001 showed.

The vast majority still looks good or best with 800×600, and a lot are adjusted to look fine or ok even at 640×480. Either they were made for it, or their webmasters followed the rule that a web page should look proper first of all on the smallest of screens—an approach that we know today as “mobile first,” but before mobile.

Also, if it looks bad on 800×600, there is no guarantee it will look perfect in 1024×768.


Also, I noticed that many pages announced to be designed for or look best with XGA did away with flexible layouts, a web design paradigm that came back a decade later under the name Responsive Design): content arranged into HTML tables with dimensions defined in percent values and therefore adapting to the available screen space gave way to exact size exact definitions in pixels, or exact pixel coordinates on the page. This stagnation in web design, caused by an increased availability of graphical WYSIWYG page editor software and professional graphic designers creating pages in Photoshop, started already in 1997; it was then deepened by the burst of the dot.com bubble in 2001, and with XGA resolution and Internet Explorer dominating users’ desktops at least from 2004 through 2008, there was hardly a reason to keep web layouts adaptive.

800×600 screenshots demonstrate this inflexibility very well. It would be a pity not to see it.

And finally, the claim that a page looks better in XGA doesn’t mean it looks worse in SVGA.


Quite often i see the pages that seem “incomplete” because the bottom part is not visible in a lower resolution. But, on second thought, isn’t it exactly the act, the necessity of scrolling, that provides a wow or comic effect?


Furthermore, statistics show that 2002 started with 60% of users accessing the web with a lower resolution than XGA.

So, 😥😌

Based on these statistics and personal observations we can rightfully stay with SVGA up until 2004. And if we want we can find reasons to show GeoCities pages in this classic resolution, synonymous with the web’s golden days, till the last screenshot.

But is it what we want?

History and nostalgia are not the same. GeoCities was seen (and ridiculed) by people in front of 1024×768 monitors since 1997. Shouldn’t it be part of history? XGA was there when GeoCities was aging, and when it was killed. Actually (looking at statistic again) they both died the same year.

Our Tumblr blog’s role is to feed the web with its past. It is tempting to conserve and show amateur web productions in smaller form, because it makes them look better, younger. 800×600 is like a beauty filter for web pages of all ages. Looking at the web through it you get sentimental.

If 800×600 is the web’s ceremonial portrait, 1024×768 is it’s hidden camera, it reveals what had to be hidden.


On a bigger screenshot you see how the amateur web was shrinking and shriveling, you can get sense of how it was becoming small and unimportant. Not only when something is obviously wrong as above, but maybe even more in the case below: where everything is still alright, but you get the feeling of the web page aging, of belonging to another epoch.


Suddenly! An answer from Jeremy Penner ‏to my almost 6(!) years old question!

… continue reading


It is my special joy to announce that this year for the first time the GeoCities Research Institute got its own advent calendar. 24 doors from One Terabyte of Kilobyte Age archive are waiting to be opened at Old Web Advent Calendar. Door by door, day by day the calendar reveals treasures from the archive. Thanks to the student of our Traditions and Revolutions project group last semester.

Surfing the archived geocities sites (it’s totally fun by the way) I found many webpages – collected by my prof Olia Lialina – all having an image of a door on them. Mostly they were used as entrance button leading to the main page.

It would be a shame if nobody would ever see those beautiful doors anymore!

So I came up with the advent calendar.
Enjoy! And Merry Christmas

—Nick Betzlbacher!



This post is to announce a new piece of performative restoration, made by Tara Donovan-Achi, a student in the Traditions and Revolutions project at Merz Akademie, Stuttgart, led by Dragan Espenschied and myself last semester.
Tara writes:

“Going through One Terabyte of Kilobyte Age Archive I came across the so-called Papercat and realised how important it was back in the days. A web cat before LOLcats! It is fascinating because it seems to break through the imaginary wall, as if there is something behind the browser. A room? A garden? Cyberspace?

Early browsers didn’t necessarily implement the GIF89a specification correctly, and this animation was originally made for buggy browsers. While it displays nicely up to Netscape version 3, later browsers have the cat disappear and leave a closed eye behind. There is also an issue with transparent pixels that obviously should be opaque white.
I decided to bring Papercat back to life, to the modern web. “

papercat.geocities.institute showcases a repaired animation, a high-dpi version and 113 wallpapers the cat was bursting through on pages saved in the 1TB archive. But that’s not all. Papercat2k16 got the paws and the tail and… everything what was behind the wall and wallpapers. Enjoy! (if possible in Chrome)

P.S.: Papercat’s earliest (September 1996) appearance in the archive:


A screenshot of Stephen King on the Net appeared this week on the One Terabyte Of Kilobyte Age Tumblr blog. The page was last updated on the 14th of October 2000. The screenshot didn’t go unnoticed: 52 users reblogged and liked it. I tagged it as “my,” ”fan,” and “clipart,” looked through the news and links section, read the promise of the author to update soon again—my everyday routine. But this time there is a history to this case that makes it special.

Half a year ago in Tel Aviv, after my talk about immersion and the early WWW, the generous Dr. Lior Zalmanson gave me a tour of the town. After a short talk he mentioned that when he was a teenager, he had his page on the GeoCities too. It’s not a rare thing to hear from people I encounter, but it’s usually the case that they don’t really remember what and where exactly it was. Lior on the contrary was quite precise in his memories. Not only that, he could remember details that made it obvious that this website was a big deal and a big part of his life. So I’ve asked Lior for an interview that ended up taking place on the 28th of October 2016 in New York.

Olia Lialina: How old were you when you started to make the page?

Lior Zalmanson: I got thirteen in December 1996, and the page was created in the beginning of 1997.

OL: Amazing!

Honestly, I didn’t have to ask, I remember you referring to it in Tel-Aviv, and it was in that moment that I’ve decided I should really talk to you. There is this arrogant phrase that I found in 2008 on a generator that made fun of the early web pages. It’s banner shouted: “Make Any Webpage Look Like It Was Made By A 13 Year-Old In 1996!” It was repeated many times in articles and blog posts, and became sort of a stamp of sarcasm they put on GeoCities pages. Now to the real question:

Were there other 13-year-olds around you who were making webpages at that time?

LZ: I think I was relatively early. I remember vaguely there were a few other computer geeks in class that have been experimenting with HTML or building pages. But I don’t remember anyone of them to have any content other than a page with their picture, their first name, saying what are their hobbies, “my favorite links.” I think I was the only one that I knew of in my class that has really done a webpage for something.

OL: Did your classmates know about your page?

LZ: I don’t think so. Maybe my best friends. The web was another universe, my page was made for people out there.

OL: You proudly state that you had a purpose. And the purpose was…


… continue reading


The history of the WWW is not a singular history, but a dispersed collection of eclectic histories and stories – many that have nothing to do with anybody else’s.

There are graphics, sounds, elements, names, tools, and events that had influence on almost everything, and there are very personal heroes that never crossed a border of a particular community or server. For many, the starry night background is the symbol of the WWW, for some it would be a Mulder and Scully portrait, for a few — a SETI button.

It is difficult to find something what would not mean anything to anybody. But I think I recently managed.


It is a collection of 10 images in TIFF (!) format that came on a CD attached to the manual “Microsoft Frontpage 97: HTML and Beyond” written by Gus Venditto in 1997.

Frontpage was a widespread WYSIWYG HTML editor. Half of the web was built with it. The manual itself was also no rarity. But have you ever seen these graphics on the web? Did you ever copy them from the CD, convert them to GIF or a JPG and put them on your page? Do you know anybody who did?

Since one year Cory Arcangel and myself are flipping through the files, asking ourselves: why TIFs? Why firemen? Who has bitten off the donut? WHAT ARE THESE SHOES? How this collection came together at all?

On the eve of the opening of our show Asymmetrical Response at Western Front in Vancouver, we decided to give these graphics a second chance. With the generous permission given to us by the webmasters of Western Front’s website, we replaced featured images on their blog with two firemen, an iguana, a flying twenty dollar bill and wide-angle view of a donut among others. Enjoy!

aol exhibition

Simon Baer, since some days a former student of mine, finished his studies with a work that contributes to several contemporary “fields of concern” – big data, surveillance, the right to be forgotten, as well as user culture, digital heritage, personal archiving. On the top of it http://cannotsleepwithsnoringhusband.online is a true online story, it speaks the language of the web, browsers, it takes place in the search engine.

It is a search engine drama. The genre we, google folk, will not have problems to get accustomed to.
… continue reading