koven j. smith dot com

This is the last in a series of posts (and sorry for the delay in getting this one up; sickness followed by the Museum Computer Network conference prevented me from getting this up sooner). Part One, with a brief introduction, is here, Part Two is here, and Part Three is here. You can read all four parts together here.

New means of producing content are only part of the equation; to claim attention from audiences both new and traditional, museums will need to experiment with different kinds of engagement in both the online and physical spaces. Because content-plus-reputation is no longer a compelling enough reason for garnering attention, museums will need to focus on types of engagements that are not easily copied. Tech writer Kevin Kelly refers to these as “generative assets,” which he breaks down into eight categories, four of which are critical for museums at this juncture in their history: immediacy, personalization, accessibility, and findability.

Immediacy

Museums need to demonstrate value by providing up-to-the-minute content and information. Immediacy here could take a number of different forms, depending on the medium and the situation. Immediacy for many users might resemble something like what TechCrunch founder Michael Arrington termed “process journalism,” in which a story is chronicled in near real-time as it unfolds, before all the facts are completely known (which, by its very nature, necessitates corrections and clarifications later on). A great recent example of this in the museum world would be Brooklyn Museum’s real-time chronicling of the CT scanning of four of their Egyptian mummies–between regular posts on the Brooklyn Museum blog and literally up-to-the-minute Twitter updates, users were made to feel that they were witnessing a story unfold right before their very eyes. The fact that the museum was willing to admit that it didn’t (yet) have all the facts was somewhat arresting, and meant that readers of the Museum’s blog felt that they were part of something developing. A simple summary of this process, delivered via a few paragraphs after all the facts were in, would not have been nearly as fascinating.

Another interesting example of immediacy would be NPRbackstory, which was created by Public Interactive, a division of National Public Radio. NPRbackstory is a service that combs Google’s “Hot Trends” data for trending search topics, searches for those topics in NPR’s own news archive. If a story in the archive matches the trending topic, a link to the story is posted in NPRbackstory’s Twitter feed. What is interesting about NPRbackstory is that the process is completely automatic; the service runs without any intervention from human beings at all, and yet it provides an immediate value by providing context and background to an emerging topic. With their vast content reserves, this is the kind of approach that museums could very easily take.

Personalization

Personalization involves tailoring content and content delivery methods based on user characteristics or selections. Personalization will continue to be the most difficult generative asset for museums to work with until their content repositories are finally deep and diverse enough to truly account for user preference.

A good example of personalization comes (again) from NPR. NPR introduced, at the end of 2008, a means by which listeners could create their own podcast streams based on preferences they select. Because NPR’s podcast archives are both deep and well-catalogued, it is possible for users to create not only personalized podcasts based on categories, but even on keywords. So, for instance, a user create a podcast feed that is updated any time an NPR-affiliated program runs an episode in which the term “MoMA” is featured. The ability to tailor this content specifically to a user’s preferences gives that user a powerful incentive to return to NPR’s archives for more content.

Accessibility

Accessibility refers to the ability to access a resource when and where it is needed. Because much digital content is free, there becomes less and less reason to physically house this material on one’s own desktop, laptop, or mobile device. Instead, many content creators and aggregators are pursuing a strategy in which all content is stored remotely “in the cloud” such that a user can access that content from any device. The most exciting recent development in this area is Spotify, a music player that is similar to iTunes, except that no content is housed on the local device–it is accessed entirely via an Internet connection. What this means is that a user’s mp3 library is always available anywhere with an Internet connection, regardless of the device.

Museums have already begun to make tentative steps in this direction with “my virtual gallery” features (a good example being the Met’s “My Met Gallery“), in which users are able to create a personal account to which they may save collection objects of interest for later perusal. The content in these personal accounts is stored entirely on the museum’s own servers, so theoretically, at least, the content is available wherever the user is able to access the museum’s Web site. This is an excellent start, and this idea should be expanded to include content outside collections such that a user, whether engaging with the museum’s information online or via a mobile device, is able to access contextual content “just in time.”

Findability

Findability is key to asserting value in the attention economy. Resources that are not easily findable may as well not exist, no matter how interesting or vital they may be. It is interesting to note in the last several years that although museums have continued to publish new information resources as they always have, albeit now digitally, the real response from the community only appears when museums actually create new means of accessing those resources that the community takes notice.

Increasing the findability of resources on the Web at least partially rests on simply having more content available, and ensuring that that content is identified such that it can be properly indexed by search engines. But this alone is not enough. Any findability strategy should be aimed at not only making resources easily obtainable, but also at ensuring that those resources are available when they are needed. Delivering information “just in time” has the net effect of increasing what Peter Samis refers to as “Visual Velcro,” or the likelihood that a user will spend more time with a given content resource.

Improving findability may also mean accepting that many of the most interesting experiments with museums’ information may happen downstream, outside their control. A good example of this recently would be with the Brooklyn Museum’s release of their collections API. Being one of the first institutions to make its collections available in this manner, there were no use cases out there demonstrating what value doing this might provide. But the Museum’s community itself provided that when a developer in Brooklyn (unasked, I might add), used Brooklyn’s collections API to build an iPhone application that would allow users to browse its collections.

But findability could have significant meaning in the gallery space as well. If we still seek to deliver information “just in time,” it is important that this information be available in the physical space as well. Lightweight finding protocols, like the QR (Quick Response) codes recently printed on wall labels at the Mattress Factory in Pittsburgh, could theoretically be resolved to any Web-available resource. A museum without the resources to put its own content on a device could easily seek out high-quality content on the Web and use QR codes to make that content available in the gallery via a visitor’s mobile device. Doing this helps to foster the increased “stickiness” between visitors and objects that Peter Samis refers to (because information is being provided when it is needed), while also emphasizing a museum’s role as a “distributor of attention.”

Conclusion

It is time for museums to finally begin transforming themselves from “buildings with Web sites” into different types of institutions altogether, in which the physical visit is but one of many possible engagements. This transformation will not be easy, as it will involve recognizing the truly disruptive impact that the Web has on museums’ traditional modes of information delivery. Competing in this environment means going beyond simply finding new ways of presenting content digitally, but also learning how to properly distribute attention to that content. By focusing on the “generative assets” of findability, personalization, accessibility, and immediacy, museums have a way forward, if they choose to follow this path.

Koven Museum api, arrington, brooklyn museum, digital domain, kevin kelly, Museum, npr, npr backstory, podcast, spotify, techcrunch

Koven Performances MCN, Museum

Koven Performances MCN, Museum

Koven Performances MCN, Museum

This post is Part Three of a series of posts. Part One, with a brief introduction, is here. Part Two is here.

By choosing to assume that the audience for online engagement is the same as that for traditional in-gallery engagement, museums are failing to nurture and develop new audiences. Instead, museums continue to produce new content based on (potentially) flawed assumptions of what their audiences might want. These flawed assumptions cause museums to spend a significant amount of time worrying about issues regarding content ownership and authority that are–at best–less important, or–at worst–completely irrelevant for online visitors, and ignoring issues of findability that are critical.

This is nowhere more apparent than on the typical museum Web site. The average museum site assumes a protracted engagement with the visitor in which the a deliberate choice has been made by the visitor to visit the site and look for information. This type of engagement is clearly modeled on a physical visit, in which the visitor enters through the front door of the building and is “captive” for a certain length of time. In this scenario, the visitor has already sought the specific museum Web site out, likely based on the museum’s reputation, and is willing to accept even unattributed content as all coming backed by the “full faith and credit” of that institution.

Some visitors may indeed seek this form of engagement from museum Web sites. The problem is that the average museum Web site gears most of its content entirely in the pursuit of this one type of engagement. With more effort devoted to research (“plans to learn”) than to production (“plans to execute”), museums may find that there are other niche types of engagements that individually represent smaller numbers of visitors, but collectively represent a significantly larger number. This is what author Chris Anderson refers to as “the Long Tail.” In the old days, the cost of producing print or in-gallery materials that would appeal to these niche audiences was simply too high to warrant even considering it. Digital production, however, enables museums to publish materials for as easily for these audiences as it does for traditional audiences. Adding up all of these possible niche markets makes for a larger number of total interactions than do the hits.

There is also the possibility that what was always assumed by the museum to be a niche audience in fact turns out to be an entirely new, previously untapped market. In fact, most disruptive technologies begin as niche markets and then evolve into something much larger. This is to museums’ advantage now that producing content is now a low- or no-cost proposition. Rather than spending years and thousands of dollars producing “perfect” publications that may never find an audience, museums can instead begin to put out smaller bits of content first to see what “sticks.” If that content finds an audience (that is, if the material begins to be linked to, referenced, and read), then the museum could decide to produce more content on that topic. The key here is that decisions around content production should be based on actual returned data (website hits, incoming links, etc.) rather than assumptions about what the largest audience wants. In this way, emerging and as-yet-unknown audiences can be turned into assets before they have a chance to look somewhere else.

Producing materials for niche audiences and adopting a research-driven content strategy means contravening many long-held production practices in museums. Most critically museums should forgo the concept of waiting for perfection before publication. Perfection is a standard far more suited for the print medium than today’s digital domain. By the time a museum has thoroughly perfected a resource, that potential audience may have already moved on and found the information elsewhere. Maxwell Anderson, director of the Indianapolis Museum of Art, recently made a similar case for immediacy over perfection:

“I arrived [at the IMA] and heard, as is so often the case, the mantra that we’re only going to put stuff online when we’ve done data clean up…Actually, what we’re going to do is we’re going to put everything online now and see how much we have to clean up. And that seems to be working better…than holding back.”

In trying to avoid “holding back” while simultaneously attempting to find and exploit new audiences, museums will have to re-configure their content-production strategies to be significantly more flexible and responsive. In the main, this will involve the removal of the traditional editorial process in favor of more direct and ongoing communication. In this model, the publication of a given resource is the beginning of a process, rather than its endpoint. Some approaches that many museums are already using successfully include:

• Allow staff to speak to visitors directly. As many museums with blogs are finding, allowing staff to speak out directly about what interests them is a relatively painless way of quickly creating volumes of interesting, colorful content. These staff postings are inherently niche-based, tending to focus on tiny area of museum practice, but often build dedicated, loyal audiences. Nina K. Simon asserts that being more overt about content authorship can also have the side benefit of increasing trust, by in essence demonstrating an author’s willingness to engage in a conversation about content he or she created with his or her community.
• Use content that already exists. Not all content has to be built from scratch. Plenty of information is already available in digital format in museums, but many museums still withhold this information until it is deemed suitable for public consumption. What Wikipedia and similar resources have taught us is that the public is remarkably tolerant of mistakes when the information is copious and findable. Put information out as soon as it is created.
• Worry less about completely owning all of your content. Museums should not be afraid to reference materials not produced by them. Museums can still provide valuable context, and acquire value over time as good pointers to interesting information. A museum that is willing to own up to incomplete knowledge, and ask its own communities to fill in those gaps, is a museum that (paradoxically, when viewed in the context of earlier paradigms) is increasing trust with its community and ensuring that it will be a “first source” of information for that community.
• Digitize your archives. This is probably the least attractive option for most museums, as the process of converting archival assets from analogue to digital carries with a relatively high price tag. But because much of this information already exists, digitization can enable the creation of large amounts of “new” content without (again) having to create content from scratch. And, like all digital production, the cost of digitization is only decreasing.

Objection to pursuing any of these strategies in museums typically takes one of two forms. The first is the concern that this kind of content represents a significant decrease in quality, and the second is that creating a too-engaging online experience will cause a drop in visitorship to the physical site. Both of these arguments represent an incomplete understanding of what disruptive technologies mean in terms of creating new audiences.

For these new audiences, there is no quantifiable drop in quality with these new production methods. With more rapid and personal content deployment, museums are actually providing more value for an audience that responds to availability and findability more than perfection. For a visitor hungry for information that can be found nowhere else, any information, even incomplete information, is better than none at all. The ability to engage directly with a curator, conservator, or educator might be far more important to this audience than comprehensiveness. Museums have to ask themselves whether by fretting over being completely authoritative if they are stymieing their content-production efforts merely to satisfy the needs of a small group of scholars. As Radiohead guitarist Jonny Greenwood said about high-quality audio formats like FLAC: “…if you even know what one of those is, and have strong opinions on them, you’re already lost to the world of high fidelity and have probably spent far too much money on your speaker stands.”

And it is highly unlikely that producing more engaging online experiences will “cannibalize” existing visitors. It is far more likely that more interesting online presences will in fact help museums to find entirely new audiences that had heretofore not had an interest in visiting at all. Sebastian Chan, Head of Digital, Social & Emerging Technologies at the Powerhouse Museum found this to be true when the Powerhouse posted images of its Tyrell photographs collection to the Flickr Commons. Powerhouse found that the images posted to Flickr received more page views in the first four weeks of availability than they had for the entire previous year on the Museum’s own Web site, and that the museum was receiving licensing requests from entirely new entities that had never contacted them before.

This is an instructive case because the Powerhouse could have just as easily not have made these images so easily available, and instead put a few highlights on their Web site, strongly urging visitors to come to the Museum to see the collection personally. Had the Powerhouse done this, these new audiences would have been far less likely to have found this collection at all. The “cannibalization” would have been reversed–an over-emphasis on the museum’s physical presence would have prevented interesting and engaging content from being made available for new audiences to discover.

Part Four should be up later this week.

Koven Museum Chris Anderson, digital domain, flickr commons, Long Tail, maxwell anderson, Museum, nina simon, Powerhouse Museum, sebchan