From: ajayshah@cmie.ernet.in (Ajay Shah) Subject: BOO REVIEW: Managing Internet Information Services Date: Sun, 5 Feb 1995 11:41:17 GMT This, and other book reviews, can be accessed through http://www.cmie.ernet.in/~ajayshah The site is functional 9-7 GMT+0530. Managing Internet Information Services Cricket Liu, Jerry Peek, Russ Jones, Bryan Buus, & Adrian Nye with Greg George, Neophytos Iacovou, Jeff LaCoursiere, Paul Lindner, and Craig Strickland O'Reilly & Associates, Inc 630 pages, paperback, $30 ISBN 1-56592-051-1, December 1994 The table of contents of this book looked mouth-watering, and I started reading it ASAP. I will abstract the TOC here. * Chapters 1 and 2, introductory. * Chapters 3, services based on finger, inetd and telnet. * Chapters 4, 5 and 6: on setting up ftp, on wuftpd, maintaining an ftp site. * Chapters 7 and 8, on creating WAIS services and indexes. * Chapters 9 through 16, on setting up gopher services * Chapters 17 through 21, on setting up a web server * Chapter 22 through 25, on email based services, with an accent on Majordomo. * Chapter 26, on ftpmail services. * Chapter 27, on firewalls. * Chapter 28, xinetd * Chapter 29 and 30, Legal issues, and intellectual property. The book is very clearly focussed on answering the question "How do I setup one or more of the abovementioned services over the Internet?". It is assumed that you broadly know what the services themselves are from a user perspective, so there is no major effort in teaching you (say) how to use NCSA Mosaic. I think this specialisation makes perfect sense -- this is a book for information providers, not for consumers. In the entire process of becoming an information provider, there are two aspects: the sysadm problem and the authoring problem. This book broadly aims to help on both fronts. It does just about everything needed to make you a good sysadm. In several cases it displays a valuable wisdom about net.software. E.g. it teaches how to install wuftpd and xinetd, both of which might be unknown to the relatively uninitiated. On the authoring front it is a HOWTO i.t.o. of files and directories and languages etc. through which you will carry out your authoring efforts. The book also has a nice accent on security issues that might otherwise get glossed over by people who are thinking in a publishing or information dissemination frame of mind. You can, to be sure, install and run every one of these services all by yourself by hanging out on newsgroups, actually reading the READMEs, FAQs and other documentation on the net. Like other O'Reilly books, however, this is aimed at making it much easier and more accessible. In this objective it succeeds very well. I maintain both Majordomo and httpd here at CMIE. The rest of the book was quite new to me from an admin viewpoint. I carefully read the chapters on ftp and WAIS and came away very pleased with the practical confidence I got out of it. The chapters on the Web will be very useful to someone starting out on the process of estabilshing a web server. I will use this book when I come face to face with forms, server-side includes, clickable image maps, etc. One nice feature of the book is pointers to further reading on the net. I generally didn't see references to RFCs -- this is a HOWTO kind of book. Their treatment of HTML is pre HTML+ and does not address tables etc. At a ground level, in authoring, I have learned the need for a great deal of paranoia is writing good HTML, and in using multiple browsers plus htmlcheck and weblint as assistance in getting there. I felt the book was a little too sanguine on the subject of code quality of HTML. The issues of copyright (and, to some extent, of intellectual property) are dealt with in a US-centric way. These chapters were useful in defining the issues, but I'd have to look elsewhere to learn the ropes in a practical way. There is a fascinating discussion of the design issues involved in plans at O'Reilly for electronic distribution of encrypted books. I would actually like to know much more about the software mentioned here at a practical level. :-) While the authors try to address the question of evaluating which services one should install, I felt they don't work on this hard enough. E.g. at CMIE we currently have a httpd and a two mailing lists using Majordomo. I anticipate a little more growth in terms of a "querying" kind of mail server, WAIS based indexing of the materials accessed through our httpd, and perhaps an ftp server. This implies that many of parts of the book are superfluous for us - but the book does not work hard enough on clarifying the role of each of the strategies. Specifically, I suffer from the strong feeling (which many people on the net share) that the time for gopher has passed with the arrival of the Web. I feel gopher should have a prominent role in books of the history of the net, for it broke new ground and pioneered some concepts. But today, I don't see any role for menu-based querying in a world of hypertext. With hypertext, you can always create gopher-style menus by just creating enumerated lists, but when you do true hypertext you can go far beyond gopher-style delivery. At a tactical level, the fees being charged for the gopher 2.x server have hurt gopher. It's not asif the fees are unreasonable from the viewpoint of most organisations, but by introducing these fees, gopher loses out on the growth and participation obtained through guerilla activities. In how many sites did web servers pop up through the activities of sysadmins without the knowledge of top management? That's unlikely to happen so much for gopher now that there is a fee of ~ $500 a year to be paid for running a gopher 2.x server. Thus I think that the Web subsumes gopher, and was puzzled at the extensive treatment that this book presents for gopher (8 chapters, 128 pages). Deciding what services to provide is a central part of thinking about internet information services; I may be missing something essential, but I feel my concerns above are the most obvious set of questions about the future of gopher, and the book does not address them. In all, I think this is a unique book on the entire question of how to serve up information resources on the net. If you are setting up to become an information provider over the net, and if you are not an old-timer who already knows his way around fluently, then you need this book. I anticipate routinely turning to this book when faced with practical exigencies of managing information services over the net. Disclaimer: I requested and received a review copy of this book from the publisher, but I have no stake, financial or otherwise, in its success. -- ------------------------- Ajay Shah, Centre for Monitoring Indian Economy, Bombay ajayshah@cmie.ernet.in (9-7 GMT+5:30) http://www.cmie.ernet.in/~ajayshah <*(:-? - wizard who doesn't know the answer.