About a month ago my colleagues and I moved back into our old building. For slightly more than a year, we’ve been housed in a collection of offices around campus as our own department was rebuilt after a devastating fire. Caused, apparently, by a workman using a blowtorch in a high wind on an adjacent building site, the fire roared through the beautiful, wide corridors, and airy, high-ceilinged offices of our building, and burned and melted everything in its path: lifetimes’ worth of research, book collections, and lecture notes.
In some ways, my colleagues will never really recover properly from the loss. Books and computers can be replaced, but some information – which wasn’t properly backed up – has gone forever, as have notes and other ephemera.
I was very lucky not to be affected by the fire at all, although I now have an inkling of what it must be like to lose so much research.
As academics, we tend to comfort ourselves with the knowledge that however vulnerable our computers may be – to crashing, to theft, to coffee spills – we can back up their contents, on other hardware, and, now more frequently, on the Internet. If mechanics fail, then the collection of cables that constitute the Internet can keep our research safe.
But as I discovered a few weeks ago, things can disappear even from the internet.
It began with a call from a friend, saying that she’d received an email from me asking for money after a mugging in Madrid. I’d read about this kind of email hacking before, and, cold with panic, I tried to log into my Gmail account. It was locked: the hacker had changed both my password and the security question, so there was no way I could access it.
While trying to answer a deluge of texts and messages from concerned friends on Facebook and Twitter, I filled out a form on Gmail, answering questions – the date when I opened the account, the addresses I email most frequently – to prove that it was mine, and about twenty minutes later, my account was returned to me.
But every, single email – in the archive, under each label, all my sent mail, and all my drafts – was gone. Five years worth of emails, charting my PhD research, trips abroad, friendships and relationships, new jobs, and the beginning of my postdoctoral fellowship, as well as attachments containing lecture notes and drafts of chapters and articles, had disappeared into the ether.
It felt like I had nothing to show for what I’d been doing for the past five years. The loss made me realise how much I had come to depend on having ready access to my email, and an account with apparently limitless storage space: I used my Gmail account as much to communicate with people as I did as a kind of virtual notice board and filing system.
Although an initial request to recover my mail was turned down by Gmail, I knew that because my emails would be saved on the company’s servers for a month after the hacking, I would be able to recover them. I contacted a friend-of-a-friend who works at Google, and within a day, my account was fully restored.
I have now backed up my account, but the experience has made me suspicious of the apparently fail-proof Internet. At around the same time as my hacking, the server of 3:AM, a literary webzine, was switched off, taking with it the website as well as twelve years of archives. After a byzantine search for the server’s administrator, 3:AM, as well as its unbelievably rich collection of stories and articles, was saved.
Reflecting on the experience, 3:AM’s editor, Andrew Gallix, comments : ‘The web is a Library of Babel that could go the way of the Library of Alexandria. … It is the ultimate Gesamtkunstwerk – “the catalogue of catalogues”… – but it also marks the triumph of the ephemeral.’
This is useful advice for academics. We need to be more careful about how we store data. The Internet is often held up as a way of making research and teaching cheaper and more efficient: whole libraries of journals can be replaced by a computer with access to JStor or Ebsco. Courses can be taught to students all over the world via the Internet. This is all fantastic, but how much thought have we given to the fact that with one crashed server, all this information can disappear as easily as when a fire destroys a library?
Stellenbosch, South Africa.
Sarah Emily Duff is an NRF Postdoctoral Research Fellow at Stellenbosch University, South Africa and is a regular contributor at University of Venus. She can be contacted at firstname.lastname@example.org.