The Flickr Foundation faces a unique task: overseeing an archive of 50 billion digital images (and counting) for the next 100 years. These images range from intimate family photographs to historic images captured by NASA scientists, they harbour the depiction and intellectual property of countless individuals. Whilst we can agree that the preservation of such an archive is a worthwhile endeavour in its own right, an archive of such scale prompts new considerations around the cultural, social, and environmental ethics of digital storage and display.
The Flickr Foundation partnered with the Oxford Internet Institute on 22nd May 2023 to convene a meeting of minds. The gathering brought together 18 scholars, researchers, artists, and archivists; stakeholders eager to address the new centennial challenge for digital preservation. The goal was to spark dialogue on priorities that might shape a future 100-year vision for the Flickr Foundation and similar digital archives. From this dialogue, three significant observations emerged:
Flickr’s centennial challenge is unique to the age of Big Data. Long-term preservation strategies vary when the numbers of objects multiply exponentially. A 2019 New York Times study shows museum collections have ballooned in the last 50 years. Consequently, part of the modern archivist’s role is prioritisation and deletion. The seeming sacrilege of destruction is, in many ways, a reaction to the proliferation of items through technological and mechanical reproduction.
Large-scale digital archives present similar challenges to storage and a necessity for curation. As Nicole De Groot warns, “Being digital does not make it zero-weight or zero-waste.” If difficult decisions must be made, this introduces a new series of questions: which digital records can we let go of? Who should get to decide this? And how should this be done in a manner that honours their original contribution?
It might seem obvious, but it is worth stating, most conserved items are not digital. The Bodleian, for instance, over its 400-year lifespan, houses 1 million digital representations compared to 13 million physical items. As one workshop participant astutely observed, “ironically, digital has the least longevity”, due to a host of uniquely digital issues, including file format obsolescence, plagiarism, copyright, and license decay. Whilst this is a technical or infrastructural issue, there is also an emotional issue at play here.
It is challenging to replicate in the digital realm the same awe and reverence we experience when walking into a physical archive or beholding historic objects. We understand, academically, that digital archives are important, but perhaps to secure their longevity, we need to make steps to imbue them with emotional connections. Creative solutions might involve drawing from alternative methods of cataloguing, curating and display, such as the wunderkammer, Chinese Scholars Objects or the humble pegboard.
Long-term planning for digital archives, such as Flickr’s, raises complex questions about data ownership and usability. The concept of informational dignity becomes central here, as digital archives will increasing include user-generated profiles and depictions of the deceased, which demand an enlivened approach to the ethical treatment of digital remains. As we move further into the age of Big Data, and as our understanding of the ‘digital human’ evolves, careful consideration must be given to ongoing access, rights to usage, deletion, and interpretation.
There are certainly positive possibilities for the usability of large-scale visual image analysis, as evidenced in the deployment of Flickr’s archive for researching climate change through analysis of flora and fauna in the background of photographs. But balancing scientific exploration and individual rights requires that dignity be at the centre of any long-term plan for user-generated archives. The consequences and correlations of future research projects are inherently unknowable, and therefore a right to be removed or ‘forgotten’ from the data should be built into any long-term plan. These rights could be extended to lineal descendants or communities, to ensure any future data deployment is reciprocal and not extractive.
As we contemplate the next 100 years, we must recognise that the fragments of our digital existence may outlive us, and their original contexts may well be lost. Digital preservation is not simply a technical problem to be solved by more storage or better file formats. In an era where much of human experience is recorded digitally, the conversation around preservation becomes essential to maintain historical continuity, understanding, and cultural flourishing. Moving towards an archive that is representative, usable, and responsible demands further discussion. As such, three core considerations for a future digital archive can be summarised as follows:
This project aims to bring together industry and academia to collaboratively develop principles for building an accessible social and technical infrastructure for preserving digital images.
