(Reviewed by David Hallowell)
Opera Software - http://www.opera.no/linux/
Commercial Software - with 30-day trial period
Beta release available for Linux. Final release for BeOS and Windows (16- and 32-bit) available. Macintosh coming soon.
The Opera web browser is nearing completion on the Linux platform and I thought I'd take a look to see what this browser has to offer. Unlike most of the other browsers available on the market, this browser is shareware. You'll have 30 days to evaluate it and then you must purchase it if you want to continue using it. As the Linux version is still in beta I couldn't find any details of what the price will be; it may be the same cost as the Windows version but we can't be sure.
So with the growing number of free browsers (as in free of charge and open source) available then why would anyone pay for a web browser? It all depends on what you're after out of a browser, so I'm going to give my views on Opera. However, you may disagree so download it and give it a try.
The first thing you'll notice is the small download size. Opera uses the Qt2.2 widget set and, if you have this installed, you can download the dynamically linked version, which is only 1.17MB in size. If you don't have Qt installed then you can download the statically linked version which is still a very small 2.67MB in size.
A quick look at memory usage shows that it uses about a third of the memory that Mozilla does. Although I use Mozilla for a lot more than web browsing, I'm writing this article in Mozillas HTML composer and using it as a Mail client too. Start up time for Opera was only a few seconds, which is a massive improvement over Mozilla and Netscape, so Opera wins hands down for download size, memory footprint and startup speed.
It's once it starts up you feel a compelling urge to go back to Mozilla. The user interface of Opera is just plain awful in my eyes. It uses the MDI style of windowing where all the application windows are contained within a main application window. As time as went on I've grown to hate MDI applications more and more because MDI applications are best suited to people who like to work with maximised windows. Personally, I never work with maximised windows and I like the placement of windows to be left up to me and the manipulation of the windows left to the window manager. MDI apps are best left to the days of Windows 3.1; in the Win32 world they still exist but aren't as common as they used to be; personally I don't want to see them in Linux.
From a user interface point of view this release also came with some strange defaults that can be changed. The default widget style was the MacOS Platinum style, which has scrollbars that look different to what you'd expect (the arrows are at the bottom of each scrollbar). This made the interface look a bit unfamiliar. There was also the possibility of the widgets to look like Windows, Motif, CDE or SGI and these all to me seemed a better idea as a default than the one they finally chose.
By default the URL bar is at the bottom although you can drag this to the top, which is what I'm more used to. However, start to load a URL and this bar turns into the status bar, which is very annoying; personally I like the URL bar at the top and the status bar at the bottom and both to be present all the time. I also find the layout of the toolbar buttons to be a bit strange, the back, forward and reload buttons are located on the main toolbar but the stop button is located on the URL bar. Although the stop button is only visible when it can actually be used, i.e. when the page is completed loading, it's replaced by some other button, therefore you may find the stop button difficult to find.
The browser does have a wide range of preferences giving you
plenty of control over how the browser operates, and it's
able to import Netscape and KDE Konqueror bookmarks (it also
claims to be able to support IE favourites, but since IE for
Linux doesn't exist, I can't prove this either way!). As yet
it doesn't import Mozilla bookmarks, but as they're stored
in the same format as Netscape bookmarks importing the
bookmarks from the
shouldn't be too difficult to add by the final release.
Page loading speed is about on a par with Mozilla, and Opera lays out all the main web pages I visit well. HTTPS is supported and, although not supported in this beta, Java and Netscape plugin support should be available in the next release.
I would have really liked this browser to be available as open source because in many respects it is a good browser with its small download size, lower memory requirements and quick start up speed while still being able to render most pages exceptionally well. However, its user interface is absolutely terrible in my opinion and, because it's commercial software, you (or some like-minded coders) can't just create a version with a better user interface. I'll be sticking with Mozilla, but if you've got a slower machine then Opera may be worth looking at. Opera has done a really good job at keeping the footprint down, although bear in mind that Mozilla has been designed to be more than just a web browser, people who hoped Mozilla would produce just an opera-like web browser have been disappointed, but I think the potential future of Mozilla is exciting as I mentioned in the last newsletter. Please note I believe it is a company's right to decide whether they release their software under a commercial or an open source licence, and you should respect the terms of the licence they use if you choose to use their software. I'm just saying that they've set such a good foundation here, if we had the source we could have an excellent browser. Alternatively they could just redesign their user interface to be more inline with what a web browsers interface is expected to be like.
Alternatives: on Linux the obvious alternatives to Opera are Netscape and Mozilla. There is also the Konqureor browser that is part of KDE2, I've heard good reviews of Konqureror and it may turn out to be an excellent browser, but as yet I've not installed it as I don't run KDE. However, hopefully I'll install it before the next newsletter unless someone else wants to review it for this newsletter.
Last but not least you've got the browsers that are based on Gecko, the Mozilla layout engine, but use a native toolkit, for example Galeon (http://galeon.sourceforge.net/) for GNOME/X11 and K-Meleon http://www.kmeleon.org/) for Windows. Because they use a native toolkit they perform better on slower machines while still offering the standards support of Gecko.
O'Reilly and Associates, Inc. 3rd Edition - May 2000
(Reviewed by David Hallowell)
Almost anyone involved with Perl programming will know what an excellent reputation O'Reilly's Perl books have. The most famous book, Programming Perl (the camel book), is the best all-round Perl book I've encountered, but they also produce a wide selection of other Perl books covering specialised areas of Perl in more detail, for example CGI Programming and database access using the DBI.
The Perl 5 Pocket Reference makes an ideal reference for programmers who have a reasonable knowledge of Perl. It provides a quick guide to Perl and makes the most important information available at your fingertips. This booklet starts off talking about the various command line options you may use to invoke the Perl interpreter and covers all the essential areas of the language. It soon becomes such a useful reference - it may become the most used of all your Perl books as it puts the information right there at your fingertips. To show what areas this booklet covers I'll list the table of contents at the end of the review.
At the end you have a list of the standard Perl modules that come with Perl with a brief explanation of their purpose. Throughout the book references to Perl manpages are frequently made to help you find further information, and at the end of the book is a list of useful URLs related to Perl. It's very hard to write a review of a pocket reference because it's very rare that you're going to read the entire contents of the book. I'd say it makes a good reference and its small size makes it easy to carry with you if you're working somewhere different and want to make sure you've got a decent reference readily available.
332 pp., £32
(includes CD-ROM with Red Hat 6.2 and clustering software)
(Reviewed by Richard Ibbotson)
Building Linux clusters is about putting computers together in a cluster and making them do something that they wouldn't do otherwise. It is of course based around the Beowulf project and much work has been done by NASA on the subject of "Beo" clusters as they are known by many physicists and computing boffins.
If you are a professional physicist or an IT expert you will have heard of the Beowulf project and of its legendary ability to make supercomputers out of the most mundane hardware. To quote from the preface...
Back in 1994, a couple of research scientists, Thomas Sterling and Donald Becker, at NASA's Center of Excellence in Space Data and Information Sciences (CESDIS) in Greenbelt, Maryland, embarked on a project to build a parallel computer out of off the shelf components. They wanted to make a low cost, yet efficient, system for processing large space science data sets. Their first system, a 16-node network of workstations, was constructed out of Intel DX4 processors. It was interconnected by a novel channel bonding method that allowed them to tie together multiple 10Mbit/second Ethernets to balance network performance without the use of then expensive network switching systems. With the addition of message passing software such as the Parallel Virtual Machine (PVM) system, they were able to build a quite effective parallel processing computer on a shoe string budget. They named the results of their work "Beowulf". Their experiment in distributed processing was an unmitigated success, principally because of the underlying system software they chose to base their efforts on: Linux.
I knew this to be the way things were before I read the book. I had at that time sent some mail to the people at NASA to ask a few questions. However, for the first-time reader of clustering software and parallel computing the idea that a few 486s might do something for you may come as something of a shock. This is not so. I have heard more than two Cray operators say to me that they can't understand why the Cray computer is still around.
Chapter one kicks off with a brilliant but shortened account of the history of computers and computing science. It explains how things were in the 1950s and 1970s and how things changed after Intel and other companies came along. As so often happens in several fields of discovery, science fiction became science fact and we now use computers that would never have been thought of even as little as twenty years ago. There are a few very useful pages on what clusters can be used for.
Chapter two goes into basic concepts and why we need clusters.
Chapter three introduces in a few brief words a topic that not many of us have read about or been involved in. Designing clusters is one of those things that most people assume is done by someone else. If you haven't built a cluster before then this is the part of the book that will interest you the most.
Building clusters at chapter four is the part I really like. The bit where you actually get your hands on the tools and build it.
Next, in Chapter five, comes software installation and configuration. This is really good and should be the kind of thing that should be shown in every modern art gallery. Very much Tate Modern.
The sixth chapter goes into managing clusters. Most people need to know this kind of thing, but I've heard of an in-joke about the fact that the Caltech janitor could take care of this kind of thing. There are some good graphical examples in this book that show GUI tools that make cluster management easy to understand. It reminds me of the Using Samba book that is also much easier to understand than the others.
The really sophisticated part of the book comes up at Chapter seven. Tools and libraries for parallel programming is very helpful but perhaps another hundred pages would have been useful?
Chapter 8, Programming in a parallel environment is also helpful. Not the sort of thing you'll find on a shelf just anywhere.
The final Chapter 9 about application examples is invaluable.
What ruined the entire book for me was that it is based around only one distribution of Linux. Both the SuSE and Debian distributions offer better clustering software and greater choice of all sorts of other things. The Debian packaging system is also much better than anything that Red Hat offer. I've come across many people who believe that it would be better if NASA were not involved in the Beowulf project.
Another thing is that the preface mentions only open source software and not free software. As we all know, the Free Software Foundation and the GNU project are very much the things that made Linux popular and continue to provide a base from which further software can be developed.
O'Reilly & Associates, Inc., June 2000
479 pp., £26.50 ISBN 0-596-00016-2
(Reviewed by James Moran)
I've read several books on XML that have been mainly a regurgitation of the specifications without much attempt to help a developer understand how the technology can actually be used. This is the first book I have read on this subject that actually bridges this gap and gives useful, working examples of how XML can be used in your Java applications to solve real-world problems. Instead of giving a code snippet that shows only how to call the XML parsing/processing APIs, the book gives full examples on how to use XML in applications from servlets for presentation to B2B applications for data collaboration between companies.
Two hot topics come together in this developer's guide from Brett McLaughlin, Java and XML. Both Java and XML are cross-platform technologies; by using Java for code and XML for transporting data, you can build truly portable applications. This title is aimed at intermediate to advanced programmers; while XML topics are explained more or less from scratch, readers will need prior knowledge of Java.
The book begins with an overview of XML and its uses, and goes on to explain how to parse XML by using the Simple API for XML (SAX 2). Next, there is coverage of how XML is validated by using Document Type Definitions (DTDs) and XML Schema, and transformed by using eXtensible Stylesheet Language (XSL). Brief coverage of Sun's Java API for XML is followed by a detailed look at the Java Document Object Model (JDOM), a new API devised by the author in association with O'Reilly, the publisher.
The last part of the book is more advanced, and covers applications of XML and Java. There are chapters on Web-publishing frameworks, XML Remote Procedure Calls (RPCs), using XML to read and write configuration data, and generating XML with Java. There is also a short business-to-business example. Appendices provide an API reference to the various specifications discussed in the book.
The strengths of Java and XML include the author's deep knowledge of his subject, and a writing style that is both clear and enthusiastic. If you happen to know a lot about Java and not much about XML, this is the ideal title. Readers who already have a good grasp of XML basics might be frustrated by the amount of introductory material.
The book was well written and easy to follow. The author doesn't waste time reiterating the same things over and over. Links for more information on each subject are given in the text.
If you are looking to implement an XML solution in Java, this book will be a great help along the way.
(Reviewed by Oliver Rose)
NB: I only downloaded the browser component, but others to handle mail, web page construction, and use AOL Instant Messenger are available. It was tested on a PII-450, 256Mb Ram, Windows 2000 with Service Pack, and 56k Modem.
It's hard to start a review of a new browser on a windows platform, without comparing it to Internet Explorer. As a result, I shall and I'll say that it is more stable, better organized, skinnable and less demanding on screen space. However, it could eat memory for America and sometimes it refused to save preferences.
In a semi-scientific test, I found that in about 5 hours of browsing, it crashed less often and with less force than the latest version of Internet Explorer. It also managed to die more gracefully than its competitor, leaving only a note in the event log and a space in the taskbar when it went. Most crashes by Internet Explorer kill off everything that it is possible to kill, cripple what's left, and leave the user requiring a restart to bring the system back to a usable state. When you use an unmetered ISP which seems to have a major problem when it comes to connecting, the fewer times you have to go through that the better.
Nice and simple this bit. The menus in Netscape are organized in a better fashion than Internet explorer. By that, I mean all of the preferences for the system are found in the Preferences Dialog, rather than being spread over the Internet Options dialog where anyone can find them.
Most modern applications are skinnable now, so the people at Netscape decided to make Netscape 6 skinnable as well. Given that they had to redesign the actual buttons in the first place, making them skinnable is the next logical step. This therefore means that it can look anyway you want it to, if you can design something that looks usable...
Bit of a dodgy category this one, because you might say "how can something that is maximized be less demanding on screen space?". The answer is that because it only uses screen space for its toolbars, the rest of it is designed for the purposes of showing pages. Of course this is only true on large screen resolutions, but it does take up a lot less space than Internet Explorer. It has been noted that Internet explorer does have a full screen mode, with an auto hide tool bar which means that some of that isn't accurate. However, as far as I know, nobody uses this mode, just in case they need to use the taskbar.
6,9,12,14,17,20... it may sound like a simple series of numbers, but in fact it is the memory usage of Netscape 6 starting up. Unfortunately, by the time you've viewed a few pages, it starts to take an extremely large chunk. Oh yeah, I thought I'd mention those numbers were scaled in 1 MB chunks, for a single instance of the browser, looking at Slashdot.
Rating: 85% - If the memory requirements drop then it may have a chance.
Found a bug: Contact http://home.netscape.com/browsers/6/feedback/index.html or through the Quality Feedback Wizard not to bugzilla.
Oliver Rose is a Student studying Software Engineering at University, insists on using Windows, and likes Java... draw your own conclusions. He is also a member of the UKUUG dnetc team, although probably the least productive member there.
Jim Elferdink & David Reynolds
Pogue Press/O'Reilly & Associates, Inc.
400 pp., £13.50
(Reviewed by Jan Wysocki)
I'll start with two confessions. First, I really wanted to review and use the companion volume MacOS 9, but I wasn't quick enough off the mark. Then, I'm not sure if this review really has a place in a UNIX journal, so I'll be brief.
I do make use of Macintosh computers at home, one runs OpenBSD, but I do use the most modern one for a little light office work. Anyway this means that I use Appleworks (nee Claris Works) a fair bit and might have occasional use for a 'Works manual. AppleWorks is an office suite that allows users to produce documents, presentations and keep records, using a variety of integrated software applications.
Since there's no Apple manual, the nearest comparison I can make is with a Claris Works 4.0 manual I've had on my bookshelf since about 1995. I can at once say that this book is an improvement. However, this is not a comprehensive "Nutshell" style reference, it's a clearly signposted tutorial for a new user. By p.25 I'd already given a nod of approval because the author had realised that a novice may not get the behaviour predicted on the page. Somebody else may have already changed the default preferences. A small point, but well made and it might just reduce a few frustrations around a family iMac. In the same vein, the author also gives a clear exposition of Mac 'troubleshooting'. If this is the only book that a novice Mac user has bought, then it'll be appreciated. At the other end of the scale you can learn precisely, but briefly, what you need to install AppleWorks on a server.
There's low key humour throughout the book. The layout is clear, but departs from the usual O'Reilly look. There is a 12 page index (still not comprehensive enough for some of my queries) and each page has a small label. These labels sit near the top of an inch and a half wide, light grey, margin on every page. Illustrations overlap the margin, but all the text is in the central white area. It may use up a little more of those northern forests, but it's easy to read. If you know roughly where a section is, you can rapidly flick through the labels instead of looking at the context or index.
I applaud the authors for concentrating on explaining the tool set in question. They don't divert into comparisons with other products, yet they are prepared to take a few legitimate shots at problems that should have been sorted before release. It's a little reassurance that they're on the reader's side.
You can probably tell that I liked this book. I recommend it to anyone who's thinking of shelling out several hundred pounds to buy some other 'integrated package' for their Mac, as a little time spent with this book may convince them that they've got all the productivity they need in their new Mac. This book will lead them through the basics of word processing databases and spreadsheets, together with drawing, painting and 'slide show' presentation. Then they'll learn how to integrate these and take them further. It's a good read.
Jan Wysocki is currently administering a network of just two Suns at Virgin Money!
Tel: 01763 273 475
Fax: 01763 273 255
Queries: Ask Here
|Join UKUUG Today!||
PO BOX 37