The Community for Technology Leaders

Web Accessibility Guidelines

Pages: pp. 5-7

To the Editor:

In "Improving Web Accessibility" (Jan. 2003, pp. 117-119), Eleanor Loiacono does an excellent job of outlining the need and methods available for addressing Web accessibility. Unfortunately, this article does not point to one additional tool: IEEE Std. 2001-2002 Recommended Practice for the Internet—Web Site Engineering, Web Site Management, and Web Site Life Cycle, which includes Web site accessibility among other best practice considerations. This standard, which became available in print in February 2003, also makes specific recommendations for five Section 508 requirements that do not have corresponding recommendations in the Web Accessibility Guidelines (1.0). IEEE Std. 2001- 2002 also addresses color blind and background/foreground contrast considerations, which have not been incorporated in current versions of the guidelines or Section 508.

In addition to improving access for persons with disabilities, following the guidelines improves visibility for Web sites in general. The alternate text for images and transcripts for multimedia facilitate indexing, search engine ranking, translation into other languages, and related access for those of us who are "temporarily able bodied."

Although not yet interpreted by the courts as applying broadly beyond US government Web sites, US government and other regulatory guidelines are already in effect. I am pleased that the IEEE Computer Society Digital Library has been actively tracking work in this area and facilitating accessibility through its use of HTML delivery of content, which is more flexible from a browser perspective than the popular proprietary alternative.

Jim Isaak, Chair, P2001 Internet Best Practices Standards Committee, Hollis, New Hampshire;


To the Editor:

It was surprising to find several errors in "Toward the Age of Smarter Storage" (Gary S. Robinson, Dec. 2002, pp. 35-41).

The article states, "Currently, hard drives can store up to 700 Mbits of data in a square inch…." However, that areal density was achieved in 1995, and the highest areal density used in current production disk drives is 100 times greater. The article also forecasts that technology improvements "…could increase disk storage densities to 40 Gbits per square inch by 2004." Since drives with areal densities of 70 Gbits per square inch were being delivered at the end of 2002, a reasonable projection for 2004 would probably be at least 200 Gbits per square inch.

James N. Porter, Mountain View, Calif.;

The author responds:

While my main objective in writing this article was to present readers with a 30,000-foot view of the evolution of storage, I had hoped that the addition of a few technical data points would be useful. Fortunately, this letter helped me to realize that content was missing that might clarify the data points used. The text should have said, "In the early years of the 21st century, hard drives can store up to 700 Mbits of data in a square inch and transfer data at up to 160 Mbits per second."

This letter also was instrumental in helping me identify and rectify another error that I had not noted earlier. On page 37, the sentence that says, "The introduction of write caching in the mid-1990s greatly improved RAID performance…" should have said "early 1990s," since EMC's Symmetrix architecture had cache in it from the very beginning.

I have asked the editorial staff to post a revised version of this article on Computer's Web site incorporating these changes so that future readers can feel confident that what they are reading is both timely and correct.


To the Editor:

I enjoyed reading Simone Santini's refreshingly nontechnical article about computing's misuse of language (The Profession, "Using Language More Responsibly," Dec. 2002, pp. 128, 126-127).

In an interesting juxtaposition, I noticed that this article faced a call for papers for Cyberworlds 2003, in which I found terms such as "cyberethics" and "cyberparenting" (which I'm guessing is not related to birth control).

It seems likely that "applet" was coined following the format of "wavelet," which is a translation from "ondelette," a French word meaning "little wave." Although I agree that "electron" is an elegant word, "chargelette" seems like it would be a legitimate alternative.

I had not previously questioned abbreviated terms such as HTML and PDA, but I agree that we should take a closer look at them. Even though it is precise and ontologically correct, the problem with Hypertext Markup Language is that it is simply too long. One way to fix it might be to try squeezing it into just one word, using a term such as "Linkomarkian." However, I'm not sure that this suggestion meets Santini's criterion of "elegance."

Andreas Siebert, Munich, Germany;

The author responds:

The benefit of this juxtaposition is that the Cyberworlds page provides an almost inexhaustible source of examples for the premise of my article.

I accept the term "wavelet" because it is etymologically sound: It is composed of "wave" and the common English suffix "-let," creating a word that means "small wave" (the French etymology is exactly the same). However, the problem with "applet" is that "app" is not a word.

A term like Hypertext Markup Language is, to all practical purposes, unusable and calls desperately for an initialism. The alternative is to give Hypertext Markup Language a "real" name, as were given to languages such as Ada and Pascal. One suggestion is "Diderot," from the 18th century encyclopedist, taking the encyclopedia as the first example of what was later both pompously and erroneously called hypertext. Another possibility is "katascope," from the Greek "kata-skopeo," to explore—but unfortunately, it doesn't offer a very elegant alternative either.

To the Editor:

Simone Santini has provided a timely article on an important subject. In addition to being depressing, the JavaBeans saga and the continual coffee terms demonstrate a poverty of language that reinforces stereotypes of the IT nerd.

We in the IT industry do have an unenviable reputation for being introverts who are poor communicators. We have our own inside jokes, and our IT guild seems designed to keep out the non-IT people—unconsciously, of course.

I spend quite a lot of time explaining new technology to senior managerial and technical staff. Every time I cover Java or Unix terminology, they look at me disbelievingly. I find myself justifying the terminology by explaining that "it's the West Coast culture," that "they don't know that we already have terms for these things," or that "it is an attempt to protect their knowledge by inventing a new language."



Like Santini, I also have noticed the use of the term "box" to refer to a computer system. It seems to me that this is an attempt at a manly display of competence, indicating that the computer holds no fears and indeed can be treated like a box.

Here in the UK, as an extension of the US, the terminology closely follows that of the Java/Unix culture. There is more resistance among senior staff, but to the kids coming out of the universities having been fed by the Internet, there is no alternative.

George Brooke, Linton, Cambridge;

The author responds:

I don't know if the social isolation of engineers is a characteristic peculiar to the West Coast of the US. However, as a relatively recent West Coast dweller, I see quite clearly that this area has all the premises for its development. Nevertheless, this characteristic, which results in, among other things, the poor quality of the language we use, certainly seems to have an eminently American, or maybe Anglo-Saxon—I'm not certain about that— origin.

When I was doing my university studies in Italy, the figure of the "nerd" or "geek," in the sense of the super-specialized engineer with no social skills and no interests outside the discipline, didn't exist. To some extent, an engineer was considered to be an intellectual and, as such, was supposed to have a well-founded opinion on everything from politics to avant-garde theater. It didn't matter what the opinion was, as long as the engineer could articulate it.

As to social skills, it is worth mentioning that the Italian term "educazione" means both educated and having good manners. The term "educato" means only well mannered, while "colto" (cultivated) is the equivalent of the English word "educated." In this case, the language is a reflection of a culture that sees education and social skills as inseparable. In such a culture, it is obvious that a "nerd," meaning someone who knows everything about a very technical and narrow field and nothing else, can't exist. However, with the emergence of isolated and unaware engineers, the poverty of the engineering language follows almost automatically.

As Europe embraces the American model, things are changing in education as in many other areas. However, I consider it a shame and a fault for Europeans to give up so easily on their rich and complex culture in favor of a more schematic and compartmentalized model.

To the Editor:

As a professor of computer science, I fought for decades against the decadence of our technical jargon, at least in the university environment, with no success. The situation in Italy is even worse than in the US because we take American words and reshape them into Italian monsters. Sometimes I wonder what would have happened here if the computer era had started in Japan.

Fabrizio Luccio;

The author responds:

I am Italian, so I am well aware of the sad lexical choices that the Italians have made. In fact, these choices are not limited to computers, but invade the entire spectrum of language from technology to politics.

I find it surprising that among the Latin languages, Italian seems to stand out for surrendering to English terminology. The term "computer," for example, is translated as "ordinateur" in French, "ordenador" in Spanish (but in Latin America, the term used is "computadora"), while the Italians invariably use the word "computer." Similarly, the mouse is called "souris" in French and "raton" in Spanish, but "mouse" in Italian.

Even when they adopt foreign words, other languages in general try to avoid declining them. The word "click," for instance, is used in Spanish but, invariably, the construction is "hacer click." Italians, on the other hand, seem to have no problem with declinations of dubious value such as "cliccare" or the horrible "chattare"—which is also inconsistent with Italian phonetics, since the sound that the Italians mean to reproduce would properly be spelled "ciattare."

I don't quite know the reasons behind this state of affairs. However, I agree that the "Anglicization" of a significant part of the Italian lexicon in the past few years goes beyond the normal process of cross-breeding between languages, mostly because the new lexicon and the rest of the Italian language are simply juxtaposed without any form of assimilation.


In his president's message, "Building a Better Computer Society" (Jan. 2003, pp. 8-10), IEEE Computer Society President Stephen L. Diamond thanked those members of the Society Board of Governors whose terms expired in 2002. This listing neglected to include Karl Reed. A professor at La Trobe University in Melbourne, Australia, Reed served two consecutive terms on the Society Board.

Computer regrets this oversight.

57 ms
(Ver 3.x)