news@UK

The newsletter of the UK Unix Users Group
Volume 11, Number 5
December 2002


News from the Secretariat

Jane Morrison

We are currently bringing together the details of the Winter Conference which will be held on 18th and 19th February 2003 in London. The full information and bookings booklet should be enclosed with this mailing.

This will be the first big event of 2003 and the programme suggests that this will be popular with members.

The LINUX 2003 Developers conference will be held in Edinburgh: please see the enclosed leaflet.

Do you need to buy some books? Remember that your UKUUG membership subscription allows you to receive a discount of 21.5 per cent on O'Reilly publications -- see the UKUUG web site for a list of titles available and prices.

The annual subscription invoices will be sent out in January, please look out for your invoice and as always prompt payment will be gratefully received!

Unbelievably it is that time of year again and I would like to wish you all a very Happy Christmas and a Peaceful New Year.

The Secretariat will be closed from 20th December to 2nd January 2002.

Please note the copy date for the next issue Newsletter (March 2003) is 21st February 2003.


Announcement: NordU/USENIX Conference

UKUUG Administration

The fifth NordU/USENIX Conference (NordU2003) will be held between February 10th and 14th 2003. http://www.nordu.org/NordU2003/


Letter: UKUUG and Open Source versus Free Software

Phil Hands

Just got the latest UKUUG newsletter [ed: that was the July issue] , and I noticed mention of the UKUUG Open Source Award 2003.

This made me wonder why the UKUUG seems to have embraced the term Open Source in apparent preference to the term Free Software.

Is this actually policy, a case of going with the flow, or is it something where case by case decisions are made (i.e. is the money for the award from someone who favours the term Open Source)?

There are a few reasons why I ask:

Personally, I have come to the conclusion that there is no more clarity provided by the term Open Source than by Free Software, moreover, the term Free Software gives one an opportunity to explain that the licences that qualify provide certain freedoms that proprietary licences are generally designed to deny.

From the point of view of the UKUUG, it seems that it (and we) have been around long enough to know that Open Source was just created as a marketing gambit, and with widespread abuse of the term (i.e. Microsoft mumbling about ``Opening their source to industry partners, governments etc.'') it might make sense to revert to the term that has a longer history.

The term Free Software is less open to abuse, because proprietary vendors are never going to try and misappropriate ``Free'', because they do not want the Free Beer misunderstanding any more than we do.

Bruce Perens, who helped ESR come up with the term, now admits it was a mistake, and has reverted to calling Free Software, Free Software.

The main proponent of the Open Source term, ESR who may have written the seminal Cathedral and Bazaar paper, but was writing it about fetchmail (a disgusting abortion of a program IMNSHO).

If the UKUUG does not already have a policy on this issue, I would like to suggest that the only time that using `Open Source' should generally be used in papers originating from the UKUUG should be in a form of words such as:

Free Software (also popularly known as Open Source software or OSS) when Free Software is first mentioned in that document. Obviously, for works written by others, who feel strongly that Open Source is the term they want to use, a more relaxed policy could be adopted, although I would prefer it if each such were first asked if they minded a global replacement of the term, with the above style of introductory note.

Phil


Services to Members

Alain Williams

What can UKUUG do for you ?

Would you like your own SSL certificate; but you cannot afford it? Would you like better control of your DNS domain, and use of a backup mail exchange server; but cannot afford it?

The UKUUG is looking to see how it can better serve its members. I think that there are several things that are difficult (and expensive) for an individual to set up, but are quite affordable when done in reasonable numbers. So, I have come up with a few ideas where we might do this; what I want from you is feedback: would you snatch my hand off if these were offered, or would you not be interested?

How much would this cost you? Nothing: it would be covered by your UKUUG subscription. However we would not want it to be abused and so reasonable usage would be expected.

Can the UKUUG afford this? The idea is that it will attract more members and the extra subscription income will pay for the services allowing UKUUG to grow and do even more of what its members want.

Possible services include the following.

SSL certificates

These are the certificates that allow you to conduct authenticated secure transactions with a web site. We would like to provide members with a certificate for one address for free.

Mail eXchanger (MX) backup

This would interest you should you receive mail directly onto your machine (i.e. you run an MTA). What happens when a friend tries to deliver mail to you machine but you are offline? The machine sending the mail will look up a secondary MX for your site; the secondary is probably your ISP. What happens when the ISP has problems? If your ISP has problems, your primary is probably unreachable as well. This service would provide a fallback for this: a well connected independent machine to receive and hold your mail for the period that your ISP has problems. This is a service that we all hope will never be used, but is invaluable on the odd occasions when it is necessary.

DNS backup

The Domain Name information for your site will probably be served up by your ISP. What happens if your ISP has problems? You disappear off the Internet. As with mail, your ISP is a single point of failure. Third-level DNS backup will provide you with some insurance against this.

Domain tag holding

Your domain needs to be registered within a top-level domain (for example .co.uk). This domain tag needs to be looked after (held) for you. Your ISP will do this but will charge you for it. If you change ISP it can be difficult to get your domain reassigned. As a tag holder for you, UKUUG should make changes easier.

Let us know what you think. Should the UKUUG offer services like these to its members? Are there any other services like this that could be added to the above list?

Please email council@ukuug.org and let us know.


Perl and XML -- XML Processing with Perl

Erik T Ray and Jason McIntosh
Published by O'Reilly and Associates
ISBN:ISBN 0-596-00205-X
216 pages
£ 24.95
reviewed by John Collins

This book describes and explains the various Perl packages for handling XML.

The reader will need a working knowledge of Perl but will not need to know much more about XML than it looks rather like HTML to be able to understand the book. In the second chapter, a fairly comprehensive introduction (described as a ``recap'') to XML is given. The horrors of Unicode and similar issues are covered quite well in the following chapter.

In the following chapters, parsers and event streams are covered, SAX, then Document Object Model and trees, then a more brief description of ``hybrid'' methods of processing XML, touching on XPATH, RSS, SOAP and XSLT, before concluding with a chapter of half a dozen brief examples.

I do not mind the writing style, although I know it will irritate some in places with references to Babel fish and Unicode providing for ``Earth begrudgingly joining the Galactic Friendship Union'' and sentences with ``kinda'' and similar words in. It is very much the O'Reilly house style along with the animals on the front which I know many people think has run its course.

The earlier chapters are pretty well-stocked with useful information and tips on the various parsers and handling techniques. I did get the feeling that a lot of readers would wonder and become impatient about apparently academic exercises ``out of nowhere'' in the early chapters before suddenly galloping through supposed real-life examples at the very end.

It was at the end of the book that I thought the authors seemed to fizzle out dramatically. Just as the reader would have hoped for some substantial real-life examples, he or she gets chunks of program and subroutines only for doing bits of things like, for example, interpreting DocBook -- which is never really described before the program to manipulate it is presented. I thought that just over a page on XSLT was far too cursory a treatment for what is currently an important part of the XML world, although, of course, it is not limited to Perl so possibly it is almost out of the scope of the book.

I think this book is overall much too short and light on detail to be particularly helpful. You can probably glean most of the better-treated parts of the book, such as the basic parsers, using ``perldoc'' once you have loaded up the Perl modules on your system. You will have to have books on XML, probably XSLT and of course Perl to do any serious work. You will need information on the ``dialects'' of XML the other applications you are working with ``speak''. Those might crowd this book off your shelf once you had got going.

In summary, this book assumes too much to be a true ``introduction'' but it is not nearly comprehensive enough to be a reference. It gives an overview of some of the concepts and a few helpful tips and signposts to get you started on XML using Perl provided you already know Perl and can download the XML modules from CPAN but you will urgently need other books to do any serious work. If you are already an XML expert with or without Perl knowledge, I do not think this book will tell you very much you will not know or be able to find out quickly elsewhere.


Web Services Essentials

Ethan Cerami
Published by O'Reilly and Associates
ISBN:0-596-00224-6
304 pages
£ 20.95
reviewed by Joel Smith

This book is aimed at developers who are new to web services. It gives an overview of web services to provide an idea of their scope and extent, but also includes sample code to enable the reader to start writing their own services.

The book focuses on the essentials of web services, and covers four main technologies: XML-RPC, SOAP (Simple Object Access Protocol), WSDL (Web Services Description Language), and UDDI (Universal Description, Discovery, and Integration). There is a high-level overview of each technology, together with a description of the relevant API and a discussion of implementation options for each technology.

Web Services Essentials is structured into five sections. Part 1 is an introduction to web services. This gives a general overview of web services, the architecture, and the web service protocol stack. Each subsequent section deals with one of the main technologies covered.

All four sections start with an Essentials chapter. This covers an overview of the specification, with numerous examples and sample code. The sections on SOAP and UDDI also cover the technologies in slightly more detail on a particular implementation.

The SOAP section has two chapters dealing with Apache SOAP as a specific implementation. Chapter 4 is `Apache SOAP Quick Start' and Chapter 5 is `Programming Apache SOAP'. Both of these are fairly self explanatory.

The UDDI section again has two chapters covering more detail of a practical implementation. Chapter 8 is `UDDI Inquiry API: Quick Reference' and Chapter 9 is `UDDI 4J'. This final chapter introduces UDDI4J, an open source Java implementation of UDDI. This gives a complete description of the UDDI4J API, and gives example code to illustrate how to search and publish UDDI data.

This book as a whole covers its material as you would expect. If you are wanting an introduction to creating web services, then it is a good place to start. It assumes a familiarity with Java and XML, and these are covered comprehensively elsewhere in the O'Reilly stable. Overall, it is a solid introduction to web services.


The Perl CD Bookshelf, Version 3.0


Published by O'Reilly and Associates
ISBN:0-596-00389-7
800 pages
£ 85.50
reviewed by Joel Smith

O'Reilly have updated their Perl CD Bookshelf. In the first update from version 1.0 to 2.0, the Bookshelf shrank from six books to five. This time the number of books jumps from five to seven. Out go Advanced Perl Programming and Perl for System Administrators. In come Perl & XML, Perl & LWP, and Mastering Perl/Tk. These are all first editions, published in 2002. Programming Perl remains in its third edition published in 2000, and the Perl Cookbook is the venerable 1998 first edition. There is a welcome return for Learning Perl, updated to the third edition. This was published in June 2001, and just missed the May 2001 release of version 2.0 of the Bookshelf, as I mentioned in my last review.

As in all versions of the Perl CD Bookshelf, Perl in a Nutshell is provided on the CD and also as a bonus paperback book, but in this version it is updated to the 2002 second edition, which covers Perl 5.8.

I appreciate that O'Reilly drops books from the Bookshelf because they are becoming out of date, but as the CD uses only around 60 Mb, I wonder if they could include an ``unsupported'' archive of the older books which are not updated in the main product. It seems a shame that Perl for System Administration and Advanced Perl Programming are no longer available when they could easily have fitted on to the CD.

Enough of the books, and on to the CD itself, and in particular the usability of the search engine. This is the same engine as used on the last version (QuestAgent Pro version 4.0.9 from JObjects http://www.jobjects.com/). It uses a Java applet interface to query a pre-generated index of terms derived from the full text of all book pages on the CD-ROM.

Thankfully the bug on the last version of the Bookshelf has been corrected, and the search engine works fine under UNIX and Windows. Unfortunately as it is the same version of the engine as before, there is still the problem with search terms involving special characters. Bearing in mind that this is a bookshelf on Perl, I should have been able to search on functions such as $arg, or ExtUtils::MakeMaker, or $_ etc. This is still not possible. To do a search on ExtUtils::MakeMaker, you would have to search on ExtUtils and MakeMaker, rather than the full module.

There is also the master index on the CD which allows you to browse through a linked alphabetical index (exactly as you would find in the back of the books), and does have entries including the special characters.

I raised the search restrictions as an issue with O'Reilly back in 2001 when I wrote my last review. The response then was:

The problem with not being able to search for special characters is due to a limitation in the search engine. Our tools group is looking into ways to do this in future editions.

Since QuestAgent Pro is now on version 6.0.3, one questions how far they actually looked. If you install the files on your hard disk, there is nothing to stop you running ht://dig or your favourite indexing software on it yourself, and creating your own search engine. However, it should not be necessary to have to resort to such methods.

To sum up, it is very useful to be able to have an electronic version of these books. Even with the limitations above, the search engine works well, allowing you to search across all the different titles. With the increase in the number of books, the CD bookshelf is still good value, even though the cover price has been increased. It works out much cheaper than buying the books individually, and at least the updated bonus paperback Perl in a Nutshell can still be read in bed.

Joel Smith offers consulting services through Dales IT Ltd. He is heavily involved with Digital Dales http://www.digitaldales.co.uk and is attempting to bring broadband to rural areas.

Mac OS X The Missing Manual Second Edition

David Pogue
Published by O'Reilly and Associates
ISBN:0-596-00450-8
725 pages
£ 20.95
reviewed by Lindsay Marshall

I reviewed some others in the ``Missing Manual'' series last issue and the comments I made there apply equally to this volume. It is well produced, well written and bulging with information on everything you can think of to do with OS X 10.2, but who is the intended audience? OS X seems to me to be perfectly easy to use, straight out the box, especially if you do not want to do any UNIX type things, and if you do want to do that then, mostly, it is just the same as any other UNIX system. Yes, there are a couple of subtle differences, but I bet that most people never come across them. I have so far not really found anything that I have a burning desire to do that I cannot work out drawing on common-sense and previous experience with Mac systems (OK, add in 30 years of working with computers too).

So who buys these books? I suspect that semi-mythical beast the Power User -- the sort of person who has to configure everything that can be configured, must always have the shortest shortcuts and enables every possible option. Me, I like the quiet life with as few options as possible, the scenic route to the commands, though I do configure things, but then usually for the sake of gaining simplicity.

I will use a book like this, because once in a while something comes up and I need to find out how to do something odd, and if I have it to hand I will look to see if there is any advice. Of course, by that time, there will probably be three new releases of the system and who knows then how fresh the information in the book is? So I will look at the book and go off to the net to see what I can find there as well. But will I use it enough to warrant buying it? Even though it is relatively inexpensive, I just do not think I would get my money's worth from it. I also do not think that this is (to quote the cover) ``the book that should have been in the box'' -- for a start it would have made the box much too big, heavy and inelegant. If the Internet did not exist then this whole series would be an excellent idea, but the Internet does exist so they are close to being a waste of paper.


DNS and BIND Cookbook 1st Edition

Cricket Liu
Published by O'Reilly and Associates
ISBN:0-596-00410-9
240 pages
£ 24.95
reviewed by Raza Rizvi

Wow, here I am again, reviewing yet another book on DNS and BIND by Cricket Liu! So what does this bring you that the other O'Reilly books do not?

Well nothing in essence but this nicely captures all those things that you are supposed to know off the top of your hostmaster head.

Building from basic questions -- what version am I running, getting BIND, starting a name server etc -- the book slowly and cleverly picks out those questions that are likely to form the bulk of matters you need to know about.

The book chapters follow the same logic as the larger reference works on DNS and BIND by the same author, namely, creating the zones files, configuring the server, handling mail, running the DNS server, security, upgrading (and this goes back to BIND 4, as well as having questions on the Windows implementation), programming, troubleshooting, and finally IPv6.

Each question follows the same sensible format: What is the problem? -- What is the solution? -- Discussion (so you actually learn 'why' rather than just 'how') -- and See Also (references to other related questions and anchors to the main DNS and BIND book).

Largely, the questions and answers are based around BIND 8 and 9, since no one should be running BIND 4 now, and the well-ordered layout and contents page make it an easy book to pick up as the need arises.

I really tried hard to think of something the book should cover that was not already in between the covers, and other than really esoteric stuff, I failed. This is a book for novice and expert alike, and really does make an essential desktop reference for all those creating, modifying, or supporting zone files.

If I had to give criticism, and this is really small fry and almost embarrassing as my only niggle, I think the book could do with a lay-flat spine.

Raza Rizvi is Technical Manager at REDNET. He still edits zone files from time to time these days, sometimes just for fun.

Managing and Using MySQL

George Reese, Randy Jay Yarger and Tim King
Published by O'Reilly and Associates
ISBN:0-596-00211-4
442 pages
£ 28.50
reviewed by Mike Smith

``2nd Edition'' ? -- but I do not remember the 1st edition! It was all explained on that page with the printing history (the bit you usually ignore). The first edition had a different title ``MySQL and mSQL'' -- ah!, I already had that. This first version only scraped the surface of MySQL (and indeed mSQL) and I thought it was quite poor -- trying to cover too much ground and ending up being pretty useless. I do not know about you but I kept forgetting to do a ``mysqladmin reload'' when I changed user access levels, that sort of thing should have been covered in big red letters -- it is frustrating when you do not notice this stuff. (Actually, you do not have to do that anymore as long as you use the GRANT and REVOKE statements.)

There is a discussion on database design first, but I like to get down to the practical stuff -- create databases, populate them, tips and tricks and the like (in fact I find entity relationship models even less interesting than network theory, but that is another review). This (i.e. the practical stuff) is indeed covered from chapter 4 onwards.

There is a chapter on performance tuning which I liked, and agreed with (e.g.the biggest tuning impact is at the application level.)

Perhaps more should have been made on backups, and some example scripts for showing how online backups can be performed would have been good. Maybe its too easy, as mysqlhotcopy does the hard work, but even a walk through of this would have been interesting.

I have never got on well with the online manual at mysql.com. I do not know why (I like the one at php.net, so I should). Anyway there is a reference section included, but as you will appreciate the book covers much more than this.

Later on we cover integration with PHP, Perl, Python and Java -- again with reference tables. There is also a chapter on writing your own functions (which you can use in SELECT statements) -- I did not know you could do that -- very cool.

In summary, this is so much of an improvement over the first version of the book. It still tries to cover a lot, delving into various programming languages, but that is good to give you an overview of the different APIs. So, yup, I like it.


802.11 Wireless Networks: The Definitive Guide

Matthew Gast
Published by O'Reilly and Associates
ISBN:0-596-00183-5
464 pages
£ 31.95
reviewed by Mike Smith

I have run a simple wireless network at home for a couple of years or so; just an Orinoco RG-1000 on the backbone and Silver Card in the laptop -- works great in the garden. I am just about to add an omni to the roof to extend the range over the whole village (hopefully) so it can be used for other purposes. If you are interested in how I get on, let me know (mike AT smithysrise.com) and I will document the results for a future newsletter. Anyway onto the actual book review...

The first two chapters provide a brief introduction, concepts and the like, and then move onto topologies.

By the end of chapter 2 we were beginning to get into the theory and low-level stuff. Now, I have read Tanenbaum, and although I am sure some people like this sort of thing, I have had my fill of it already. (I always remember Manchester encoding for some reason though.) So, I skipped forward a bit, to chapter 12 to be precise -- back up to a higher level. It was funny: the first sentence read ``Whether you've made it to this point by skipping Chapters 3-11, or read all the theory ...'' -- obviously I fall into the ``user'' category these days -- ah, well.

I was a bit disappointed with Chapter 12 -- which was about installing a couple of cards (Nokia and Orinoco) under Windows. I used the somewhat sporadic wireless connectivity at UKUUG's Linux Developers Conference this year (which was excellent, by the way). Now, some people noticed I happened to have an inferior ``Xtra Pricey'' operating system installed (I have lots of excuses lined up, so let us just sidestep this issue for a moment). However it was embarrassingly easy to configure -- even with the additional PPPoE and PPTP layers. So, I suppose there is not much to say on this OS, but a little more info on tuning, monitoring, and tweaking would not go amiss. Enough said, let us move on.

The Linux chapter was better as there is more to get your teeth into. I remember struggling with drivers two years ago, but it has obviously got a lot easier these days.

There is a good discussion on architectures, and having only one access point, one question in my mind was answered. That of whether the wireless medium itself can be used as the backbone between access points (yes, it can).

No book on wireless networks would be complete without a section on security -- there are two chapters in the bit I skipped initially; one on WEP, covering why it is weak; and one on 802.1x and what it gives you. I did not notice anything on authentication, such as Cisco's LEAP (or Microsoft's new PEAP), so maybe this is an omission.

There is also a chapter on network analysis. I am pleased it was included -- covering the use of Ethereal, and AirSnort amongst other things -- followed by a short chapter on tuning. With an appendix on the MAC (OS 9, not X unfortunately) the end of the book got better.

In summary, I did not think there was actually very much of real use to me, at least, in this book. Some bits are good for a beginner, other bits for the gory details. Nevertheless, if you like wireless networks and if you use them I am sure there will be something for you, and its certainly worth having on the shelf.


TCP/IP Network Administration, 3rd edition

Craig Hunt
Published by O'Reilly and Associates
ISBN:0-596-00297-1
746 pages
£ 31.95
reviewed by Mike Smith

I had not previously read the 1st or 2nd editions, so this was new to me. It starts at a very basic level and is geared towards the practical side of networks, and that is what I like. We begin with OSI of course, and how TCP/IP fits into it (or does not fit, as the case may be). Then onto routing, masks, subnets, ports, etc.

There is information on CIDR, which I found interesting, as well as the reasons why IPv6 has not taken off. The section on BIND is adequate, but you could never hope to match the information in DaB of course.

There is a small section on security, covering encryption and firewalls, but not a great deal of material. There are many other sources for this type of information so its not really a big weakness -- for many private networks the measures discussed are sufficient.

I did not really consider the chapters on services (like sendmail) appropriate, although this has been in since the very first edition (and why sendmail, not qmail, exim, etc?) Things have got worse, because the 3rd edition now also includes a chapter on Apache. There is coverage of SAMBA and NIS, but I can see that being more relevant in book like this.

I know there are other O'Reilly titles on Load Balancing and Caching technologies, but I think it would have been more appropriate to cover these subjects, at least in overview, rather than the services I talked about above. People like Radware are doing some really interesting stuff in this area too.

I did not see anything on 802.1q either.

With so many more ISPs these days, and organisations now entering into peering arrangements, I would like to have seen more on this area. An example of configuring named for BGP is given, but I would like to know so much more!

Another omission, IMHO, was on tools and techniques for the management of networks. There are many commercial proprietary and Open Source tools. I did not spot anything at all on SNMP, for instance.

There is a chapter on troubleshooting, but nothing on tuning. Sliding window sizes, locking ports to 100BaseT and other techniques should be covered these days, but you will have to use Google if you want any tips on these.

Its okay, and a good start for most people I suppose, but maybe it is about time someone wrote an ``Advanced TCP/IP Network Administration'' book, instead of reissuing this one.


XML Schema

Eric van der Vlist
Published by O'Reilly
ISBN:0-596-00252-1
379 pages
£ 28.50
reviewed by Lindsay Marshall

See the combined review below.


XML in a Nutshell 2nd Edition

Elliotte Rusty Harold and W Scott Means
Published by O'Reilly and Associates
ISBN:0-596-00292-0
634 pages
£ 28.50
reviewed by Lindsay Marshall

Yet another Nutshell book. What is there left to say about them? We all know that they are, with rare exceptions, consistently good, that they are well produced, and that they are a little pricey if you do not take advantage of your UKUUG discount or get your employer to pay. We also know that they often tend to be agglomerations of manual page material linked by some explanatory material and that if you get the one that relates to what you are actively working on then you refer to them constantly. You probably know that I do not like the new icons for pitfalls and tips.

So what about XML in Nutshell? Second edition. If you read my review of the first edition then I bemoaned the lack of coverage of XML Schema. This is fixed in this edition with a fair bit of material devoted to the subject and with much other material brought up-to-date with the latest developments. This does not make for an easier read however. The trouble with XML and books about it is that the basic technology itself is so simple that there seems to be a feeling that it must be surrounded by unimaginably complex layers of other stuff. It is standard--it is, just like C++: everything has to be tied down and the last bolt on the bed tightened to the correct torque, even if it means that you cannot actually lie down on it for the lumps of metal. Do not get me wrong, I am not against standardisation, just the kind of standardisation that sees complexity as a goal in itself. I am actually quite happy to have gray (or perhaps grey) areas in my standards -- so long as I know they are there I will avoid them. Basic XML is fine, but add in the dozen or so different ways of specifying what your XML means and how to reference it and process it in to other forms then you have a rococo edifice rivalling the most ornate wedding cake. How many people need to do all this stuff? Most people, it seems to me, use XML as a nice way of shipping data between systems and not much more than that. (OK the SOAP guys like to push the boat out -- or perhaps it is a rubber duck -- but you know what I mean). Whoever sees the stuff anyway?

Anyhow the book covers all the nooks and some of the crannies and is not taken in by the hype.

And so to XML Schema. I sense a terrible miasma of fear round the word Schema whenever it is used in these books. The bottom line is that it is peculiar because it is singular and most people do not really know what the plural of the word is: ``Schemas'' just sounds wrong, and ``Schemata'' scares people away with its classicism, assuming, that is, that they have even heard of the word. That aside, if you work with XML, at some point you are going to have to get to grips with XML Schema in order to create the schemata that describe what you are doing. This book provides a fast track to getting to grips with what is what, though, as with most tracks that are fast, it is a little hair raising at times and left me feeling slightly faint at times (I am still not entirely sure what xs:gDay is useful for nor why ___31 will include Feb 28th in a non-leap year?). Loads of examples, so many in fact that you tend to open the book to XML rather than text. There are examples in different scripts too, though the typography for these is somewhat muddy. Lots of descriptions of the many different ways to achieve the same ends and their pros and cons, and of the bits where they got things wrong in the standard -- the author does not try to hide any warts. A big chunk of the book is manual page and some of the material overlaps with that in the Nutshell book -- unavoidable and forgivable I think. Serious Schema sculptors will have this book to hand; I know that I shall be using it a lot, assuming that it does not vanish from my office?


iPhoto: the Missing Manual

David Pogue, Joseph Schorr and Derrick Story
Published by O'Reilly and Associates
ISBN:0-596-00365-X
350 pages
£ 17.50
reviewed by Lindsay Marshall

See the combined review below.


Learning Unix for Mac OS X

Dave Taylor and Jerry Peek
Published by O'Reilly and Associates
ISBN:0-596-00342-0
156 pages
£ 13.95
reviewed by Lindsay Marshall

See the combined review below.


Office X for Macintosh:the Missing Manual

David Reynolds, Tonya Engst and Nan Barber
Published by O'Reilly and Associates
ISBN:0-596-0033203
728 pages
£ 20.95
reviewed by Lindsay Marshall

See the combined review below.


Mac OS X Pocket Reference

Chuck Toporek
Published by O'Reilly and Associates
ISBN:0-596-00458-3
150 pages
£ 10.50
reviewed by Lindsay Marshall

Let me start by saying that I love OS X -- all the things that are good about UNIX, plus all (well, most) of the things that are good about Mac OS, and 10.2 Jaguar is most decidedly the bee's knees. I probably use OS X more now than any other flavour of UNIX and I switch between terminal and GUI interactions all the time. Of course, having all those years of UNIX experience clearly makes things easy, and many Mac users being converted to OS X may well find it hard to get to grips with using a more raw UNIX interface. (Though again OS X wraps up most things in slick GUIs if you want to go down that route).

So, what of these books aimed at OS X users? The Pocket Reference does what it says on the cover and it does it pretty well -- lots of potted information about the main interface and applications of the system. I learned loads of neat interface tricks looking through the book, but I will almost certainly forget them all before I get the chance to use them -- I tend to be fairly simple minded when using systems and often find the shortcuts too much of a hassle. I do not know that I would actually go out and buy this book, even though it is relatively inexpensive (in technical book terms anyway). It is thin and you can extract its essence pretty quickly.

I started in on Learning UNIX for Mac OS X expecting it to be like most other books of this kind: manual pages with decoration added to them, and for the most part I was not disappointed. There is also a lot of what seems to me extraneous stuff, like how to read your mail using Pine. Why would anyone want to do that on OS X (or on any system come to think of it!) which has a couple of really quite knacky GUI mail programs in MS Entourage and Apple's own Mail? For someone with UNIX knowledge there is very little in here that you will not know already, though I must admit to having learnt about a couple of special OS X commands that I had not come across before. For novices this not a bad book and flags potential pitfalls clearly and well. The book, as might be expected, deals only with the UNIX interface as it comes with a vanilla OS X installation -- there is little about the possibilities of extending it with the extra GNU utilities that have been ported to the system. (Well, not to 10.2 yet which is a bit of a pain!) The extras are certainly useful but might just be too confusing for a beginner.

The whole ``Missing Manual'' thing that O'Reilly has been doing is a great marketing idea, but I am not sure that it is much more than that. On the evidence of the two books in the series covered here, the reason the manual was missing in the first place was that you really did not need it. iPhoto is a great example. It is without a doubt the easiest piece of digital camera software I have ever used. Its interface is simple and it does what the vast majority of people need without any complexity. I had been using it for ages before I met this book and I can honestly say it told me almost nothing that I had not figured out for myself with little effort and no cost. What is there to say about such a simple interface? Sure, you can weave lots of stuff about digital photography in there, but that is not iPhoto. Big-ish print, lots of white space, loads of pictures -- looks great and is easy to read, but it is mostly padding. I did learn some things about the program, but it was about a part of the program I have never felt the need to even try and now I am even more certain that it is of no use to me. If you feel reassured by the presence of a manual then this book is fine but for most experienced users it would be a waste of money.

Finally, the biggy -- all 714 pages of it. Bloated, just like the programs it describes. Most people use Office and never need a manual. They have their favourites tricks and techniques and they can do pretty well most of what they want to do without problems, and if they need something else they will ask around till they find the answer, or just do something else. What they will not do (IMHO) is flick through a great thick book looking for the answer to their problem. (And if they do, they probably will not find the answer anyway. I spent some time trying to find out something about graphs in Excel and could not come up with the information) Far too much of the book is devoted to using features of Office that no sane person should be allowed near: Creating a web page in Word, Multimedia effects in PowerPoint, that kind of thing. Again, you cannot knock the presentation or the writing; as with almost all O'Reilly's, it is well done. It is just that I really can not see the point of all that ink on all that paper.


News from O'Reilly and Beyond

Marshall McLuhan vs. Marshalling Regular Expressions

Andy Oram explains why the success of ``Mastering Regular Expressions, 2nd Ed.'' should help assuage the panic Marshall McLuhan envisioned. http://www.oreillynet.com/pub/a/network/2002/07/08/platform.html

Getting loopy with Python and Perl

How do Python's loop constructs stack up to Perl's? Find out in this article by Aahz, one of the featured speakers at the Open Source Convention. http://www.onlamp.com/pub/a/onlamp/2002/06/27/pythonandperl.html

Google Needs People

Peter Morville talks about the recent stir Google created when it launched Google News, a service Google claims was generated without human editors. Peter is a co-author of Information Architecture for the World Wide Web, 2nd Edition.

Securing Linux

Michael Bauer explains why it is both possible and worthwhile to secure Linux for use as an Internet server platform. http://www.onlamp.com/pub/a/onlamp/2002/10/03/securinglinux.html

O'Reilly Beta Chapters Are Back!

Visit our Beta Chapters site for a preview of upcoming books. http://www.oreilly.com/beta/

Building Your Personal Anti-Spam Strategy

Spamfire developer shows how to fight back. http://www.macdevcenter.com/pub/a/mac/2002/11/01/spam.html

Protecting Privacy with Translucent Databases

Could translucent databases have helped Yale protect its data? http://www.oreillynet.com/pub/a/network/2002/08/02/simson.html

O'Reilly Bioinformatics Conference

The second O'Reilly Bioinformatics Technology Conference is slated for February 3-6, 2003, at the Westin Horton Plaza in San Diego, California. http://conferences.oreillynet.com/cs/bio2003/create/e_sess

Making the Python Cookbook

Putting together such a collective work presented some unique challenges. We recently spoke to the book's two editors about the community effort involved in creating the Python Cookbook. http://www.onlamp.com/pub/a/python/2002/08/01/cookbook.html

XML and Web Sites

John Simpson points to helpful resources for using XML to build web sites. http://www.xml.com/pub/a/2002/10/30/qa.html

Printing from XML: An Introduction to XSL-FO

Printing from XML: An Introduction to XSL-FO--Dave Pawson provides a simple introduction to creating printable page layouts with W3C XSL Formatting Objects. Dave is the author of XSL-FO: Making XML Look Good in Print. http://www.xml.com/pub/a/2002/10/09/xslfo.html

Running Zebra on a UNIX Machine

Iljitsch van Beijnum looks at whether Zebra, the host-based routing software, can be used as an alternative to a real router. http://www.onlamp.com/pub/a/onlamp/2002/11/07/zebra.html

Perl's True Success is in the Telling

A look at how the ``Perl Success Stories'' phenomenon began, as well as where to send your tales. http://www.oreillynet.com/pub/a/oreilly/perl/2002/11/04/perlsuccess.html

SquirrelMail, a Web-based Mail Server

As remote computing increases in popularity, it is becoming harder for regular users to send email. While running mail servers from customer laptops is one solution, a good web mail package may be even easier. In this article, Glenn Graham thinks SquirrelMail may be your best choice. http://www.onlamp.com/pub/a/onlamp/2002/10/24/squirrelmail.html

Building Your Personal Anti-Spam Strategy

Spamfire developer shows how to fight back. http://www.macdevcenter.com/pub/a/mac/2002/11/01/spam.html

CVS Third-Party Tools

CVS is an essential part of many developer toolboxes, but it has its limitations. Jennifer Vesperman has put together a list of add-ons and interfaces to the popular source control system. http://www.onlamp.com/pub/a/onlamp/2002/11/21/cvs_third_party.html


Contacts

Charles Curran
Council Chairman; Events; Newsletter
Oxford
07973 231 870
charles.curran@ukuug.org

James Youngman
UKUUG Treasurer
Manchester
james.youngman@ukuug.org

Sam Smith
Website
Manchester
sam.smith@ukuug.org

Alasdair Kergon
Events
Reading
alasdair.kergon@ukuug.org

Alain Williams
Watford
alain.williams@ukuug.org

Roger Whittaker
Schools; Newsletter
Borehamwood
roger.whittaker@ukuug.org

Jane Morrison
UKUUG Secretariat
PO Box 37
Buntingford
Herts
SG9 9UQ
01763 273 475
01763 273 255
office@ukuug.org