Weeknotes: 31.2014 – Cultural Oversharing
The Internet didn’t invent oversharing. That accolade goes to Dawson’s Creek. I still remember the first time I looked deep into James van der Beek’s televised eyes and felt compelled, suddenly, to talk far too much, far too openly about my feelings.
Nevertheless, ever since its conception, and particularly with the subsequent development of social media, the Internet has been helping us to go that extra mile. We have developed digital platforms where we can tell absolutely everyone absolutely everything about ourselves, and it can be hard to resist.
This week, coming off of the back of two projects – an evaluation of a regional film archive’s online services, and a review of a prototype smartphone guide – I began to wonder if cultural organisations aren’t falling into a similar oversharing trap as they increasingly embrace the use of digital technology.
I’ve also been fuelled in my thinking by reading this fantastic and provocative post by Ed Rodley; an appeal to museum’s to open up collections metadata online. His argument is a convincing one, and I’ll be coming back to it in a moment. First, though, some scene setting…
As a rule, collecting institutions possess a vast quantity of data, from catalogue entries to images, interpretative text, video, audio, and so on. Before the web, this all had its place. A small amount went into carefully curated galleries and exhibitions, and the rest got ferreted away behind the scenes. Now, though, the Internet and mobile technology have made it possible for cultural organisations to share every last scrap of data they have easily and quickly with the public wherever they may be in the world.
For Rodley, this new accessibility represents a limitless opportunity to ‘create and spread the digital DNA of our shared cultural heritage’. He argues that by being ‘promiscuous’ with our data, we can ‘connect museums to maker communities’, reimagining the museum-visitor dynamic and giving cultural audiences greater freedom and autonomy. As an intellectual appeal, this is certainly persuasive; however, it is not without its problems and should not be embraced lightly by collecting organisations.
The problem I want to focus on in particular is the part audiences play in open data schemes. Just because we can give our visitors access to data doesn’t necessarily mean that they want it or know how to use it. In fact, offering it to them wholesale can be problematic, as we found was the case for the regional film archive we have been working with.
For the archive digitisation is currently a hot topic. Its website gives it the opportunity to show and share digitised versions of its film holdings. Digitisation is expensive, however, and so as it stands only around a tenth of their collection is held digitally, with less than this available to view online. The result is that when users search their online catalogue, they are presented with a mixture of different results: some digital films to view instantly; some simple textual catalogue entries.
This comprehensive presentation is a deliberate strategy and felt to be a positive offer for users who are given an overview of everything in the archive, regardless of its availability. Where material isn’t online to watch instantly, a polite notice explains the logistical and financial reasons for its absence.
In speaking with the archive’s users, however, we found that this combination of different data types, allied to the archive’s honesty in explaining where and why holes in its digital collection existed, negatively affected how users perceived the organisation and its collection. Visitors to the site were looking for film to consume immediately, and as such they didn’t appreciate having text catalogue entries scattered throughout.
Rather than communicating the real breadth of the its collection, the archive’s openness in fact made its service feel incomplete – therefore inadequate – and difficult to use. The suggestion was that not everything was ‘digital’, therefore not enough was digital. Users perceived it to be simpler and more reliable to use rival archives where ‘everything’ was ready to view online and so could be consumed in the manner they anticipated. The fact that the particular archive we were working with had in fact got over 400 hours of film available to view instantly, license and get creative with was lost.
For the archive, this came as a shock. Despite the best of intentions, for the majority of its audience it was in effect providing too much data. In a sense, it was oversharing.
Turning to the second project mentioned above there was a similar pattern in our findings. The smartphone guide in question had been conceived of as a way to offer interpretation to visitors to one specific permanent gallery at the museum. Each of the objects in that gallery had metadata attached to it, and each also had links to other objects dispersed in other corners of the museum. In building the guide, it was a principle to share with users all of the information that they could need to understand objects and develop an experience of the gallery that would extend out into the museum as a whole. Broadly, if it was in the museum’s CMS, it would appear on the guide.
Again, speaking to guide users we found that they were struggling to cope with a surplus of data. Much of the information on the guide was misconstrued, ignored, or viewed as irrelevant. Whilst catalogue metadata duplicated information available in the physical gallery space, interpretative text tapped from the CMS was too long to be of use or interest to guide users absorbed in the objects displayed around them.
Users hoped that the guide would offer them something distinct, stories, an experience – not just more information. However, in the physical context of a visit, surrounded by artworks, chatting with friends and browsing with varying degrees of focus and direction, a vast, almost raw pool of data was too unstructured a resource to be truly helpful.
What both these examples show us, is that data in and for itself isn’t necessarily useful, and that the ways in which it is packaged and presented can have an enormous impact on whether it benefits, liberates or simply disengages audiences.
All this isn’t to contradict Ed Rodley’s argument. Aware of these issues, he warns against the perils of a skeumorphic approach to digital; creating online analogues of traditional museum structures like catalogues and displays. He even states explicitly that ‘it’s not about “putting the collection online”’. It’s also worth noting that the projects discussed here are rather different to those cited in Rodley’s blog. The smartphone guide in particular – in being designed for use in a specific gallery setting – is subject to a whole range of contextual challenges.
Nevertheless, there is a real risk that the present enthusiasm for moving online and opening up data could express itself in a wide-ranging, expensive and ultimately fruitless exercise in oversharing. Without understanding their audiences – what they need, and why they are offering them data – ‘promiscuous’ museums might well earn themselves a cold shoulder in return, rather than the warm embrace they desire.
Does the responsibility to make use of the Internet’s capacity for sharing data outweigh considerations about how or if that data will end up being used? Should it be left to audiences to decide how much data they want to see, and to filter that data in its manifold forms personally, whilst we simply open the doors to it?
Or is there value in showing restraint, hiding selected bits of information (like textual collections metadata) and foregrounding others in order to help users make simpler choices about what to use? Can we provoke better interventions and better making among cultural audiences if, instead of adopting an open data mantra and insisting on absolute, comprehensive accessibility, we place constraints on the data we decide to share? Is there more value in offering users curated, if limited, online experiences, and keeping the metadata behind closed doors?
Finally, speaking of value, is the cost of preparing and publicly hosting vast datasets justified by the often small numbers of users with the drive and technical savvy to understand and use it?