|
|
|||||
Scientific Communication SIG |
|||||
| Home | Activities | Our mission | Newsletter | Online resources | Contact us |
Issue 12(2), April 2005
Beyond research: advanced scientific content needs advanced technology behind the scenes
Editorial: Overextending metaphors can lead to communication breakdowns
E-mail alerts on topics of interest
Feedback from the rechartering survey
SIG networking lunch at the annual conference: come talk to me!
Contact and copyright information
By Bret Freeman (bfreeman@vasont.com)
Publishing scientific research is not for the fainthearted. A scientist’s goal is the pursuit of truth, and with that comes a great responsibility. Any statement or theory that is presented must be supported with hard evidence that can stand up to scrutiny. Scientists who publish research in a scientific publication affect their fields forever, whether they are confirming a long-accepted assumption or challenging a modern theory.
The editorial process of publishing this scientific information also entails a large responsibility. Editorial departments that deal with scientific content face some unique challenges. In this article, I will identify these challenges and show how one large corporation used content management technology to successfully address them.
The first challenge of managing scientific content is the sheer volume of information. Scientific information is known for being wordy and complex; thus, the publications that report this information are rarely small and simple. For example, the McGraw-Hill Encyclopedia of Science and Technology (discussed later in this article) consists of approximately 15 600 pages. Managing and organizing this large amount of content manually would be extremely cumbersome and difficult. Furthermore, publishing has expanded from just print to include electronic media such as CD-ROM, PDF, and the Web, further increasing the volume of content that editorial departments must manage.
As the volume of content increases, so does the risk of inaccuracies and mistakes. Scientists have a responsibility to publish statements that are supported by carefully compiled evidence. In this environment, typos, spelling mistakes, and inaccurate data simply cannot be tolerated. Scientific publications are used for sharing information and educating professionals, so inaccurate or out-of-date information is useless, if not outright dangerous. For example, what if a new species of deadly snake was discovered in the rainforest and nature guides were not updated? What if a medical treatment was found to be dangerous or ineffective?
To be most useful, the information must not only be correct, but also referenced, cross-referenced, and indexed correctly. Scientific publications are often particularly time-consuming for editorial departments to manage because of the complex referencing, indexing, footnoting, and other details involved in putting together the final product or products. When a scientist is using a publication such as a journal or encyclopedia, they need to see how the facts in a particular study or article connect to other studies, as well as to other authors and their hypotheses and theories. If those connections or links are broken or inaccurate, then once again the reference material may become useless.
Finally, the accuracy must extend beyond just the text. Another characteristic of scientific communication is that it often includes more tables, figures, and equations than other forms of information. Scientists must often present data in all these various forms to communicate their facts or support their theories. And as was the case for the other challenges mentioned thus far, the calculations must be correct, referenced at the correct locations within the text, and updated for the information to be useful.
One way that editorial departments are addressing these challenges is by taking advantage of content management technology. In the current technology community, content management is defined in many different ways. To make things less confusing, here is a quick overview of the various definitions out there:
To achieve the best results, you must thoroughly research your needs and the corresponding properties of the content management system. For maximum content reuse, the system should be able to handle all the different kinds of content within your documents—text, images, equations, and tables in McGraw-Hill's Encyclopedia—as reusable chunks. Working at the component level is a key feature for efficiently reusing and repurposing large amounts of content while maintaining its accuracy. The system should also offer advanced link management capabilities for creating complex cross-references and indexes. This means the system should have the intelligence to know what content is linked to other content at all times, and should never allow you to delete a piece of content that is linked to another piece of content without taking appropriate action (e.g., removing links to the deleted content). Finally, ensure that potential vendors have experience working with your type of information (here, scientific material) and ask them for client references. Finding a system to manage your organization’s scientific material will dramatically improve the quality of your editorial process, freeing up time to focus on the actual scientific research and on communication of this research.
Editorial departments that work with large amounts of scientific content have found that single-source content management systems work best to address their unique needs. This is due to the systems’ ability to publish to multiple channels, protect content accuracy, manage complex components (e.g., equations, tables, and figures), and ensure that no links (e.g., cross-references, footnotes, indexes) break or are omitted. One well-known organization whose revenues depend on high-quality scientific content is the McGraw-Hill group of companies. When McGraw-Hill’s editorial department ventured into the world of single-source content management, it dramatically improved efficiencies and content integrity.
Since James McGraw and John Hill merged the book departments of their two companies to form the McGraw-Hill Book Company in 1909, McGraw-Hill has become synonymous with excellence in information products and services. McGraw-Hill’s education division has become recognized around the world as a leader in developing educational materials and professional information products.
One of McGraw-Hill’s best-known publications—the Encyclopedia of Science and Technology—is considered a classic. It is the most authoritative and comprehensive guide to both well-known and new science that is currently available. At 20 volumes, and with 15 600 pages and 12 000 digital images, the information included in the Encyclopedia had become unwieldy to manage and update with the systems currently in place. Thousands of large, interrelated SGML files held the Encyclopedia’s content for its print and CD-ROM versions. However, even though these files used the same document type definition (DTD) and tagging scheme, many contained similar content.
Maintaining separate SGML files for the print and CD-ROM versions of the Encyclopedia created twice the work for the editorial department whenever additions or changes to the Encyclopedia’s content were required. For example, when an article appeared in both the print and CD-ROM version of the Encyclopedia, McGraw-Hill had to create two separate SGML files for that article. In addition, it was difficult to manipulate these numerous SGML files because they were not logically connected in any way. As a result, authors had to write and run custom scripts each time they wanted to process these files, which was a cumbersome and highly error-prone process. McGraw-Hill also wanted to produce a Web version of the Encyclopedia, which would have further complicated the process. The company recognized the need to develop more efficient ways to manage the Encyclopedia's content across three different media—print, CD-ROM, and the Web.
Manually created cross-references posed another challenge. A topic number identified the articles in the SGML file, but these numbers were written on index cards. Cross-referencing information was even more difficult because the authors maintained a list on paper with hundreds of thousands of link IDs to articles, tables, figures, and equations. When editors had to insert a reference to one of these components, they had to manually look it up in the list. The authors were using TeX, a typesetting system that is commonly used for scientific communication, to represent equations in the encyclopedia, but it was difficult to manage and integrate the TeX files within the SGML files.
McGraw-Hill acknowledged that it faced numerous hurdles in managing and creating the content for its print and CD-ROM encyclopedias, and that more difficulties would be added when it began the Web version. They decided that the solution was a technology application that would streamline the process of updating the Encyclopedia's content and provide enough flexibility for authors and editors to access the vast body of information and extract content for multiple projects, without having to expand the existing staff. As a result, the company began searching for a content management system that could handle multi-channel publishing and solve these challenges.
The selection process for a content management system turned out to be relatively easy for McGraw-Hill. Roger Kasunic, Vice-President of Editorial, Design, and Production of McGraw-Hill, was impressed by the Vasont team and their understanding of the editorial process. Their knowledge made him feel much more comfortable with putting McGraw-Hill’s content in their hands.
During the analysis and initial implementation phase, the Vasont implementation team assisted McGraw-Hill in troubleshooting SGML errors and fixing discrepancies between the files so that the resulting information was “clean” and ready for entry into the new system. After the data had been loaded into the Vasont system, the team implemented a global change capability so that updates could be applied simultaneously between multiple products that shared the same content. For example, when an author made a change to an article in the print version of the Encyclopedia, that change was instantly reflected in all other versions of the encyclopedia.
Kasunic described several other key features of the product. For example, Vasont let the editors and authors view the content in multiple existing versions of the files so they could compare versions and resolve any discrepancies. The software even allowed editors to examine the list of author names to ensure consistent spelling, designation, and use of middle initials.
Another challenge that the Vasont team solved was how to incorporate McGraw-Hill’s paper-based index of topics and articles into a software repository. This capability enabled editors and authors to quickly link to related articles. Because Vasont can also dynamically build and maintain embedded links, this simplified the once arduous task of inserting the corresponding references. Now, the software lets authors easily pick from a correct, constantly updated list that is directly available within the system.
Vasont is designed to block the deletion of a component when references to it exist, until the references themselves are deleted or redirected. The system provides a concise list of the references, which makes it easy to manage links and references effectively. Vasont also solved the challenge of using TeX to create equations: the production staff can still use this external application to create the equations, which Vasont then integrates into the content.
Within 12 months, McGraw-Hill saw a full return on investment in terms of improved efficiencies and content integrity. The content management system provided the following benefits:
Based on these results, the company has expanded its use of the system. McGraw-Hill has implemented an application service provider (ASP) version of Vasont to manage and publish the contents of its Current Medical Diagnosis and Treatment publication, which comprises nearly 2000 pages of text and graphics. The ASP version enables authors and production staff to access the system through a Web browser interface.
Bret Freeman is Vasont Systems’ (www.vasont.com) Eastern Regional Sales Manager for Vasont® Content Management Solutions. Bret has more than 10 years of experience in the information management and publishing industry and has delivered numerous presentations and training courses to various groups in the United States, Canada, and the United Kingdom. He has extensive knowledge of content management strategies and complex multilingual outputs.
by Geoff Hart (ghart@videotron.ca)
As scientific communicators, we rely heavily on metaphors and other image-heavy text to communicate complex concepts. This can be a powerful tool for simplifying complexity, but unfortunately, significant problems arise when we lean too heavily on those images and forget about the consequences of carrying a metaphor too far. The problem lies in our attempt to make the complex simple: some concepts really are too complex to express simply, and oversimplifying them can create vexing communication breakdowns.
Paul Berg, interviewed in the April 2005 issue of Discover, illustrated the kind of problem that can occur when a word or image acquires a connotation that conceals the real meaning and the underlying complexity associated with that meaning: "For example, the word cloning. What do you think resonates in the minds of the general public when a scientist says he wants to clone stem cells? Well, right away, they associate it with cloning people. But we're all agreed we shouldn't clone people... But the word cloning just triggers fears. What do you see? The Boys from Brazil, Star Wars."
A recurring example involves a cliché that has become part of the repertoire of popular science writing about genetics. One of the fascinating recent discoveries in molecular genetics is how strongly some genes have been conserved in different species despite thousands of years of evolution. Unfortunately, evolution is a complex science, and genetics even more so. Inevitably, this makes writing about both at the same time difficult. To simplify the task of communicating this complex interaction between evolution and gene conservation, authors usually try to demonstrate how similar humans are to other species by comparing our respective genomes. Thus, in studies of primates, you'll often hear the claim that chimps and humans share 99% of the nucleotide sequences in our respective genomes. This number sounds high, and indeed, each of us would be deliriously happy to score this well on a university genetics text—and it's this visceral understanding of the number that makes the comparison so effective.
But the initial impression generated by that clear image fails under closer examination and proves to be dramatically misleading. In this case, it's not the number that I dispute, but rather the effect of citing it. Here's the problem: there are something like 3 billion nucleotides in the human genome. A 1% difference that seems insignificant on a test becomes very significant indeed once you understand that it represents a difference of 30 million nucleotides. Of course, nucleotides aren't genes, so let's consider genes instead. Current estimates suggest that there are something on the order of 25 000 genes in the human genome, which means that the aforementioned 1% difference represents a difference of 250 genes between humans and chimps. Worse yet, a phenomenon known as alternative splicing means that each of those 250 genes may be responsible for the production of two, three, or more proteins that have important and sometimes even crucial regulatory effects on our bodies.
Clearly, there will be no hybrid chimp–human embryos anytime soon, despite that 99% similarity between us.
Another familiar example from evolution has been the anthropomorphism that organisms evolve to adapt to their environment. In fact, classical evolution tells us that organisms unable to adapt die before they can reproduce, and only those that can adapt to changes in their environment will survive to pass on their genes. It is the genes that survive through the generations and the species, not individual organisms, that evolve. Richard Dawkins famously captured this image in his book The Selfish Gene, which clearly and convincingly describes how genes seem to "selfishly" compete with each other for survival, sometimes at the expense of the organism's own goals and desires. Dawkins and his colleagues revolutionized the study of evolution by demonstrating the importance of this process.
Yet genes clearly have no consciousness, despite the implications of the word "selfish". Forgetting this can mislead us into believing that all organisms are guided solely by their genes. While this may well be true for lower organisms, and is demonstrably true to some extent for even the highest organisms, this image can lead us to miss an important point: that humans may well be the first species on Earth that can think beyond the simple urging of our genes and make decisions (such as genetic engineering) that directly contradict the supposed urgings of our genes. Moreover, larger processes that are more difficult to quantify, such as social structures (e.g., religion, laws) and learned behaviors (e.g., the teaching of our parents), can exert behavioral influences every bit as powerful as those of our genes.
There are many other misleading word choices I encounter frequently. Consider, for example, the "no artificial chemicals" blazon on "organic" products. The word "organic" clearly communicates the fact that the product contains none of the really scary artificial (man-made) compounds that lead to the 300-word ingredient lists for even the simplest Kraft products. Yet none of us believes that "natural" chemicals such as arsenic, lead, cyanide, ergotamine, and digitalis are safe simply because they're natural. Similarly, there is much talk in agronomy about the merits of "organic" fertilizers versus "chemical" (inorganic) fertilizers—as if organic fertilizers are not, themselves, composed of chemicals. The simplicity of these and other wordings does make an important point clearly, but that's not good enough.
As scientific communicators, we must beware the temptation to simplify "facts" to the point that they become misleading. The high similarity between chimps and humans fails this test, and as a standard for comparison, it's tempting to propose (as I have done in the past) that it should be retired from our vocabulary. Similarly, those of us who must write about evolution and genetics learn to avoid oversimplifying evolution as either a process of conscious change by individuals or as malicious and selfish control by sentient genes. Those of us who write about chemistry must be careful to use precise terms such as "inorganic" or "artificial" rather than "chemical" and equally careful to use emotionally loaded words such as "natural" judiciously.
But there's a better solution than merely to avoid using simplistic wordings. Instead, I would propose an easy strategy for preserving the power of simplicity without sacrificing clarity: introduce concepts simply, but just as any good manuscript begins with an introduction and moves onwards to elaborate on that context, so must we build on that simple introduction and use it as a tool for delving deeper and revealing the true underlying complexity. That's a more challenging communications task, of course, but it's also much more satisfying when we succeed.
[Based on an article in Tara Calishain's wonderful Research Buzz newsletter (www.researchbuzz.org/sciencegov_offers_alert_service_.shtml)]
Interested in keeping up on the latest developments in a particular area of science? The U.S. federal government's science portal, Science.Gov (www.science.gov), has introduced an e-mail "alert service" that lets you specify the areas you're interested in so you can receive ongoing updates on the latest information. Here's how it works:
E-mails should arrive every Monday morning if there are any "hits" for your search results.
by Jean-luc Doumont
[Previously published in IEEE Prof. Commun. Soc. Newsletter 45:5, 13, September/October 2001]
Getting the message across: that's how participants of my training programs almost always define effective communication. Few, however, can readily provide a definition of the word “message”, though most, after some thought, will agree that a message is to information what conclusions are to results: it covers the interpretation of the information more than the information itself and, as such, it is normally both audience-oriented and purpose-oriented. These definitions are not without consequences, among which the following two deserve particular mention.
First, and as a kind of “principle zero” of effective communication, a document or presentation should have messages in the first place. While such a statement should be obvious, we can probably all remember documents or presentations that left us with the unpleasant question of “what's the point?” The document may have contained a lot of information, but unfortunately lacked any explicit message.
Second, and probably far less trivial, effective documents or presentations place the message up front and the supporting elements afterwards, rather than the other way around. As I proposed in my last column (Columbo, not Poirot, in the February 2005 issue of this newsletter), the apparently commonsense storytelling approach, with conclusions at the very end, works poorly in professional contexts. Stating the message first works better for full documents or presentations, but also for individual sections or paragraphs.
Technical writers usually agree that effective paragraphs start with a topic sentence, which orients readers as to the topic and possibly the structure of the paragraph. A paragraph without a topic sentence is often misleading at first reading and does not allow readers to retrieve information easily through subsequent skimming. Such paragraphs, which lack a unifying theme, are unfortunately frequent in the Materials and Methods section of scientific literature or research reports.
I would further argue that an effective topic sentence states not only the topic but also the message—that is, the one thing that the audience must remember about the topic. In other words, it states not only the what, but also the so what?
Paragraphs that start with the topic but not the message are frequent in the Results section of scientific literature or research reports. In chronological fashion, the first sentence refers the reader to a figure by describing it (the what, as in “Figure 5 shows the evolution of temperature as a function of time”), further sentences make observations, and the very last one concludes (the so what?, as in “In other words, the temperature increased roughly linearly with time”).
Not only is the so what? infinitely more interesting for the audience—it also usually contains the what, making a separate statement of the latter uselessly redundant. In the above example, the paragraph can more usefully state the message in the first sentence, with a simple reference to the figure, as in “The temperature increased roughly linearly with time (Figure 5).” Subsequent sentences would elaborate on (discuss, detail, illustrate, ...) the point, in theorem–proof fashion.
Turning chronological paragraphs into message-first ones requires more than just turning the last sentence into the first one: it requires a change of logic, perhaps conveyed chiefly through the various signal words. A chronological structure, leading to conclusions as an end point, is essentially a logic of therefore. In contrast, a message-first structure is more typically a logic of indeed.
Dr. Jean-luc Doumont teaches and provides advice on professional speaking, writing, and graphing. For more than 15 years, he has helped audiences of all ages, backgrounds, and nationalities structure their thoughts and construct their communication.
by Geoff Hart, Manager, Scientific Communication SIG (ghart@videotron.ca)
In the previous issue, I posed a series of questions leading up to the rechartering exercise that STC will be asking all communities to perform, including our own Scientific Communication SIG. I received very little feedback, but here's what has been said thus far:
by Geoff Hart (ghart@videotron.ca)
As in previous years, there will be no formal SIG business meeting at the annual conference. I've scheduled several of these over the past 5 years, and the turnout has always been fewer than three SIG members. Although I've had some pleasant conversations as a result, it's been a poor use of my time from the perspective of the SIG. Given how many other things I'll be doing at the conference this year and the disappointing turnout in the past, I've decided not to schedule a formal SIG business meeting. (For example, as manager of the Writing and Editing stem, I'm likely to be running around like a headless chicken for much of the conference. Amusing to watch, perhaps, but not very productive in terms of our SIG.)
Instead, I'll be hosting a table at the SIG networking luncheon (currently scheduled for Tuesday at 12:15, but check the conference schedule when you arrive). I'd love to meet you and learn your opinions of our SIG and what needs the SIG could or should fill. So come see me at the SIG networking luncheon, and we'll turn that into an ad hoc business meeting if there's enough interest from SIG members. If you can't attend that lunch, try catching me in between duties. We can always arrange to meet for coffee and conversation at some more convenient time.
by Geoff Hart (ghart@videotron.ca)
For roughly 5 years now, I've been kept quite busy working on the Scientific Communication SIG. I started out editing and publishing the newsletter, fell into the SIG manager role when nobody else was willing to take on this responsibility, put up a primitive Web site when it became clear nobody else was willing to do this, and even took on the role of managing our e-mail discussion group.
This has all been very satisfying work, but I'm increasingly busy now that I'm working as a freelancer, and I simply can't continue to sustain this level of commitment. More to the point, several key roles such as managing the SIG and developing our Web site simply aren't getting the attention they deserve. For that reason, I will, albeit with regret, be resigning all of these roles at the end of this calendar year. My goal is to leave you with a replacement who will ensure that all SIG members get their money's worth from the money allocated to the SIG. Of course, if nobody is willing to take over some or all of these key roles, there won't be a Scientific Communication SIG next year to join.
For our SIG to remain viable, we need several volunteers willing to share the load:
So: It's been a pleasure wearing all those hats these past 5 years, but now it's time for some fresh blood. Or, if nobody steps into these roles, it's time to quietly wind down the SIG's activities and move on to new things.
If you're willing to take on any of these roles, please contact Dana Chisnell (dana@usabilityworks.net), STC's advocate for SIGs and other communities, and offer your services.
"A new study shows that large doses of Vitamin E do not protect against heart attacks and cancer, and might actually raise the risk of heart failure. The study was published in this month's Journal of Things that Scientists Told You to Do Last Month That Turned Out to Be Harmful This Month."—Dennis Miller
—
"[In Silent Spring, Rachel] Carson's achievement was to take the science of DDT poisoning and the intricacies of ecological studies and write them as a story that speaks to us."—Jack Hitt, A gospel according to the Earth
—
"The task of science is both to extend the range of our experience and to reduce it to order, and this task presents various aspects inseparably connected with one another. Only by experience itself do we come to recognize those laws which grant us a comprehensive view of the diversity of phenomena. As our knowledge becomes wider, we must even be prepared therefore to expect alterations in the point of view best suited for the ordering of experience."—Niels Bohr, Atomic Theory and the Description of Nature (translated from the Danish)
—
"A man said to the universe: 'Sir I exist!' 'However,' replied the universe, 'the fact has not created in me a sense of obligation.' "—Stephen Crane, writer (1871–1900)
—
"The road to wisdom? Well it is plain and simple to express: Err and err and err again, but less and less and less."—Piet Hein, poet and scientist (1905–1996)
—
"The rung of a ladder was never meant to rest upon, but only to hold a man's foot long enough to enable him to put the other somewhat higher."—Thomas Henry Huxley, biologist and writer (1825–1995)
—
"Some sense of judgment, some degree of participation in science is a necessity if alienation is not to dominate the response of most people to this world of swift technical change. Some view of the experiences that underlie chains of order is needed if authoritative conclusions are to make sense. Evidently the experience cannot be complete; thoroughness defines the professional and not the general reader. It is the very richness of the experiences of science that limits the expertise of every working scientist to a rather narrow domain."—Philip and Phylis Morrison, The Ring of Truth
—
"When one door closes another door opens; but we so often look so long and so regretfully upon the closed door, that we do not see the ones which open for us."—Alexander Graham Bell, inventor (1847–1922)
—
"… science is not done in a vacuum. It is done in a social context, and the results of science have important implications for society, even if it is simply providing a general understanding of how we humans fit in the cosmos. Thus, simply producing new knowledge, without making any attempt to help disseminate it and explain it, is not enough. I think one cannot expect every scientist to spend time on the effort to explain science. But in a society in which the science is of vital importance and also in which many forces are trying to distort the results of science, it is crucial that some of us speak out."—Lawrence M. Krauss, Questions that Plague Physics
The Exchange is published on behalf of the Scientific Communication special interest group of the Society for Technical Communication. Material in the Exchange can be reprinted without permission if credit is given to the author and a copy of the reprint is sent to the editor. Please send comments, letters, and articles to the editor.
Editor, publisher, and SIG manager:
Geoff Hart (ghart@videotron.ca)
Webmaster:
Geoff Hart (ghart@videotron.ca)
© 2005, Society for Technical Communication (901 North Stuart St., Suite 904, Arlington, Virginia 22203-1822 U.S.A., 703-522-4114, 703-522-2075 fax, www.stc.org)
Return to the main STC Web site: www.stc.org