The Times Reader 2.0 is an Adobe Air application, which means that you need to install the Adobe Air platform in order to use it. The netbook, running Windows 7, came preinstalled with Air (and, in fact, also with the Times Reader software). Linux, however, obviously needed to have Air installed manually, before I could install the Times Reader application. When you go to the Times Reader website, a flash applet will load and automatically launch the Air installer. You’ll need your root password to install it, of course.

Installing the Times Reader in Linux

Once Air is installed, the applet will automatically trigger the installation of the Times Reader program. I did not encounter any issues installing it on Fedora 13 using this method, and everything was installed in the /opt directory.

The first time you launch the program, you are going to be pleasantly surprised by its look and feel. The only complaint I have is that the Reader feels a bit cramped on my netbook. It seems that the program was not designed to be used on the 10.1″ that is common on many netbooks. When connected to the internet, the application will display banner ads (so far, I have only seen ads for NY Times products), and these ads take up enough space on the screen to make it feel like there isn’t much ligitimate content on the screen, as well as to considerably lengthen the number of page-flips required to get through an article.

For comparison, here are some screenshots showing the Reader with and without advertisements (the advertisements do not show up in offline mode):

The Times Reader has a beautiful interface...

...but the advertisements cram everything in if you are on a netbook.

Once you get past the sizing issues, or if you just aren’t using the Reader on a netbook, the interface is intuitive and easy to use, with a navigation menu on the left-hand side presenting buttons for each of the New York Times’ paper sections. Readers of the traditional hard-copy will feel right at home, and I find that I look forward to seeing the Science Times link pop up in the Reader as much as I used to await the Tuesday morning delivery for that section (which brings up another advantage of the Reader software: I don’t have to wait for a late paper, and I don’t have to go outside in the cold to retrieve it from the sidewalk).

In addition to the sections that hard-copy readers are used to, there are additional sections only available in the Reader: Most Emailed, News in Video, News in Pictures, and Latest News. Admittedly, I don’t often use these sections; at most, I’ll load the Latest News section once or twice a week at the end of the day to see if anything major happened that I missed.

Tthe Reader will automatically store 7 days of content; that, along with a search function, lets you easily find things you remember reading previously, or locate specific stories you’re interested in:

I don't think your hard copy comes with a search feature...

Ultimately, the user interface for the Times Reader 2.0 is very well done, and it has a polished feel to it, providing an intuitive, comfortable interface that shows you what content is available without getting in the way. However, the Reader definitely gets points off for not performing perfectly on the netbook, especially since it came pre-installed on my HP Mini. This tells me that either they tested it, noticed the problems, and decided to accept the shortcomings, or that it wasn’t tested at all. Either way, since the netbook will be my primary platform for the Reader, I simply wish it fit better on the screen.

Once you get past the novelty of reading the newspaper via an application on your computer, it all comes down to how well the application presents the content, and the Times Reader 2.0 does this very well. As I discussed in the User Interface section, the reader presents a simple navigation panel, showing the sections from the day’s paper. The dominant right-hand panel is dedicated to displaying the current story, or summaries of the highlighted section.

You don’t appear to lose any content when going from the physical version of the paper to the Reader. In fact, I very often see notes at the end of articles that the version delivered to my Times Reader contains content not in the print version. With the Reader application, there’s no space limitation, so editors likely have more freedom to make articles complete, as opposed to making articles fit.

The Reader experience will largely depend on how comfortable you are reading lots of text, for a sustained period of time, on your computer screen. Personally, I don’t have a problem reading the entire paper on my netbook’s display; but if you find it difficult to read long passages on-screen, the Reader may not be for you.

No review of the New York Times Reader would be complete without atleast mentioning the Crossword. I personally don’t usually do the crossword puzzles, but the Times Reader’s implementation is actually very well done.

A great interface, but the Advertisement cuts off the puzzle!

Again I have the recurring complaint that the application does not fit perfectly on my netbook, which resutls in the bottom section of the crossword being cut off (you can scroll to it). Aside from that, the Reader’s version of the crossword has several benifits as compared to the traditional version, including “Reveal” and “Check” buttons that become accessable 60 minutes after you begin. Like the rest of the features of the application, you can revisit any crossword downloaded in the past 7 days, and print to hard copy should you have the need (or want to share).

This all being said, I personally find that the Reader application is far more convenient than the traditional paper version. Having the electronic delivery makes the paper available from the moment I wake up to 7 days later, and it means that the newspaper is always with me (as long as I have my netbook).

One thing that shocks me is the price; as of this article, the Reader is available (on its own, with no home delivery option) for $3.45 a week. For subscribers who already recieve home delivery, the Reader is completely free. Personally, I feel that these prices are actually too low, and that the New York Times is pricing the Reader service below its value.

Once you use the Times Reader for a few days, reading the digital copy becomes almost as second-nature as flipping through a physical newspaper. People who enjoy the experience of newspapers — peeling out each section, folding and unfolding the broadsheet — won’t find that experience replicated here; but those who want to get their daily dose of news available on-demand will find that the Reader is an excellent tool, that may even replace the paper at their breakfast table.

Ultimately, the Times Reader 2.0 has, for me, replaced home delivery as method getting the daily paper.

BTGuard is a paid proxy service intended for bittorrent users. For $6.95 a month you can use BTGuard to hide your IP address from other downloaders and bypass any traffic shaping your ISP performs to limit your bittorrent usage. I’ve given this service a try, with mixed results. Read on for more.

BTGuard provides a Socks5 proxy, through which you pass your bittorrent (or other data) traffic before it goes out to other members of a bittorrent swarm. As a result, the tracker, as well as anyone else on that torrent, will see the BTGuard IP address rather than your own; your identity effectively stays hidden from the public.

When you register for a BTGuard account, you set a username and password, which you use to identify yourself to the proxy. This is called Socks5/A, or Socks5 with Authentication. Consequently, the service will only work with programs that support this proxy scheme; most bittorrent clients do, with the notable exception of Transmission (now the default client for Ubuntu Linux).

It is possible to use the proxy for more than just bittorrent traffic; but as of yet I have been unable to find a web browser with Socks5/A support. This includes Firefox, Opera, Konqueror, and Internet Explorer. However, any program that has support for this protocol should work, and the primary intention of the BTGuard service is for bittorrent anonymization.

In my tests, I used various clients, including Deluge, Azureus, and uTorrent (running via Wine).

Setup
Setting up your client to use BTGuard (after you have procured an account) is simple, and instructions are provided on the BTGuard website for the most popular clients. The only information you have to fill in the proxy address and port number, as well as your username and password. Once you configure your client to use the proxy, you’re good to go.

Speed and reliability
Unfortunately, the speed of the BTGuard service isn’t quite up to par. While my download cap from my ISP is quite high (I’ve often hit 2MB/s, sustained), when routed through BTGuard this falls to approximately 100KBps on average, with peaks around 300KBps and lows of 50KBps. These speeds are sure to disappoint serious downloaders with fast connections, but for someone who only occasionally grabs a file or two it may be acceptable.

Another problem with BTGuard is reliability; more than once I noticed my connection dropping entirely for short periods (about 30 seconds) before it would start up again. I ran all tests with several Linux ISO files downloading at once, so I can say with certainty that the strength of the bittorrent swarm was not the issue; this is also a problem I only encounter when using the BTGuard proxy service.

Overall, using BTGuard will most likely mean a slower, less reliable connection (at least for now, or until they upgrade their systems to meet with demand), and it’s something to consider before spending $7/month on the service.

One of the largest questions about BTGuard is simple: Can you trust them? When you use a proxy, you’re sending all your data through someone else’s servers; as a result, they can view and log everything you do. Unfortunately, there isn’t any readily-available information on exactly who is behind BTGuard. On their website they explicitly state that they do not keep any logs, but you have to take them at their word.

Using BTGuard to become anonymous changes your bittorrent security paradigm from “security through obscurity,” or becoming lost in the crowd, to “security through trust” of BTGuard’s systems. It’s up to the individual to decide whether or not they want to trust BTGuard.

Speed

The number one thing BTGuard needs to do to improve their service is make it faster; 100KBps on average is simply too slow for bittorrent traffic. As more users sign up for the service, their “tubes” will become increasingly full, and individuals will see their download speed decrease; hopefully BTGuard will vigorously combat these growing pains and provide a faster service.

Security
BTGuard needs to make it clearer that you can trust their service. More legal information, more information about who exactly is running the service, and a better privacy policy are must-haves before many people will consider them trustworthy.

Pricing
Currently, the BTGuard service costs $6,95 a month, which I think is fair for what they provide (although speed remains an issue, and some people will likely be turned away from paying for a service as slow as it is now). However, there is no long-term pricing; you can sign up for one month, or a recurring monthly payment, but there is no price difference. A 12-month discount price, for example, would be a nice addition; perhaps $75 instead of the current $83.50.

BTGuard fills an important niche, and many bittorrent users will likely be willing to pay for such a service, but it’s currently plagued by some significant problems that will turn many potential users. If the company is able to increase available bandwidth, deal with its growing pains, and make themselves more trustworthy, it’s likely that BTGuard will become a powerful tool in the fight for privacy on the internet.

Know the laws in your country and locality. I do not suggest, condone, or practice the violation of national or international copyright law. The BTGuard service explicitly forbids using the proxy to commit illegal activities, including copyright infringement. No laws were broken or copyrights infringed during the writing of this article or review of this service.

Griffith is an application for the Gnome desktop that organizes your movie collection, much in the same way Alexandria is used to organize your book library. Simply by entering the name of a movie, Griffith will query various databases on the internet and download cover art, plot details, and cast information.

Griffith is probably included in your package management repositories; for Ubuntu, you can install it by running:

But you can find more information about installation, including a Windows version, at the Griffith download page.

The nicest part about using Griffith to organize your movie collection is that it does all the work for you. By clicking on the “add movie” button, you’ll see a window for a new entry:

You can see here that I’ve entered the title of a movie, and on the right side there is an option to select the database which Griffith will query for the movie’s information. Griffith supports over 25(!) different databases on the Internet, but the International Movie Database seems to do a great job at providing all the information needed for a movie.

Once you click “Get from Web,” Griffith will go out and find the details for your movie. Most likely, it will return a list of movie titles that are similar to the one you entered, and ask you to select the correct movie. Normally this will be the first movie listed, but if, for example, there were several remakes of a title (“Hamlet” would be a good example), it also lists the year in which the movie was produced, letting you select the appropriate item. Once you tell Griffith which movie you want, it will fill in all the information for you, automatically:

You can repeat this process for part or all of your movie collection; when you’re finished, you’ll have a Griffith window filled with your movies. You can select the movie you want to view, and tab between all the information downloaded by the program. Griffith will even let you grab a larger version of the cover image from Amazon.com, and then help you print it out.

Adding personalized information
Now that you’ve gone through the hassle of adding your movie collection to Griffith, and you’ve had all the information about your titles entered automatically by the program, you can start manipulating your collection to suit your needs. For example, you might choose to rate movies, which Griffith allows you to do on a scale of 1 to 10. The little thermometer icon will indicate how well you’ve rated a movie. You can also tell Griffith about the media on which a movie is stored; is it a DVD? What region encoding does it use, and what quality is the disk in?

Search
Searching through your movies is easy, and you can filter by pretty much any type of information Griffith stores: from title, to genre, to director, to year. You can also indicate whether you’ve loaned out a particular movie, and to whom; so you’ll never wind up arguing with a friend again over whether or not she still has your copy of Back to the Future.

Exporting your database
What fun is cataloging all your movies, if you can’t show off your massive collection to your friends? Griffith’s export function is simply amazing. Not only does it let you export to PDF, HTML, XML, CSV, or iPod formats, but it lets you fine-tune exactly what data you want to include. Personally, I would want to use the HTML export function the most, and when I selected this option I was amazed at the control Griffith gives you over the output:

Not only can you select various themes for your HTML movie list, but you can choose which information to include, and what you want to leave out. You can also set how you want Griffith to split the entries, how titles should be sorted, and which movies you want to leave out altogether.

Movie buffs as well as casual collectors will love Griffith’s ability to generate an entire, well-documented database simply from knowing the title of a movie, and the single push of a button.

Google’s Picasa software is a powerful tool for managing your photo collection. It provides a simple interface for organizing your pictures, as well as some easy-to-use, yet powerful image manipulation tools. However, until recently the Windows version of Picasa has had some features that put it above the Linux version; some of these features, like Picasa Web Album integration, are at the core of the program, without which Picasa isn’t very useful. With its newest release, the Linux version is being brought up to par with its Windows counterpart.

Picasa 2.7 is still in beta, so it isn’t included in many repositories just yet. However, Google provides easy-to-install packages in RPM and DEB formats on their download site.

The major new feature in Picasa 2.7 is integration with Picasa Web Albums, but we’ll get to that later; there are a plethora of other upgrades that make this new version worth getting excited about.

Save/revert edits
Picasa has some great image manipulation features. You can apply basic filters, as well as fine-tune an image based on fill lighting, highlights, shadows, and color temperature. There are also several effects you can use, such as sharpening, tinting, saturation, and others.

Picasa now allows you to save these edits either to the original file, or to create a new image with the modifications. More importantly, however, is the “revert” option, if you decide the original was better than the modified version.

View images by folders
One of my favorite features is the ability to find and sort files by their location in the filesystem, rather than sorting by albums in a library. This allows you to use your existing filing method to sort your pictures, rather than forcing you to create an entirely new system to use Picasa. A small change, but a very useful one nonetheless.

Better camera importing
Along with the ability to use your existing folder structure for picture imports, Picasa now allows you to import images from your digital camera using this filesystem structure. Google has also added better RAW support; you can read more about these features on Google’s website.

And now the feature we’ve all been waiting for – web album integration – is finally here! And I have to say that the implementation is as straight-forward and simple as it could possibly be.

Once you’ve logged into your Picasa Web Albums account, you’ll see your account information in the upper right corner of the screen:

Now, you’ll be able to create web albums, upload images, and manage existing albums via the Picasa desktop interface. Right-clicking on an image gives a contextual menu that will allow you to upload the image to your web gallery:

Clicking this item will open the Web Album manager, which lets you choose the option of adding the image to an existing album, or to create a new album.

All in all, the new interface for managing, uploading, and modifying Picasa Web Albums is a great improvement, and one that has been sorely lacking in the Linux version of Picasa for some time. With this new feature, Picasa for Linux becomes, in my opinion, the easiest to use photo management and web gallery software available for linux.

There are quite a few imaging programs for Linux. FSpot and Gphoto are two that come immediately to mind; and while they’re good programs, I have to say that Picasa has really began to mature, and is starting to impress me as a possible complete replacement for these other image cataloging programs. Of course, unlike its alternatives, Picasa is a closed-source application (not to mention, it runs via Wine on Linux), so there are several licensing as well as philosophical issues to be dealt with before Picasa becomes the default image manager for any Linux distribution.

If you’re thinking about trying out this new version of Picasa on your production machine, it’s not a decision you should take lightly. There are quite a few known bugs listed on the Google Picasa for Linux page, and even though most of them fall into the “vaguely annoying” catagory, it’s probably best to wait until the official release until you upgrade. If you just have to have the latest version – and you’re especially excited about the web album integration – I don’t think it’d be too harmful to try out this beta. Just remember that there will be bugs!

While it may not let you go where no man has gone before, Celestia is an amazing desktop application that lets you travel anywhere in the known Universe.

You can view any object in the Solar System, travel to distant stars, and even leave the Galaxy to view planets only recently discovered by NASA and global space agencies.

Celestia is cross-platform, which means you can run it on Windows, Mac OS X, or Linux. It’s also open source, so if you’re using a different operating system, or if it’s not available as a binary package for your distribution, you can compile it yourself.

Windows and Macintosh users should head to the Celestia download page, while Linux users should check and see if it’s available through their package manager. In Ubuntu, you can download Celestia using the following command:

By default, Celestia uses a QT interface, which will work great in a KDE setup (such as Kubuntu). Gnome users will probably want to install the Gnome front-end, using the command:

There are also loads of add-on packages available for Celestia, which increases the size of the database of objects Celestia is able to view. These include various spacecraft and satellites, deep space objects, and even fictional objects such as space bases in Star Trek and Babylon 5. For a complete list of add-ons, check out the Celestia Motherlode site.

Once installed, you’re ready to start navigating your way around space in Celestia. The first thing you’ll probably want to do is check out Earth, or maybe some of the other planets in our Solar System. By default, Celestia comes with the graphics for objects in the Milky Way, such as each planet’s moons, various asteroids, stars, and comets. The Gnome front-end has a really handy tour guide feature, which will let you select some of the more popular attractions in space:

Selecting “Go To” will take you on a journey through the millions of miles to whichever object you have selected. I traveled to the comet Borrelly, and was able to view this beautiful image taken by the Deep Space 1 spacecraft:

Next, I was curious about how things looked from the most distant parts of our Solar System, so I traveled over to the ex-planet Pluto, to check out how things looked from that icy rock way out in space. Celestia gave me a beautiful view of Pluto’s moon, Charon, which you can see in the upper-left corner here, and you can see all the inner planets labeled in the distance:

To travel back, I simply right-clicked on Saturn and selected “Go-To”; stars in the background whizzed by as Jupiter and one of its moons, Titan, approached, giving me this wonderful view:

I could have spent hours traveling around the planets and asteroids in our Solar System, but I wanted to get into Outer Space and see what else Celestia could do.

Celestia comes with a very handy Star System Browser, which will show you a list of known stars, and you can sort by distance, known planetary systems, or brightness. Personally, I find stars with planets to be the most interesting, since they’re the ones that will likely have life, so I selected the Wolf 562 system. Once Celestia traveled to this distant star, I was able to see its planetary companion (romantically named “b”), and plot its orbit around its mother star.

Obviously, there are hundreds, if not thousands, of other stars objects to explore. That’s the fun of Celestia! You’ll never run out of objects to view, or new things to find. And for more cool pictures, check out the Celestia Screenshot Gallery!

I heard on the news that there will be a lunar eclipse today (Feburary 21, 2008). Unfortunately, I’m not in a position to see it; but with Celestia, I can get a 3D simulation of any eclipse (Solar or Lunar) and of any planet, at any point in history – or even the future! Celestia’s eclipse finder lets you enter a date range, and select which eclipse you’d like to view:

Even though I won’t be able to see the Lunar eclipse today, I can still check out the view of the Earth, Sun, and the Moon lined up, using Celestia:

I haven’t really discussed the details about what type of information Celestia shows you. Obviously, you can set it to label stars, orbits, planets, and other bodies; it will tell you the surface temperature and luminosity of any object, its distance from any other object, and what class planet or star something is. This is all really great information for astronomers, and I have no doubt that this software will serve as a useful tool for anyone in that profession or field of study. But Celestia is great even for those who are just interested in space and the Universe. Celestia lets travel millions of light years without ever leaving home!

While Linux is praised for its ability to run on older hardware, modern distributions such as [K]Ubuntu and Fedora eat up lots of disk space, memory, and processing power. And, while there are many smaller distributions that are designed to run on antiquated machines, most of them require years of Linux experience, lest the user risk becoming bogged down in problems. Fluxbuntu’s aim is to be a “lightweight, productive, agile, and efficient” operating system; this review takes a look at Fluxbuntu and whether it lives up to the challenge of creating a user-friendly experience on a tight resources budget.

Fluxbox, unlike Ubuntu, uses a text-based installation program that might scare away some users that haven’t experienced a text-based install. That being said, it’s really a simple, straight forward process, and it’s necessary to keep resource requirements low. The entire install CD comes in at a slim 306MB, which makes it relatively practical for dial-up users to download (just over one day at 3.5KB/s, compared to 57 hours for a 700MB ISO).

The installation program will detect your hardware for you, ask how you want to partition your disks (or allow you to manually setup your partitions), configure your network, install all the components, and allow you to pick a username. If all goes well, the installation should be as simple as the graphical installations which require much more memory and processing power (not to mention download time) to run.

“Fluxbuntu” is a portmanteau of “Fluxbox” and “Ubuntu.” While it is based upon Ubuntu, and uses the Ubuntu repositories for software distribution, it uses the Fluxbox window manager and ROX file manager to build a desktop desktop environment. Both of these programs are featherweights, fitting with the Fluxbox goal of building a system that uses minimal resources. Not to mention, the artwork is simply beautiful:

Those used to Gnome or KDE are likely to feel a little uncomfortable with Fluxbox at first. There is no “start” menu – you access the menu by right-clicking on the desktop. The Fluxbox panel at the bottom serves to indicate the current workspace, switch between open windows, and display a clock. The Fluxbox Configuration program is used to change various aspects of the user interface, although its cryptic options will probably frighten most new-comers (see the Limitations section below).

While Gnome uses Nautilus as a file manager and to display desktop icons, Fluxbuntu uses the light-weight ROX program. Compared to Nautilus, ROX has only a minimal feature set and a rather confusing interface, but there’s no argument that it’s a snappy, stable piece of software. Again, users coming from KDE or Gnome will feel out of place using ROX at first, but once you learn the interface it’s easy enough to use.

Web Browsing
In an effort to keep down the bulk, Fluxbox has tossed Firefox from its default set of applications, and substituted Kazehakase as the default web browser. It has some important features that you’d expect from any browser: tabbed browsing, bookmarks, and integrated search. It also has some more advanced features, such as mouse gestures and integrated RSS feeds (via the bookmarks bar), which makes it feel like you’re not using a completely stripped-down application.

It does, however, lack a pop-up blocker, which is just asking for trouble if you do any serious web browsing. Users that need more than the basics will probably wind up installing Opera, which manages to keep itself relatively light while providing a full-browser experience.

Messaging

Fluxbuntu ships with Pidgin (formerly Gaim), which is also the default on Gnome and has a home on many Windows desktops. It supports a load of protocols (AIM, Jabber, Yahoo, MSN, etc…) and is an incredibly mature piece of software. There’s not much to say about Pidgin that hasn’t already been said, since it’s such a standard application, and it’s nice to see that it was included in Fluxbuntu. It isn’t an incredibly heavy piece of software, so it fits in with the light-weight scheme of the operating system.

Mail
Correction: This section previously incorrectly identified the mail client in Fluxbuntu as Sylpheed.

Evolution, included by default in Ubuntu, is far too heavy to be the mail client for Fluxbuntu, so the developers have chosen Claws-Mail in its stead. I personally used Sylpheed, the ‘grandaddy’ of Claws-Mail, as my primary mail client several years ago, and can attest that it is a well-written, fully-functional email program. Users not looking for a complete Personal Information Manager will find it to be easy to use, and powerful enough for their day to day needs.

Claws-Mail has a host of features, such as mail filtering, an address book, and spam filters which make it amazingly powerful for its size. Thunderbird users will be impressed with its speed, and there aren’t any noticeable features missing. Claws is mature enough that some users may even consider switching to it as their default email client on other desktops!

While OpenOffice.org is an incredibly powerful office suite, it’s far too cumbersome to run on the dated hardware that makes up Fluxbox’s target audience. That being said, the makers of Fluxbox have included OpenOffice.org Writer with the distribution, probably as an effort to increase compatibility in a Microsoft Word-dominated world. Along with OpenOffice.org Writer, Fluxbox installs Abiword, which is a complete word processing program that manages to keep its footprint surprisingly small.

This is another application that users of other distributions might consider installing on their more powerful machines, due to its extensive feature-set and limited resource use.

Gnumeric is also included with Fluxbuntu, but oddly (see the Limitations section below) it is not placed in the menus – only when I went to install it from the repositories did I discover it was there. This leaves me to wonder what other programs were installed that I haven’t yet found. Unfortunately, those looking for a complete office suite will have to download various applications from the Ubuntu repositories.

Although there’s a lot of good things going for Fluxbuntu, you can’t use it for more than a few minutes before running into some serious limitations that make it largely unsuitable for beginners or even intermediate users.

A major limitation is the distribution’s lack of installed-by-default applications. While it makes sense to limit the number of programs included on the installation disk, most people intending to use Fluxbuntu as a desktop operating system will find themselves installing a slew of applications the first few days. As stated above, Fluxbuntu lacks a complete office suite, yet they chose to included two word processors. There is no PDF or image viewer, both of which take up little space (xpdf, xview) and are vital for any desktop.

Another concern is that some of the menu options point to terminal-based applications. Most disturbing of these is the Help command, which points to an Info file. While advanced users might see this and immediately understand how to use the manuals, most will experience a moment of shock the first (and likely, last) time they open this window. For all the time that the Fluxbuntu team spent on artwork – and I must admit, they did an excellent job making Fluxbuntu look amazing out-of-the-box – it’s disappointing to see that something as important as a basic User Guide was not included. Also of note is that most everything in the System menu opens a command-line based program (for example, Aptitude, which is used to add/remove programs), which is not as bad as the Help option, but still makes Fluxbuntu unsuitable for the average end-user.

I decided not to write about the multimedia functions in Fluxbuntu, since any machine on which it is installed will probably be too slow to playback DVDs or movies (and, in fact, will likely lack a DVD drive entirely), but I have to wonder about the choice of mp3/audio player. Instead of going with a product under current development, included in the distribution is the antiquated XMMS. While XMMS was once the end-all-be-all of simple MP3 players on Linux, its glory has been replaced with a selection of other players, with better support (Audacious and Beep immediately come to mind).

Finally, I can’t at all understand the situation with the Fluxbuntu menu. As I stated above in the Word processors section, I only found out that Gnumeric was installed by default after attempting to install it myself; it is not included in the menus at all. Furthermore, the menu includes two entries for applications – “Apps” and “Applications.” While most everything is included under “Apps,” Pidgin is listed under “Applications – Network” rather than “Apps – Net,” while Open Office is listed in “Applications – Office” rather than being placed alongside Abiword in “Apps – Editors.” While this isn’t a particularly big deal (although the complete exclusion of programs is a rather large problem), it smacks of an incomplete desktop. And – since there is no menu editor installed – most users will likely leave this as-is, rather than edit the menu configuration file.

Fluxbuntu bundles a relatively complete desktop experience in ultra-light package. Those with significant Linux experience under their belts might find the operating system to be a miracle that brings their older machines out of the graveyard and back to life. Another good place for Fluxbuntu – and one where I think it is currently more appropriately suited than the desktop, due to its limitations – might be on a home server, where you don’t want all the extras included in Ubuntu, but still want a GUI interface instead of a commandline-only machine. Fluxbuntu certainly has a niche to fill but it has a long way to go, and if you aren’t comfortable using the command line and editing the occasional configuration file, it’s probably best to stay away from Fluxbuntu.

This week Canonical, the company behind Ubuntu Linux, announced a partnership with Parallels, maker of the Virtualization products Parallels Workstation and Parallels Desktop for Mac. Consequently, the Parallels Workstation virtualization software is now available to download and install in Ubuntu Linux, completely supported by Canonical, and done entirely through the Add/Remove programs interface. This makes four different virtualization programs — three of which are installable via the package repositories — that run on Ubuntu Linux. (See the Correction: in the Installing VirtualBox section for more information)

This article compares four virtualization products available for Ubuntu Linux: the free, open source x86 emulator Qemu; the closed-but-free versions of VirtualBox and VMware-Server, and the commercial Parallels Workstation.

Installing Qemu
Installing Qemu is simple, with a twist: Qemu supports the KVM, or the Kernel Virtualization Machine, which enables it to use certain acceleration techniques that the other programs don’t use. As a result, we’re going to install a few extra packages along with the Qemu package itself:

The qemu-launcher is a graphical front-end to Qemu which we will see later; kvm and kqemu-common are the packages that allow Qemu to work with the Kernel’s Virtualization Machine. As a side note, the KVM will be installed by default and be fully supported in the next version of Ubuntu, Hardy Heron.

Installing Parallels
Installing any of these products is easy, but you may need to enable some extra repositories to install Parallels. To do this, go to your System Menu, and in the Administration submenu select “Software Sources.” Click on the “Third-Party Software” tab, and make sure that the line for “partner” software, “http://archive.canonical.com/ubuntu gusty partner” is selected:

Saving those changes, the package manager will update its lists, and you can now install Parallels, which comes in the logically named “parallels” package:

Installing Parallels will be a fun reminder of the Windows world for those who have become accustomed to Linux licencing – there’s a EULA (End User License Agreement) that must be clicked-past before you can install the program. Finally, once installed, you need to run Parallels and sign up for a 15-day free trial key — or purchase a key — assuming you don’t already have one.

Registration is quick and easy, requiring just a username and an email address; but it is another additional step you must take because of the software’s proprietary and commercial nature.

Installing VMware
To install VMware in Ubuntu, you’ll need to enable the “partner” repositories, just as you do for Parallels. Like Parallels, you’ll need to get a key from the VMware website, which you will be prompted for during installation. You’re required to submit a slew of personal information, but the key is given to you immediately (you don’t have to check your email for it, if you understand what I’m getting at). Unlike Parallels, this key is free for personal use, and does not expire.

Installing vmware can be done either through the graphical package manager (in the Third-Party section), or with the command:

The install is a bit lengthy compared to the other programs here, and the files are rather large (on the order of a hundred or so Megabytes), but all-in-all it’s a breeze as long as you have a fast connection.

Installing VirtualBox
Correction: The Open Source Edition (OSE) of VirtualBox is, in fact, available through the Ubuntu repositories, but that is not the version used here. Thanks to Daniel of ibeentoubuntu.blogspot.com for pointing this out!

VirtualBox is the one program that isn’t available via the Ubuntu repositories, but that doesn’t mean that installation is difficult. In fact, it’s as simple as going to the VirtualBox downloads page and downloading the appropriate package for your system. The menu for VirtualBox binaries is extensive, including multiple versions of Ubuntu, Debian, and Fedora, as well as some other distributions. For Ubuntu, select your version and architecture, and open it with the gdebi installer:

The package is a light-weight (compared to VMware) 17MB, and installs, thankfully, like any typical Ubuntu package: without prompting for any registration keys or personal information. This article will be looking the closed (but free) binary version of VirtualBox, not the Open Source Edition copy also available from the downloads page.

One caveat with VirtualBox: it requires that any users who will use the program be added to a special “vboxusers” group, so you’ll probably want to add yourself; and then you’ll have to logout and login for these changes to take affect. A minor annoyance, but an annoyance all the same.

VMware Server
Upon startup, the VMware Server Console will ask whether you want to connect to a local or remote server; in this test, I am using server on my local machine, but VMware can also be run over a network. Creating a new virtual machine is done via a wizard, which is simple enough for a beginner to use, but has enough customization options that an advanced user won’t feel restricted.

One problem I encountered, however, was that even after selecting Windows 2000 as my guest operating system, the wizard still recommended I use an emulated SCSI drive instead of IDE. I followed its advice, and the Windows 2000 install program wasn’t able to recognize any hard drives; it took me a while, but with the help of some kind people in the Ubuntu IRC chats I was able to correct the problem. Still, I was surprised that VMware made this mistake.

Running a Virtual Machine inside VMware is as simple as selecting the machine you want from the “Inventory” column, and clicking the “power on” button. Unlike its competitors, the Virtual Machine runs inside the VMware Server Console, instead of in a separate window. The Console also has the ability to run in a Kiosk-like mode called “Quick Switch”, which could be useful if you are using a dedicated machine for virtualization.

Adding additional hardware, or editing existing hardware, is simple and easy to do in VMware, and in my case the program auto-detected all of my physical devices correctly, and was able to map them to the guest OS without any problems.

VirtualBox
VirtualBox’s interface is written in QT, so it may look and feel a little out of place on a Gnome desktop. Anyone who’s ever used a KDE application in Gnome knows that it will function perfectly fine, however, even if it doesn’t quite look exactly right. VirtualBox doesn’t add itself to the Gnome menus by default, however, unlike the other three applications here (Parallels and VMware add themselves to the System Tools menu, while Qemu-Launcher places itself in Applications), so I loaded it up using the terminal command “VirtualBox”.

The VirtualBox user interface is the cleanest of all the applications I’ve looked at. Its wizard for adding a new Virtual Machine is elegant and effective, while its menu for editing an existing machine is the most user-friendly of all four programs.

VirtualBox’s own Preferences window is the most minimal of the bunch, allowing you to configure certain basic options such as the key combinations for focus grab/release, and the default language; unlike Parallels and VMware, it does not allow you to set a maximum amount of memory to allocate for the entire application as a whole.

Parallels
Like the other Virtualization clients, Parallels’ interface is sleek, polished, and user-friendly. Like VirtualBox, it’s written in QT, so it will feel at home on KDE, while working just fine in a Gnome environment.

Like the applications we’ve seen already, Parallels comes with a Virtual Machine Wizard which will take you through all the steps to create a new Virtual Machine, from the type of operating system you plan on running, to the amount of memory to allocate, to the size of the virtual hard disk you want to use. Modifying an already existing hardware configuration is as simple and straight-forward as in VMware.

Qemu-Launcher

Note: After writing this review I was told about the Qemulator program, which behaves similarly to the other applications’ interfaces. I will be adding a section on this program soon.

In contrast to the Virtual Machine Wizards and the sleek interfaces of the other programs, the Qemu-Launcher, which is your best chance at a GUI when it comes to using Qemu (another option, especially for KDE users is qtemu), is sorely lacking any user-friendly features, and will surely intimidate the beginner.
The Qemu Launcher is divided into five main tabs. The main tab is where you select your Virtual Machine, its hard drive, how much memory to allocate to it, and whether you want to boot off the CD-ROM (for installation of an operating system). The Hardware tab allows you to select things like the type and number of processors to use, what kind of video to emulate, and whether you want to use full-screen or windowed mode.

In the Emulator tab you can enable logging, but most importantly you can enable Acceleration. With the additional packages installed, you want to select “Full Acceleration” for the best performance. You can also set the process priority, which will “nice” the process to give other programs priority (or, optionally, to give the virtual machine priority over other processes). A higher number here means a lower priority, the highest being 19.

Aside from the difficulties in configuring a new virtual machine in Qemu, I’ve found the launcher to be somewhat of a hindrance. Since it is simply a front-end for Qemu’s command-line interface, and it is incomplete at best, there are times when it doesn’t work as expected. For example, if you set amount of RAM to be dedicated to the virtual machine too high and Qemu can’t allocate enough memory, clicking the “Launch” button will do absolutely nothing. It won’t give you an error, but instead, will do nothing at all. The only way to figure out the problem is to launch qemu-launcher in a terminal, and look for the error to be written there.

Once your virtual machine is setup, the aspect of virtualization that really matters is how well things run. I tested Windows 2000 Professional, using 360MB of RAM and 2GB hard drive files, on a Pentium 4 2.6Ghz host system. To make things as fair as possible, I restarted my computer after each trial, and didn’t use any other functions while testing the virtualizers.

VMware Server
VMware was incredibly impressive in its performance; installing Windows 2000 Professional took only 15 minutes, and booting into the new operating system after a fresh install took under a minute.

Once inside Windows 2000, I found applications to be as quick and responsive as if they were running natively on real hardware. Not once did I hit swap, which would have really slowed things down.

VirtualBox

VirtualBox proved itself as strong a contender as its heavyweight challenger, VMware. Installation took exactly the same time (15 minutes) from start to finish, with a few seconds margin of error for user-interaction.

Once booted, Windows performed like a champion inside VirtualBox, and felt even snappier than in VMware. I did notice some flickering inside the virtualization that wasn’t present in VMware, especially when maximizing/minimizing windows, or doing anything involving transparency (for example, dragging an icon about the desktop), but nothing that would hinder the user experience. VirtualBox is the only application reviewed that allowed me to set a size for video memory, so I left this at its default of 8MB.

Parallels
Surprisingly, the commercial Parallels application did not do as well as its contenders in this part of the test. From beginning to end, installing Windows took twenty-seven minutes, or 180% of the time it took in VirtualBox or VMware. This was surprising and disappointing, and once inside Windows I found its performance to be lacking, as well.

Booting up took considerably longer than in either of the other applications – to be sure, I ran this test several times. Unlike VirtaulBox and VMware, applications were not as responsive one would expect on real hardware; it was obvious that the operating system was running within a virtualized environment. Along with maxing out my CPU, Parallels inexplicably began using my swap partition although there were several hundred megabytes of real memory free on my system. While I can’t necessarily attribute this to a problem in Parallels, it does greatly slow things down, and is rather unexpected. Video performance was lack-luster; menus flicker as they scroll, and there is a noticeable delay when drawing windows. Overall, Parallels wasn’t impressive in its performance, to the point that I wonder if there is a bug in its default configuration on Ubuntu.

Qemu
Note: A few people have mentioned in forums that you should start your Qemu virtual machine (or you can just start the launcher itself) in a terminal, in order to make sure that full KVM support is working. I did that in these tests, and didn’t get any errors from KVM/kqemu so I am assuming that it acceleration was working correctly (set to “Full”).

Qemu was interesting in that, instead of using my physical processor as the processor for the Virtual Machine, it emulated a Pentium II. This may have something to do with the fact that Qemu, unlike the other applications here, is not a true Virtualization platform; it is an x86 emulation platform, capable of full virtualization.

Qemu took just under 25 minutes to install Windows 2000, coming in well behind VirtualBox and VMware, but beating out Parallels. While video performance was acceptable, there was a noticeable lag in opening applications inside the virtualized Windows platform. Qemu also was the only application in which I wasn’t able to get sound to work inside the guest operating system (the other applications set this up out of the box). Lastly, my biggest complaint by far is that Qemu continued to eat up CPU time even when the virtual machine was idle; I didn’t experience this problem with any of the other applications. These problems severely hinder Qemu’s ability to compete.

This article looked at four different products for virtualization in Linux, specifically Ubuntu Linux. The findings were interesting – the only product that requires the purchase of a licence for personal use, Parallels, actually performed the worst of the group. Qemu did well for a completely free-as-in-speech application, although VMware and VirtualBox blew the competition away in terms of performance.

Of these applications, VMware and VirtualBox definitely come out the clear winners. For individual use, where the ability to run a single Windows desktop inside of Linux is all that is needed, VirtualBox turns out to be my preference. It’s much smaller than VMware, which comes packed with extra features that aren’t necesary for the end-user. If your needs are different, however, you’ll probably want to give VMware a closer look. Overall, the maturaty and stability of all four applications was impressive enough to call virtualization on the Linux desktop “complete.”

Its official release is just under three months away, and Ubuntu 8.04, or Hardy Heron, promises some great improvements over the already user-friendly Ubuntu 7.10 (also known as Gusty Gibbon).

This is a look at the fourth Alpha release of Hardy; including many of the applications that are now included by default and the major changes that will improve stability and usability. Among these are the addition of Firefox 3 and Remote Desktop on the applications side, and a new method for systems control known as Policy Kit, which enables the administrator to unlock certain functions for normal users.

Default Bittorent Client: Transmission
Transmission replaces the Gnome Bittorrent Client as the default BT client in Hardy. This is great news for its large number of fans, though Deluge and Azureus users might feel a little shafted.

Transmission is well-known for its light-weight, simplistic approach to torrenting; and it has just recently received a user interface overhaul.

Remote Desktop
Hardy installs by default with a Remote Desktop utility, allowing users to connect to your X session either locally or over the web.

You can set various options, such as required encryption, the port specifications, and whether users may control your session (or just observe). This is a great addition, as it will make remote technical support far easier for both parties. Like sshd, I expect that it will quickly become a necessary accessory to all Linux distributions.

Vinagre VNC Client
Along with allowing users to remotely access your Hardy desktop, Ubuntu will now include the Vinagre VNC by default upon installation.

This will allow access not only to remote Ubuntu machines running Remote Desktop, but to Windows XP and Vista machines with Remote Desktop enabled.

Firefox 3
The long-awaited revamp of Firefox, version 3, will be included by default in version 8.04 of Ubuntu. Firefox 3 is faster than its predecessors, uses less system resources, and include better integration into the Gnome desktop.

More information about Firefox 3 can be found on its official Mozilla wiki site.

Brasero Burning Program
Brasero is the new default burning program for Ubuntu. Its interface is sleek and easy to use, and certainly gives k3b a run for its money.

Brasero includes a Nero-like start screen, has capabilities for making 1:1 disk-copies, audio CDs, burning disk images, and data CDs. The interface for audio and data is drag-and-drop, with a file manager on one side and your CD/DVD destination on the other, reminiscent of Norton Commander. On the outside, its interface is simple, and it will automatically detect many options for you; however, its slew of configurable options will be a comfortable addition for the more advanced Ubuntu users.

World Clock and Weather Applet
Replacing the standard, run-of-the-mill clock-and-calendar applet is a new applet that not only displays the time and date, but also includes a weather forecast.

This does away with the need for two applets (time/date plus weather), and has an elegant interface which users are sure to love.

Simpler Administration Access for Users
One of the biggest complaints of users new to Linux (fresh from Windows) is that many functions require root privileges in order to work. While this is important for security and stability of the system, sometimes it can get in the way of productivity.

New in Ubuntu 8.04 is the Policy Kit, which allows the user to set certain functions as “unlocked,” allowing the use of selected administrative tools as a normal user.

Another administration improvement is the Run-As dialog which prompts the user for credentials when asking to run a program as another user:

But most impressive is the Authorizations panel, which allows complete control over what system functions each user is able to have.

While it’s still in its early stages, and far from polished, this is a great addition for those who aren’t used to the Linux way of doing things. Administrators will be pleased, since there is now a GUI to quickly allow (or disallow) users and groups to perform tasks such as the mounting of external devices and rebooting the machine.

A New Theme for a New Version
There has been much speculation as to what the new theme for Ubuntu will look like; work on it isn’t complete, and the Alphas still use a variation on the Human theme. Therefore, there can’t be much said about the new theme, but for more information on its development you can visit the

GVFS
GnomeVFS is being replaced by GVFS, which will do away with many of the bugs experienced in Ubuntu 7.10 and previous releases. For example, it should now be possible to restore files from the trash, a simple feature that has been sorely lacking in Gnome since its inception. Among other improvements, GVFS will also allow better integration with the new Policy Kit, as well has enhancing performance of various file-system operations.

Xorg 7.3
There isn’t much to say here, except the new version of Xorg will help with multi-monitor setups and graphics card detection. Both have been problems in the past for Linux, and the emphasis here is on auto-detecting the user’s hardware, and building an appropriate xorg.conf file; the less tinkering with text files, the better, in my book.

Prefetch Speeds Things Up
Prefetch is now included, which will speed up both boot-time, and the time it takes to launch common applications. Developed as a Google Summer of Code project, Prefetch is both a patch to the Linux kernel and a userspace daemon.

Kernel Version Update
The current version of Ubuntu, 7.10, runs Linux 2.6.22, while Hardy’s Alpha4 comes with 2.6.24. What’s the difference? Well, for one thing, Kernel-based Virtualization is now officially supported inside Ubuntu. Another improvement is the addition of power-management to the 64-bit edition of the kernel, which is a blessing for those x64 users with laptops and notebooks.

Above is the output of uname running in Hardy; note also that, upon opening a terminal, the user is told about the “sudo” command – another authorization addition to the user interface, aimed at making it easier for newer Linux users to adapt.

Currently in alpha, Hardy Heron is slated for release on April 24th of 2008, with the first beta release scheduled for March 13. Most users will want to wait for it to become stable, but for those who simply have to have it now, upgrading from Ubuntu 7.10 is simple. Know that this is probably a bad idea if you can’t afford to destroy your system. If you aren’t comfortable with many, many bugs in your soup, wait a few months for everything to be worked out. That being said, here’s the process:

Option A) Download the ISOs
You can simply download a LiveCD/installer from the CD-Image Release site and go from there. This is probably your best choice; you can create a separate partition, or choose not to install it at all. But for those who really like to live on the edge, there’s always:

Option B) Upgrade from 7.10
Upgrading to Hardy Heron from Gusty Gibbon is as simple as upgrading to Gusty from Feisty. Start with the command:

The “-d” flag tells the update-manger to look for development versions, of which Hardy is currently one. It will run, and you’ll see this option available:

Think about it for a little while, ponder whether it is really worth it – and then go ahead and click upgrade. You, fine dare-deviler, are on your own from there.

File backups are a key element for every user and on every computer. Whether it be an office setting or a home desktop machine, backups are essential – your hard drive will fail at some point; you will need a backup, and you’ll be kicking yourself if you learn this lesson the hard way.

While it’s always been possible to use tools such as cron and rsync to backup files in Linux, there has never previously been an easy-to-use program that end-users and beginners could feel comfortable configuring, until now. TimeVault allows you to backup your system with snapshots; this means that if you lose a file (or an entire file system), or want to return a document to a previous state, it is as simple as selecting the file through TimeVault’s Snapshot Browser and clicking “Revert.” Though it is currently in Beta, I have found TimeVault to be a complete solution for backups, and as of yet haven’t discovered any single major problem.

TimeVault is a relatively new project, and as such it isn’t available through most distributions’ repositories. This makes installation a little bit harder, but not much. Here’s a step-by-step walkthrough of installing TimeVault, tailored to Ubuntu Linux (but will work on any platform, especially those that are debian-based):

Installing TimeVault
First, go to the TimeVault Download Site and select the most recent package appropriate for your distribution. For Ubuntu and Debian users, download the TimeVault 0.7.5 Beta Candidate DEB file; Other distributions will have to use the tarball version, which will not be covered here.

Select “Open with” and choose the GDebi Package Installer (or, it can be opened from the commandline with gdebi-gtk command). Once the download is complete, the package installer will open, and you will see this window:

Clicking “Install Package” will install TimeVault, as well as any of its dependencies (shown in the “Status:” line). Following installation, you will be informed that it is necessary to log out and log back into Gnome in order to use the Nautilus features included in TimeVault. But before you do this, you should setup TimeVault to start when Gnome starts:

Startup Notifier
Go to your System menu, then Preferences, and open the Sessions control panel. Here, in the “Startup Programs” tab, you will see a list of applications that start when you log-in. TimeVault does not add itself by default, and though it is not entirely necessary, it makes sense that you add TimeVault to this list. To do so, click the “Add” icon, and you will see this dialog:

For the fields, insert this information:
Name: Timevault Notifier
Command: /usr/bin/timevault-notifier

Comment:The Timevault Notifier Tray Tool

The “Name” and “Comment” fields can be different, but make sure the command line is correct, or it will fail to work properly. Now, TimeVault Notifier has been added to your startup programs list; just make sure it is checked off, and close the window:

Now you can logout and log back into Gnome, and your TimeVault Notifier will appear in your system tray.

In order for TimeVault to start working, you have to configure it by telling it what to backup, and to where. TimeVault offers some basic configuration options, which can be accessed by right-clicking on your TimeVault Notifier tray icon, and selecting Preferences. (Note: If you choose not to have TimeVault-Notifier activate at startup, it is accessible through the Applications — System Tools menu)

TimeVault’s configuration is divided into four tabs: General, Include, Exclude, and Expire.

General
The General tab allows you to configure important aspects of your backup system, such as whether you want to have your snapshots automated, whether you want to be notified when files are backed up, and more importantly, where you want your backed-up files to be located. Select this under the “Storage” menu; obviously, an external hard drive, a network mount, or a remotely-mounted internet drive are the best options.

Here, I created a folder on my external drive called “My Backups” and enabled automation.

The “Size Options” let you disable backing up of very large files (I set 128MB as my limit), and you can have TimeVault reserve space on your backup drive, in case you start to run out. The time increment is the amount of time TimeVault will wait each time it detects a new file to be backed up; so, if it detects a change to one set of files, and then to another set, I have told TimeVault to wait 5.0 mins before continuing on.

Include Directories

The Include tab is where you tell TimeVault what folders and files you want backed up.

Here, I have selected to backup my documents, binary files, and photographs; by default, TimeVault will include your /etc folder, which contains the configuration files for your system. This can be very useful if you make a disastrous change to your computer at some point; you can simply revert to the old, working version.

You will probably want to tell TimeVault to create a “Baseline” copy of each directory you are backing up; this will force it to create a backup of the files in their current state. When you do this, they are added to the Pending Snapshots window, shown below.

Excludes

TimeVault comes pre-built with a list of files to be excluded. Most of these, such as /proc and /dev are folders used by the system,
that you wouldn’t want to backup. You should also probably add your backup folder to this list,
so you never run the risk of recursively backing up your backups (sounds bad, doesn’t it?)
If you look closely, you will see that your .mozilla folder is excluded from backups
(I am assuming this is because your web cache and cookies are constantly being modified,
so taking snapshots of this directory would be a waste of resources, and largely useless).

Unfortunately, this makes it difficult to backup your bookmarks.html file for firefox, since TimeVault only
allows directories to be included in the back-up list (see Limitations below);
the only workaround would be something like running:

inside your firefox profile directory.

Expire
The Expire settings allow you to choose how often you want snapshots to be deleted. By default, TimeVault will remove a snapshot over a week old, if it has ten later snapshots (modifications), or uses over a particular amount of space. These settings should be fine for most users, but you may want to change it depending upon your computing habits; if you tend to constantly edit and save a file, but still want to be able to access the original versions of it, you may want to increase the value of the second row, for example.

The lower options allow you to control how TimeVault handles merging snapshots that were taken during a particular time-frame. You can select to have it keep only the most recent shot of a file per hour (the default), or longer, if you wish.

Now that TimeVault is properly configured, you should run your Baselines, to create a current snapshot of the files and directories you’ve included to be backed up. As stated before, this is done in the Includes tab of the Preferences window. Once you add these tasks, you can open up the “Pending Snapshots Window” by clicking on the TimeVault-Notifier icon in your system tray:

You can see here the operations TimeVault is waiting to perform, at what time they will be executed, what files will be backed up, and how large they are. Since this is its first run, the Event column will all be “Baseline,” but once TimeVault begins to notice files being edited, it will mark the event as “Changed.” You can also right-click on a job to abort it, or use the Delete icon.

Notifications
If you have notifications enabled, you will see popups like these every so often, letting you know what operations TimeVault has performed:

It’s a matter of personal taste whether you want to leave these enabled; during the base-line creation they can be constant, and somewhat annoying, but once the application is only monitoring your system for changes, it’s a nice reminder that your files are tucked away, safe and secure.

Snapshot Browser
The Snapshot Browser is the interface used to view previous snapshots, revert to old versions, and restore files. Access it by right-clicking on the TimeVault icon in your tray, and selecting “Snapshot Browser.” You may wish to use the Root option if you want to restore certain files, such as those in the /etc directory. You can scroll chronologically using the top bar, and view the backup reports for each snapshot session:

Nautilus Integration
In directories backed-up by TimeVault, you will see an icon that will launch the Snapshot Browser, showing you all previous snapshots of the files contained inside that folder.

Backup File Structure
While it’s not really something meant to be seen by the end-user, it’s worth mentioning the structure of TimeVault’s output; these are the files you will see if you browse your Backup directory. There are three folders:

Catalog – this contains an SQLite database, used by TimeVault to track backups, and is really of no interest to end-users.

Pending – this directory is typically empty; it’s likely the temporary directory used while files are being transfered into the Internal folder.
Internal – this is the folder of interest; it is where all your backups are kept. They are sorted first by owner (UIDs, or User IDs, are the unique numbers assigned to each user on your Linux system). uid-0 is root, and will contain items such as the configuration files found in /etc. Other UIDs are other users; your personal files are probably in a uid of 1000 or greater.

Each UID directory contains another folder, which is the day on which the snapshots are created. This follows the same lines as the interface in the Snapshot Browser. Going one level deeper will show you the files that have been backed up…

Filenames Include Hashes!
While all your files are in their original, unchanged format, they have been renamed with a hash of its contents preceding the original file name. So, you can view your backed-up files from here, but they will be a little harder to find. Better would be to use the Snapshot Browser, selecting the file you wish to see and clicking “Open.” While I understand this method of naming, read on in the “Limitations” section why I believe this might not be the greatest solution:

Can’t reschedule times
One of the biggest gripes I have with TimeVault is that it doesn’t allow you to reschedule your backups for a later date, or specify a time (preferably during which you’re not using the computer) to do backups. While this is usually only a minor annoyance, it can become frustrating when TimeVault starts copying files when you’re right in the middle of something intensive. A nice feature would be able to postpone a backup in the “Pending Snapshots” window.

No encrypted backups
One of the things I really want in a backup program is the ability to automatically encrypt and password protect the backed-up files. Timevault doesn’t offer this, and based upon the manner in which it backs up files, I don’t think it will be a feature at any time in the future. The only workaround I can think of here is to use an encrypted filesystem, which is less than ideal in most configurations.

No Backups of Single Files
As far as I can tell, TimeVault will only allow you to backup directories, not single files. While this is what you want to do most of the time, it can cause problems. Backing up your firefox bookmarks list, for example, requires creating a directory, and putting a link to the bookmarks file inside the directory (as done above). This is something I’d really like to see fixed soon.

No quick restore without Timevault
Due to the way that TimeVault names files (including the hashes in the copied version’s filename), there’s no easy way to restore files using your file manager. Instead, you must use the Snapshot Browser to select an old version of a file to which to revert; not a terrible limitation, but in the event of a system meltdown, it would save some time.

TimeVault finally offers a complete, easy-to-use, intuitive backup system for Linux. While advanced and experienced users have been able to schedule backups using rsync, cron, and other tools, new users will find Timevault a comfort; knowing that their files can be easily and safely backed up, and reverted to an older state if necessary. The interface is relatively intuitive, and although the configuration could be a bit simpler, beginners should have no problem setting up TimeVault to keep their files safe.

I’m eagerly looking forward to the next release of TimeVault, and I hope they continue to add new features; backup software is an essential part of any operating system, and TimeVault gets the job done well in keeping user’s files safe.

Installation
Installation is as simple as running the following command on debian-based distrobutions (such as Ubuntu):

Or, you can visit the Alexandria Homepage to see more information on installing Alexandria on your own distrobution.

Interface
The main interface for Alexandria is a simple, sleek window, separated into two columns; on the left side is your library collections, allowing you to sort books in whatever manner you please, be it by location, genre, owner, etc. The right column displays the books contained within the selected library.

Adding an Item
Adding an item to your library is as simple as clicking the “Add Book” icon, and entering its ISBN. Alexandria will search several sites, including Amazon.com, Barnes & Noble, and several non-English sites. When it finds your item, it is added to the library with all the relevant information, including its cover image, author, publisher, and binding.

For items without ISBNs (mainly older items), you can manually add the book’s information, and select a cover image manually from a website like Amazon.com, or scan in your own image.

Managing Items
Once an item is added to the database, you can then rate the book on a scale from one to five stars, with the highest-rated items receiving a “heart” icon, indicating that it is one of your favorites. You can also add notes, or indicate whether you’ve loaned the book out, including the date of the loan, and who it has been loaned to; so you never lose track of your books.

Searching your Collection
One of my favorite features of Alexandria is its search function. The program allows you to search by author, publisher, title, or other characteristics. This is particularly useful if you have a large collection, and you’re looking for a specific item; or to see if you’re missing a title by a favorite author.

Exporting
Aleandria allows you to export to various formats, including Tellico, KDE’s physical media collection software (which also supports music and videos, among others). But my favorite is the ability to export to a static web page, to show off your library to friends. The webpage displays book covers, titles, publishers, and ISBNs. This is a great feature to share with friends that are as interested in reading as you are; you can view each other’s libraries and borrow books – and then indicate that the book has been loaned out using Alexandria’s interface.

Web View

Conclusion
While Alexandria has some limitations that don’t make it as powerful as Tellico, it’s more than adequate for small to mid-sized home libraries. It’s a great way to keep track of your books, and show off your collection to friends and family. I haven’t had any stability problems, and I look forward to the new features that are currently in the works!

Links
Alexandria Homepage
Tellico Homepage