Welcome to Orcmid's Lair, the playground for family connections, pastimes, and scholarly vocation -- the collected professional and recreational work of Dennis E. Hamilton

Recent Items
 
Re: Video Schmideo
 
OLPC: Bring on the How-Tos
 
I will Facebook no more Forever
 
The Seven Deadly Signs of Blogstipation
 
Publiversity: Public and Alternative Higher Educat...
 
IIW 2007b: Internet Identity Roundup
 
One Laptop Per Child: Labor Pains
 
WHS Fascination and the Social Grid: Distributed P...
 
OOX-ODF: Oh the Outrage, the Stridence, the Indign...
 
Democracy 2007: The First American Civil War

Happy Birthday, CACM!

The Communications of the Association for Computing Machinery (ACM), a.k.a Comm. ACM and C.ACM (pronounced ka-kum), is 50 this Month, resulting in a special anniversary issue: Comm. ACM 51, 1 (January 2008).  The content is behind a digital-library subscription wall, but I don't want to let a few nuggets of this retrospective issue go unobserved.

Update 2008-01-09: I've just learned that Communications of the ACM is being published on-line.  This anniversary issue is the trial edition.  I don't know if the issues will be made available to the general public, but the editorial content of the January 2008 issue appears to be on-line.  I have added appropriate links below.

Looking Backward

The Forum (letters) department has just one submission, an unusually quiet condition.  Executive Editor Diane Crawford explains:

Edsger Dijkstra wrote a Letter to the Editor of Communications in 1968, criticizing the excessive use of the go to statement in programming languages.  Instead, he encouraged his fellow computer scientists to consider structured programming [a term not used by Dijkstra in the letter - dh].  The letter, originally entitled "A Case Against the Goto Statement," was published in the March 1968 issued under the headline "Go To Statement Considered Harmful."  It would become the most legendary CACM "Letter" of all time; "Considered Harmful" would develop into an iconic catch-all.  Dijkstra's comments sparked an editorial debate that spanned these pages for over 20 years.   In honor of the occasion, we republish here the original letter that started it all. 

[with content of Dijkstra, Edsger W.  Go To Statement Considered Harmful.  Communications of the ACM 11, 3 (March 1968), pp. 147-148.  The original manuscript is archived as EWD 215 (PDF file).  EWD 215 lacks the final two paragraphs and the two references of the published paper.  These tie in some related work that does not alter the main point of the letter.]

The newer CACMs'  careless references are demonstrated here with Corrado Böhm being named Bohn in the reprint.

Looking Forward

The Communications is being redesigned and refocused again, inspired by the successful model of Science magazine.   Incoming Editor-in-Chief Moshe Vardi ends a recap of the different editorial regimes and different editors-in-chief with a look at CACM: Past, Present, and Future (PDF file), explaining the motivation for the redesign and reconfiguration.  

The centerpiece of the anniversary issue is the Voices section, a look into the future with this editorial introduction: 

"As difficult as it is to look even a few years ahead, these voices have managed to anticipate the needs of future users, even as they advance the the field's deepest scientific principles.  A notable theme is how much computing ultimately depends on trust, whether in an e-commerce transaction, a robot's behavior, a software agent's instructions, a network's architecture, a human's intentions, or the scholarship of a paper.  See how computation connects us all."

The thirteen contributors cover a broad territory, some paying homage to how they were personally inspired by their first encounters of CACM.  All of the voices deserve a reading.  Here, I want to feature the short essay by Carnegie Mellon's Jeannette M. Wing, "Five Deep Questions in Computing"  (PDF file).  Here Professor Wing identifies the computer science at the heart of all computing by the questions it undertakes:

• P = NP?
    
• What is computable?
    
• What is intelligence?
    
• What is information?
    
• (How) can we build complex systems simply?

The first question is about  a family of functions (the NP) requiring dramatically more complex (and costly) to calculations than a subset of the family (the P).  We have so far been unable to resolve with mathematical certainty whether the complexity is simply because we simply haven't found a simplification (or demonstrated its impossibility) that allows the question to be answered affirmatively.  The second question is about computations that are not possible at all, not simply too costly to compute.  These take us back to Alan Turing and the foundation of computation theory.

The remaining questions impinge more directly on the relationship of computation to human affairs, and vice versa.  I recommend Professor Wing's articulation of them to careful reading.

It is good that we continue to ask these things, because it is not clear that we understand the questions much less what it could possibly be like to answer them.

---

I don't look that favorably on the notion of computation as a natural science.  I'm also concerned that the popular embrace of "computational thinking" presumes far too much.  I think Wing's treatment of the topic (PDF file) is refreshing.  I fear it makes Computer Science be special in ways it is not.  I have the nagging feeling that computer scientists need to get out more.  Meanwhile, addressing the five deep questions seems like a promising way to deepen our grasp of how computation relates to and illuminates our understanding of human intellectual and linguistical activity.

This event is a reminder that the 50th anniversary of my first line of code is in May of 2008.  My first contact with anything like computer science was not until 1959 when someone at the University of Washington computer laboratory had an off-print of a paper on compiling code via an intermediate two-address notation.  I didn't encounter issues of CACM until a visit to the Seattle Central Library stacks in 1961.  I joined ACM immediately, attending my first ACM National Conference in Los Angeles that year.

Although my first full set of CACM issues were those from 1961, I later acquired all of the early CACM and Journal of the ACM issues on microfiche.

[Update 2008-01-10T02:15Z I've added information from a just-arrived e-mail announcing the inauguration of the on-line edition of CACM.  I don't know how freely-available the on-line articles will be beyond the initial trial with this anniversary issue.  I have downloaded the PDF version to my local machine where it is automatically indexed for full-text searching.  I also smoothed my summary of P = NP? a bit.] 

posted by orcmid at 1/09/2008 12:40:00 AM