Profiles in computing: Allan L. Scherr

Reviewer: Paul E. Ceruzzi

These profiles are part of a special issue of the Communications of the ACM devoted to the 40th anniversary of the founding of that organization. (Other papers in the issue include a detailed history of the computing program at the Office of Naval Research, in many ways the ACM's godparent, and a series of personal reminiscences by various actors in the ACM's history.) Karen Frenkel, features writer for the journal, has selected three individuals from the 17 past winners of the ACM's Grace Murray Hopper Award, which since its inauguration in 1971 has recognized computer scientists who have made substantial contributions to the field before their 30th birthday. Each profile is based on an informal conversation between Frenkel and the individual and has a similar structure. Each begins with an account of how the person entered the field of computer science and then describes the origins of the award-winning project (which had to follow closely, given the age restriction of the award). The person then makes a few remarks on the current state of computing and how computer science has changed and concludes with a few comments on some current topics such as the Strategic Defense Initiative, personal computers, and artificial intelligence. The profile of Donald Knuth is, as usual, revealing and informative, although it does not have the depth of the interview he did for a recent issue of the Annals of the History of Computing [1]. At several places in the profile Knuth mentions the uncompleted volumes of his The Art of Computer Programming, with a clear statement that he intends to complete the remaining four of the seven planned volumes. However, both Frenkel and Knuth avoid the question that must be on many reader's minds, namely, How can Knuth finish the series, given all that has happened in computing since volume 1 appeared in 1968__?__ It is hard to imagine him sitting down and writing the next volume in such a way as to maintain the conceptual unity of the series. Yet that is what the profile implies. Brian Reid was honored for his development of SCRIBE, a document specification language for the first laser printer and the ancestor of many desktop publishing programs in common use today. The account of Reid's background as a computer hacker who was attracted to computing as a way to avoid the Vietnam-era draft is refreshing for its candor, although it is sure to offend some readers. It is as good an example as one could hope for to show that computer science is still in many ways an immature discipline. Allan Scherr is the polar opposite of Reid, and Frenkel no doubt chose to profile him for that reason. Scherr went directly from MIT to IBM, where he has worked for the past 22 years. He received the award for his thesis on time-sharing systems, which at IBM became the basis for the Time Sharing Option and later the Multiple Virtual Storage system for the 370-series computers. Frenkel's profile of Scherr reveals a man at the center of large projects, managing hundreds of people scattered across the globe and fashioning programs having thousands of line of code. For many (including, one suspects, both Knuth and Reid), such a job would be a nightmare, but Scherr thoroughly enjoys the work—so much so that he continues to do it even though, as an IBM Fellow since 1984, he is now free to do anything he wishes to. Taken together, the three profiles are interesting to read and do convey some of the diversity of computer science. They do not go much farther than that, nor do they fit in well with the issue's theme of conveying a sense of the ACM's history. The profiles, along with the other pieces in this issue, can serve as raw material for a genuine history of the ACM, should someone choose to write it someday.