A recent writing gig involved me nosing around in search of the old haunts of Max Harris, a brilliant poet who was big in Adelaide in the early ‘40s. I tried to trace the guy’s youthful haunts — where did he booze, lunch, boogie, pontificate? — only to find that every promising tip-off would lead not to a clandestine dance hall or dingy café heavy with the preserved essence of yesteryear but to a building site or just-constructed gleaming luxury condo. Whenever this would happen, I’d scribble ‘404 ERROR’ down in my notebook as kind of dorky shorthand. 404, 404, 404, 404. It seemed fitting to use web terminology as a marker of impermanence. After all, to rip down a cobwebbed bookstore and replace it with a multi-story car park requires time, money, planning, gumption: it’s possible, certainly, but it isn’t a cakewalk. To pull down a website takes a lot less: a few clicks, maybe, or a swift tug on a server’s power pack.
When Tumblr, for example, went down for 24 hours in early December, those without much vested in the service LOL’d and lamented a future without earnestly vignetted photographs of sunsets overlaid with aliased 72pt Helvetica. Within the geek community, though, the Great Tumblr Outage should have precipitated a long overdue “Oh, shit” moment — the recognition that the web is fundamentally and probably irremediably unstable. But it didn’t.
In his “Open Letter to Tumblr,” Patrick Rhone of Minimal Mac argued that the outage was “disheartening at best and irresponsible at most… What good is the trust I have spent time forming with my audience if they can’t trust that this site will be there for them to enjoy at a time that suits them?”
The use of “irresponsible” and “trust” speak volumes. We adopt many web services because they’re convenient (and free!), but it’s only after becoming dependent on those services that we recognize why they were provided for free in the first place: after all, it’s only by eliminating the inconvenience of paying users that startups can snag attention and secure the freedom to alter, downgrade, or cancel their services at will. By then, of course, we’re trapped in an unstable relationship, and our only means of recourse is to wail as loudly as possible, “You broke my heart!”
The big lesson that should have come out of the Tumbleocalypse was that we trust too easily. Did any of us listen? Nah. Instead, we’re signing our friends up to Dropbox to score 250 megs of bonus storage space and sending our most important documents to “the cloud.” We trust Dropbox because we trust others who use Dropbox: web designers, tech writers and professionals who, we believe, would never gamble with an unproven, flaky, or suspect service. Without this kind of trust-by-proxy, free web services couldn’t survive at all. Can you imagine anybody in their right mind signing up for a Facebook account today without a good friend by the sidelines whispering, “Don’t mind all that privacy whaffle. I know these guys mean well.”
Cloud storage is convenient, of course — ask anybody who’s experienced the horrors of manually synching PC to iPhone — but we downplay the risks involved in outsourcing control of the data we own. We so badly want to live in the future that we’ve lost the ability to question what living in the future might actually mean.
In late December, Yahoo! killed (“sold”) Delicious, the social bookmarking service it acquired in 2005. It remains to be seen whether the site will shut down or reemerge, but what’s significant is the effortlessness with which large-scale communities can be razed online. We push our lives into the internet, expecting the web to function as a permanent and ever-expanding collective memory, only to discover the web exists only as a series of present moments, every one erasing the last. If your only photo album is Facebook, ask yourself: since when did a gratis web service ever demonstrate giving a flying fuck about holding onto the past?
An acquaintance of mine recently applied for an administrator position at Wikipedia. When we got to talking about the encyclopedia, I mentioned the possibility of Wikipedia going down — not for a minute, or an hour, but forever.
“Unlikely,” he told me, but I pushed the issue. What if the Wikimedia Foundation could no longer support the encyclopedia’s hosting costs? What if the community was to lose interest slowly and let the quality of articles slip, until the site was deemed no longer worthy of preservation? What if those in control of Wikipedia decided to run advertising or push a crackpot ideology? These things can happen gradually, little by little, and so long as Wikipedia is, indeed, managed by mere mortals, anything can happen. Didn’t a little part of us, back in ’97, think that Geocities would last forever? Every web service is dependable, until the point at which it isn’t.
Those who believe that “the cloud” can act as a storage platform for our collective memories believe that everything that was available to us yesterday will be just as available to us tomorrow. Where exactly does this conviction come from?
The web is like any other sprawling city, and maybe worse: it’s so damn rickety it’s a minor miracle it hasn’t collapsed entirely. When you link, you do so trusting that the data to which you direct your readers won’t just up and disappear into the virtual ether. Except that, inevitably, it will — the short history of the web has established that much. We live somewhere, we leave, it becomes forgotten, and then we come back years later to find our old haunts brutally 404’d.