TechPortal

Your daily source for Tech news, views, gadgets, and lots more…

RAR is a proprietary archive file format that supports data compression, error recovery, and file spanning. It was developed by a Russian software engineer, Eugene Roshal (hence the name RAR: Roshal ARchive), and is currently licensed by win.rar GmbH as well as being an acronym for ‘Rational And Reliable’.

The file extension RAR is used by RAR for the data volume set and .rev for the recovery volume set. In previous versions, if a RAR-archive was broken into many smaller files (a “multi-volume archive”), then the smaller files used the extensions .rar, .r00, .r01, .r02 etc.

Comparing to ZIP file format, file extension RAR provides a number of advanced features: more convenient multipart (multivolume) archives, tight compression including special solid, multimedia and text modes, strong AES-128 encryption, recovery records helping to repair an archive even in case of physical data damage, Unicode support to process non-English file names and a lot more.

File extension RAR supports data compression, spanning and error recovery. Though less common, RAR files are also used to denote a Resource Adapter Archive, which is the specified format for the deployment of resource adapters, or connectors, on IBM’s WebSphere Application Server.

RAR files typically take longer to compress, though have a higher compression ratio than other popular archiving formats such as File extension ZIP or GZIP. The exact improvement in compression largely depends on the types of file being compressed. As well as data compression, RAR files support spanning, which means that archives can be split across multiple volumes. In this case, the first file in the archive will retain the RAR file extension, with subsequent volumes numbered in the form File extension R00 up to File extension R99. The entire archive can be browsed by opening the original RAR file. This was a particularly useful feature for compressing large volumes of data. Some RAR files may also contain a recovery volume, which allows the contents of an archive to be restored in the even of minor corruption. RAR archives also support password protection and strong 128-bit AES encryption algorithms.

Currently, RAR files can only be created using commercial software such as WinRAR or Squeeze, both of which are only available on Windows. The command-line version, RAR, can be used to create and extract archives on Windows, Mac, Linux and UNIX-based operating systems. However, RAR file can be opened and the contents extracted by several applications such as WinZIP, WinAce and Stuffit, courtesy of the freely available UnRAR source code from RARLAB which allows developers to incorporate RAR decompression into their applications.

Via its own research Microsoft has determined that its Internet Explorer 8 browser (expected to launch next week) is faster than both Google’s Chrome 1.0 and Firefox 3.0.

The company claims that under its own test suite, IE8 is capable of loading web pages three times faster than Firefox, and one third quicker than Chrome.

Microsoft has hit some rough spots since the release of Vista, and its once unassailable browser market share has been on a perpetual downward trajectory. In an attempt to win back both developers and users who have begun to stop focusing their development efforts on Internet Explorer users, the company has even produced a 14-page report and video to prove their claim.

Lab tests are great, however typically their results are very difficult to reproduce in real life. In this particular test, conducted in January 2008, Microsoft loaded the 25 largest websites, at faster rates than other browsers.

The company also kept add-ons to a minimum: by running tests with no browser plugins, add-ons or third party toolbars, and a minimum of three in some instances, the browser was able to load faster. Under normal, everyday use this would be next to impossible.

The page load times were measured using “visual cues” or a combination of visual cues and the page interaction to be sure the page had actually loaded. The company felt relying on the done notification would have provided results that were inconsistent on some sites, such as those which contain AJAX.

Under normal circumstances this information really wouldn’t be news unless you were a developer, however in this instance the use of the browser has tanked dramatically, and Microsoft is grappling in effort to get the attention of both the industry and its customers.

Source :APCMag.com

When one chooses the Gentoo way patience is most certainly a virtue — but you can speed it up a bit with this simple and effective tweak. Speed up Gentoo? Read on!

I should probably feel ashamed — I’ve been using Gentoo since *counts on fingers*… almost since it began. Back when Daniel Robbins was leading the charge, and it quickly became the world’s most popular source distribution.

Because that’s the other side of the double-edged source-distro sword: optimised performance, complete configurability — and anywhere from minutes to hours for any application you install. Now, granted I have Gentoo on a RAID 0 stripe, but oh my gosh how did I ever miss this simple tweak?

The following line added to /etc/fstab will speed up your compile times, reduce disk thrashing, and make you attractive to the opposite sex:

none /var/tmp/portage tmpfs size=1512M,nr_inodes=1M 0 0

Just make sure you set ’size=’ to a value less than your total RAM. Here I’m using 1.5G for my 2G system.

What does it do? It knocks out the slowest component of your system, your hard drive, from the compiling process mapping /var/tmp/portage — where Portage stores all the temporary files during compilation — to a tmpfs in memory. And tmpfs, bless it, is dynamic and only uses as much as it needs — so the 1.5G setting I use is only its maximum allowable size. If nothing is being compiled, no memory is used.

The speedup is nice but dependent on what’s being compiled, and the speed of your drive (less benefit for 10k RPM drive users, for example). Only caveat: some large compiles may need more space (Open Office for example needs at least 3G) and when it does the kernel will page tmpfs out to swap, slowing the whole process down. Only a handful of applications may need this in the hundreds you compile however. Still, if you have less than 1G, you don’t want to be allocating most of that, and you may need to disable it (just umount /var/tmp/portage) for the super-large packages.

tmpfs has many other uses, but this one never occurred to me until I read it in passing in a forum post. Simple and effective, all the best tweaks are.

Source: APCMag.com

When one chooses the Gentoo way patience is most certainly a virtue — but you can speed it up a bit with this simple and effective tweak. Speed up Gentoo? Read on!

I should probably feel ashamed — I’ve been using Gentoo since *counts on fingers*… almost since it began. Back when Daniel Robbins was leading the charge, and it quickly became the world’s most popular source distribution.

Because that’s the other side of the double-edged source-distro sword: optimised performance, complete configurability — and anywhere from minutes to hours for any application you install. Now, granted I have Gentoo on a RAID 0 stripe, but oh my gosh how did I ever miss this simple tweak?

The following line added to /etc/fstab will speed up your compile times, reduce disk thrashing, and make you attractive to the opposite sex:

none /var/tmp/portage tmpfs size=1512M,nr_inodes=1M 0 0

Just make sure you set ’size=’ to a value less than your total RAM. Here I’m using 1.5G for my 2G system.

What does it do? It knocks out the slowest component of your system, your hard drive, from the compiling process mapping /var/tmp/portage — where Portage stores all the temporary files during compilation — to a tmpfs in memory. And tmpfs, bless it, is dynamic and only uses as much as it needs — so the 1.5G setting I use is only its maximum allowable size. If nothing is being compiled, no memory is used.

The speedup is nice but dependent on what’s being compiled, and the speed of your drive (less benefit for 10k RPM drive users, for example). Only caveat: some large compiles may need more space (Open Office for example needs at least 3G) and when it does the kernel will page tmpfs out to swap, slowing the whole process down. Only a handful of applications may need this in the hundreds you compile however. Still, if you have less than 1G, you don’t want to be allocating most of that, and you may need to disable it (just umount /var/tmp/portage) for the super-large packages.

tmpfs has many other uses, but this one never occurred to me until I read it in passing in a forum post. Simple and effective, all the best tweaks are.

Source: APCMag.com

Just when Microsoft wants you to forget about alternatives to Vista, along comes the Debian based project that allows you to not only install Linux on a Windows system, but do it right under its nose.

Just when Microsoft wants you to forget about alternatives to Vista, along comes the Debian based project that allows you to not only install Linux on a Windows system, but do it right under its nose.

Linux is renowned for being a compatible operating system, catering to all manner of protocols, filesystems, and co-habitation with other operating systems including Windows.

But usually you need to install it from a clean boot. Now, thanks to Debian coder Robert Millan it’s possible to not only install Linux alongside Windows, but through it as well.

While the Ubuntu team initially announced exactly this concept earlier this month along with a downloadable prototype, Millan appears to have completed a finished version replete with a promotional (if not controversial) website to boot — goodbye-microsoft.com.

By simply clicking the ‘install Debian’ link the installer will download and, when run, install Debian through your Windows desktop. Aside from the obvious ease of use this sort of install mechanism provides, Millan also listed the following benefits in his announcement post to the Debian lists:

- Migrating to Debian on hardware without CD drive (or USB boot
capability).
- Migrating to Debian for users who have no idea how to burn an ISO and/or how to configure their BIOS for CD boot.
- Migrating a system to Debian immediately, at any time, wether
we have a boot CD at hand or not.
- Advocating Debian to Windows users. Rather than give them a CD (expensive), give them a piece of paper with an URL.

It all sounds good, but how does it work?

The idea of booting Linux from a Windows partitions isn’t entirely foreign — for a long time we’ve had distributions capable of installing to a FAT16/32 partition and boot directly from it (see loadlin), but we’ve long since moved away from those days.

And in fact this new method isn’t too dissimilar. Here the appropriate Debian images are download and configured, and installed as a loopback ext3 filesystem on the NTFS Windows drive — for those unfamiliar, this means a self-contained filesystem created as a single large file on the NTFS partition — then mounted as a loopback device within Linux when the kernel boots.

Which is the second part of the magic — the installer finishes by using GRLDR from GRUB4DOS to load the loopback image directly after being booted from the Windows boot manager. So effectively both Windows and Linux are loaded from the NTFS partition, although to Linux it’s a world within a world. No partitioning, no data moved, and aside from updating the boot loader, uninstalling would be a matter of deleting the image file.

It’s definitely a nifty trick, easy to do, and from the average Windows user’s perspective, transparent. It should probably explain however that the Windows NTFS partition becomes home to both OSes, lest they later decide to update or even remove Windows in favour of Linux and nuke their Debian install in the process.

It’s also not entirely clear how one could extend the disk space for the ext3 container file, though of course Linux can read the parent NTFS filesystem (and potentially delete said container file, now wouldn’t that be interesting).

On that note it’s probably best to wait until this mechanism has had the pants tested off it before you forward it around to your soon-to-be-converted friends, but it is none the less an ingenious use of loopback filesystems and a Linux install that hits Windows users where it matters most — at the simple click of a button.

It also bodes well for future installs of Linux regardless of distribution. The Ubuntu system, while still being developed, apparently goes a step further and pulls in necessary information from Windows itself, streamlining the process for the new Linux user even more.

If you want to give either a try, the Debian installer is here and the prototype Ubuntu one here.

Source :APCMag.com

Just when Microsoft wants you to forget about alternatives to Vista, along comes the Debian based project that allows you to not only install Linux on a Windows system, but do it right under its nose.

Just when Microsoft wants you to forget about alternatives to Vista, along comes the Debian based project that allows you to not only install Linux on a Windows system, but do it right under its nose.

Linux is renowned for being a compatible operating system, catering to all manner of protocols, filesystems, and co-habitation with other operating systems including Windows.

But usually you need to install it from a clean boot. Now, thanks to Debian coder Robert Millan it’s possible to not only install Linux alongside Windows, but through it as well.

While the Ubuntu team initially announced exactly this concept earlier this month along with a downloadable prototype, Millan appears to have completed a finished version replete with a promotional (if not controversial) website to boot — goodbye-microsoft.com.

By simply clicking the ‘install Debian’ link the installer will download and, when run, install Debian through your Windows desktop. Aside from the obvious ease of use this sort of install mechanism provides, Millan also listed the following benefits in his announcement post to the Debian lists:

- Migrating to Debian on hardware without CD drive (or USB boot
capability).
- Migrating to Debian for users who have no idea how to burn an ISO and/or how to configure their BIOS for CD boot.
- Migrating a system to Debian immediately, at any time, wether
we have a boot CD at hand or not.
- Advocating Debian to Windows users. Rather than give them a CD (expensive), give them a piece of paper with an URL.

It all sounds good, but how does it work?

The idea of booting Linux from a Windows partitions isn’t entirely foreign — for a long time we’ve had distributions capable of installing to a FAT16/32 partition and boot directly from it (see loadlin), but we’ve long since moved away from those days.

And in fact this new method isn’t too dissimilar. Here the appropriate Debian images are download and configured, and installed as a loopback ext3 filesystem on the NTFS Windows drive — for those unfamiliar, this means a self-contained filesystem created as a single large file on the NTFS partition — then mounted as a loopback device within Linux when the kernel boots.

Which is the second part of the magic — the installer finishes by using GRLDR from GRUB4DOS to load the loopback image directly after being booted from the Windows boot manager. So effectively both Windows and Linux are loaded from the NTFS partition, although to Linux it’s a world within a world. No partitioning, no data moved, and aside from updating the boot loader, uninstalling would be a matter of deleting the image file.

It’s definitely a nifty trick, easy to do, and from the average Windows user’s perspective, transparent. It should probably explain however that the Windows NTFS partition becomes home to both OSes, lest they later decide to update or even remove Windows in favour of Linux and nuke their Debian install in the process.

It’s also not entirely clear how one could extend the disk space for the ext3 container file, though of course Linux can read the parent NTFS filesystem (and potentially delete said container file, now wouldn’t that be interesting).

On that note it’s probably best to wait until this mechanism has had the pants tested off it before you forward it around to your soon-to-be-converted friends, but it is none the less an ingenious use of loopback filesystems and a Linux install that hits Windows users where it matters most — at the simple click of a button.

It also bodes well for future installs of Linux regardless of distribution. The Ubuntu system, while still being developed, apparently goes a step further and pulls in necessary information from Windows itself, streamlining the process for the new Linux user even more.

If you want to give either a try, the Debian installer is here and the prototype Ubuntu one here.

Source :APCMag.com

One of the greatest strengths of open source – freedom of choice and open code for anyone to work and develop on – is also one its weaknesses. The Portland project may be about to change all that.

Without getting too deep into the philosophical and developmental debates, one of the greatest strengths of open source — freedom of choice and open code for anyone to work and develop on — is also one its weaknesses. In a world where everyone can do their own thing, standardisation and consistency aren’t usually invited to the party. The LSB (Linux Standard Base) was one key step in helping to solve this, and now Portland is another.

Many years ago now, from two completely separate toolkits, the popular Gnome and KDE desktops were born. Built as they are for themselves, interoperability between applications of one on the desktop of the other wasn’t a priority for a rather long time — it’s only been in the last couple of years for example that you could install applications that will work in KDE’s system tray the same way it will with Gnome’s, for example. While the applications always ran, as long as you had GTK or QT installed, they weren’t generally aware of their host desktop if it wasn’t native.

And this was especially true when it came to updating menus. If the functionality was included at all, the author of a Gnome application for example would only bother about updating menus under Gnome, leaving KDE users to scratch their heads and manually create a new entry.

Generally, it’s been up to the package maintainers of distributions to fix this, to create install scripts that properly add a program to the menus of the installed desktops (KDE, Gnome, Fluxbox, XFCE and others). But the onus then is on the distribution package maintainers to do this for each and every application where applicable, and it doesn’t help at all if users install programs not part of the distribution’s package portfolio.

So back to Portland — a great idea that’s sorely needed, Portland provides a set of tools primarily for use at application installation that standardise the process completely, so no matter whether you use Gnome or KDE, an application can simply call the same Portland tools to update the user’s desktop, and Portland will do the rest.

While, generally, this isn’t going to remove the entire workload of package maintainers — as distributions generally have their own configuration formats to maintain — it’ll certainly help automate the process.

Best of all, however, is if everyday developers start using the Portland tools — this way users can install an application not provided by a distribution and find, miraculously, that their desktop is aware of it and they can get right on and use it, just like under Windows. It’s such a simple thing, but it’s been missing from Linux for too long. If there’s one thing we can learn from Microsoft, it’s that consistency matters.

Actually, it’s more just common sense: people want to use their desktops, so don’t make them work for it. Linux as a desktop has come a long way — Ubuntu is a testament to this — and projects like Portland that are one more step in the yellow brick road.

You can check out Portland’s releases and documentation at the official homepage.

Source : APCMag.com

One of the greatest strengths of open source – freedom of choice and open code for anyone to work and develop on – is also one its weaknesses. The Portland project may be about to change all that.

Without getting too deep into the philosophical and developmental debates, one of the greatest strengths of open source — freedom of choice and open code for anyone to work and develop on — is also one its weaknesses. In a world where everyone can do their own thing, standardisation and consistency aren’t usually invited to the party. The LSB (Linux Standard Base) was one key step in helping to solve this, and now Portland is another.

Many years ago now, from two completely separate toolkits, the popular Gnome and KDE desktops were born. Built as they are for themselves, interoperability between applications of one on the desktop of the other wasn’t a priority for a rather long time — it’s only been in the last couple of years for example that you could install applications that will work in KDE’s system tray the same way it will with Gnome’s, for example. While the applications always ran, as long as you had GTK or QT installed, they weren’t generally aware of their host desktop if it wasn’t native.

And this was especially true when it came to updating menus. If the functionality was included at all, the author of a Gnome application for example would only bother about updating menus under Gnome, leaving KDE users to scratch their heads and manually create a new entry.

Generally, it’s been up to the package maintainers of distributions to fix this, to create install scripts that properly add a program to the menus of the installed desktops (KDE, Gnome, Fluxbox, XFCE and others). But the onus then is on the distribution package maintainers to do this for each and every application where applicable, and it doesn’t help at all if users install programs not part of the distribution’s package portfolio.

So back to Portland — a great idea that’s sorely needed, Portland provides a set of tools primarily for use at application installation that standardise the process completely, so no matter whether you use Gnome or KDE, an application can simply call the same Portland tools to update the user’s desktop, and Portland will do the rest.

While, generally, this isn’t going to remove the entire workload of package maintainers — as distributions generally have their own configuration formats to maintain — it’ll certainly help automate the process.

Best of all, however, is if everyday developers start using the Portland tools — this way users can install an application not provided by a distribution and find, miraculously, that their desktop is aware of it and they can get right on and use it, just like under Windows. It’s such a simple thing, but it’s been missing from Linux for too long. If there’s one thing we can learn from Microsoft, it’s that consistency matters.

Actually, it’s more just common sense: people want to use their desktops, so don’t make them work for it. Linux as a desktop has come a long way — Ubuntu is a testament to this — and projects like Portland that are one more step in the yellow brick road.

You can check out Portland’s releases and documentation at the official homepage.

Source : APCMag.com

Businesses are turning to Linux en masse as they face massive budget cuts driven by the economic crisis, according to a new report.

When the global economic crisis unfolded last year, Linux advocates were quick to predict the rise of their favorite operating system.

However, these arguments were met by seemingly logical counter-arguments that foresaw otherwise. One of them even went as far as predicting the downfall of open-source projects.

Well, contrasting forecasts like these can be treated in equal footing until the numbers finally come out.

This month, some numbers finally came out and have been basically favouring Linux.

In a recent global survey conducted by IDC (but sponsored by Linux vendor Novell), results have revealed how IT decision makers feel about a Linux adoption in their organizations in the midst of the global recession. A white paper entitled “Linux Adoption in A Global Recession” summarized the results of the survey.

A glimpse of the white paper’s introduction reveals a clear indication of what most CIO’s, IT Directors, IT Managers, IT Staff, IT Consultants and the like, who were the usual respondents of the survey, thought of the adoptation concept. The introduction states:

“Economic downturns have a tendency to accelerate emerging technologies, boost the adoption of effective solutions, and punish solutions that are not cost competitive or that are out of synch with industry trends.

“IDC research finds that Linux users are clearly satisfied about their choice to deploy Linux, and during trying economic times, the potential for those same customers to ramp up their deployment of Linux is strong”

Among the 330 organizations surveyed, those from Asia Pacific appeared the most aggressive in increasing Linux adoption. Of the 134 respondents coming from the region, 73% expressed interest in increasing adoption on server deployments and 70% expressed interest in increasing adoption on clients.

For comparison, 53% of the total number of respondents expressed interest in increasing adoption on server deployments and 48% expressed interest in increasing adoption on client deployments.

This interest in Linux adoption, according to IDC, stems from the fact that while most organizations aim to reduce spending because of the the economic downturn, they don’t have the intention of implementing a directly proportonal reduction in deployments. In order to achieve this, they turn to Linux and its low-cost nature.

The respondents also indicated that the top 2 reasons that would enable them to accelerate new deployments of Linux would be cost reduction and stronger interoperability with Windows and other operating systems.

This raises the interesting question: Will Microsoft try to stick to its guns and promote the value of its product and service ecosystem against ‘free’ Linux in tough times, or will it discount to retain market share, as it has done with Windows on netbooks?

Despite all of its successes during the global recession as revealed by the research, the study did not fail to recognize the fact that Windows stood in the way of Linux’s long-term growth potential; i.e., beyond the economic crisis.

In fact, Windows is still identified as the platform holding the majority position on both PCs and the x86 server hardware, one of the architectures where Linux has strongly benefited due to the crisis.bull

A free downloadable copy is available for those who are interested in viewing the IDC white paper.

Source: APCMag.com

Businesses are turning to Linux en masse as they face massive budget cuts driven by the economic crisis, according to a new report.

When the global economic crisis unfolded last year, Linux advocates were quick to predict the rise of their favorite operating system.

However, these arguments were met by seemingly logical counter-arguments that foresaw otherwise. One of them even went as far as predicting the downfall of open-source projects.

Well, contrasting forecasts like these can be treated in equal footing until the numbers finally come out.

This month, some numbers finally came out and have been basically favouring Linux.

In a recent global survey conducted by IDC (but sponsored by Linux vendor Novell), results have revealed how IT decision makers feel about a Linux adoption in their organizations in the midst of the global recession. A white paper entitled “Linux Adoption in A Global Recession” summarized the results of the survey.

A glimpse of the white paper’s introduction reveals a clear indication of what most CIO’s, IT Directors, IT Managers, IT Staff, IT Consultants and the like, who were the usual respondents of the survey, thought of the adoptation concept. The introduction states:

“Economic downturns have a tendency to accelerate emerging technologies, boost the adoption of effective solutions, and punish solutions that are not cost competitive or that are out of synch with industry trends.

“IDC research finds that Linux users are clearly satisfied about their choice to deploy Linux, and during trying economic times, the potential for those same customers to ramp up their deployment of Linux is strong”

Among the 330 organizations surveyed, those from Asia Pacific appeared the most aggressive in increasing Linux adoption. Of the 134 respondents coming from the region, 73% expressed interest in increasing adoption on server deployments and 70% expressed interest in increasing adoption on clients.

For comparison, 53% of the total number of respondents expressed interest in increasing adoption on server deployments and 48% expressed interest in increasing adoption on client deployments.

This interest in Linux adoption, according to IDC, stems from the fact that while most organizations aim to reduce spending because of the the economic downturn, they don’t have the intention of implementing a directly proportonal reduction in deployments. In order to achieve this, they turn to Linux and its low-cost nature.

The respondents also indicated that the top 2 reasons that would enable them to accelerate new deployments of Linux would be cost reduction and stronger interoperability with Windows and other operating systems.

This raises the interesting question: Will Microsoft try to stick to its guns and promote the value of its product and service ecosystem against ‘free’ Linux in tough times, or will it discount to retain market share, as it has done with Windows on netbooks?

Despite all of its successes during the global recession as revealed by the research, the study did not fail to recognize the fact that Windows stood in the way of Linux’s long-term growth potential; i.e., beyond the economic crisis.

In fact, Windows is still identified as the platform holding the majority position on both PCs and the x86 server hardware, one of the architectures where Linux has strongly benefited due to the crisis.bull

A free downloadable copy is available for those who are interested in viewing the IDC white paper.

Source: APCMag.com