Skip to main content

Museums and the Web

An annual conference exploring the social, cultural, design, technological, economic, and organizational issues of culture, science and heritage on-line.

TourML: An Emerging Specification for Museum Mobile Experiences

Robert Stein, Indianapolis Museum of Art, and Nancy Proctor, Smithsonian Institution, USA

http://code.google.com/tap-tours

Abstract

In addition to launching a successful tour, how can museum professionals ensure that the content we create will outlive the technology and vendors we select today? A successful metadata standard for mobile tours would provide a number of advantages for museums. A mobile standard would allow museums to migrate the content and structure of a tour among the variety of tools available today and tools that may be available in the future. Similarly, a mobile standard would allow the same tour to be used on a number of current and future mobile devices. In addition, museums could choose authoring tools which best fit their staff and workflow needs while still deploying to whatever mobile device is best for their content – even if these tools are provided by different vendors. Perhaps most important, a successful mobile standard would ensure that museum content outlives any particular technology we use today.

Keywords: mobile content, content strategy, sustainability, specifications, content management systems, schemas

1. Sustainable Mobile Content

The 2010 Horizon Report for Museums highlights "mobiles" as one of two technology trends on the near-term horizon, stating that "Mobile technology has developed at a staggering pace over the last few years, and today affords many more opportunities for museums..." (Johnson, 2010) And the recent explosion of mobile technology as an important way for museums to distribute content is undeniable. Dozens of new tools and companies have emerged in the last 24 months to address the needs of museums that are planning, producing and launching new mobile experiences. A recent Pew Internet survey indicated that 40% of American adults had access to the Internet from a mobile phone in 2010 (Smith, 2010), and studies from Gartner suggest that by 2013, mobile phones will overtake PC's as the most common method for accessing the Internet worldwide (Gartner, 2010).

With 4 billion mobile phone subscribers worldwide, it's clear that mobile devices and content will be an important means of access for museum visitors today and in the future. More recent anecdotal evidence suggests that these trends have accelerated rather than subsided, and that an increasing number of museums are contemplating how they might deliver content via mobile devices. The Museums and Mobile Survey 2011 indicates that over half of large museums (annual attendance of more than 50,000) already have mobile experiences, and almost 70% of all museums say that their institution will "definitely" have in-house mobile content development within the next five years (Tallon, 2011).

While the promise of mobile technology seems self-evident, the more perplexing problem of how to manage the content museums produce for these devices is daunting. How will museums ensure that the mobile content they produce today will be available and accessible five years from now? When the mobile vendor selected to produce a museum's first mobile tour is no longer viable two years later, how can the museum retain and reuse this valuable content? Is there any hope of sharing content between museums with related collections, programs, and exhibitions? These questions are among many critical – and currently unaddressed – issues related to today's mobile content explosion.

The creation of common and open software tools for mobile tours and other mobile experiences – tools that can be shared and referenced by museums and vendors – would offer an effective way to answer many of these questions and would provide a mechanism to ensure that content created today could be easily re-purposed and adapted to future generations of mobile platforms. Building consensus among museums and vendors for a description of mobile content and building tools to aid in the adoption of this platform are necessary steps to achieve the goals of content sustainability and cross-collection sharing that museums desire. A successful solution of this kind would provide a way to integrate and interoperate among a number of content-creation systems and mobile interfaces, allowing both vendor-provided and custom-developed application software to use the same set of content.

The key element in achieving such a compatibility among mobile platforms is the existence of a specification – a common language – describing mobile content and the experiences they provide. In the summer of 2009, the Indianapolis Museum of Art (IMA) proposed a simple draft specification called TourML (tûrmoil) (Stein, 2009) which offered a working, but preliminary, example of what such a common language might look like.

If successful, a content specification like TourML would provide a number of important advantages for museums:

  • it would allow museums to migrate the content and structure of a tour or other common mobile experience among the variety of software tools and mobile devices available today and in the future
  • similarly, this specification would allow the same tour to be used on a number of different mobile devices at the same time
  • museums could choose authoring tools which best fit their staffing and workflow needs, integrating with whatever mobile device was best for their content, even if these tools were provided by different vendors.
  • In order to ensure a high level of input from the community, Robert Stein (IMA) and Nancy Proctor (Smithsonian Institution) organized two half-day workshops, inviting museum staff members, academics and software vendors to join a preliminary effort to formulate just such a standard. The first meeting was held at Tate Modern in conjunction with the Tate Handheld Conference in September 2010. The second meeting was hosted by the Museum Computer Network conference in Austin, Texas, in October 2010. A third workshop is scheduled to be held in conjunction with the 2011 Museums and the Web conference in Philadelphia in April 2011.

In all, nearly 100 members of the museum community will have played a significant role in these workshops, resulting in multiple subsequent revisions to the TourML specification. Notes from all meetings are available from the Museum Mobile Wiki, (Stein and Proctor, 2010) and the resulting TourML specification is available under an open-source license from the project's Google Code Website (Moad and Stein, 2009).

2. About TourML

TourML is an XMLSchema (http://www.w3.org/XML/Schema) that describes the structure, content and intended experience of a museum mobile tour by naming its constituent elements according to a commonly accepted or 'standard' language. Two terms are used here as shorthand: (i) 'tour' refers both to handheld museum tours and other mobile experiences that become common enough to be described by standardized terminology in the museum field; and (ii) 'instances' are individual tours or other mobile experiences. The primary goal of TourML is to make it easier for the different mobile devices and software systems that museums might use to be compatible with content that has been 'tagged' or described by this common language. The formal description of an individual tour or other mobile experience – an XML instance – is produced from Content Authoring Systems that support TourML. This XML instance documents the content and its constituent parts using standardized terminology as specified by TourML. This makes it easier for mobile platforms that are compatible with the TourML XMLSchema to import or ingest TourML-tagged content in an automated fashion, understand what role each constituent part plays in the overall mobile experience, and deploy it all correctly.

Tours and mobile experiences built using the TourML specification more easily support the sharing and migration of content between different mobile devices or software systems (perhaps because they are updating their handheld devices, or changing platform vendor), and are less likely to require rebuilding from scratch to do so. TourML's goal is to ensure that at least 80% of mobile content can be ported directly from one platform to another without requiring manual re-assembly by museum or other staff. If the platforms and the content all use the same standardized 'language', this kind of compatibility and even interoperability becomes possible.

There are 3 core elements of any TourML instance, in addition to supporting and 'framing' elements for the tour:

  1. the media 'assets' for the tour
  2. the 'stops' in which assets are experienced
  3. the 'connections' for moving from one stop to another

Each element may have a number of attributes described by the specification that can provide more information on how the content from the TourML document should be displayed or played. For example, authors can specify a code for a stop that users must enter into a keypad on the handheld screen in order to play that stop. Media assets may be linked to a geographic position on a map, and tour authors may indicate which images in a slide show should be played in a particular order and for how long. The semantics and terminology required to describe these kinds of display characteristics of a tour is a topic of ongoing discussion within the community working on the TourML specification. It should be expected that a more detailed refinement of these principles will be an outcome of the next few community design workshops for TourML.

Asset

The term, 'Asset' is used in the TourML specification to refer to single media entity, e.g. images, videos, audio tracks, and text. There are several additional types of Assets defined in the specification: VideoAssets, ImageAssets, and TextAssets are a few examples. An Asset is the only required core element for a valid TourML document. A TourML instance must include at least one asset with each tour, and should describe all the Assets associated with that tour. TourML does not require that a tour contain any Stops or Connections.

Stop

'Stops' are defined as a collection of Assets that are meant to be experienced together. This might be limited to a single video, or it could be composed of a set of images for a slideshow with a background audio track and a header image. The Stop refers to its related Assets and contains a set of instructions that describes how the assets should be played and used together. The Stop element allows the tour's author to reuse the same assets in different ways among multiple stops by specifying different playback and display instructions in each stop. This permits each asset to appear many times in the tour without requiring multiple copies of it in the final product (app or mobile website).

Connection

A 'Connection' is a one-way relationship used to define navigation or flow from one Stop to another. Connections describe how a user can move between Stops. A 'two-way' relationship or the ability to move back and forth between two Stops should be represented in TourML as two separate Connections, e.g. one from Stop A to Stop B, and a second from Stop B to Stop A. By defining Connections, authors are building a directed acyclic graph (http://en.wikipedia.org/wiki/Directed_acyclic_graph) or tree that represents the way in which a tour is intended to be experienced. For example, a 'Stop' that asked users if they would like to learn more about a topic they just watched a video about, would use a 'Connection' to another 'Stop' that contains 'Assets' about that deeper content.

3. A TourML Example

The following code is an example of what a TourML XML document looks like. This XML can be produced easily from many common Web software packages, and is consistent with the accepted practices of data and Web professionals.

An Example of the Title and Description Element of a Tour

<tourml:TitleDescription xml:lang='en'>
<tourml:Title>A sample museum tour</tourml:Title>
<tourml:Description>
This tour shows you the highlights of our museum.
</tourml:Description>
</tourml:TitleDescription>

Notice that the XML description elements use syntax that is similar to HTML, and is fairly 'readable' as plain English. This content can be parsed and transformed by popular Web tools to display and present a tour in any number of different page designs.

An Example of a TourML Asset

<tourml:Asset tourml:id='img-1' xsi:type=
tourml:ImageAssetType tourml:width='400'
tourml:height='300'>
<tourml:Source tourml:uri='./image1.png'
tourml:format='image/png'/>
</tourml:Asset>

Notice that the Asset description tells the software running on the mobile device, the "mobile client", what the defined width and height of the image being used should be. The Source element embedded in the ImageAsset above describes the name and location of the image file. An Asset may have multiple Sources; for example, to provide different language versions of the same ImageAsset.

An Example of a TourML Stop

<tourml:Stop tourml:id='stop-2' tourml:code='200'>
<tourml:TitleDescription xml:lang='en'>
<tourml:Title>
Stop 2: The Director's Favorite Painting
</tourml:Title>
</tourml:TitleDescription>
<tourml:AssetRef tourml:id='img-1'
tourml:usage='primary' />
<tourml:AssetRef tourml:id='icon-1'
tourml:usage='icon' />
</tourml:Stop>

This Stop description demonstrates how to incorporate an image asset into a tour stop. First, the stop, identified as "stop 2", is given a "code" of 200, which would allow a user of the mobile tour to navigate to this stop by entering the code in some way. The title of the stop is provided in English, but may also be translated into as many languages as is desired. Two Assets are referenced in this stop: one asset ("img-1") is intended to be the primary image used by the stop, and the second ("icon-1") is to be used as the icon representation of the stop in tour navigation.

An Example of a TourML Connection

<tourml:Connection tourml:srcid='stop-1'
tourml:destid='stop-3'/>
<tourml:Connection tourml:srcid='stop-1'
tourml:destid='stop-4'/>

Connections let authors of mobile tours describe the intended sequence and connections between stops that should be experienced in some order. This example demonstrates how one stop might provide links to other tour stops that should be experienced next. In this case, two stops, "stop-3" and "stop-4", are offered to the users as choices for how they might want to continue after "stop-1."

A more in-depth discussion of the TourML specification, as well as a full list of details and attributes available with TourML, is available from the open-source Google Code project for TAP (http://code.google.com/p/tap-tours).

4. Standards and Partnerships

As a result of the mobile explosion described above, museums are investing significant financial resources on an annual basis to produce new content for mobile consumption. The majority of this content will be just as relevant ten years from now as it is today. However, chances are good that the mobile platforms for which this content is designed will become obsolete far sooner. This situation recalls similar challenges of a mismatch between enduring content and technical change that has become all-too-familiar for museums. Many kiosk and interactive Web applications developed just a few years ago have proved to be very difficult to maintain over time, and repurposing content from these applications is often impossible. This project evolved in response to the lessons learned from those failures and seeks to avoid similar mistakes as museums embark on many new content efforts for mobile devices.

In the past, museums have depended on commercial partners to ensure the preservation and sustainability of their interpretive content. Vendors, however, are likely to be focused on the more immediate concerns of creating a strong user experience with features that distinguish them from their competitors. The nature of commercial competition often makes it difficult for vendors to lead the charge for portability and preservation. In addition, without a consensus in the community around content definitions, tools that succeed in ensuring portability and preservation are all but impossible. Previous successes in defining content specifications and standards, such as those supporting object collection metadata (LIDO, CDWA Lite, and Dublin Core), have been led by the content producers. In short, the onus for consensus and collaboration around standards falls squarely on the shoulders of the museum community. Partnering with commercial vendors to encourage adoption of these standards, thereby ensuring that these tools help to enhance the vendors' business, is the only sure way to secure the viability of such an effort.

Through the ratification of the TourML specification, museums can spearhead the adoption of this specification by the vendor community. In explorations of the feasibility of this effort, many commercial vendors have been very receptive to TourML as a potential specification for mobile content, and have been actively engaged in the mobile workshops that have been held. Some vendors have already integrated early support for the draft TourML specification into their products. Ideally, a healthy relationship with and involvement of the vendor community in this process will result in a viable and sustainable specification that can truly realize the benefits we seek.

5. Epilogue: Next Steps Towards a Mobile Toolkit for Museums

In this project we propose that the TourML specification become widely adopted and integrated by museums into a wide range of mobile platforms and systems. Long-term project goals include the creation of a mobile toolkit of software that can be used by museums without requiring special technical knowledge on the part of museum staff, and that can be modified and customized to integrate with many in-house content authoring systems.

In addition to promoting the creation of a mobile content standard, the project believes that it is important for museums to have access to simple and easy-to-use content authoring tools that facilitate the creation of mobile content. Thus, a mobile content management system called TAP, which is also available under an open-source license, was created in 2009 (Moad and Stein, 2009).

TAP provides a way for non-technical staff inside the museum to assemble a wide range of mobile experiences without needing to know any of the underlying details of Web or mobile technology. In addition to supporting the creation of content, TAP, in its current state, also provides user interfaces for Web-based mobile tours and simple native applications for an iPod-based tour.

The project believes that freely available tools and standards are important to the museum community because they promote the adoption of best practices, facilitate collaboration, and encourage avenues for future content sharing. IMA's initial work with the TAP authoring tool includes support for an early version of the TourML specification and offers a functional, but incomplete, proof-of-concept, demonstrating how such a system might work. Since its release, the TAP system has been successfully adopted and deployed by a few independent developers as well as other major museums, and has served as inspiration and example code for a number of other implementations. Recent applications and tours launched by the Museum of Fine Arts, Boston, The Minnesota Historical Society, and the Dallas Museum of Art have drawn from the work of the project, and use elements of TAP and TourML, either in whole or in part. The project's commitment to collaboration will allow the work of those museums to be integrated back into the TAP toolkit, improving it for a future software release.

Though the initial efforts on TourML and TAP show great promise, there is still an enormous need for investment in the project to ensure that these resources can be effectively put to use by the community of museums. We hope that by beginning with an open and collaborative standards initiative by and for museums, vendors, and all other stakeholders in the museum mobile community, we can help foster a community effort that will more efficiently and effectively develop a mobile toolkit to respond to the fast-changing demands of mobile platforms today.

6. References

Gartner Inc.. (2010). Gartner Highlights Key Predictions for IT Organizations and Users in 2010 and Beyond. January 13, 2010. Consulted January 27, 2011. http://www.gartner.com/it/page.jsp?id=1278413

Johnson, Laurence F., Alan Levine, Rachel S. Smith and Holly Witchy (2010). Horizon Report: Museum Edition. Austin, TX: The New Media Consortium, 2010. http://www.nmc.org/pdf/2010-Horizon-Report-Museum.pdf

Moad, Charles W. & Robert J.Stein (2009). TAP-Tours, Google Code Project Site. 2009. Consulted January 27, 2011. http://code.google.com/p/tap-tours/

Smith, Aaron (2010). Pew Internet & American Life: Mobile Access 2010. July 7, 2010. Consulted January 27, 2011. http://www.pewinternet.org/Reports/2010/Mobile-Access-2010.aspx

Stein, Robert J. (2009). TourML (In Progress). 2009. Consulted January 27, 2011. http://wiki.museummobile.info/museums-to-go/products-services/tourml

Stein, Robert J., and Nancy Proctor (2010). Museum Mobile Wiki: Standards. 2010. Consulted January 27, 2011. http://wiki.museummobile.info/standards

Tallon, Loic (2011). Museums & Mobile Survey 2011. January 2011. Consulted January 27, 2011. http://www.museums-mobile.org/survey/

Cite as:

Stein, R., and N. Proctor, TourML: An Emerging Specification for Museum Mobile Experiences. In J. Trant and D. Bearman (eds). Museums and the Web 2011: Proceedings. Toronto: Archives & Museum Informatics. Published March 31, 2011. Consulted http://conference.archimuse.com/mw2011/papers/tourml_an_emerging_specification_for_museum_mobile