image courtesy of Quartz
unclaimed pixels on the Million Dollar Homepage. image courtesy of Quartz

A couple of years ago, I wrote a series of posts on the atemporal nature of ruins and abandoned spaces. I moved from physical ruins to the nature of abandoned digital spaces, and the conclusion that I came to was that although both physical and digital “spaces” (not totally comfortable with that word but it seems the easiest and most understandable one so I’m going with it) are both atemporal – because a state of ruin/abandonment is intrinsically atemporal – physical ruins are atemporal with an orientation toward the future, while digital abandoned spaces are oriented toward the past. This is because physical ruins visibly decay, and that results in a kind of forward-thinking memento mori; we imagine our own future death and ruin through the present ruin that we see. By contrast, digital “ruins” – in the form of webpages – are marked as abandoned by the fact that they have stopped changing. They are locked forever into the last thing that was done to them, a snapshot of a past present (assuming they continue to exist at all). To borrow my brief summary of what the hell atemporality is in this context:

Atemporality most simply refers to the idea that our experience of time is not necessarily as linear as we like to present it; that we don’t just move in a straight line from A to B in time but that we often experience aspects of the past, the present, and the future simultaneously, simply by virtue of our nature as remembering, imagining creatures — as I wrote in my last piece on this topic, we remember the future, imagine the present, and experience the past. Moreover, this phenomenon is intensified by technology and especially by technologies of documentation and sharing.

However, that analysis left out something sort of important: links.

Sites on the web are defined by their webness – they hardly ever exist in isolation. You can reach them directly through the URLs you enter, but I think most of us access webpages through hyperlinks way more frequently. The pages almost always contain links of their own. Links are in many ways the markers of the presence of pages in the most fundamental sense.

So what does it mean when those links decay?

The technical term for this is link rot, and Quartz did a rather interesting case study recently that examined what link rot actually looks like, using the Million Dollar Homepage. For those who don’t know (I confess that I didn’t), the Million Dollar Homepage was a project started by a college student named Alex Tew, who essentially paid for his entire education by selling 10×10 pixel units of his webpage for $100 a unit. Basically, we’re talking about a (somewhat nightmarish) webpage that consists entirely of link ads.

That was eight years ago. In the sense that I talked about digital space being frozen in a past present, the MDH is exactly that – nothing has changed about it in terms of its appearance. It’s a perfect snapshot of the moment the last ad was posted. But it has changed. We can see decay. Because if one of the fundamental features of almost all webpages is the presence of links, there’s no rule that says that those links will still go anywhere eight years after the fact. And indeed, 22% of the MDH now appears to consist of doors to nowhere.

Therefore, we can understand the appearance of abandoned digital space as past-oriented atemporal. But the fact that there exists a literal process of rot means that the properties of a webpage are future-oriented atemporal – to the extent that we notice at all, it invites us to imagine the dissolution of our webbed pathways, the vanishing of entire sites, or at least their relocation. When a site goes offline, we might notice it when we can’t get to it anymore, provided we go there regularly – or if it’s a large, frequently used site like Facebook or Twitter – but otherwise, like those species of insect in the rainforests that no one ever discovers before they go extinct, websites probably disappear every day without anyone really noticing. Until you click on a link and nothing is waiting for you at the other end.

Complicating the picture, however, are databases like that feature projects that essentially back up large portions of the web. When the online zine that published one of my short stories vanished without warning, I was able to go to the Wayback Machine and retrieve an archived version of the page that I could link to on my author site; my story is still accessible if anyone wanted to read it. But it’s clearly an URL. It’s still accessible, but it’s also still clearly a webpage that isn’t officially there anymore in its original form. The Memento Project seeks to address this; it proposes an approach to archiving that would make web content available via the original links that led to it – this subtly alters the nature of the ruin of these abandoned pages, though it doesn’t erase that nature.

Archived websites also often don’t look the same – image links are frequently broken. They aren’t the pristine snapshots that I wrote about above. They’re in a state of ruin – a static state, for the most part, but ruin nevertheless. Even archived, it’s not uncommon for other links to go nowhere, if didn’t save them (this is also leaving aside the very real possibility that the servers on which the archived content is stored might themselves be destroyed; this past November Archive’s scanning center in San Francisco caught fire and scanning equipment was lost).

This all introduces yet another question: if not all atemporality works in the same way, can we say the same of abandonment? Are archived websites – visibly in a state of ruin or not – really abandoned? Do we need a more complex set of definitions for what abandonment means in this context?

One of the things that Wallace Koheler, a researcher at Valdosta State University, discovered was that link rot appears to stabilize over time; the longer a link stays active, the more likely it is that it will remain active. This suggests that, like ruined physical space, not every digital space will experience ruin in the same way, but it also suggests that, unlike physical space, digital space may essentially stop its process of ruin after a certain point.

I’m still not sure what to make of all of this, but, again, it suggests a more complicated picture than the one I presented two years ago. There’s clearly more work to be done here.


Sarah is still there on Twitter – @dynamicsymmetry