Wednesday, May 20, 2009

Reference Publishing Meets Digital Objects: Wolfram|Alpha, Google and Yahoo Pursue On-The-Fly Curated Content

While the tsunami of buzz surrounding the Wolfram|Alpha reference service (by their own claim not a search engine) seems to indicate a desire for novelty at least as much as interest in its actual merits, the service is one of a few major announcements this week which indicate a shifting attitude towards online publishing that is catching up with the realities of today's publishing technologies. Wolfram|Alpha offers a simple "white box" query interface with semantic parsing of requests that access a fairly limited, curated set of reference data feeding through display functions such as tables, charts and graphs. The W|A team is careful to note that these images and data displays are not search results but useful publications unto themselves - hence a bit of static about their terms and conditions, which emphasize that query results are W|A's intellectual property.

Wolfram Alpha is an interesting reference tool for people wanting to chart and graph contrasting points of data, but it's hardly alone in the movement towards more robust on-demand content. Recently Google announced at its Searchology event a range of enhancements to its emerging Universal Search capabilities, including search options that enable one to embed relationship trees, videos, reviews and other displays that relate to a query - in addition to already embedded rich content such as maps. For example, the image to the right shows a relationship tree for people relevant to Apple CEO Steve Jobs as well as relevant video clips. Included on this menu of options is a feature called "rich snippets," which enables publishers to encode content that's related to a particular item from their Web sites that appears in a search result using a microformat specification provided by Google. Examples of this feature in use are fairly thin so far, but it holds out the promise for a wide range of content sources to be placed in context with content returned from Google searches. Google's open approach to helping publishers to develop search-embedded display applications for their content returned from queries, as opposed to Wolfram|Alpha's "it's our content" approach, is far more likely to accelerate the development of rich content applications cued by queries into a wider array of databases.

The team at Yahoo has been looking at this emerging landscape for enriched queries and is trying to steer somewhat of a middle course between the Wolfram|Alpha approach of tight curation of sources and applications and the content available on the open Web. As noted in SearchEngineLand recently. Yahoo is ceding the "all the world's information" indexing battle to Google and is instead focusing on doing a better job of curating specific types of Web sources more effectively and serving them up through a variety of display objects. Yahoo's Search Monkey display capabilities, similar to the "rich snippets" microformats announced by Google at Searchology, already help to power rich content in Yahoo search results, and will be folded into broader use of digital objects that get served up via Yahoo queries.

This is all a way of saying that search was never really about "just search" to begin with. Search results are and always have been content in and of themselves, a collection of content sources that are arranged to enable people to determine what's the most relevant information on a given topic. In other words, search is an editorial function, albeit one that's highly automated, but it performs much the same function as an editor working on a news article or an encyclopedia entry - except that it is done on an on-demand basis. We've seen many efforts through recent years to enrich search results with more robust graphics and related content, making a given search result more like a reference compendium rather than just a listing of links. But what we seem to be moving towards at a faster pace as of late is the realization that the digital objects served up by search engines are increasingly likely to be the objects where people get their answers and insights in full, rather than trudging off to various links to get more in-detail answers.

Now, this is usually where some of my good friends in publishing start to howl about the evils of search engines, but realistically this kind of aggregation is happening whether publishers want it to happen or not. The only question is how they want their own content to participate in this automated just-in-time editorial environment. I believe that the most constructive answer for publishers is to embrace the increasingly object-oriented environment of search warmly and to recognize that there are opportunities abounding in getting more of the right content in front of the right audience at the right time through enhanced search services. For example, instead of having to compel someone to click on a link to read a news story on your own Web site, you could have either a lede paragraph or an entire article come up in the search results page. That article could have your own embedded ads, or links to a subscription or micropayment monitoring service that would enable the publisher to expose premium content in a search context.

However it's done, query results on a search engine represent the point of highest demand for much of today's content. Getting the right content into those results with the right monetization scheme gives a publisher a potential jump on the competition that hopes that someone will click on their link into their Web site. Destination Web sites serve an important purpose, but in the world of distributed online content aggregation, but relying on them solely is a little bit like saying that one should only buy newspapers at a publisher's printing plant. Search engines and other content technologies that allow on-demand contextualization of content for an audience are the newsstands of today, leaving publishers with but one choice: do you want to hide your content behind the counter or do you want it where people can see it? The serving up of rich content through digital objects asks the question more loudly and with more and better answers to the "how" of meeting this challenge, but it's the same challenge that's been with us for many years.

The most important innovation that publishers can embrace over the next several years are the technologies that enable them to have cross-platform digital objects that are easily monetized and licensed for monetization through a broad array of partners adept at on-demand contextualization of content. While the Wolfram|Alpha platform offers an interesting view of how a limited range of sources could be curated into a useful reference service, ultimately it's a model that is far too limiting to allow most publishers to succeed. A handful of content-serving graphs and charts is useful for only a few types of information sources. Publishers need a robust array of content-serving objects, ones that enhance their own content and that allow it to trigger the integration of other content sources more easily for enhanced value.

Search engines and social media tools have empowered a new generation of editors and curators who have the power to put a publisher's content in its most valuable context more quickly and more effectively than traditional distribution media. Hopefully the efforts by Wolfram|Alpha, Google and Yahoo begin to make publishers think more actively how their content can be served up more automatically in more contexts through their object-oriented publishing technologies.
Post a Comment