Delete. Erase. Trash. Wipe. Shred. Tape over. Sanitize. Undo. The words used to specify the removal of a record is as diverse as the formats that it can take shape in. Files are deleted from the drive. Audio and video wiped from tapes. Words erased from paper. For more certainty, the paper can be shredded. Tapes burned. Hard drive dismantled, platters grinded.
Paper does not erase very well. Like a CD or DVD, a user will typically write once and then move on to the next paper or disc. While writable drives such as hard drives and usb drives are coming down in price, they are valued for their read, write, and erase ability. This document was written and replaced in RAM at every keystroke. And while hard drives do continuously replace iterations of working document through an autosave function, in this particular case the word processor is on a distributed computing network. The file is autosaved across several hard drives with a trace of every sentence fragment sent to the system by copying itself across servers.
Confronted with an unwanted word written in pencil on paper, an eraser is employed, the evidence – a mash of rubber and carbon – brushed off the page. On a computer the material experience of deleting a record can be as effortless as holding the appropriate keys for undo. The physical ability of the storage media we use everyday – RAM, hard drives, web storage – is represented by icons and drive letters, a percentage for available physical space. Likewise, erased content is represented by a request for confirmation, an empty folder for the file, a blank space for the word.
As effortless as it seems to delete a file, undo a word, or even wipe a tape, the physical proximity and possession of the recordable medium no longer provides a tangible guarantee. The comfort of the possibility of effortless deletion has proven disruptive when the computer is plugged into the network, files systems scattered and distributed, users clicking in and sharing out. Material presence of information never guaranteed the author control of the content, but the language of expressing and representing control over information is changing. While arguably etymologically similar, the words delete and erase are associated with largely two different activities. Search “how to erase” and the results will be about scrubbing content off storage media. Search “how to delete” and the results are about closing accounts, revoking privileges to packets of information. Comparing the two search terms on Google’s Insights for Search tool reveals that since 2007 requests for methods of deleting information have widely increased in proportion to requests for methods of erasing.
In response to the uncertainty of whether a file has been completely deleted and responding to legal requirements, institutions write “data sanitation” guidelines, search engines deliver on policies to remove records after x number of days, and social networks permit users to remove posts. This shapes our communication to the present, but not the future or past. For as the deleted record has become a matter of social consensus within our civil, everyday life, the physical matter of record in the hands of a forensic investigator, a media archaeologist, or an individual with criminal intent is different. We are agreeing to forget without material verification.
This project is about verifying and testing the hypothesis that a meaningful change is taking place in our relationship to the materiality of deletion. My proposal is to foreground the various material surfaces, tools, and processes involved in record deletion across a spectrum of media. The project can develop along several lines towards an exhibition. Preferably, I would have the research method rooted in the exhibition. To make the formation of a database driven exhibition itself, a method of research.
It would be a website in the fashion of a “how to” guide for deleting documents and erasing media. While the questions would be structured in the binary of medium and message, each entry in the catalogue of methods would be ranked by physical verifiability, chance of recovery, and perhaps medium usefulness after deletion, along with the usual “helpfulness” rating that may be applied to the typical questions and answers platform website. The research of methods of deletion will be guided by the current search terms used for seeking “how to” guides. From that point historical analogues would be traced and catalogued. The result would be a survey of deletion practices of historical media that may illuminate how we came to accepting a difference between deleting content and physical erasure of the medium.
While the exhibition could end there, as an open database it would theoretically evolve as users searched, added, linked, and engaged the evolving cosmos of media platforms. If successful in this second stage, the exhibition would continue to bring new light to the materiality of information. It may even go a different direction than expected. Certainly, recognizing that most web sites don’t gain an audience in two months, the focus of this project will be to draw conclusions on the preliminary experiment of pushing deletion practices of new and old media into the hard categories accepted as the norm for a culture acclimating to serving and saving content in the cloud.