As Peter G. Neumann discussed in his "Inside Risks" column "Optimistic Optimization" (June 2004) there is clearly a problem in the IT industry involving the failure of developers to consider the long-term implications of their decisions. But is the issue more that developers fail to identify them up front or to identify them at all? My experience is the latter, and the solution is to foster better modeling and collaboration skills among developers.
Developers should then apply these skills at the most appropriate times in the most appropriate wayssometimes up front, sometimes other times. There is clearly more than one way to mitigate this risk. Neumann hit on only one of them.
Implications not identified up front are typically not addressed at all until it is much too late. The message of the June column was that if you do not address the long-term implications up front, everything that follows may be suspect. However, Ambler is correct in saying that continual reassessment is appropriate. I have no disagreement with that.
Peter G. Neumann
Menlo Park, CA
A friend of mine is the son of the J.L. Austin cited in the "Technical Opinion" column "Toward Universal Literacy: From Computer Science Upward" by Hartmut Wedekind and Erich Ortner (June 2004). He said the following about the column: "How to Do Things with Words is the title of a book that has nothing whatever to do with literacy. My father was always referred to as J.L. Austin rather than John Austin and was certainly never taught by Wittgenstein." I can't help wondering whether the column was intended as a spoof?
Toronto, ON, Canada
John Langshaw Austin (19111960) was a prominent member of a philosophical school based in Oxford known as Ordinary Language Philosophy. Its members referred explicitly to their intellectual forebear Ludwig Wittgenstein (18891951) of Cambridge. In particular, Wittgenstein's influential book Philosophical Investigations laid the foundation for the analysis of everyday language, a cornerstone of the concept of literacy. In this sense, Wittgenstein was indeed a teacher of Austin.
Once something becomes indispensable, even the Internet, as in "Has the Internet Become Indispensable?" by Donna L. Hoffman et al. (July 2004), concerns about its availability, security, reliability, and pricing emerge, followed by the issue of regulation.
Internet services today are classified by the U.S. Federal Communications Commission as "information services" (telephone service, for example, is a "telecommunications service"). There is thus only minimal regulation of the Internet. Some ISPs (such as cable companies) view the Internet as an entertainment service and are under no obligation to provide any particular level of service to their users and sites.
The solution to how to protect the public interest while generating only minimal government intervention is to define Internet services as "common carrier services." Eli M. Noam of Columbia University described it this way: "[The originators of common carriage] intended to guarantee that no customer seeking service upon reasonable demand, willing and able to pay the established price, however set, would be denied lawful use of the service or would otherwise be discriminated against. In return for reduced discretion, a carrier obtained certain benefits, including limited liability for the consequences of its own actions."
Common carriage does not mean all customers are treated equally. Different levels of service (such as bandwidth) could be based on price. But, for any given level of service, all customers (including information providers) must be treated equally. Common carriage also does not imply universal service, regulated rates, or other such features.
Note, too, that this might affect the current terms of service used by many ISPs attempting to forbid a range of unwanted online behavior, including: bans on flaming; harassment, sending bulk unsolicited commercial email (spam); reverse engineering; bandwidth hogging; malicious conduct; and worse. Some items (such as bandwidth limitations) could be built into the contract, while others (such as control over content) might not be.
Fountain Hills, AZ
I made it four pages into the article by Donna L. Hoffman et al. (July 2004) only to find the authors were misusing the word. Food is indispensable. Water is indispensable. A Brand-X double-tall decaf latte with a hot twist, while perhaps a favorite part of one's morning routine, is not. Something being part of a daily routine does not make it indispensable. Routines can and do change with no ill effect on anyone involved. Were the coffeeshop to close, most people would be able to do without or with something else. That's not a feature of something indispensable.
Technical specialties often usurp common words to convey special technical meaning, but no special technical meaning was involved here. This was just a twist of language to imply something not proven by the article.
George P. Schell in "Universities Marginalize Online Courses" (July 2004) seemed to conclude that the expansion of institutions of higher education into online course offerings is doomed to failure. Not only do they fail to give the faculty associated with such courses appropriate incentives, in many cases they provide positive disincentives to developing and teaching online courses. The same disincentives also apply to any distance-learning initiatives (such as closed-circuit TV).
Schell ascribed this approach to the emphasis by university administrations on the traditional academic values of research, publishing, and, in recent years, attracting outside funding. Traditionally, teaching per se has been relegated to a secondary role, so any novel teaching initiatives fall even further outside the domain.
Schell further noted that preparing and teaching distance and online courses is far more labor intensive than traditional on-grounds courses, something seldom recognized by deans and department chairs.
Moreover, while many academic subjects might reasonably be offered at a distance (correspondence courses are nothing new) years of experience lead me to believe that some subjects cannot be usefully taught or studied at a distance, including: the physical sciences; mathematics; computer science; languages (at least if oral comprehension and speech are goals); music; and the visual and performing arts.
I see no way to usefully automate the homework assignments (good physics problems do not lend themselves to multiple choice), and grading large numbers of problem sets is incredible drudgery. Nor do I feel that difficult and nonintuitive subject matter can be imparted without face-to-face interaction between student and instructor. Laboratory exercises also do not translate well to online venues.
Some university administrators hope, in the name of cost-effectiveness, to expand enrollment beyond the campus without correspondingly increasing the number of instructors. I cannot be alone in feeling that applying business models to the delicate and difficult task of educating the late-adolescent-college-student population is counterproductive.
I suspect that many academics, especially in science, mathematics, and IT, are inclined to resist such attempts, on the enlightened grounds that it is a poor educational strategy, as well as on the more prosaic one that it would tend to reduce the demand for our services and therefore adversely affect our earnings potential.
Julian V. Noble
George P. Schell's article (July 2004) suggested that earning a college degree is not as difficult as it used to be. But will employers accept an online degree, even from an accredited college or university, as the equivalent of a traditional degree? Can any online education substitute for a traditional university?
Some Web-based programs promise a bachelor's or master's degree in days or weeks. Some "universities" even lack campuses or accreditation. Ultimately, such claims contribute to doubts about the credibility of all online courses. Most students enrolled in them may not only waste money and effort but are also likely to be rejected by potential employers.
Schell wrote: "For online courses, distance is not measured in miles but rather in the lack of face-to-face contact." The loss of social interaction among teachers and students in online courses is a significant drawback. A key reward of education is learning communication skills not based entirely on technology, even if it involves chat rooms, instant messaging, email, and teleconferencing. Face-to-face discussion sparks new thinking, teamwork, and leadership.
Online learning is incapable of handling such academic activities as course projects that involve laboratories and libraries. The issue of quality control is yet to be addressed, leading to yet more issues (such as how to evaluate student performance). What content might these programs consist of? What are their standards? Perhaps these question help explain Schell's assertion: "The long-term viability of online courses in the U.S. is in serious doubt if the viability is tied to the perceived academic value of developing them."
One has to ask whether an online degree will ultimately count for or against one's qualifications for employment. And how might graduates of Internet-only programs convince prospective employers that their online study or degree carries the same weight as a traditional degree?
Vancouver, BC, Canada
I am an industrial consultant amazed by the numbers of new graduates who are unprepared for the needs of industry, as Peter J. Denning discussed in his "The Profession of IT" column ("The Field of Programmers Myth," July 2004). My specialty is databases, and many new graduates not only do not understand how to deal with them but are unaware of the pitfalls of their own ignorance.
I can report from experience that a chemical engineering education (all my degrees are in chemical engineering) better prepares its graduates for industry than does a computer science education. (I practiced chemical engineering for several years before switching to computer science.)
The whole situation is intensely frustrating to me.
Regarding Ian Perry's Forum comment on counting lines of code (Aug. 2004), what we should instead be measuring is the elegance of code. Elegance is the metric of maximal comprehension and clarity (minimal obfuscation) from minimal lines of code.
Code is like algebra. We all learned to simplify formulae using the rules of algebra to aid understanding. The same principle works for code. To attain the most elegant code requires the code's evolution through many drafts, in the same way academic papers are written. The problem is that business haste often gets in the way.
West Yorkshire, England
©2004 ACM 0001-0782/04/1000 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2004 ACM, Inc.
No entries found