The publishing industry is an interesting beast: Its front-end moves rapidly to get content out to readers, but its back-end processes to deliver that information are so tightly packed that there's not a lot of drive to make sweeping changes in those technologies or processes.
“They have to be on this schedule, so traditionally they have been slow to change their operations,” says Rachel Lovinger, associate experience director at interactive agency Razorfish. At next week’s Semantic Web Media Summit taking place in New York City, Lovinger will be among the speakers discussing topics such as whether the industry’s tolerance for change is growing, given its need to find more innovative ways both to reach audiences and be more profitable.
“The problem the media industry is facing is not going to resolve itself,” she says (see some of the thoughts fellow speaker Michael Dunn, VP and CTO at Hearst Interactive Media, has on that topic here). “The panic has died down a little but there is still this need to innovate and to grow.”
Developments such as schema.org count as interesting responses to that need, and also validate the requirement for content providers to adopt structured data and for major search engines to pull together and support that adoption, she thinks. Its debut, however, has likely created some more confusion for publishers trying to figure out whether and how semantic web technologies might fit into their strategies, given initial questions about schema.org’s compatibility with W3C markup standards such as RDFa.
“I think there’s more movement to bringing [the schema.org] effort in line with other structured data approaches,” Lovinger says. But it wouldn’t be surprising if its emergence caused some publishers to put plans on hold while waiting for these issues to be worked out.
If they even had such plans to start with. In the time since Lovinger wrote the 2010 report, Nimble: Publishing in the Digital Age (which The Semantic Web Blog covered here), she has probably seen more early adopters pushing forward rather than a multitude of new content parties publicly clamoring to bring semantic web technologies onboard. The BBC, for instance, plans to do for the Olympics 2012 coverage what it did for its World Cup 2010 coverage, using metadata and Linked Data to organize and manage the site dynamically. “The people I see using it are still a lot of the same that were doing so when I was writing the report,” she says.
And at Razorfish, it is generally also more a case of its own internal team seeing semantic web technologies as a solution to a client’s problem than client demand for implementing such technologies per se. That said, clients certainly aren’t closed to the idea, she says; “it’s more [the case] that we are seeing possible applications that could be of use to people.”
Among them is one she considers a low-barrier, high-results way of leveraging semantic technologies, as it doesn’t require deep integration with the whole publishing stream. “There are some interesting tools that people can use to just do research. For people who track stories or topics or responses to stories, there are some advanced monitoring tools that use semantic technology,” she says. She mentions products such as Tattler, which promotes its offering as much to journalists and researchers as to communications professionals such as marketers and public relations representatives.
“People can start using that to provide better and more accurate results, and they don’t even have to change the actual publishing system. It just becomes part of the research and monitoring process and you get better feedback on the content you’re creating,” Lovinger says.
Using semantic technologies to boost advertising revenue also should be top of mind, too. Some semantic advertising tools, she thinks, primarily target advertisers themselves as the main market. But publishers might want to be more aggressive on their end about pursuing these opportunities, so that they can have more control over content monetization. “Semantic advertising has lot of promise because we have all this information about the behavior of a user, ...information about how a person has behaved since they entered a site, and you could use that to provide advertising that’s more relevant and more interesting,” she says.
For broad structured data adoption to become prevalent among publishers, however, there’s also a question about when content management systems themselves catch up to the semantic web. Drupal 7, the open source CMS that supports RDFa added a module to enable the collections of schemas available at schema.org just days after its announcement. But-most commerical CMS products have longer development cycles and are slower to adopt capabilities that aren’t in high demand by the user base. “If the user base says put RDFa ahead of everything else they’re developing, then they will – but the demand has to come from the publishers,” she says.
That may leave us with something of a chicken and egg scenario, perhaps. Lovinger is as interested to find out how it's all trending as any conference attendee. "The need to change may have caused some shifts and I'm curious to see if that’s true," she says. "There’s enough motivation to change things now that it might overcome any resistance."