Metadata: Really Cool…When You Don’t Have to Mark It
Earlier this month, the announcement that Siri’s creators had successfully placed an order for pizza with voice commands made a splash in the technology community because the pizza ordering process is a fairly complex process for dumb machines. The success of the pizza order is one realization of the Semantic Web envisioned by Tim Berners-Lee, a vision that sees computers intelligently communicating with each other to automate complicated tasks. If the voice-commanded pizza order is any indication, the future is bright for increased productivity through digital assistants.
Much of the promise of the Semantic Web is built on a platform of metadata, which is used to identify data in a machine-readable fashion. For example, metadata can be used to differentiate between types of doctors so that someone asking their digital assistant for a nearby doctor isn’t directed to a doctor of veterinary medicine.
Metadata can be useful wherever a bit of machine intelligence is needed, including in an organization’s business-critical content. Take, for instance, online policy documents that are peppered with important terms defined in a glossary: it might be useful for key terms to be automatically linked to the definition of those terms, or a window with the definition to appear when your audience hovers over the term. For another example, financial reports may need public companies automatically linked to stock data.
In the above examples, glossary terms and public companies need to be marked as such for the system to identify where linking or hover behavior needs to be added. The problem is that leaving the marking responsibility to your subject matter experts (SMEs) is an undue burden. First, manual marking of metadata is an inefficiency that wastes your SMEs’ time. Second, the potential exists for metadata-appropriate content to be overlooked, which means that your audience may not receive the full benefit of your metadata-enriched processes.
Automated marking of metadata is one way in which Quark’s content automation expertise can help organizations enrich content processes — documents can be scanned for glossary terms, public companies, or other such important words or phrases and have the necessary metadata inserted at key points automatically using Quark software. Organizations can place the power of the automated metadata marking into their SMEs’ hands so that they can see where the metadata is being added and supplement or remove the marking as needed. Alternatively, the metadata can be added to content during the publishing process without any human intervention. Either way, your published content will be all the more rich for the intelligence added through metadata.