We all know that it's tax season, but perhaps we should declare it taxonomy season, too? With so much focus on optimizing Web sites for search engine rankings and paid search listings, little attention is devoted to the importance of implementing robust taxonomies to define standardized and consistent metadata for one's content collection. The major Web search engines recognize the need for tagging content found on disparate Web sites in order to improve contextual matching and personalization algorithms, but is a top-down generalized search engine approach going to be able to tackle the task of developing a taxonomy for everything on the Web?

Last week I attended a seminar on taxonomies that was sponsored by Verity, Inc. Taxonomies are a hot topic. Maybe not exciting, but definitely hot. The room was filled with IT and information professionals who are faced with the tricky task of implementing a taxonomy across their institutions' content collections. With the help of a knowledge expert or an information specialist/librarian, creating a robust and dynamic taxonomy doesn't have to be difficult, particularly if tools and prebuilt taxonomies are available. However, all panelists at this seminar agreed that the trickiest aspect of building and implementing an organization-wide taxonomy involves pulling together and gaining consensus from the people in the various departments whose data, image files, sound, video or other types of content need to be assembled and categorized.

Not surprisingly, Verity specializes in software tools for mapping content into categories--what they call thematic mapping--and providing prebuilt taxonomies, usually build on an industry standard, such as the MeSH from Medline for medical information, or Factiva's company and industry taxonomy. Today's headline story about Verity's work on homeland security was referenced in the seminar, in large part because Verity had built a custom taxonomy in conjunction with the Department of Homeland Security for the project.

A couple of key take-aways from the seminar, which apply to both individual and institutional Web site publishers, as well as Web search engine companies:

--To achieve a quality result in creating and using a taxonomy to tag content, a combination of machine tools (i.e., software) and human review is necessary. One speaker put it this way: "people for quality; tools for quantity."

--A major benefit of having an underlying taxonomy is the ability to create browse categories that enhance text search. Note to search interface designers: user tests show that the best approach includes a text search window as a first step, followed by related browse categories that can be used to refine a search. Alternatively, navigation paths could be devised based on the browse categories that apply to the initial text search.

When the subject is taxonomies, it is easy to get bogged down in terminology: what is a taxonomy vs. a thesaurus vs. an ontology. The Verity speaker did a superb job of differentiating between the terms. I won't try to replicate her definitions. Instead, I'll close with this thought: As institutions increasingly apply system-wide taxonomies to their internal and externally-shared content collections, we will be well along the path toward the Semantic Web, where it will be far easier to determine context of individual sites on the Web and to identify communities of related content. Especially if some standards are adopted.

WebProNews notes the debut of a new tool called Newsmap developed my Marcos Weskamp and his associates. Newsmap provides a graphical representation of headlines for various news sectors that are color-toned and sized based on presumed importance - similar in some ways to heatmaps developed for financial services such as SmartMoney's Map of the Market. The tool is more and experiment than anything else at this point - you'd have to be a pretty patient person to make use of some of the font sizes and angles used to represent relative relevance - but it's a very interesting exercise in automated news layout, more sophisticated in concept than those used by Google, MSN and Yahoo! for their news services. Will there be a time when we pull up an automatically generated news page on our desktops or I-Ink PDAs that has a newspaper-like front page and section layout with banner headlines for major items, etc., based on relevance to our specific needs and interests? Editors out there, don't go aghast at me, since this kind of tool may help in layout as much as replace it as a human art, but this kind of nascent concept should be watched closely for far more practical application.

DM Review picks up on business content soultions provider OneSource's new quality control initiatives using multiple techniques, including automated monitoring, parsing and tagging of news, content vendor management, corporate Web site reviews, primary research and rapid data correction. As OneSource finds its content increasingly a part of institutional portals that need this information as up to date and accurate as possible for immediate action, the initiative places it in a good position to leverage its strategic use of Web services integration to make those portal presences a seamless and highly reliable part of their clients' operations. It comes also at a time when D&B-backed rival Hoover's is stumbling with data quality issues and when flashy online alternatives such as Eliyon are failing to prove that they have industrial-strength content integrity. As content distribution technology advantages fall away from aggregators, cost-effective quality control is one of the primary strengths on which any enterprise-level distribution strategy must rest - and perhaps becomes the core function for content that may come from licensed sources, public sources and client-supplied sources. Many have promised strong QA for business information, but few have delivered to their clients' needs and expectations, so OneSource is placing themselves in a strong position for improving their market penetration on the top end. Now if only they can do something about the bottom end of the market...

Holy escalation, Batman, when is this all going to end? It's getting down to hand-to-hand combat in the search wars, as The Mercury News picks up on Microsoft's plans for a new weblog search service slated for later this fall and Redmond's nascent answer to Google News, dubbed NewsBot. No real specifics on the weblogs service, other than that it's going to be selective (yes, that's good...), but of equal interest is the maturity of NewsBot, which offers both preset pages and a search tool, similar in scope and effect to Google's well-established beta news service and Yahoo!'s new news service. Overall results in quick tests are pretty good in terms of search relevance, we'll be testing it in more detail over the next few weeks. While I am sure that Google has some concerns about this increased competition, if I am a major news portal, I have to be very worried about these developments. Automated news aggregation is poised to become the default "front page" for a new generation of maturing online readers, providing a level of objectivity in news gathering that editor-guided news services cannot replicate easily. Newspapers still presume the printing press as the central argument of their editorial powers, but the quickly evolving powers of Web search and content presentation tools are becoming the central news gathering metaphor very quickly. Not time yet to short your New York Times or Gannett stock just yet, but it's a quickly evolving landscape that's just begun to get interesting.

As noted by CNET News. Google has launched yet another Beta in the escalating search wars frenzy, this time taking on search results personalization. In true Google fashion, the interface provides a relatively simple way to specify how much personalization one wants in search results - specify categories within a simple taxonomy that interest you, then when you get search results you may slide a simple control from left to right to specify how much personalization you want included in the results. As you slide the control, new items will appear in the results list, with some friendly-looking "Google balls" indicating that they were added to the results from the personalization controls. In some informal search tests, the results do change, but since we cover a lot of areas at Shore that are both specific and broad, I found myself checking off lots of categories, resulting in personalization that really didn't do much to help the quality of results. The taxonomy-driven approach to setting up a personalization profile seems to be somewhat awkward and contradictory to other Google philosophies. Getting concierge-like results in content services requires responding to actual interests, not potential categories of interest. It's all about customer service, not library science - ask Amazon. Much better to have a little box or such when one completes a search that says something like "include this search in my profile", which would have the added benefit of tying into text ads more effectively. While more effective than far more crude personalization services found at Yahoo! and other portals, this is one Beta that's still in need of some additional thinking.

Shore Senior Analyst Jack McConville reports on how the Caplin move will favor their own Siphon distribution technology, and how MarketAxxess and ICAP are leveraging their joint capabilities for the treasuries markets. Registration required. more...

Digital Rights Management (DRM) systems have been notoriously user-unfriendly, with no perceived value to reader. But new applications are evolving. In an announcement this week, Elsevier has licensed RapidRights from Cadmus Communications to manage their sales of electronic reprints of health science journal articles. Sales of article reprints from major peer review journals is an attractive business, and Elsevier has prestigious titles, including The Lancet and the American Journal of Cardiology. Another publisher, the American Medical Association, receives significant revenue from print individual articles from its flagship title, JAMA (Journal of the American Medical Association). These articles are used as sales and customer service collateral by pharmaceutical and medical equipment manufacturers. The process is clunky, as is all print sales collateral. However, as described in DRM Watch, the RapidRights approach is based on the purchase of a fixed number of copies, and the download simply decrements a counter of the copies. No registration is required, and no plug-ins --both major drawbacks of other DRM approaches, and the format is standard PDF.

Medical content is rapidly becoming more electronic and readily accessible to medical professionals, with a PDA becoming standard equipment for younger doctors. Simplicity is the key to making this work, and this application meets that criteria. And the purchasers, those medical products companies buying reprints, will welcome faster and easier distribution!

As noted in CNET News and other outlets, Sony will release an eBook reading device next month using a display developed by Royal Philips Electronics using technology developed by partners E Ink and Toppan Printing. The device, dubbed LIBRIe, will retail at around $375 according to CNET, and will use Sony Memory Sticks (of course) to store between 20 to 500 eBooks, depending on the account. The key factor in this new product is the E Ink technology that allows for a display that has the resolution, reflective properties and overall look of newsprint. The device is somewhat larger than a paperback book and sports a QWERTY keyboard, presumably riding on top of a Palm OS similar to its Clie PDA products. I don't have one in my hands yet, but I am sensing from all of the clips I've been going through is that this very well could be a "gotta have" gadget for technoholics and readers alike. The form factor is a little wierd, but being able to show people the latest and greatest in display capabilities in an affordable (PDA/premium cell phone range) package in a screen that's far more readable than the typical PDA and far more usable and private than a tablet PC (comes with a handy flip-close cover) has something to be said for it. As we've been telling folks in our logs and research, eBooks are a quiet revolution underway, and this highly affordable and techno-sexy unit may be just the thing to pump up the volume for eBooks - and steal Microsoft's march in the process.

The European Union's $613 Million judgement against Microsoft and requirement to unbundle its Media Player is just the beginning of the EU's actions against Microsoft: according to PC Pro, competitors in both instant messaging and mobile phones markets have accused the company of using Windows XP to strengthen its position. It's all too likely that Microsoft's heavy handed tactics for eBooks will also come under scrutiny as its Reader software gets bundled into its new OSes as it did with Internet Explorer. Content vendors cozy up to Microsoft as a friend that can help them to penetrate the desktop market more effectively, but at the end of the day these vendors are digging their own grave for high-margin profits. The expensive licensing for Microsoft products and other monopolistic I.T. investments that don't pay back in productivity gains proportionate to these investments is handcuffing institutional implementors who would otherwise have money to spend on innovative ways to consume content. With more and more content revenues dependent on such innovation, it behooves professional content vendors to promote the most open market for underlying I.T. infrastructure possible, so that they can become the leaders in vContent that their clients require them to be.

Borrowing from the headline: All Eyes on Google from the March 29 issue of Newsweek, it's fair to say that Google is the talk of the town these days. Search engine marketing and contextual ads served by Google and Yahoo/Overture are hogging the spotlight at the expense of other advancements in Web publishing and online monetization models. But there are other developments in online advertising that merit attention. One notable example is the news that Procter & Gamble (P&G) is expanding its custom Web publishing initiatives with a new health-care related Web magazine called HealthExpressions.com.. This Web site follows P&G's earlier site, HomeMadeSimple.com. While these early publications are chock-a-block with advertisements for P&G products (which makes it very clear who is sponsoring the site), the content-centered advertising vehicles illustrate a trend in advertising--a trend toward advertisers creating their own context in which to place their ads. Offline (i.e., print) examples are appearing on the scene, too, such as a Wal-Mart magazine planned with Time, Inc, which will be targeted for the Wal-Mart shopper demographic. So, while Google, Overture, Kanoodle, and others are busy improving their technology and bid-rate models for placing ads on contextually-relevant Web sites, some top advertisers are taking matters into their own hands and creating the content that is contextually relevant to the demographic they seek. Through this Web-centric publishing and marketing strategy, consumer product companies like P&G can create direct relationships with their customers and learn what products and promotions click with various consumer segments [pun intended]. What can B2B publishers learn from these developments? Perhaps most important, P&G's content-heavy Web sites demonstrate how advancements in institutional publishing technology, along with general acceptance of the Web as an information, news, and entertainment medium, are fundamentally changing the models for distributing content and associated advertisements. Trade publishers beware!

As noted in our weblog two weeks ago, Factiva has been readying a new single-box search interface for its premium content, which was announced today billed as iWorker, a technology that maps keyword queries into Factiva's multi-tier taxonomy to further refine results. Translation: our back end technology works rather differently than typical search engines, which is good in some ways but also indicative of the difficulties of mixing content from unstructured sources from the culled and categorized sources found in a typical premium aggregator's database. Along with iWorker comes a new Factiva-specific search box widget that can be imbedded in one's browser toolbar, similar to tools developed by Google and AtHoc. Good stuff, and again a step in the right direction, but it's indicative of how far behind most premium aggregators are these days in providing content value. Taxonomies are great, but they're not going to justify the cost of a premium database of 9,000 sources in the long run when public search engines are learning how to integrate premium sources into their own algorithms. Aggregators must accept that the ultimate premium content solution will always go beyond the content that they can license to their users. A painful fact, but one that will be coming home to roost soon enough anyway.

A recent Salon article [PREMIUM] chronicles the sad but all too familiar story of one of the legions of authors who struggle to make a living writing fiction and general interest books. Given that it's written by a pretty capable author, it turns out to be an engrossing and touching story: a promising first book falls short of the mark commercially, dooming the author to years of semi-success and heart-wrenching anguish, finally culminating in - gulp! - getting a real job to pay the bills. Hey, it happens. In the meantime, as outlined at the eBooks in the Public Libraries Conference [PREMIUM WEBLOG COVERAGE - REGISTRATION REQUIRED], publishers are beginning to explore the E-P-E cycle for releasing titles: start electronically, go to print when volumes warrant, cut back to electronic and on-demand printing when they start to fade. So much more sensible - and less frustrating for authors who want to make progress in the marketplace. Instead of trying to get PR attention in the general media, then, authors will find themselves using libraries as the minor leagues needed to gain the notoriety that they need to get their much-desired stardom. Publishing is about to undergo some very welcome and overdue transformations thanks to eBooks and library outlets, providing a correction of supply and demand that better meets the needs of both the marketplace and authors eager to make their mark. It may be that editors will be able to get back to the real work of getting talent developed for commercial exploitation, rather than sweating which mega-hit is going to make the quarterly earnings look flush.

In our Content eCommerce premium weblog, Shore Senior Analyst Janice McCallum reflects on how Norton Firewall's proclivity for blocking virtually any useful advertising, including contextual text ads, is eliminating valuable contextual content in many instances. more...
In our eResources Marketplace premium weblog, Shore President and Senior Analyst John Blossom covers the recent OeBF conference on eBooks in the public libraries. Libraries appear to be the lever that eBooks have been looking for to take off, offering many intriguing possiblities for content commercialization that can easily spill over into other sectors. more...

The USC Annenberg Online Journalism Review covers a recent comprehensive report by the Project for Excellence in Journalism, an industry effort to profile every aspect of American journalism on an annual basis. As PEJ Director Tom Rosenstiel sees it, "...journalism is in the middle of an epochal transformation, as momentous probably as the invention of the telegraph or television." It's a transformation that puts journalism at risk in the view of PEJ, as the eyeballs are surging towards an online model that is not yet producing the revenues required to support top-notch journalism operations. The stats show that online news is getting youngsters' attention, something that TV and print just haven't cracked, so there's no doubt where the future of news is going. But how quickly? While the study shows that only 15 percent think of online outlets as their primary source of news, 26 percent looked at online news within the last day - about half of today's online population. Will news outlets be able to bridge the revenue gap effectively in era of rapid transition? As noted in our earlier news analysis, there is less and less to hold together traditional news operations as other kinds of content aggregation - including user desktop integration via RSS-fed weblogs - become increasingly powerful. The issue may not be whether news organizations as they exist today will be able to make the transition as much as whether they will be anything like they exist today. It's not just the technology that's changing but the very nature of how news is formed. The social networking and hyperlinking provided by Weblogs, for example, provides much of the source validation provided by traditional journalism, an organic kind of quality validation that precludes many traditional editorial controls. It may not be that journalism quality is declining in the face of online forces as much as it is that online journalism is redefining how quality is achieved, and for what purposes. News formation is a key focal point for vContent today, a concentration of the very human aspects of content with the truly revolutionary aspects of humble but world-changing technologies.

London's The Times reports on yesterday's New York Post scoop that AOL parent Time Warner is mulling over a spinoff of the profitable but shrinking online service. The Times article speculates that possible suitors for a purchase may include Yahoo! and Barry Diller's InterActiveCorp. AOL suffers from any number of ills well analyzed elsewhere, but from our own perspective it's interesting to see how the marriage of media and online content interests continues to be rocky at best. The Web is not something that works best with forced selections of interest: its underlying premise is that it's okay for anyone to create any kind of content for anyone and for people to get it any way they want with near-zero distribution costs. Traditional media packaging falls apart in this environment. AOL's relentless commercials blare forth from televisions trying to sell us the Web as if it were a box of laundry detergent. Sorry, folks, but we all know that there's nothing in the box anymore. Now that AOL has abandoned content creation as an activity, all that's really left is an incredibly dumb user interface and a connectivity business that was commoditized years ago. Its only truly valuable content assets are its email and instant messaging accounts - certainly valuable to Yahoo!, MSN or others, as these are sources of valuable personal content that are difficult to migrate without acquisition. Fearing Microsoft, TW will probably go with Yahoo! The remaining online assets could be merged in with its cable assets as a complementary home portal for its cable modem subscribers, similar to the alliances that Yahoo! has struck with SBC and other cable operators. Unless one controls the distribution technology, one does not have a distribution business. Unless one controls the content, one does not have a publishing business. Time Warner appears to be facing these facts head on now and moving on to search for the soul of a 21st century media company in its more tangible assets.

Dialog, a Thomson company, has announced the fruits of 3 years of labor in two guises this month. One, SmartTerms(TM), an internally-developed taxonomy of company names, industry names, geographic locations and subject terms; and the other an integrated platform for company reports, analyst reports, market research reports and news (i.e., Dialog Profound and Dialog NewsRoom). These are encouraging moves by the company that provides access to hundreds of databases via multiple services some of which require separate passwords and use different search interfaces, which were inherited through acquisitions. In his article in Information Today, Matthew McBride posits that perhaps the "technology tide [is] turning for Dialog?" Indeed, with these announcements, it is clear that they are investing resources in adding value to their database collection through improved functionality.

However, two key issues confront Dialog as it tries to maintain its position in the era of search engines and Web distribution. First is alternatives for contributing publishers. With publishers increasingly providing direct access to their content via the Web--whether it be free or for-fee--it becomes more difficult for them to justify sharing revenue according to the royalty models that were established before direct distribution was feasible. Second is pricing. Dialog adds fees on top of document charges. Again, with new, more efficient distribution models of aggregation developing on the Web, will customers continue to pay a premium for access via Dialog?

[More complete coverage of this event is provided on our eResources premium weblog. Registration required.]

The Open eBooks Forum (OeBF) conference on eBooks in the public libraries held yesterday in New York City was one of those events where you can feel the surge of a movement beginning to realize its true strength in real-time. After several years of false starts, eBooks are starting to take hold in public libraries and are now providing library patrons in numerous major cities growing access to electronic materials. Numerous success stories at the conference pointed to rationalizing technologies, improving availability of content and an economic environment that has forced libraries to find more cost-effective methods of servicing patrons as leading factors in eBooks lending growth. Will this be the breakout year for eBooks? Probably not, as balky rights managment, entrenched library staffs and antiquated cataloging systems still pose significant challenges to a complete eBooks takeover any time soon. But where some see a continuing evolution of eBooks it's clear that this is a quiet revolution in the making, with irresistable forces beginning to compel libraries to embrace eBooks far more rapidly than may have been imagined. Not the least of the pressure for change is coming from the publishing industry, which looks to the strengthening success of rights-protected eBooks in libraries as a test bed for experimenting with new ways of deploying and monetizing premium content in models that can help them to define profits between the trusty but aging "one book, one user" model and the melee of open Web access. Community libraries offer publishers an environment in which they can work within a comfortably familiar distribution model to work out the details of how eBooks and other premium rights-protected content can best serve users. And then? The unmentioned factor at the conference was the Web itself, where eBooks already enjoy healthy sales. Content ecommerce portals could easily extend current commercial models to include lending and other mixed-use models that could tie into local library cards, corporate IDs or other forms of access subsidization. Just as corporate librarians were caught in the downdraft of technological and economic change that rendered many of them redundant, public and institutional librarians feeling the pinch of budgets and patrons going "Web first" for answers may find the publishers whose revenues depend on them finding the answers not hesitating to get content to their patrons by any means necessary if libraries fail to get their products to their markets in a manner that will sustain their profits. The power of local communities as a component of the Web's strength is likely to keep public libraries in the mix for a long time, but competition for servicing their patrons without their services will only increase unless they decide to go from evolution to revolution fairly soon.

The New York Times reports on MFS Investment Management's decision to provide "execution only" pricing from investment banks and brokers who typically justify their premium commission fees for institutional trading based in part on the value of the research reports that are provided to them. Robert C. Pozen, the current nonexecutive chairman of MFS, was quoted by the Times as saying "We are valuing their research at zero.'' Mutual funds houses find themselves struggling with repercussions from accountability scandals which place pressure on their bottom lines, even as U.S.-based investment banks and brokers adjust to higher accountability and segregation for developing their research products in the wake of their own deceptions, so both sides are stuggling to come up with new execution models. The bloom has been off the rose for broker research for quite some time and investment banks will be hard pressed to stop this trend once it starts. As surely as the brass bull stands on Broadway, the herd mentality will push this trend along very rapidly. What options for broker research? This certainly doesn't bode well for outlets like TheMarkets.com, a portal supported by numerous investment banks to provide common access to broker research and encourage trades with participants. With broker research and analytics freely available from numerous vendor sources and electronic crossing networks driving margins on institutional trades ever lower, it will take a far more sophisticated approach to packaging content as a valuable part of a transaction to reverse this trend. U.S. Fair Disclosure rules make this ever harder to accomplish, but the future of premium financial content for securities clearly lies in defining its specific and contextual value at the time of execution. This push from mutual funds may be just the pressure required to send both investment banks and financial content vendors back to the drawing boards.

Cnet's article on competition in the Web search engine segment provides a very good overview of new players that are gaining attention due to specialized capabilities, such as personalization, improved display of search results, local search, and search specialization by content-type or industry. Red Herring [registration required] categorizes the emerging (and re-emerging) specialty search companies a bit differently, but combined, the two articles provide a fairly thorough picture of various businesses that focus on Web search technology. While there is a need for many types of companies that attack search relevancy from different angles, the piecemeal approach we're seeing of some companies focusing on display, some companies focusing on behavioral patterns, some on localized search, some on contextual relevancy makes it clear that consolidation of efforts is required in order to provide a better result for consumers and advertisers. Presumably the heightened press about the up-and-comers will facilitate Google's job of identifying which companies to buy--whether or not they have IPO cash available. But, even with an acquisitive Google, there remains room for specialized search companies that can provide a deep understanding of research or business applications for particular functional areas or other well-defined communities of users. In some cases, the specialized search technology can also serve the purpose of supplying a contextual advertising platform for targeted advertisers. iPhrase and IndustryBrains are two such companies that were mentioned in the above-referenced articles. Interestingly, both were cited for specialized search and contextual advertising for the financial information segment.

Sometimes you start in one direction and go in another, only to come back to where you started: MetaCarta is a company that originally had ideas back in the dot-com boom for relating content to geographic locations for major consumer portals, but some research funds from the U.S. government's DARPA defense research arm helped them to nurture their capabilities by focusing on more strategic needs. After numerous successful governmental installations, it appears as if their capabilities have matured to the point where it may be time for them to take a stab at the consumer market again. Directions Magazine provides some compelling examples of MetaCarta's ability to relate search terms to specific geographic coordinates using their own language processing algorithms and Web services-based search modules. Certainly useful for tracking down bad guys in obscure places and finding reports and research that relate to oil exploration sites, but how about for tracking down hard-to-find-widgets in Manhattan or pumping in Krispy Kreme coordinates automatically into a GPS-equipped PDA? Geographic context has been exploited in broad terms already in Web formats and with handhelds via online news outlets, directories and guides, but pinning down content to contexts that define the intersection of very broad content sets and very specific locations is still a growing opportunity in content markets. Getting content down to very finely defined geographic contexts is one key avenue for vContent aspirants, one of many "war benefits" from government-funded efforts that's benefitting both individuals and institutions in the private sector.