copyright notice
link to the published version: IEEE Computer, January, 2015

accesses since December 1, 2014

THE $149 Gamble

Hal Berghel


This month marks the first time in 47 years that a printed copy of Computer is not being shipped as an included member benefit. This is not a good thing.

By the time that this issue of Computer appears in print, many of you will have noticed that it didn't for you! As we all saw in our 2015 Computer Society membership renewal form, a mailed hardcopy of Computer will no longer be offered as a member benefit. Optional print copies will now cost members an additional $149. To paraphrase the words of the legendary New Zealand philosopher-politician Fred Dagg, “We don't know how impropitious are these circumstances.”

I, for one, am of the opinion that this is an exceedingly bad idea. In my view, including a print copy of a flagship publication as a member benefit is part of the cost of doing business for any professional society worthy of the name. The present situation is the worst of both worlds: not only is the print copy withheld, but the additional membership charge for the optional print copy is insultingly prohibitive. $149 for 12 issues of any publication takes it out of the category of newsstand periodical and places it in the rarified air of things literary or pornographic. The members deserve a rethink on this decision.


I’m not a print Luddite. I’m not against online publishing. To the contrary, I envision the ultimate extension of future digital publishing to be an interwoven fabric of thought threads rather than a slowly expanding repository of static documents-cum-metadata – the latter being subsumed under what we now call digital libraries. As things now stand, information content continues to be driven by the information provider, rather than the information consumer. Even if the information retrieval is nonlinear (as with hyperlinks), the traversal remains prescriptive. For many years I’ve argued that this is an unacceptable restriction that ensures sub-optimal information uptake. There should be a way to distill information content into non-prescriptive presentations that are closer to the information consumer’s interests than authors’ brain dumps? On my account, the information consumer would attach information from one source to that from other sources and digitally re-assemble the information into a nnew thought frame that is more relevant to the consumer. Of course, a digital challenge that looms large is to retain links back to the original sources so that the information authorship metadata is always available and the thought threads become multi-directional from any node. (see A Cyberpublishing Manifesto).

Years back I developed a few prototypes of such a system using the alternate data streams built into earlier versions of Microsoft Windows file managers. If you're familiar with the ADS data structures you can imagine how bi-directional authorship chains might work. (Hal Berghel and Natasa Brajkovska, “Wading into Alternate Data Streams,” Communications of the ACM, 47:4, April, 2004) One can't implement such a model with a straightforward application of a cut-copy-paste desktop metaphor. Neither can you accomplish this with the conventional storage and indexing technologies present in modern digital libraries.

If my vision is reasonable, then the migration to digital print is inevitable. In the past six years, the publishing philosophy behind Computer has been directed toward enriching the digital content, including convergence formatting via PDF so that the layout of all content would be compatible with both print and digital formats. Some of the online content was also multimedia enhanced such as Chuck Severance's Computing Conversations videos. Software simulations and voice over slide presentations are now a staple of Computer. All of these features are noteworthy digital enhancements and innovative uses of bandwidth in service to Computer Society members. Kudos go to EiC Ron Vetter and the Computer staff for all of the value that has been added to the digital fabric of Computer.

So, If I'm committed to new digital publishing technologies, especially those which extend collaboration by creating new data structures in support of thought frames and non-prescriptive, non-linear information traversal, why would I defend the inclusion of a hardcopy publication as a Computer Society membership benefit? The answer gets at the heart of what it is to be a professional and a professional society. For want of a term, let's call this brand effusion.


Peter Denning has spent many years ruminating about what it is to be a profession (Peter J. Denning, “The Profession of IT: Who are we?,” CACM, 44:2, February 2001; Peter J. Denning and Dennis Frailey, “The Profession of IT: Who are we – Now?”, CACM, 54:6, June, 2011). He has this topic pretty well nailed down at this point. According to Denning, the four hallmarks of a profession are 1) a durable domain of human concerns, 2) a codified body of principles, 3) a codified body of practices, and 4) standards for competence, ethics and practice. He has over the past few decades validated this scheme against the profession of computing. By most accounts, the profession of computing and IT satisfies these criteria. But what is it to be a society serving such a profession.

Let's drill down a bit into Denning's analysis. He distinguishes between a discipline and a profession. Disciplines are fairly well-defined areas of scholarship. Traditional university areas of study are disciplines in this sense. Disciplines are also to be contrasted with crafts, trades and guilds, members of which share an affinity and perhaps an organizational membership, but are not bound together by a well-defined, widely-accepted body of knowledge that would qualify as a discipline in a diversified, well-rounded university or college. On this account, computer science and computer engineering would be disciplines within the profession of computing, while sundry tech support areas would fall within sub-professions, crafts, trades and/or guilds where licensing and certification rather than university degrees are the coin of the realm. The focus of a professional computing society worthy of the name must be on the profession as a whole, and not just one of the disciplines, trades, crafts, or guilds that make it up. Such inclusive societies must be mindful that their constituencies may have very different expectations in terms of services. Societies must also understand that if it is to be successfully inclusive, there must be some over-arching service or brand that all members can relate to. That includes symbolic, information-rich communication vehicles that is shared between all members.


I was on the ACM Publications Board through most of the evolution of the ACM Digital Library. The ACM DL – one of the first –if not the first - complete offering of its type for a professional society– sought to simultaneously satisfy several membership demands. First and foremost, it attempted to provide quicker delivery of scholarly research to interested members via networking. By the time of launch in the early 1990’s, most academic and institutional sponsors of the two professional societies were connected to the Internet, so the timing was right.

Second, it sought to reduce the cost of information delivery. By the time of launch, most of the ACM leadership felt that digital delivery was inevitable, and that such delivery would significantly reduce the marginal cost of publications to the point where it would be economical to bundle digital collections to consortiums of libraries, universities, etc. at a fraction of the individual subscription cost. Two very attractive consequences were also anticipated: first, the effort would ultimately prove to be cost-saving for the association, and second, as a result of the new cost structure the association would be able to offer more publications to serve increasingly smaller niche audiences. In addition, there were collateral advantages: SGML-derivative document structures could render easily for both print and digital output via Postscript/PDF/LaTeX and HTML (well not so much in LaTeX – but that's another story); and the reviewing system could be automated by means of the same digital infrastructure as the production system . Finally, the entire repository would be indexable and searchable virtually without limit. This experience has since been replicated many times by professional societies worldwide with widespread success.

The reasons for the success of such digital libraries are now pretty obvious to all onlookers: separation of production costs from the subscription base, lower overall marginal costs of production, amortization of expenses over an subscription base that can expand after production, ability to economically deliver information content to ever narrowing constituencies (i.e., with volunteer editors and reviewers, the cost of production to the organization is essentially in the layout of the content rather than the content), integration of peer review with the production process, and so forth. Digital Libraries translate into more cost-effective information delivery to members, pure and simple.

As great as digital libraries and online publications are, they don't satisfy our criteria of being information-rich communications vehicles sharable within affinity groups. Only a subset of a societies' membership relies on the digital library and digital push products as primary information feeds. Simply put, researchers and innovators in the included disciplines rely on digital libraries and websites far more than other members. So neither the DL nor the websites are ideal candidates to carry brand identity.


Part of what it is to be a professional is to network with professionals in related areas. And the success of such networking requires self-identification with the group. Sociologists explain this in terms of social identity theory. Our self-image is a function in part of the many groups with which we identify. Part of such self-identification and shared experience involves shared communication and association with a brand, and that’s where the print version of Computer comes in. In organizations that I’ve been associated with, print copies of Computer and CACM have always been in more-or-less continuous circulation. It is part of the professional hand-off process to potential members, colleagues and interested students: Computer gives an idea of the issues that are relevant to the society.

Computer remains the most visible brand of the Computer Society. Think of it as an organizational logo with content. The fact that it will still be available in digital form via IEEE Xplore, isn't the same: you can't circulate a copy in Xplore with an earmark or post-it attached to kindred spirits. You don't get attracted to an affinity group through indexable and searchable databases. Facebook is ubiquitous. SQL is not. Members of organizations associate themselves with objects of common interest that are portable across social situations. Computer qualifies. Digital libraries don't. Neither do membership cards. Digital libraries are the ideal vehicle for technical research publications, not for casual reading by colleagues. It is both inconvenient and impersonal to share mutually interesting information from opposite sides of a Web paywall. The barrier is too high to be effective for bonding.

Now for the coups de grace of my argument – Berghel's digital epidemiology hypothesis: Reading the print edition of Computer is 97.6% safer than reading it on mobile platforms because hardcopy is a poorer habitat for bacteria and viruses. Eliminating the print copy as a member benefit may lead to a sudden increase in E.coli dispersal in high tech offices globally. If there's a sudden outbreak of MRSA in Silicon Valley, don't say I didn't warn ya! (The validation of my hypothesis is left to the reader – but remember to wash your hands after reading – unless you're reading the print copy, that is!)

(If you're interested in my vision of digital publishing, see “A Cyberpublishing Manifesto,” CACM, 44:3, March, 2001 – (online); H. Berghel and D. Berleant, “The challenge of customizing cybermedia,” Heuristics 7 (2) Summer/Fall 1994), pp. 33-43 (online) extends this model to browser design that maximizes information uptake.)