• Home
• About
• Writers
• Landscape
• Wizards Index
• Fraud
• Wanna Be Like Larry?

Trapped by a Business Model, Not Technology

The headline “Reuters CEO sees Semantic Web in its Future” triggered an immediate mouse click. The story appear on O’Reilly’s highly regarded Radar Web log.

Tim O’Reilly, who wrote the article, noted: “Adding metadata to make that job of analysis easier for those building additional value on top of your product is a really interesting way to view the publishing opportunity.”

Mr. O’Reilly noted that: “I don’t think he [Devin Wenig, a Reuters executive] should discount the statistical, computer-aided curation that has proven so powerful on the consumer Internet.”

Hassles I’ve Encountered

Reuters comment about the Semantic Web did underscore the often poor indexing done by publishing and broadcasting companies. In my experience, I have had to pay for content that was in need of considerable post-processing and massaging.

For example, if you license a news feed from one of the commercial vendors, some of the feeds will:

• Send multiple versions of the stories “down the wire”, often with tags that make it difficult to determine which is more accurate version. Scripts can delete previous versions, but errors can occur, and when noticed, some have to be corrected by manual inspection of the feed data.

• Deliver duplicate versions of the same story because the news feed aggregator does not de-duplicate variants of the story from different sources. Some systems handle de-duplication gracefully and efficiently. Examples that come to mind are Google News and Vivisimo. Yahoo’s approach with tabs to different news services is workable as well, but it is not “news at a glance”. Yahoo imposes additional clicking on me.

• Insert NewsXML plus additional tags without alerting downstream subscribers. When this happens, the scripts can crash or skip certain content. The news feed services try to notify subscribers about changes, but in my experience there are many “slips betwixt cup and lip.”

Now the traditional powerhouses in the news business face formidable competition on multiple fronts. There are Web logs. There are government “news” services, including the remarkably productive US Department of State, largely unknown Federal News Service , and the often useful Government Printing Office listserv. There are news services operated by trade associations. These range from the American Dental Association to
the Welding Technology Institute of Australia. Most of these organizations are now Internet savvy. Many use Web logs, ping servers, and RSS (really simple syndication) to get information to constituents, users, and news robots. Podcasts are just another medium for grass roots publishers to use at low or without cost.

We are awash in news — text, audio, and video.

Balancing Three Balls

Traditional publishers and broadcasters, therefore, are trying to accomplish three goals at the same time. I recall from a lecture that the legendary president of General Motors, Alfred P. Sloan (1875 – 1966) is alleged to have said: “Two objectives is no objective.” Nevertheless, publishers like Reuters and its soon-to-be owner are trying to balance three balls on top of one another:

First, maintain existing revenues in the face of the competition from governments, associations, individual Web log operators, and ad-supported or free Internet services.

Second, create new products and services that generate new revenue. The new revenue must not cannibalize any traditional revenue.

Third, give the impression of being “with it” and on the cutting edge of innovation. This is more difficult than it seems, and it leads to some executives’ talking about an innovation that is no longer news. Could I interpret the Reuters’ comment as an example of faux hipness?

Publishers can indeed leverage the Semantic Web. There’s a published standard. Commerical systems are widely available to perform content transformation and metatagging; for example, in Beyond Search I profile two dozen companies offering different bundles of the needed technology. Some of these are known (IBM, Microsoft); others are less well known (Bitext, Thetus). And as pre-historic as it may seem to some publishing and broadcast executives, even skilled humans are available to perform some tasks. As good as today’s semantic systems are, humans are sometimes need to do the knowledge work required to make content more easily sliced and diced, post-processed and “understood”.

It’s Not a Technology Problem

The fact is that traditional publishers and broadcasters have been slow to grasp that their challenge is their business model, not technology. No publisher has to be “with it” or be able to exchange tech-geek chatter with a Google, Microsoft, or Yahoo wizard.

Nope.

What’s needed is a hard look at the business models in use at most of the traditional publishing companies, including Reuters and the other companies who have their reports in professional publishing, trade publishing, newspaper publishing, and magazine publishing. While I’m making a list I want to include radio, television, and cable broadcasting companies as well.

These organizations have engineers who know what the emerging technologies are. There may be some experiments that are underway and yielding useful insights into how traditional publishing companies can generate new revenues.

The problem is that the old business models generate predictable revenue. Even if that revenue is softening or declining, most publishing executives understand the physics of their traditional business model. Newspapers sell advertising. Advertisers pay to reach the readers. Readers pay a subscription to get the newspaper with the ads and a “news hole”. Magazine publishers either rely on controlled circulation to sell ads or a variant of the newspaper model. Radio and other broadcast outlets sell outlets to advertisers.

These business models are deeply ingrained, have many bells and whistles, and deliver revenue reasonably well in today’s market. The problem is that the revenue efficiency in many publishing sectors is softening.

Now the publishers want to generate new revenues while preserving their traditional business models, and the executives don’t want to cannibalize existing revenues. Predictably, the cycle repeats itself. How hard is it to break the business model handcuffs of traditional publishing. Rupert Murdock has pulled in his horns at the Wall Street Journal. Not even he can get free of the business model shackles that are confining once powerful organizations and making them sitting ducks for competitive predators.

Semantic Web — okay. I agree it’s hot. I am just finishing a 250-page look at some of the companies doing semantics now. A handful of these companies are almost a decade old. Some, like IBM, were around when Albert Einstein was wandering around Princeton in his house slippers.

I hope Reuters “goes semantic”. With the core business embedded in numeric data, I think the “semantic” push will be more useful when Reuters’ customers have the systems and methods in place to make use of richer metatagging. The Thomson Corporation has been working for a decade or more to make its content “smarter”; that is, better indexing, automated repurposing of content, and making it possible for a person in one of Thomson’s more than 100 units to find out what another person in another unit wrote about the same topic. Other publishers are genuinely confused and unstandably uncertain about the Internet as an application platform. Buggy whip manufacturers could not make the shift to automotive seat covers more than a 100 years ago. Publishers and broadcasters face the same challenge.

Semantic technology may well be more useful inside a major publishing or broadcasting company initially. In my experience, most of these operations have data in different formats, systems, and data models. It will be tough to go “semantic” until the existing data can be normalized and then refreshed in near real time. Long updates are not acceptable in the news business. Wait too long, and you end up with a historical archive.

Wrap Up

To conclude, I think that new services such as The Issue, the integration of local results into Google News, and wide range of tools that allow anyone to create a personalized news feed are going to make life very, very difficult for traditional publishers. Furthermore, most traditional publishing and broadcast companies have yet to understand the differences between TV and cable programming and what I call “YouTube” programming.

Until publishing finds a way to get free of its business model “prison”, technology — trendy or not — will not be able to work revenue miracles.

Update February 13, 2008, 8 34 am Eastern — Useful case example about traditional publishing and new media. The key point is that the local newspaper is watching the upstart without knowing how to respond. Source: Howard Downs.

Stephen Arnold, February 12, 2008

Comments

Copyright © 2008 Beyond Search · ArnoldIT.com · Disclaimer · Revolution News theme by Brian Gardner · Log in