Two industry experts presented on DAM and taxonomies at the latest meetup for people who take content seriously. In this special one-off event, participants got to experience an eclectic corner of London in an even more eclectic venue.
About the venue: Rotherhithe Picture Research Library
The Rotherhithe Picture Research Library is an extensive collection of visual media – photos, drawings, paintings, maps, video, and even costumes – that media producers use to study eras and areas when conducting background research for their films and plays.
What makes this venue and collection unique is the approach that one Managing Director, Olivier Stockman, takes in managing the collection: the index is completely analog.
Stockman explains his philosophy behind the decision. He wanted to create an environment of discovery. The idea that someone would do an online search and settle on a single answer belies the richness of the material.
Researching a topic for a film, for example, could involve looking at streets and architecture, typical household items or clothing from that period, or typical work and holiday activities. Call this way of research the equivalent to the slow food movement, where one is expected to take the time to savour and digest what’s before you. But more on that later.
Digital asset management (DAM): Theresa Regli
The first speaker was Theresa Regli, one of the top DAM (Digital Asset Management) consultants in the industry and a new transplant to London. Hers was more of a conversation than a formal presentation, where she answered questions about how DAM systems work, and some of the challenges around managing digital assets. Here are some highlights.
About DAM systems
DAM systems are, in some respects, the new kid on the block, though the functionality is growing in sophistication quite rapidly. The need to manage digital assets grew out of organisations such as museums and corporations having large numbers of images that weren’t being stored in ways that were useful for finding and using them later on.
Digital assets aren’t just images
The notion of digital assets are expanding from static images, such as drawings and photos, to items such as video, 3D renderings, and other properties that contribute to virtual reality environments. This could range from gaming companies looking to manage all of the minutiae that gets combined in a multitude of ways during the development of games, to multinational corporations managing virtual reality apps that let you see furniture at home before you buy.
- The IKEA virtual reality app that shows their furnishings in your home.
Connecting digital assets with physical assets
In the more interesting projects that Regli has worked on, there has been a need to connect the digital assets with physical ones. For example, one multinational had an extensive collection of physical objects from their century-old corporate history, and what was displayed online had to be keyed to its physical location in a warehouse.
Categorisation and data modelling
Whether you’re a company trying to organise your website images or an organisation with complex digital asset needs, Regli warned of the dangers of thinking that a technology will fix what is essentially a categorisation problem. Before pouring data into a DAM system, the organisation must do the up-front work of thinking through the business problem to be solved, analysing the assets, and then creating a categorisation system – a taxonomy or ontology – that forms the foundation of the data modelling to be done within the system. Regli says it may be a hard conversation, but it’s a disservice not to tell clients that buying the system without having the right complement of people to do the preliminary and ongoing work will be wasted, an expensive exercise.
- The DAM Maturity Model (PDF)
Treasure hunting in analog
With Regli’s words of wisdom ringing in our ears, participants engaged in a treasure hunt through the stacks of the library. Regli provided a handful of topics to find, and participants could choose which topic to locate. The familiarisation on where the stacks were and how to look through them went fairly quickly, and several people chose “hops” as their topic. Hops played – and to an extent, still plays – an important part in the British economy.
Soon the oversized, loosely-bound packets of photos appeared on the desks. One of the photos found was of families picking hops in Kent. This discovery led to a discussion about how families who wanted to take a vacation, but really couldn’t afford one, would go to Kent for a week and pick hops. It seems that Stockman’s discovery method proved itself that evening.
Creating successful taxonomies: Andreas Blumauer
Wrapping up the triple-bill of DAM activities, Andreas Blumauer discussed the organisation at the heart of any digital management: taxonomies. Organising content for presentation is not as simple as it seems. Presentation needs to happen in context, and the relationships between entities are what provides enough context to give us a better understanding of a topic. Indeed, Blumauer introduced himself using an example of relationship categorisation to demonstrate the principles.
Using recognised standards like SKOS (Simple Knowledge Organisation System)
There are a great number of factors that make a taxonomy successful, with a few of them standing out in Blumauer’s presentation. It’s important to keep in mind that a taxonomy is not meant for presentation, as is an information architecture. The taxonomy is meant for storage, to create classification, thereby contributing to knowledge.
First, effectiveness depends on the taxonomy being understood by the systems, search engines, and so on. This means using recognised standards. SKOS (Simple Knowledge Organisation System) is developing standards for knowledge systems – and the W3C is working to ensure that there is alignment between the ISO 25964-1 standard and SKOS.
Mapping to create context
Second, effectiveness depends on mappings to create context. Using a simple example, Blumauer demonstrated how connecting terms and labels creates a wider understanding of a topic.
A simple hierarchy is:
– – Stemware
– – – Champagne flute
Non-hierarchical connections would include:
– Champagne flute is used for Bellinis
(And then Bellini gets connected back to champagne flutes)
– Champagne flute is a champagne coupe
– Champagne is served at Tony’s cocktail bar
(And then Tony’s Bar gets connected back to champagne cocktails)
Mapping is business dependent, so it’s important to build a solid foundation and then to maintain the taxonomy. Nothing stays static, and new connections need to be made on an ongoing basis.
Include semantics in content architecture
Third, it’s important to connect the content lifecycle with a four-layer content architecture that includes a semantic layer. The semantic layer contributes to the success of search – semantic search, recommendation systems, analytics and in presenting the right content within content management systems: dynamic content publishing and automatic content authoring.
- See Andreas Blumauer’s presentation: Taxonomies put in the right place
About the speakers
Theresa Regli worked for many years as a taxonomist and then as a DAM consultant for The Real Story in the US. She is now based in London, where she helps organisations turn content into digital assets, simplify complexity, and realise their potential in the digital world.
Theresa started as a journalist, transitioned to web development, and taxonomy, and then became director of content management at a systems integration firm. She has advised over 100 businesses on their digital strategies, including 20% of the Fortune 500. She is the author of the definitive book on managing digital marketing & media assets, Digital & Marketing Asset Management.
Andreas Blumauer is managing partner of the Semantic Web Company, and has experienced with large-scale semantic technology projects in various industry sectors. He is also responsible for the strategic development and product management of PoolParty Semantic Suite. Andreas has been a pioneer in the area of linked data and the semantic web since 2002; he is co-founder of SEMANTiCS conference series, and editor of one of the first comprehensive books in the area of the semantic web for the German speaking community. Andreas holds a master’s degree in Computer Sciences and Business Administration from the University of Vienna/Austria.
Join one of our meetups
London Content Strategy meetup
A relaxed atmosphere where content professionals can learn about best practices and emerging trends, and network with your counterparts in related fields.
Content, Seriously meetup
This meetup is for in-depth presentations, short workshops, and interactive sessions for professionals who may need a deeper understanding in a particular area. Suitable for content people and managers tasked with managing content.