donner@bree.watson.ibm.com (Marc Donner) (06/30/90)
The USENIX association intends to publish books and monographs on the general topic of computing systems. The intended audience for these books is the community of system designers, builders, users, and scholars. Our intent is to publish material of lasting interest and importance, with an emphasis on actual systems. Subjects may include design, implementation, history, and analysis of real systems. While we are inspired by UNIX and UNIX-inspired systems, we do not expect to limit our attention to such systems in any way, as we see ourselves responsible to the entire systems community. We see several specific needs that we would like to satisfy and for which we solicit manuscripts. The needs fall in two areas - books in traditional styles and formats about topics important to the systems community and things new or unusual. Among things new or unusual, we are interested in exploring at least these ideas: Significant systems - many significant systems are documented, if at all, only in reference manuals or user guides. Journal publications often concentrate on narrow specific details, as is appropriate for focussed technical audiences. What is lost is the broad description of the design and its evolution, with consideration of the success and failure of specific features and lessons learned. Code - We are interested in exploring the possibilities of publishing code to read. A truism among the programming community is that one learns to write good programs by reading good and bad programs. Sadly, there is little code available to read. The recent interest in public-domain code and open systems has increased the quantity of high-quality source code available. Many open questions in the publication of code remain to be explored. The conventional codex form, long accepted as appropriate for literary works and texts, may not be the right one for programs. Very few experiments have been made with this form, something that we hope to encourage. The audience for published code includes serious students of systems, including both the undergraduate and advanced levels, and practitioners involved with development, modification, and analysis of actual systems. Important technical reports - many important technical reports, issued in small numbers by industrial organizations, research labs, or university departments, are not disseminated as widely as they merit. This is often because the originating organization doesn't have the resources or the will to publish them more widely and because the material is deemed inappropriate by commercial publishers because of its narrow scope or limited size. Many technical reports are too large for journal publication and too small for conventional book publication. We hope to provide a means of publication and distribution of the best of these. To submit a manuscript or proposal for consideration for the Monograph Series, send a copy to Monograph Editor USENIX Association 2560 Ninth Street, Suite 215 Berkeley, CA 94710 or send electronic mail to monographs@usenix.org
smb@ulysses.att.com (Steven Bellovin) (07/11/90)
In article <1990Jun30.013509.24393@arnor.uucp>, donner@bree.watson.ibm.com (Marc Donner) writes: > Code - We are interested in exploring the possibilities of publishing > code to read. > ... The conventional codex > form, long accepted as appropriate for literary works and texts, may > not be the right one for programs. It isn't. I'd hate to try to read any substantial piece of code without (at the least) grep to help me out. Years ago, on other (blue) machines, I relied on elaborate cross-references to read listings, but bemoaned their inadequacies. If you're going to publish programs, publish them electronically.