We must protect abandoned sites from a 'datapocalypse'

WIRED: ‘The Facebook of its time’ may seem outdated now, but GeoCities is still worth preserving, writes DANNY O'BRIEN

WIRED:'The Facebook of its time' may seem outdated now, but GeoCities is still worth preserving, writes DANNY O'BRIEN

THIS WEEK was the 40th anniversary of Arpanet, the predecessor of the modern internet.

It was also the week that GeoCities, which in 1994 became one of the first widespread public expressions of involvement in the internet, died.

We know of the 40th anniversary because of the diligent work of archivists and historians, who have interviewed, collected and published the recollections of the obscure technicians whose early experiments turned into a global communications medium.

READ MORE

Those records, of course, were not digital. The first plans for Arpanet were scribbled on pieces of paper or tapped out on typewriters.

But the struggle to preserve GeoCities, which held the first early experiments of hundreds of thousands of web users, has proven to be just as challenging as collecting those dusty records. Whatever marvellous changes Arpanet heralded and the internet has brought, perfect recollection is not one of them.

Forty years is one of those points where the contemporary world begins to fade into history. A young person at the beginning of such a period – like students Vint Cerf and Stephen Crocker, who helped to invent the internet – would be grey-haired and preparing to retire. Many of those in positions of authority would have died.

The internet, in some ways, was lucky to have succeeded long before that tidal mark.

One of the best books documenting its history is Katie Hafner and Matthew Lyon’s Where Wizards Stay Up Late. It written in the 1990s when its origins were a mere 26 years old and many of the pioneers now departed were still alive – such as JCR Licklider, who developed the first ideas for the internet in a 1962 specification for an intergalactic computer network, and Jon Postel, the man behind domain names.

While the internet was created by a few brilliant individuals, GeoCities was created by hundreds of thousands. It was, as someone said to me recently, “the Facebook of its time”, a simple mechanism for creating your own website. And that was at time when almost none of us had any idea how to create a webpage.

If you weren’t using the internet in the 1990s, you won’t remember GeoCities. If you were, you’re probably still trying to forget it.

The products of its simple, page-composing service were garish, clumsy and unaware of the developing social standards of this newly popular world wide web. PEOPLE WOULD SHOUT IN ALL CAPITALS or create huge flashing fonts for the few words they could think to say.

You couldn’t have a custom domain, or even decide on a sensible web address. Instead, personal webpages were arbitrarily ghettoised into fake “neighbourhoods” such as SiliconValley, BeverleyHills, and CapeCanaveral.

Having a GeoCities web address quickly became tangled up with class snobbery – like having a MySpace webpage is now.

When Yahoo, GeoCities’ eventual owner (it bought the site for an amazing $3.57 billion (€2.4 billion) in stock), announced it was closing the service, that snobbery was one of the reasons it perhaps expected little reaction. Like bulldozing a slum, who would care about those decade-old abandoned websites?

But it turns out that even abandoned buildings hold history. The Internet Archive, the San Francisco non-profit organisation that records the web for posterity, began collecting its pages, as did several other independent efforts by technologists.

They were inspired by what technology archivist Jason Scott has called a “datapocalypse”: when thousands of users’ personal sites are destroyed in one go by a third-party closure.

Yahoo at least gave some public warning. AOL Hometowns, a similar project run by America Online, gave very little warning last year when it disappeared for good.

Scott wrote at the time about how the internet’s technologists had failed those who had put so much work into these sites. “We’re talking about . . . hundreds of thousands of man-hours of work, crafted by people, an anthropological bonanza and a critical part of online history, wiped out because someone had to show that they were cutting costs this quarter.”

Why have we failed? Because for all our enthusiasm and optimism about providing a space and opportunities for everyone to contribute to the public internet, we have failed to match those new abilities with tools that let them preserve their contributions.

This mismatch still goes on. How will you save your Facebook posts when Facebook is no longer around? Imagine a time when it won’t be entirely ridiculous to ask: “Where are the Twitterers of yesteryear?”

Scott is right. It’s damning that the anniversary of the internet coincides with the visible disappearance of one of its first mass flourishings.

We date the beginning of Arpanet from the first packet of data sent by the first routers, the refrigerator-sized interface message processors (IMPs).

I’ve stood beside an IMP in Silicon Valley’s Computer History Museum, delighted that it was so carefully preserved and in its rightful place so soon.

But if our 40-year history as internet geeks has been so celebrated, we certainly have a responsibility to those who committed their past to that same internet, only to watch it vanish in less than a decade.