by Peter H. Salus
Peter H. Salus is a member of ACM, the Early English Text
Society, and the Trollope Society, and is a life member of the American
Oriental Society. He has held no regular job in the past lustrum. He
owns neither a dog nor a cat.
My "Top 10" list for 1999 is at the end of this column. Resurfacing One evening in June 1986, I was sitting in the lounge of the Atlanta Hilton reading Jon Bentley's Programming Pearls. I was really enjoying it and I've continued enjoying the book until the present. But now there's a new edition of Programming Pearls. It's about 60 pages longer than the original (Bentley says there are three new chapters; I didn't bother comparing the sections), and that's a very good thing. In fact, with new editions of the Knuths a year ago and the publication of Kernighan and Pike's new book, there's a wonderful library of excellent programming instruction available. I'd personally add in Software Tools and Plauger's three volumes of essays. But it would be tough to read those and not emerge knowing something. Perhaps Bill Gates will buy sets for each of his programmers . . . There's also a new edition of Radia Perlman's Interconnections (also Addison-Wesley). I liked the first edition in 1992; this one is bigger and better. The 389 pages of the 1992 version have waxed to 537. Perlman has done a terrific job. Why might some folks work over Level 3 rather than Level 2? See page 518. But Perlman's last chapter, where this appears, is genuinely outstanding. Studded with sidebars headed "Real-World-Protocol," I loved every one. But the best was:
When my son was three I saw him in the hallway crying, holding up his hand, saying, "My hand! My hand!" I took his hand lovingly and kissed it a few times and said, "What's the matter, honey? Did you hurt it?" He sobbed, "No, I got pee on it." I can't top that. It's a wonderful book and full of useful information well presented. Windows? I got a copy of Iseminger's Windows 2000 Quality of Service and was eager to look at it. Had Microsoft "solved" the problem of machines going down while the user was working? Had the Raptor of Redmond figured out how to save work when the machine collapses? No! Microsoft has again redefined the universe, even though the ATM community has long had a definition of QoS that is widely used; and despite the fact that at the beginning of RFC 1633 (June 1994), Braden, Clark and Shenker state: The Internet, as originally conceived, offers only a very simple quality of service (QoS), point-to-point best-effort data delivery. Before real-time applications such as remote video, multimedia conferencing, visualization, and virtual reality can be broadly used, the Internet infrastructure must be modified to support real-time QoS, which provides some control over end-to-end packet delays. This extension must be designed from the beginning for multicasting; simply generalizing from the unicast (point-to-point) case does not work. Iseminger tells us: "Microsoft has built a set of components into Windows 2000 for network bandwidth management called Quality of Service." Once more, as so frequently in the past, the PR folks on the eastern side of Lake Washington are redefining English. As near as I can tell, a crusty wag of my acquaintance is right: "Microsoft uses 'service' in the veterinary sense." There are, I'm happy to say, several books on QoS that are both useful and accurate. I can't decide whether Windows 2000 Quality of Service is a joke or a tragedy. Network Matters If you're into VPNs, then you are into the tunneling protocol. Shea's compact volume L2TP is really good. Written when RFC 2637 was a draft RFC, it does a good job of explaining the place of Level2 in the stack and has a first-rate step-by-step guide to implementation. Shea's chapter on IPSec is quite good, but if you really need the nitty-gritty of RFC 2401, turn to Kaufman and Newman. They have produced a volume with just the right amount of detail and have included (for the truly hard-core) the entire RFC as an appendix (pp. 173240). There's also a fine list of references/further readings. I was quite wary when I received Clark's Designing Storage Area Networks. First of all, I've never been concerned with SANs; second, I've never been enamoured of Fibre Channel technology. This may well be because all the Fibre Channel standards are ANSI; none is ISO, despite the fact that ANSI X3-230 on the interface is 1994. But I found Clark's exposition excellent. There's also a new edition of Kofler's Linux book. It's well done, but more important than the Linux install, etc., which is now pretty straightforward, Kofler has got really good sections on KDE, GNOME, and the GIMP. It's too long and heavy (772 pages plus two CDs), but there's really a lot in it. The CDs contain Red Hat 6.0 and a bunch of applications and tools. The Year's Top Ten 1. Brian W. Kernighan & Rob Pike, The Practice of Programming (Addison-Wesley) 2. Charles N. Thurwachter, Data and Telecommunications (Prentice Hall) 3. Ken Arnold et al., The JINI Specification (Addison-Wesley) 4. David S. Bennahum, Extra Life (Basic Books) 5. Dorothy E. Denning, Information Warfare and Security (Addison-Wesley) 6. Paul E. Ceruzzi, A History of Modern Computing (MIT Press) 7. John McMullen, UNIX User's Interactive Handbook (Prentice Hall) 8. Brian Tung, Kerberos (Addison-Wesley) 9. David Wood, Programming Internet Email (O'Reilly) 10. Bruce Sterling, Distraction (Bantam Books) In a normal year, Hammel's GIMP (SSC) would have been listed. In another universe, there would have been another Rich Stevens, or the revision of a Rich Stevens. But his untimely death has deprived us of those possibilities. In the past twelve months the world has lost Jon Postel, Mark Weiser, and Rich Stevens. I think the entire Internet community30 years old and with over 160 million people worldwideis poorer now. Books reviewed in this column:
Jon Bentley
Radia Perlman
David Iseminger
Richard Shea
Elizabeth Kaufman and Andrew Newman
Tom Clark
Michael Kofler
|
|
Last changed: 13 Dec. 1999 jr |
|