Denver, April 13-17 2010
A slightly-edited version of my institutional reportback
One of the gratifying things about Museums and the Web was finding that as well as data enthusiasts, digital managers and social media experts, there were a good number of curators, librarians and other professions at the conference. It seemed like proof to me that the web is becoming ever-more intimate with the everyday practice of museums, and that the boundaries between old museum professions and new are not so much being broken down as producing new hybrids: geeks who are intimately familiar with collections, and curators for whom web 2.0 is second nature.
The conference was multi-stranded, and many topics and ideas of interest were repeated; therefore, rather than sequentially describe every session I went to, I’ve tried here to group things together, bringing out highlights and connections.
Aaron Cope, a former Flickr developer discussed the introduction last year of Flickr galleries, a feature which allows users to select and annotate others’ photos. A restriction on the number of photos was deliberate, forcing users into making small, considered selections rather than longer lists, and also moving users away from promoting their own photos and towards the intellectual understanding of others’ work. Aaron described this as the discovery of a ‘curatorial muscle’: without fetishising the word ‘curation’ or suggesting that the role of museum curators has been supplanted by the crowd, everyday web users are creating the kind of connections and interpretations that can’t be done by machines.
Joe Dalton from the New York Public Library discussed their Flickr commons project: in adding a large amount of material to the Flickr commons, how to merge existing subject headings into Flickr tags seemed problematic. The answer was to break down subject headings by delimiters into tags that made sense, update the archaic language of some subject headings, and use only what made sense.
Natasha Waterson of the National Maritime Museum presented their use of Flickr as a platform for the Astronomy Photographer of the Year competition. Though Flickr offers a ready-made place to submit photographs, making the costly build of a separate submission mechanism unnecessary, it also has its own rules, oriented more towards community than competition, which have to be respected; working directly with Flickr itself was also helpful. The quality of the content was very high, and an important lesson learned was that niche participation (a limited number of people engaging in astrophotography) can have popular appeal (many people fascinated by pictures of the stars). The reward of being on display at Greenwich itself was a pull, and the physical displays also included comments from Flickr users.
Paula Bray from Powerhouse and Ryan Donohue from George Eastman House presented Common Ground, a global day of meetups at cultural heritage institutions involving Flickr Commons users who had been active in enriching the Flickr commons. Slideshows were presented and projected, curated by Flickr users rather than the institutions.
Nate Solas from the Walker Art Centre showed how on the Arts Connect Ed site, research into what people searched for showed that they were effectively using search as a roundabout way of browsing the collection. The introduction of a browsable interface to the collections and facet browsing not only increased the specificity of what people were searching for but also made the site much more indexable by search engines; a reminder also that search starts outside the site.
Martha de Vit and Edith Schreurs of the Van Gogh Museum in Amsterdam presented the mobile iPhone application they built in collaboration with Antenna Audio around the publication and exhibition of Van Gogh’s letters. They found it gained a wide an international audience, with downloads peaking around the touring exhibition visiting new locations. Contrary to assumptions, multimedia applications didn’t interfere with appreciation of the actual exhibition; but it’s also important to remember that building mobile ‘gadgets’ is not a simple process.
Projects & evaluation
Richard Morgan from the V&A presented the V&A’s Search the Collections project. Drawing lessons from the overall project experience, he recommended making new projects like projects you’re already good at, being realistic to the point of pessimism, and planning for dependencies. Search the Collections was broken down into smaller sub projects (eg, IT building virtual servers) and built iteratively (in an almost Agile fashion), bringing working prototypes to each meeting of the project board. Richard also made the point that it’s important to distinguish between creating a thing, and creating an organisational capacity: a thing can be outsourced more easily, while a capacity is much more important in the long term.
Allegra Burnette of MOMA, Dana Mitroff Silvers of SFMOMA and Charlotte Sexton of the National Gallery made a three-way presentation on websites post-launch. The post-launch phase, they argued, is as critical as the development phase, and should be planned for. A website redesign shouldn’t be considered as a standalone web project but as an initiative of the entire organisation, because important decisions about audiences have to be made; a relaunch may even highlight the lack of a shared vision for the organisation.
The combined launch of a new site and a new CMS will not deliver an efficiency miracle: indeed, everyone on the web team could be working harder after project staff are lost. It’s tempting to make the pitch for the budget to implement a new CMS on this basis, but along with separating content from presentation, next-generation CMSes demand more data and more complex metadata like tags. A new CMS is a lot for non-technical staff to learn, and is not going to make anyone engage with the website by itself, and so iterative rollouts and simplified interfaces for infrequent users should be applied. The role of the web team in providing systematic oversight and documentation does not disappear either.
A redesign won’t necessarily provide more traffic, either: content and the programme drive traffic. Analytics can be time-consuming but important. Post-launch user-testing is essential: you should have the capacity to test and tweak the final design. There’s a lot of invisible admin in bug and feature tracking, and post-launch budgets are typically 10% of the development costs. Ultimately the launch of a new site is only a line drawn in the sand of a larger strategic process, but when the camaraderie of the launch build-up dissipates, disillusion can set in.
Liam Wyatt, Vice-President of Wikimedia Australia and a small group of volunteer wikipedians came to the conference for an invitation-only workshop to thrash out some the tensions between cultural heritage institutions and Wikipedia. The results of the workshops, with opportunity for further discussion, were disseminated through an conference session and discussed in an un conference session (an unconference involves self-organised mini-sessions and discussion on shared topics of interest).
Rather than a homogeneous entity, Wikipedia is made up of overlapping communities. The Wikimedia foundation owns the framework, and volunteers form national chapters and project groups. One way for cultural heritage institutions to engage with Wikipedia is to consider it a community rather than a resource and look for local groups to engage with. Volunteer wikipedians aren’t that different from museum volunteers. Adding a note to the talk page of an article, finding an appropriate subject-based project, or a local editor to form a relationship with are all ways to begin improving the interface between Wikipedia and the institution.
Museums are primarily interested in objects, while Wikipedia is all about subjects; this is one of the reasons why museum content isn’t an immediately perfect fit for Wikipedia (though with the correct licensing and accompanying notice, museums’ own texts can be incorporated directly into Wikipedia articles). Liam’s desire to forge better relationships between museums and Wikipedia comes from a recognition of some content deficits in the encyclopaedia itself. As he put it ‘Wikipedia is good at Pokemon, and not so good at what museums do’. Wikipedians are primarily interested in content, and museums should look to meaningfully enhance articles, rather than garnish links to their institution. Deep links to objects and their interpretations as citations are desirable: they increase the quality and usefulness of Wikipedia itself. Adding links to the ‘notes’ section as footnotes, for instance, is preferable to adding links to the ‘external links’ section at the bottom of the page.
This special Wikipedia page, created by Liam, discussing meaningful and useful ways to work between Wikipedia and the cultural heritage sector is worth reading:
Social media & community
Wayne La Bar presented a social media site for the Liberty Science Centre built on the Ning platform (there was some discussion of Ning during the conference, as they have just announced a switch from a free to a paid-for model) for a forthcoming cooking exhibition at the liberty science centre. The aim was to involve the public in planning the content of the exhibition, and some users were successfully integrated into the exhibition planning team
Mia Ridge from the Science Museum presented their Cosmic Collections project, a data mash-up competition challenging coders to build new websites based on a collections data API. Though opening up data doesn’t always mean that people will use it, the competition was successful and produced sites that are still live. The possibility that no-one will enter always exists, said Mia: acknowledge that, plan for it, and then go ahead.
Jeff Gates from the Smithsonian discussed the opportunism of social media. The Smithsonian’s long-running email enquiry service, Joan of Art, seemed to be a rights-bound silo of inaccessible material until Joan of Art herself started tweeting, and the best of her answers were re-posted in a blog.
Ailsa Barry of the Natural History Museum presented Nature Plus, a system to tie together physical and virtual offers at the new Darwin Centre. Given a barcoded ticket, visitors can pause at several interactives, manipulate and collect information, which they can then retrieve from the website and use to set up a personal page and profile. This was a project in which many others inside the NHM also saw opportunities for marketing, development and education. The system has 10.5k users, with a high level of loyalty.
Erika Dicker from Powerhouse discussed the museum’s curators’ blogs, set up by the Powerhouse web team, but with all content controlled by curators themselves. She also surveyed museum curators and found that 60% of them were using one form or another of social media. Some curators use blogging as a means of raising their personal profile, others are uncomfortable with the medium, still others find obstacles to it. It’s also worth asking whether curators have access to the right tools to make the most of opportunities.
Gail Durbin from the V&A ran a workshop around the V&A’s intention to make a transition from contributions to their websites to a community around the museum. Many user-generated content projects involve a contribution, after which there is a dead-end. The V&A is looking to replace its 30-odd different sign-ons with a single sign on across all sites, and the ability for users to create profiles for themselves. Gail noted that Flickr, a photographers’ site, allows you to say where you’re from and your relationship status, information not immediately germane to photography, but which might begin to foster a sense of community; workshop participants were asked to come up with a comparable, unobvious question to ask V&A users to add to their profile. Then, looking at two things close to the V&A: Body Art, and drawings and sketches made in the galleries, we were asked to consider how we might best build a community around the activity or subject.
Though the conference was mostly about news ways of doing things, rather than new ways of configuring computers, there wasn’t a lot of talk about platforms and CMSes. Open source CMS Drupal was mentioned more than once (some of the conference attendees went on to Drupalcon in San Francisco) as well as the Python framework Django, and the PHP framework Symfony. I had an interesting chat with the developers of GLAMkit (http://www.glamkit.com/) a Django-based framework developed specifically for the needs of galleries, libraries, archives and museums.
The Brits didn’t sweep the board at the Conference’s Best of the Web awards, but they did much better than an average Oscars. Royal Observatory’s Solar Stormwatch won the Innovation award; Culture24 won the Long-lived award; the V&A’s Search the Collections won the Research award; and Dulwich Picture Gallery’s Dulwich OnView won the Small award
Full awards listing: http://conference.archimuse.com/forum/congratulations_mw2010_best_web_winners