Cross-media publishing, or repurposing, has been talked about ever since the exploding of the Web in the mid '90s. Every year, we hear claims from vendors that cross-media publishing is a reality. Usually, these vendors are talking about the ability to publish to both print and the Web without manual conversion effort. As a simple example, Microsoft Word contains a filter that creates HTML from Word documents. As a more complex example, Quark's avenue.quark creates XML documents from QuarkXPress pages that are intended for the Vignette StoryServer Web publishing system.
But now, cross-media publishing has taken on a greater meaning. Content brands need to be visible in as many venues as possible that attract the right audience. This means not only a profusion of devices, such as Palm Pilots, WAP phones, and television set-top boxes, but also the many Web sites that attract the desired demographic.
For example, a business magazine may want to publish stories to Internet devices, but also have its content appear where business people look on the Web, e.g., on an office supplies e-tailer or an online brokerage. The magazine's production group will need to send each of these venues a file. To further complicate things, several of the venues may need their own versions of the content; for example, the Palm Pilot version will probably need to be shorter than the print or Web browser versions.
Cross-media publishing started in the production area—typically a group of people who converted print-format pages to HTML—and gradually infiltrated the editorial area as Web adjuncts of publications added Web-only content or launched eparate Web editorial operations. But it can't go on this way. Editorial operations won't be able to proliferate at that rate to meet the needs of a growing number of distribution channels.
The solution to this problem was first expressed as "Create Once, Publish Many," by various people, including Paul Zazzera, Time, Inc. CIO. This means that there should be some technology that lets publishers push a button and have the content magically come out in all the right formats, all automatically.
This was a nice vision, but it can't work. Different distribution channels call for different versions of content, and those versions can't be created without editorial in- tervention. This engenders two important concepts: create once, produce many; and the idea of the content master.
The Content Master
The idea of the content master is well known in the music and film industries. In pop music, a band creates a master of a song, and then uses that to create various versions, such as the dance club version (drums and bass mixed louder, more echo), and the AM radio version (shorter length, more audio compression), etc. In the film industry, there might be a director's cut, a theatrical release, and versions for videocassette (different aspect ratio) and airplanes (R-rated scenes edited out). Each of these versions requires manual production work.
The publishing industry is slowly realizing that it should adopt the same ideas in order to serve a growing number of distribution venues. Yet editorial processes are not changing to become scalable in this manner. There are many reasons to create content masters. Most importantly, content masters can be archived in a content management system for later production and distribution. A content master can also serve as the "content of record," for purposes of capturing editorial intent and dealing with legal liability issues.
For centuries, the print version of content served as the content of record—simply because there was no other version. Everyone assumes that this has to be the case in the present era, but if you think about it, it's not necessary. In the print world, especially that of magazines and newspapers, a single process combines editing for the sake of editorial intent with editing for the sake of copyfitting. It's possible to capture editorial intent by saving an edited piece of content before it becomes subjected to the layout ax, but if that content is never published, there's no point in doing so.
You compromise editorial intent whenever you fit copy to a page layout. But on the Web, and in other formats, there's no need to do this. As anyone who has written for both print and the Web knows, there's great freedom in being able to make a Web article as long or short as you want, and in knowing that production won't screw up your article for the sake of copyfitting.
For this reason, it makes more sense to have the master version of an article be the same as the Web version, not the print version. The print version could be treated as a derivative, just like a Palm Pilot or WAP phone version.
This idea has significant implications for editorial processes and technology. It relegates the creation of print layouts entirely to the print production process on the back end, instead of spanning the entire editorial process from front to back as it often does now. It also allows content creators to author content in a media-neutral way, for later conversion to multiple output formats.