Living With Data

Your article “Dying for Data” [October] provides a good overview of the general problem but does not discuss the challenges of data integration that exist between potentially thousands of data domains. For standards to be truly useful, all the applications across every institution need to talk to each other using the same standard language.

If a nationwide system, let alone a global system, of “interoperable medical records is to be realized, it will mean getting every hospital, every nursing home, every pharmacy, and every one of the hundreds of thousands of physicians who belong to solo or small group practices,” as stated in the article, to adopt a single standard. Yet there are many standards in the health care industry, and it is not realistic to expect one national, let alone global, standard to dominate the market.

Why is this doomed to failure? Because no one data standard can meet every requirement. Different organizations and institutions will have unique needs that no single data structure can possibly address. Such standards require a very precise match between source and destination tags. You cannot make even the smallest change to the standard vocabulary to meet a particular or unique requirement that your organization, department, systems, or processes may have. In essence, there can be no regional dialects within a standardized language. This “one size fits all” ultimatum has not been successfully adopted across other industries, and if experience is any teacher, it is not likely to be adopted globally in health care.

The current trend in dealing with interoperability is to manage it not on a syntactic but on a semantic level. Called the Semantic Web, the idea is to capture the shared meaning expressed by the data between common processes. The Semantic Web includes protocols such as RDF and OWL, but also a new protocol called the Metadata Semantic Language (MSL). It is this new protocol that is of interest to the integration problem.

The MSL is formatted as a separate layer to XML—it is an integration layer, or a metadata layer. Data structure and metadata naming conventions are enhanced with a separate layer of semantic information that encodes the definition of each element of data, including its relationship with other elements. Approaches based on the Semantic Web are the only possible hope of achieving data interoperability in health care and should have been discussed at some level in the article.

Steve Perry

Plymouth, Mass.

Your article makes the point that one of the primary goals of the U.S. National Health Information Network (or any other national electronic health care record system) is universal access—instantaneous and reliable access to health records for all physicians, hospitals, insurance companies, and any other health practitioners. However, the article curiously omits any discussion about how individuals will access their own medical records.

In today’s paper-dominated system and in my experience, access to personal records is at best a hassle (for example, asking for copies at every physician visit) and at worst nearly impossible (getting an electronic copy of a CAT or an MRI scan, for instance). But a national electronic system has the potential to make it immensely easier for individuals to access their own data.

If the National Health Information Network is to fully achieve all of its benefits, it must ensure that individuals have complete, reliable, and secure access to their own medical histories, not just those who provide health care services.

Michael Lexa

IEEE Member

Houston

On Being Systematic

As to Robert Lucky’s “Unsystematic Engineering” [Reflections, September], I believe the need for systems engineering extends well beyond high-technology engineering. In every aspect of our complex society, we hear references to “multidisciplinary activities,” “stake holders,” “unintended consequences,” “What are our options?”—all phrases commonly used by systems engineers. One hears these words used by city planners, wilderness managers, environmentalists, highway engineers, and others. Our society abounds with numerous examples of systems engineering.

San Francisquito Creek courses through expensive suburban neighborhoods, borders on Stanford University, separates two counties, and floods into a low-income area as it enters San Francisco Bay. It is also a viable creek for salmon. There is an ongoing effort to develop a plan for managing the creek that satisfies all these concerns. I participated in a seminar on restoring the watersheds of a northern California river system involving farms, forests, and rural residential areas. The purpose of the restoration is to improve the habitat for wild salmon—which has been compromised by the commercial activities of our society. We had a lively discussion when I raised the question of metrics for evaluating the efficacy of a plan; most participants were unfamiliar with systems engineering and such features as metrics.

For sheer complexity, consider planning and managing the system of transportation in any of our large metropolitan areas. Perhaps more manageable is forestry practice. A simple system would be to decide which is more productive: two 20-acre [8-hectare] clear-cut parcels or one 40-acre [16-hectare] clear-cut parcel. This list is endless and endlessly complex.

In 1991, a group of engineers from the aerospace industries and related industries formed a society for systems engineering, the International Council on Systems Engineering, INCOSE. I and a number of other early contributors tried to interest INCOSE in these “other” systems. We also tried to interest the practitioners of these systems in INCOSE. For a while we had some participants from the U.S. Department of Agriculture and from other federal facilities that were diversifying. But it never came together.

Like Robert Lucky, I realized, for example, that the young agricultural scientist or engineer would not find much of a career pursuing systems engineering in the name of agriculture. There would be no support from peers and managers for attending INCOSE conferences to learn about the practices of systems engineering. It was a foreign culture to them and foreign to the mainstream of INCOSE.

Yet I perceive that many professionals in all these fields believe that there is a real opportunity and benefit to our society if systems engineering were to become a widely recognized discipline applicable to every aspect of public planning and policy. I suspect, however, that systems engineering in these other applications will come to fruition only when our politicians and high-level policy planners have recognized the need.

Fred Martin

IEEE Senior Member

Philo, Calif.

Ready, Set, Launch

I was disappointed that, in your otherwise interesting article on Cape Canaveral [Spectral Lines, October], you were less than open about the causes of the increased number of hurricanes inflicting the place.

The article says, “Hurricane-related problems have become more frequent at Florida’s Kennedy Space Center in recent years, as the east coast of the United States transitions from a decades-long period of historically low storm activity.”

The relationship between anthropogenic climate change—global warming—and weather-related events is fully and deeply established in the scientific literature. Without a rational shadow of doubt, NASA’s problems with hurricanes are related to excess carbon dioxide.

I do know that there is, particularly in the United States, a well-funded and concerted effort to downplay the importance or the facts of global warming, but I do hope the IEEE, as a rational and independent professional organization, is free from the adverse political pressures that are sometimes applied to civil servants. It is far too important to be otherwise.

David Hirst

IEEE Member

Brighton, England

Software-Defined Radio

The concept of software-defined radio [Tools & Toys, October] is not entirely new. Using a single piece of hardware for different electronic circuits can quite easily be implemented with field-programmable analog arrays. In fact, FPAAs are well suited for technologies like the software-defined radio: take an FPAA, some memory to store the circuit designs for demodulation, and some digital logic and, theoretically, you could have a device that performs as good as the software-defined radio—and one that is easier to use, as it is all hardware.

There are, of course, some limitations to FPAAs, like the difficulty to emulate some important physical behavior of components. However, improvements to FPAAs occur on a daily basis.

We shall see which technology will be the first one ready to be adapted by the end-user market. Software-defined radio developers, you are warned.

Emmanuel Lesser

IEEE Student Member

Antwerp, Belgium

Strongly Attracted

Perhaps the best way of avoiding the issue of a shallow gene pool [“When Engineers’ Genes Collide,” IEEE Spectrum Online, October] is finding a mate who has “chemistry” with you. By this, I mean not just a physical attraction but an attraction that goes deeper and is defined by your own innate needs as is directed by one’s own genes.

The lack of this ability to “hear” the calling of one’s genes could be a result of an oversexed society that views a societal singular image that is all too stereotypical as more appropriate than the unique-faceted image that is developed by encouraging uniqueness and individualism—which is the key to any species’ survival.

The problem may be truly a social disease, where society at large is accepting a commercialized view of who we should be and paying less attention to the arts and the individualistic attitude that art inspires. Although I suppose another argument could be made that this is an anomaly constant with an evolutionary shift.

Mike Tocher

IEEE Member

Nipomo, Calif.

Fuel Efficiency

In “Stricter U.S. Gas Standards Stalled” [News, September], the 5 and 10 percent gasahol numbers are political numbers. When less than 15 percent ethanol is mixed with gasoline, fuel economy suffers and emissions rise. Studies in the 1970s showed that a mix with 15 to 22 percent ethanol results in increased fuel economy and reduced emissions.

Meeting the heavy CAFE (corporate average fuel economy) standards is now readily obtainable in engines using direct fuel injection. With carburetors, it was required to mix fuel with air. Air contains nitrogen, which inhibits combustion. The formation of nitrogen oxides is endothermic and reduces the cycle efficiency by up to 30 percent. Separating nitrogen from the air feedstock is easily done at the aspiration pressures of an internal combustion engine.

When excess oxygen is present, complete combustion becomes possible, resulting in SULEV (super-ultra-low-emissions vehicle) emissions and requiring substantially less fuel for the same output horsepower. The caveat is that elimination of a catalytic converter and building of a smaller block results in lower manufacturing costs and, hence, lower sticker prices for a greener, more fuel-efficient vehicle, with similar or better performance.

Robert Balch

IEEE Member

Woodland, Calif.

A Quicksand of Clicks

I caught the article about “click fraud” [“When Good Clicks Go Bad,” Technically Speaking, October] in my Google alerts for “direct navigation.” Don’t know much about the click fraud problem other than a lot of it is in Russia and overseas, not in the United States. I can tell you that most major parking services and the aggregators (Yahoo and Google) are heavy hitters when it comes to click fraud. They have spent millions of dollars setting up parameters to prevent this type of activity.

The problem of click fraud has been actually estimated to be less than 5 percent of all click-throughs, although there is no way of being sure. One thing is certain: if click fraud were rampant, Google and Yahoo wouldn’t be making billions a year in pay-per-click ads.

My purpose for writing, though, is to inform the author and your readers that “direct navigation” is not the result of links from “link farms” and other forced redirects, but from the simple act of the user typing in the browser search bar (or URL locater) the exact phrase of the product/service/info they want, and then adding a .com, .net, .org, .info, or other extension behind it. In other words, if you wanted to find out information on the spinach E-coli problem, you might want to type in your browser search bar “produceindustry.com.” (Warning: this author owns that domain, so you would end up on my parked page.)

Direct navigation, or “browser searching” or “item surfing,” is the process of using the domain name itself to find a product, not a way to lure people through preset links on Web pages. The process is a very much understood and valuable procedure in the investment industry, which includes some powerful players paying high prices for domains that describe their products and services. They know people who need wheelchair cushions will type in “wheelchaircushion.com” (another of this author’s domains) into their browsers to see what Web site comes up.

If the domain is owned by a domain speculator, it will be parked at a parking service that provides the owner with a free Web page and advertisers’ links. If visitors click a link to find out more information on the products they’re looking for (say, a wheelchair cushion), then the parking service gets paid by the ad aggregator (such as Google or Yahoo), and the parking service splits the revenue with the domain owner (me).

The whole point of legitimate domainers—those who invest in keyword domains by buying them and parking them or building them into relevant sites—is to make money by legal means, so that they can stay in business, promote their own products and services, and maybe sell the domain later to a company that finally “gets it.”

Any company that doesn’t own the keyword generic domains of the products they sell is not going to have a great online presence in the next two years. Most people (and retail buyers) by then will understand that if they want impact sprinklers, for instance, probably the most reliable supplier of impact sprinklers will be the company that owns “impactsprinkler.com.” They will bypass the search engine mess and just type that domain directly into their browser bar to see what comes up. That will be the quickest way for Internet users to find what they want.

Stephen Douglas

Manhattan Beach, Calif.

Terms of Appointment

The article “What’s Up, Postdoc?” [Careers, September] states that “British academics hold fixed-term appointments and are reevaluated at the end of the term, which can lead to their losing their positions.” This is not true. What is true, as stated in the article, is that tenure was abolished in UK universities during the reforms of the 1980s.

As a result, most UK academic positions may be terminated by either side on, say, three months’ notice, typically to take effect at the end of a semester (to try to prevent a faculty member from leaving courses in disarray). There may also be a probationary period of a few years, typically three. Details vary by institution.

Nevertheless, the usual expectation for most faculty members is to continue in their positions indefinitely unless transgression or dire performance forces them out.

Also, the lecturer grade in the UK (roughly equivalent to assistant professor in the United States in terms of seniority, responsibilities, and performance expectations) is formally regarded as a “career grade,” with many lecturers staying in that grade for much more than the six years that would be normal for a U.S. assistant professor, because, for whatever reason, they are not promoted to senior lecturer (roughly equivalent to U.S. associate professor).

Regular faculty appraisals (frequency varies among institutions) were introduced in UK universities as a result of salary reforms in the 1990s, and in some cases have been used as a basis for removing certain people.

Perhaps that is what your author was thinking of, but appraisals are a long way from fixed-term contracts. There are also some fixed-term faculty positions available in special circumstances, such as to fill a temporary vacancy caused by a regular academic being awarded a sabbatical or a multiyear fellowship, but these are not the norm.

Postdoc research-only appointments—much more common in the UK than in the United States in the EE field—are almost all fixed-term but are not regarded as faculty positions and usually carry no teaching or supervisory duties. A UK postdoc position is a typical stepping-stone between a Ph.D. and a faculty appointment.

Clive Woods

IEEE Senior Member

Baton Rouge, La.