Selasa, 30 Oktober 2012

Linux Certifications

The Linux Certification Landscape
Currently, the Linux certification landscape may be divided into three groups: primary, secondary and upcoming. Primaries are those certifications or programs that concentrate exclusively on Linux; by now all have achieved some degree of market penetration and success. Secondaries are those that cover Linux in the context of other certification programs. Upcomings include a couple of interesting offerings that have been announced but aren�t yet available.
I�ll provide information about all three groups, plus comments on a former primary that is no longer active. After that, I dig into details on the primaries (which provides a major focus for this story, and this entire StudyGuide).
Primary Linux certifications and programs include:
  • Red Hat Certified Professional Program: numerous training courses and two certifications that focus on Red Hat Linux: Red Hat Certified Technician (RHCT) and Red Hat Certified Engineer (RHCE).
  • Linux Professional Institute (LPI) Program: three levels of vendor-neutral certification with related exams (LPIC1, LPIC2 and LPIC3), of which the first two levels are currently available (and the third is under development).
  • CompTIA�s Linux+: A single entry-level, vendor-neutral certification.
Those familiar with Linux certification may notice that a former primary player is missing�the SAIR-GNU Linux program. Like the LPI program, SAIR-GNU�s design included three tiers for credentials, including Administrator (LCA), Engineer (LCE) and Master Engineer (MLCE). According to reports, the program has been discontinued in the wake of mergers and management changes.
LPI supports orphaned LCAs in two ways (higher-level SAIR-GNU exams were never made available, so it�s the only credential that counts):
  • They can attain LPIC1 certification by passing any single LPIC1 exam (instead of the customary two).
  • They can take LPIC2 exams (instead of having to complete two Level 1 exams to meet normal prerequisites) and pursue LPIC2 certification directly.
Either way, this gives LCAs a path to more advanced certifications, with little or no loss of status or need for extra work.
Secondary Linux certifications include two elements, both of which incorporate Linux coverage into other, more general programs:
  • HP offers Linux coverage in two credentials. HP Accredited Integration Specialists (AISs) can elect to follow AlphaServer + Linux or ProLiant + Linux tracks. Each of these tracks requires candidates to meet Linux prerequisites, including the SAIR-GNU LCA, LPIC1, RHCT or completion of Red Hat�s RH133 Linux System Administration and RHCT course. HP Accredited Systems Engineers (ASEs) can follow these tracks at a more technical level, provided they obtain more advanced certifications (the RHCE is the only prerequisite still available). Visit for more information.
  • IBM also offers two certifications with a Linux focus. The eServer Certified Systems Expert-iSeries program includes a Linux Solution Sales V5R2 credential. Although it has a technical component, this certification includes no additional Linux prerequisites. The eServer Certified Systems Expert-xSeries program also includes a Linux credential. To qualify, candidates must already hold a Linux certification, such as LPIC1 or RHCE, and take one additional exam. Visit, and scan the �Select a certification� pull-down menu for mentions of Linux for more information on these offerings.
My scans of other cert programs show no other vendors or organizations that currently incorporate Linux in their offerings or offer other Linux-oriented credentials or programs. Read on, however, for news about some upcoming programs that will add to the landscape once they launch.
Upcoming Linux programs include two items, one of which is delayed:
  • UnitedLinux is a partnership that involves leading Linux and operating-system vendors Connectiva, The SCO Group, SuSE and Turbolinux. They�re building the UnitedLinux Certified Professional (ULCP) program with LPI. In fact, it�s essentially an add-on to LPI credentials, with an additional exam for LPIC1 and LPIC2 on the UnitedLinux distribution. (Add exam #103 to go from LPIC1 to United Linux Certified Professional, or ULCP; add exam #203 to go from LPIC2 to United Linux Certified Expert, or ULCE). Despite announcing the program�s launch in the first quarter of 2003, e-mail from General Manager Paula M. Hunter indicates that they expect the program to debut no later than September 2003, but that no firm release date is set. For information, see
  • Novell claims a December 2003 launch for its Certified Linux Engineer (CLE) credential. In addition to stressing Linux setup, management, maintenance and troubleshooting skills, it will also focus on various Novell services for Linux. The CLE service flyer mentions a practical exam, which means they�ll test hands-on skills and knowledge. It also recommends candidates obtain LPI Level 1 certification. Those who certify as Novell 6 Certified Novell Engineers (CNEs) (new or upgrade) can take the CLE exam for free. Details are sketchy, and the program is still under development. For information, visit and follow links to CLE coverage and documents.
In the sections that follow, you�ll learn more about the details involved in the primary Linux certifications and programs. They�ll be discussed in their order of market penetration as measured by size estimates of the certified populations involved.
Red Hat�s RHCT and RHCE
As the biggest and oldest Linux professional certification, RHCE is arguably the premier Linux certification available today. Both RHCT and RHCE rely on performance-based testing. (The majority of each of these exams occurs in a laboratory setting, where candidates must demonstrate hands-on skills and knowledge of Red Hat Linux to pass, with only minor use of standard multiple-choice tests on the RHCE.) Much of their value derives from the levels of proficiency that exam candidates must possess to pass these lengthy and demanding exams. (See Table 1 for details.)



What is Parabola GNU/Linux?

Parabola GNU/Linux is a libre software project aiming to provide a fully free as in freedom distribution based on the packages of the Archlinux distribution, with packages optimized for i686, x86_64, and Loongson 2F (mips64el) CPUs. Parabola aims to keep its package and management tools simple. The primary goal is to give the user complete control over their system with 100% Libre software. Parabola is listed by the Free Software Foundation as a fully free software distribution.
Development is focused on a balance of simplicity, elegance, code-correctness and bleeding edge free software.
Its lightweight and simple design makes it easy to extend and mold into whatever kind of system you're building.



News: New install medium 2012.10.17

There is a new install medium available created by Esteban Carnevale (alfplayer). We plan to release new installation media monthly.
The live system can be downloaded from Download and be used for new installs or as a rescue system.
We have updated packages, fixed bugs and done the following visible changes:
  • First medium with Linux-libre 3.6 (3.6.2) with Atheros AR8162 (ALX driver) support, reported on issue201.
  • The script boot parameter works again, reported on Archlinux Bugs.
  • When booting via PXE and NFS or NBD the ISO will be copied to RAM to ensure a more stable usage.
  • The live medium contains usb_modeswitch and wvdial which e.g. allows to establish a network connection using an UMTS USB dongle.
  • Furthermore the newest version of systemd and netcfg are included.
  • systemd is used to boot up the live system.
  • initscripts are no longer available on the live system but are still installed by default on the target system. This is likely to change in the near future.
  • EFI boot and setup has been simplified.
  • gummiboot is used to display a menu on EFI systems.
  • The following new packages are available on the live system: ethtool, fsarchiver, gummiboot-efi, mc, partclone, partimage, refind-efi, rfkill, sudo, testdisk, wget, xl2tpd.

Ryan Merkley: Online video -- annotated, remixed and popped

Jumat, 26 Oktober 2012

GNUs trick-or-treat at Windows 8 launch

 Thanks to  the Free Software Foundation (FSF)

for the E-mail. We support 100% the FSF.



 GNUs trick-or-treat at Windows 8 launch

Posted by libby at Oct 26, 2012 11:25 AM |
Yesterday morning, the Free Software Foundation crashed the Windows 8 launch event in New York City. A cheerful GNU and her team handed out DVDs loaded with Trisquel, FSF stickers, and information about our new pledge, which asks Windows users to upgrade not to Windows 8, but to GNU/Linux.
Check out these great photos of the fun, and don't forget to sign our pledge!

Special thanks to our volunteers, Linnea the GNU and Tristan the photographer, for helping make this action a huge success!
Document Actions

Senin, 15 Oktober 2012

Ubuntu 12.10 (Quantal Quetzal) Beta 2


Amazon results in Unity
The second beta release of Ubuntu 12.10 has been made available for download.
It arrives ahead of Ubuntu 12.10′s final release in just over 2 weeks time.

What’s New?

Unusually for an Ubuntu release at this late stage of development, an awful lot has changed since the release of Beta 1 - though most of it cosmetic.
Unity has been updated, tweaked and ‘finessed’; the Dash has been given slick new animations during Previews; and pair of controversial new features have made their debut…

 Ubuntu being an open source initiative is bound to be a constantly changing environment. There have been around 15 releases of Ubuntu so far and the 16th is all set to roll out in April of 2012. The upcoming entrant nicknamed the "Quantal Quetzal" has been reviewed and previewed by many Linux geeks and has been found to be very efficient.

All its Alpha and Beta releases have ensured the fixing of majority of the bugs arising during user testing. We expect the final version to be released soon to incorporate a new look and feel in the Ubuntu Line-Up.

 ubuntu 12.10

Linux Deepin 12.06 Final Release Out !


Linux Deepin is one of the most active Linux distributions in China. The developers of LD endeavour to provide its users with an operating system of high stability and efficiency, in order to fulfil our goal to "Keep newbies free from pain and save time for the experts". With the efforts from both the community and the company that work behind the project, LD is becoming easier to use every day. We would like to say thanks to all of you who join us and support us. Please follow our blog ( ). Linux Deepin is released twice a year. Last major release was Linux Deepin 11.12.


Deepin Gnome Shell

Deepin Gnome Shell is a desktop environment based on the native Gnome Shell. It has been tweaked by the Deepin developers.

New plugins

In Linux Deepin 12.06, three new plugins are added:
  • Hide Message Tray: As the name indicates, it can hide the Message Tray.
  • Kimpanel: Enables users to search in Gnome Shell for applications, files or contacts.
  • Panel Settings: Lets users decide where the panel should be (top/bottom) and whether it should autohide.

Enhanced message indicator

The message indicator of Gnome Shell is replaced by one from Ubuntu, placed at the right-top corner of the screen.

Other enhancements include:
  • Deepin Dock: Application groups; instant preview; Closing applications Traditional system tray
  • Better mouse experience: Application groups and workspace is placed on the left.
  • Integrated searching engines including Google, Bing, Yahoo!, making it easier for you to search under Gnome Shell.
  • Enhanced working space navigating

System improvement

  • Fixed the bug of .zip file garbled name
  • Starting Gnome Shell with special effects is now supported on VirtualBox 4.1.6/VMWare 8.0.3/8.0.4.
  • Read/Write support of exFAT file system.
  • More hardware drivers integrated
  • More fonts preinstalled to provide better CJK display
  • Gnome Fallback mode now features a system tray.
  • Opening plain text from other systems without errors.
  • The Linux Deepin version of WUBI - DeepWin lets you install Linux Deepin under Windows.

  • The latest official AMD graphic drivers (fglrx) contain bugs with 2D rendering, so the background of Deepin UI is unable to display normally. For this reason, we have disabled fglrx in Jockey-gtk. If you still wish to install fglrx, you may search for it in Deepin Software Center.
  • If a bug with flash occurs, you may remove its incorrect config files by running
    sudo rm ~/.pulse*

Featured applications

This version of Linux Deepin is the first stable version to ship with Deepin Music Player and Deepin Media Player.

Deepin Music Player

Deepin Music Player is designed for Linux users. It is free software licensed under GNU GPLv3. The current version is 1.0.

Deepin Music Player features
  • Playlists
  • Equalizer
  • Music management
  • Auto-download album covers and artist pictures
  • Simple/normal mode of interface
  • Lyric searching with multiple engines
  • Desktop/window mode of lyric display
  • Local file searching
  • Theme selection support Common formats supported
  • Other: Player history, crossfade, system tray, hotkeys

Deepin Media Player

Deepin Media Player is a frontend of MPlayer2, featuring:

  • Theme selection
  • Most common formats support
  • Screenshots
  • Video preview
  • Auto-searching and downloading titles
  • Online playback. Just enter the URL of the video.

Other featured applications include Deepin Screenshot tool and Deepin Software Center

Deepin Screenshot: a lightware screenshot tool
  • Automatic window identification
  • Selection of area
  • Editing screenshots
  • Saving to clipboard
  • Delayed shots

Deepin Software Center features:
  • One-click install/upgrade/uninstall software
  • Integrated update manager
  • Score/comments on software
  • Software Searching
  • Theme selecting
Linux Deepin 12.06 is still packed with Deepin Software Center 2.0, which is the same as Linux Deepin 11.12. The Deepin Software Center 3.0 will be released in the coming release 12.06.1. At the same time, the user manual is updated to 2.0, supporting the English language and Simplified/Traditional Chinese.

Development related: Deepin UI graphic libraries

Since Linux Deepin 12.06, the Deepin UI will be preinstalled. The Deepin UI can be applied to develop easy-to-use Linux applications, giving a boost to desktop development. Deepin UI enables a developer to
  • Manage the layout of applications the way he wishes, instead of restriced by the layouts of GTK+.
  • Build a theme-selecting engine which is fast and easy to use. Instead of piecing pictures together to build a theme, one may build a theme with the engine by selecting one simple picture.
  • Use a variety of widgets and graphic effects that are ships with Deepin UI.
  • Use a variety of application modules, so one does not need to waste time on such things as dragging window, rounded corner, Gaussian blur, gradient effect, compelling windows and keystroke identification.
Deepin Music Player and Deepin Media Player are applications built with Deepin UI.


Desktop backgrounds

Linux Deepin 12.06 ships with 8 pictures as desktop background.

Machine Learning for Hackers

Case Studies and Algorithms to Get You Started From O'Reilly Books, This is a great Book to hence Your Knowledge.

If you’re an experienced programmer interested in crunching data, this book will get you started with machine learning—a toolkit of algorithms that enables computers to train themselves to automate useful tasks. Authors Drew Conway and John Myles White help you understand machine learning and statistics tools through a series of hands-on case studies, instead of a traditional math-heavy presentation.
Each chapter focuses on a specific problem in machine learning, such as classification, prediction, optimization, and recommendation. Using the R programming language, you’ll learn how to analyze sample datasets and write simple machine learning algorithms. Machine Learning for Hackers is ideal for programmers from any background, including business, government, and academic research.
  • Develop a naïve Bayesian classifier to determine if an email is spam, based only on its text
  • Use linear regression to predict the number of page views for the top 1,000 websites
  • Learn optimization techniques by attempting to break a simple letter cipher
  • Compare and contrast U.S. Senators statistically, based on their voting records
  • Build a “whom to follow” recommendation system from Twitter data

Sabtu, 13 Oktober 2012

Anonymous: Operation V (Nov. 5th 2012)

Updated Debian 6.0: 6.0.6 released

 Debian GNU/Linux

The Debian project is pleased to announce the sixth update of its stable distribution Debian 6.0 (codename squeeze). This update mainly adds corrections for security problems to the stable release, along with a few adjustments for serious problems. Security advisories were already published separately and are referenced where available.
Please note that this update does not constitute a new version of Debian 6.0 but only updates some of the packages included. There is no need to throw away 6.0 CDs or DVDs but only to update via an up-to-date Debian mirror after an installation, to cause any out of date packages to be updated.
Those who frequently install updates from won't have to update many packages and most updates from are included in this update.
New installation media and CD and DVD images containing updated packages will be available soon at the regular locations.
Upgrading to this revision online is usually done by pointing the aptitude (or apt) package tool (see the sources.list(5) manual page) to one of Debian's many FTP or HTTP mirrors. A comprehensive list of mirrors is available at:

About Debian

WHAT is Debian?

The Debian Project is an association of individuals who have made common cause to create a freeoperating system. This operating system that we have created is called Debian.
An operating system is the set of basic programs and utilities that make your computer run. At the core of an operating system is the kernel. The kernel is the most fundamental program on the computer and does all the basic housekeeping and lets you start other programs.
Debian systems currently use the Linuxkernel or the FreeBSDkernel. Linux is a piece of software started by Linus Torvaldsand supported by thousands of programmers worldwide. FreeBSD is an operating system including a kernel and other software.
However, work is in progress to provide Debian for other kernels, primarily for the Hurd. The Hurd is a collection of servers that run on top of a microkernel (such as Mach) to implement different features. The Hurd is free software produced by the GNU project.
A large part of the basic tools that fill out the operating system come from the GNU project; hence the names: GNU/Linux, GNU/kFreeBSD and GNU/Hurd. These tools are also free.
Of course, the thing that people want is application software: programs to help them get what they want to do done, from editing documents to running a business to playing games to writing more software. Debian comes with over 29000 packages (precompiled software that is bundled up in a nice format for easy installation on your machine) — all of it free.
It's a bit like a tower. At the base is the kernel. On top of that are all the basic tools. Next is all the software that you run on the computer. At the top of the tower is Debian — carefully organizing and fitting everything so it all works together.

It's all free?

You may be wondering: why would people spend hours of their own time to write software, carefully package it, and then give it all away? The answers are as varied as the people who contribute. Some people like to help others. Many write programs to learn more about computers. More and more people are looking for ways to avoid the inflated price of software. A growing crowd contribute as a thank you for all the great free software they've received from others. Many in academia create free software to help get the results of their research into wider use. Businesses help maintain free software so they can have a say in how it develops -- there's no quicker way to get a new feature than to implement it yourself! Of course, a lot of us just find it great fun.
Debian is so committed to free software that we thought it would be useful if that commitment was formalized in a written document. Thus, our Social Contract was born.
Although Debian believes in free software, there are cases where people want or need to put non-free software on their machine. Whenever possible Debian will support this. There are even a growing number of packages whose sole job is to install non-free software into a Debian system.

You say free, but the CDs/bandwidth cost money!

You might be asking: If the software is free, then why do I have to pay a vendor for a CD, or pay an ISP for downloading?
When buying a CD, you are paying for someone's time, capital outlay to make the disks, and risk (in case they don't sell them all). In other words, you are paying for a physical medium used to deliver the software, not for the software itself.
When we use the word "free", we are referring to software freedom, not that it's without cost. You can read more on what we mean by "free software" and what the Free Software Foundation says on that subject.

Most software costs over 100 US dollars. How can you give it away?

A better question is how do software companies get away with charging so much? Software is not like making a car. Once you've made one copy of your software, the production costs to make a million more are tiny (there's a good reason Microsoft has so many billions in the bank).
Look at it another way: if you had an endless supply of sand in your backyard, you might be willing to give sand away. It would be foolish, though, to pay for a truck to take it to others. You would make them come and get it themselves (equivalent to downloading off the net) or they can pay someone else to deliver it to their door (equivalent to buying a CD). This is exactly how Debian operates and why most of the CDs/DVDs are so cheap (only about 12 USD for 4 DVDs).
Debian does not make any money from the sale of CDs. At the same time, money is needed to pay for expenses such as domain registration and hardware. Thus, we ask that you buy from one of the CD vendors that donates a portion of your purchase to Debian.

What hardware is supported?

Debian will run on almost all personal computers, including most older models. Each new release of Debian generally supports a larger number of computer architectures. For a complete list of currently supported ones, see the documentation for the stable release.
Almost all common hardware is supported. If you would like to be sure that all the devices connected to your machine are supported, check out the Linux Hardware Compatibility HOWTO.
There are a few companies that make support difficult by not releasing specifications for their hardware. This means you might not be able to use their hardware with GNU/Linux. Some companies provide non-free drivers, but that is a problem because the company could later go out of business or stop support for the hardware you have. We recommend that you only purchase hardware from manufacturers that provide free drivers for their products.

I'm looking for more information.

You may want to check out our FAQ.

I'm still not convinced.

Don't take our word for it - try Debian yourself. Since hard disk space has become less expensive, you can probably spare about 2GB. If you don't want or need a graphical desktop, 600MB are sufficient. Debian can be easily installed on this extra space and can coexist with your existing OS. If you eventually need more space, you can simply delete one of your OSes (and after you see the power of a Debian system, we are confident you won't delete Debian).
As trying a new operating system will take some of your valuable time, it is understandable that you may have reservations. For this reason we compiled a list of pros and cons of Debian. This should help you decide whether you think it's worth it. We hope you'll appreciate our honesty and frankness.

How do I get Debian?

It's most popular to install Debian from a CD which you can buy for the price of the media at one of our many CD vendors. If you have good Internet access, you can download and install Debian over the Internet.
Please see our page about getting Debian for more information.
If you haven't yet, you may want to first look at the Linux Hardware Compatibility HOWTO.
Don't forget to take a look through the packageswe offer (hopefully you won't be intimidated by the sheer number).

I can't set it up all by myself. How do I get support for Debian?

You can get help by reading the documentation which is available both on the web site and in packages you can install on your system. You can also contact us via the mailing lists or using IRC. One can even hire a consultant to do the work.
Please see our documentation and support pages for more information.

Who are you all anyway?

Debian is produced by almost a thousand active developers spread around the world who volunteer in their spare time. Few of the developers have actually met in person. Communication is done primarily through e-mail (mailing lists at and IRC (#debian channel at
The Debian Project has a carefully organized structure. For more information on how Debian looks from the inside, please feel free to browse the developers' corner.

Who uses Debian?

Although no precise statistics are available (since Debian does not require users to register), evidence is quite strong that Debian is used by a wide range of organizations, large and small, as well as many thousands of individuals. See our Who's using Debian? page for a list of high-profile organizations which have submitted short descriptions of how and why they use Debian.

How'd it all get started?

Debian was begun in August 1993 by Ian Murdock, as a new distribution which would be made openly, in the spirit of Linux and GNU. Debian was meant to be carefully and conscientiously put together, and to be maintained and supported with similar care. It started as a small, tightly-knit group of Free Software hackers, and gradually grew to become a large, well-organized community of developers and users. See the detailed history.
Since many people have asked, Debian is pronounced /ˈən/. It comes from the names of the creator of Debian, Ian Murdock, and his wife, Debra.

Syllable Desktop 0.6.7 Released


System Structure

Syllable is an operating system with a high degree of compatibility with the POSIX standards and de facto conventions in Unix systems and Linux distributions. This is a double edged sword. It makes it much easier to get existing third-party software to run on Syllable, but it also burdens us with ancient methods from the beginning of computing that have partly lost their meaning. One example is the layout and naming of files in the system and in additional software packages. We have always tried to strike a balance between following conventions and inventing more modern and user friendly methods. With a decade of experience, we have gathered the confidence to make a number of extra steps towards more user friendly structures, with as little disruption to compatibility as possible.

Instead of trying to shield people from the workings of their system, we want them to be able to go into the directories of the system or a separate package just with a filer, and be able to make some sense out of the files there and find what they are looking for. To that end, we are introducing a series of new measures to rename and move subdirectories that often form the starting point for searches for information.

The system tools Packager and Builder were further developed to support these restructurings, among other things to improve the use of symlinking for package registration. Packager is included in the system, but Builder needs to be installed as part of the Developer's Delight package, if you want to compile software yourself. When Builder is installed on Syllable 0.6.6, an older version is used that is still compatible with the older system structures. The Syllable branch in Builder/packages/ is also still largely compatible with Syllable 0.6.6, so that some packages can still be built on older systems. After the release of Syllable 0.6.7, Builder and its recipe tree will be updated to the latest versions, using the new structures.

Graphical programs that, besides the desktop menu, also need to be available from the command line are now kept in the search path locations and symlinked from /Applications/. Previously, this was usually done the other way around, but that meant that the program in /Applications/ couldn't be moved.

Packages are in the process of moving to clearer names for their main package directory. Instead of the traditional, short, lowercase names that are confusing to common people, common typographical standards are now being used. Resource packages have project names, which are proper names, so they are capitalised. Generic terms remain lowercased. When names consist of multiple words, CamelCase is used to avoid spaces. When uppercase abbreviations would lead to words of only capitals being concatenated, hyphens are used for more clarity. Abbreviations are expanded when they don't take too much space; roughly twenty characters maximum. Exceptions can be made to keep names recognisable to people who already know the short names, and to keep them roughly in the same place in alphabetical order. Around half the resource packages in the system are currently renamed.

The internal structure of resource packages is in the process of being heavily reorganised, doing away with the traditional Unix naming that is confusing and has lost its meaning over time. The structure of the symlink pools has changed accordingly. Subdirectories are being moved and renamed thus:
etc             -> settings
bin            -> programs
sbin            -> system-programs
libexec        -> framework/executables
lib            -> framework/libraries
include        -> framework/headers
share/aclocal  -> framework/AutoConfigure
share/pkgconfig -> framework/PackageConfigure
lib/pkgconfig  -> framework/PackageConfigure
share          -> data
share/doc      -> documentation
man            -> manuals
share/man      -> manuals
info            -> manuals/info
share/info      -> manuals/info
Subdirectories that are not immediately meaningful to common users are tucked away in the framework subdirectory, a name which should at least signal to users that it should not be tampered with without appropriate knowledge. The documentation directories are split between an unmanaged and a managed (structured manuals) part, to make a distinction in how they should be used and the degree to which a user could tamper with them.

Since there is no standard for the location of REBOL and Boron libraries, they can go into framework/REBOL.

The Syllable-specific initialisation scripts were consolidated into one subdirectory. This leaves room for adding the common Unix script classes for stop, reload, restart and status commands:
early-init      -> tasks/setup
init            -> tasks/start

An extra subdirectory applications was introduced for resource packages that contain graphical programs that should appear in an applications menu. This was prompted by the port of the VICE emulator. This and other packages need to be in /resources/, but want to make entry points in the Syllable menu like /Applications/ do.
Support in the menu needs to be implemented separately.

A few packages need extra, internal compatibility symlinks because they don't fully support this reorganisation. On the other hand, such links can also be used to solve interdependencies between packages.

The symlink pool /resources/index/ and Packager have compatibility facilities to support most older binary packages.

Moved bin/ to programs/.
Most common people would probably associate “bin” with a dustbin, and wonder why every package has its own trashcan. We'd have to explain that it means “binaries”, then explain what binaries are and what a binary number system is, and then explain that bin/ directories don't just contain binaries, and that binaries can also be found elsewhere... bin/ has only been a fitting name for a short period in the earliest Unix history, before the advent of scripts. Since the invention of libraries, there are also binary files elsewhere.

Moved sbin/ to system-programs/.
The “root” super user in Syllable Server has already been renamed to “system”. This is also planned for Desktop. system-programs/ will then be a matching name for programs that need to be executed by the super user.

Moved share/ to data/.
share/ originally meant files that can be shared over a network between multiple machines, because they're data instead of architecture dependent binaries. This name is not very meaningful on Syllable, because disk space is not an issue anymore for these files and we build self-contained systems that can be operated off-line from a network. share/ directories are symlinked to make them available as an aggregate within the system, but most resource subdirectories are shared that way.

init/ and early-init/ were already consolidated into tasks/. The remaining subdirectories are not meaningful to common users, because they contain support files for programs/ and applications/, so I folded them into a new subdirectory framework/ to signify their function and move them out of sight.

Moved lib/ to framework/libraries/.
Libel? Liberation? Makes it clear what we're talking about. It's still not the same as a public library, but at least it's clear that it belongs to the framework of a resource package, and libraries is the term used everywhere in system and programming documentation, so it would be confusing for techies to change it more than this.

Moved libexec/ to framework/executables/.
libexec/ clearly has a confused personality, so this makes that explicit. It can contain both support programs that don't need to be in the search path and libraries, which are both executables, so it clearly belongs to the framework.

Moved include/ to framework/headers/.
Include which in what? We're talking about compiled language programming headers for the libraries, and that's still a very technical term, so we tuck it away in framework/.

Moved share/aclocal/ to framework/AutoConfigure/.
This contains M4 macros for the GNU AutoTools, AutoConfigure and AutoMake, that many packages use for source code configuration. Very technical and applies to the libraries, so it's part of the framework. Besides, M4 macros are scripts, so it's not even proper data/. Not even techies can know what aclocal/ means until they find the explanation. Having a directory named local inside the directory share/ is also pretty disingenious. This is not a universally used configuration system, so we name it after the clearest component, that also has a relation to the original name.

Consolidated lib/pkgconfig/ and share/pkgconfig/ into framework/PackageConfigure/.
This is for another software configuration system named pkg-config, and contains data files describing properties of the package. Again, this is mostly meant for other packages to use the libraries in this resource package, so it's part of the framework. We use the clear CamelCase version of the name of the configuration system. Packages in the wild are confused about whether this is data or belongs with the libraries. Consolidating those two locations into one simplifies our search path for PackageConfigure files.

Moved lib/REBOL/ to framework/REBOL/.
Programming languages tend to have their own hierarchy under lib/ for their support files.
There is no standard place for REBOL and Boron libraries, so we are free to change this.
Having it immediately under framework/ instead of under framework/libraries/ shortens paths.

Moved etc/ to settings/.
etc/ must be the dumbest computer term ever introduced. It seems the designers were completely out of inspiration. I imagine it was scribbled on a napkin over lunch at the end of the other brilliant ideas for the Unix filesystem layout, and later implemented without further thought. While it's understandable that these people didn't count on common people needing to understand their system in the early 1970's, the naming is even of exceptionally poor quality in a technical context. What's really stupefying, though, is that these design choices have been defended as high culture by Unix types ever since.
Note that eventually, we shouldn't have changing data in program directories, so in most cases we direct this settings directory to the global and unfortunately standard /etc/.

Unlike earlier restructurings, the locations of these directories are critical to the operation of the software packages. Many components have accompanying search paths, that I have adapted in different places in the system. However, packages don't always use these and may try to find components in fixed locations or relative to themselves. Many packages configured with the GNU AutoTools support build options to change these locations. As much as possible, Builder now uses those options to perform these moves, so the packages are aware of them. However, many package configurations are partly broken, especially regarding this because few systems are interested in changing these locations. Such packages may need to be repaired by adding compatibility symlinks from the traditional subdirectories to the new ones.

This can actually be turned into an advantage. Quite a few resource packages act as platforms for other packages. Those other packages contain extensions in the same location as in the platform package. In other systems, these files would end up in the very same directories. In Syllable, the platform packages can't find those modules because packages are built into their own, separate directories. However, they are registered in our symlink pools to create the same view as in other systems. By using the traditional subdirectories in platform packages to symlink into the symlink pools, they can be pointed to the extensions in other packages.

Moved ^/lib/ to ^/libraries/.
This is the subdirectory relative to a program executable, that can contain private dynamic libaries for the program. Changed it to match the other renaming.

Some of these changes still have the old counterparts in the system, so that most old resource packages and applications will still work on systems with the new structure, until they are obsolete.

Reorganised the init scripts.

We've been porting more Unix server packages over time. They may need setup and teardown procedures that are usually categorised in five types: start, stop, reload, restart and status. We have init/ and early-init/ subdirectories within ported packages. init/ corresponds to the start script in Unix systems, which is actually a better name, because common users have no idea what “init” means. There's no real equivalent to early-init/, and we use that mainly to set variables in users' shell environments. Eventually, we would have to support the other script types, as well. Having them each in a different subdirectory within a package would be a mess.

Created a new (optional) subdirectory tasks/ within resource packages. Builder and Packager work together to symlink the content into /resources/index/ and other symlink pools.
Moved the init/ subdirectory to tasks/start/.
Moved early-init/, having no naming identification with init/ anymore, to tasks/setup/.
This name also better reflects the fact that setup scripts are called at every shell start, not just at system start-up or user log-in.
Future scripts can be added under tasks/ as tasks/stop/, tasks/reload/, tasks/restart/ and tasks/status/.

Reorganised, unified and integrated the documentation in ported packages.

Third-party packages have at least six locations for documentation. We collect loose documentation files in a documentation/ subdirectory. Many packages also have freeform documentation in a share/doc/ subdirectory. Traditionally, “man” pages are in man/ but GNU has been moving them to share/man/ lately. They prefer “info” files, anyway, and have first defined those in info/ but are now moving them to share/info/.

This is a mess for users. We want them to be able to go into a package directory just with a filer, and find documentation they are looking for. We were already moving share/man/ and share/info/ back to man/ and info/ because the share/ subdirectory is not very meaningful in Syllable. We build a self-contained system that typically doesn't share system files over networks, and we symlink several subdirectories to make them globally available within the system, not just the share/ hierarchy.

Any files in share/doc/ are now merged into documentation/. The man and info pages should not go in there, because they're structured databases that are meant to be read in their respective viewers. man/ and share/man/ are now moved to manuals/, a much clearer name. Because they have a structured content, it's possible to also move info/ and share/info/ to manuals/info/. The latter was pioneered by GoboLinux. Builder and Packager work together to accomplish this. As much as possible, Builder uses packages' configure options so they are aware of the moves.
Remaining files are moved afterwards.

The GCC libraries and headers installed in the system now have priority over a separately installed full GCC package. This protects system integrity while allowing to install versions of GCC older than the system libraries. This is a cleaner way than manipulating the GCC package, which was done before.

There are several legitimate reasons for installing a GCC package that is not in sync with the system libraries. The GCC package may be newer than the system release. It may be older, or an old GCC may be required to compile some program. If the versions diverge too much, this can disrupt the working of the system.
Originally, I removed the shared libraries from the GCC packages, but that may not be enough. Then I removed the complete lib/ subdirectory, but GCC does not appreciate that in all cases. Currently, the GCC packages have their include/ and lib/ subdirectories linked to the versions in the system. However, that may not work when the GCC version is too different from the system. Also, it makes the complete /system/index/include/ and /system/index/lib/ structures appear in /resources/index/, where they may conflict with other versions of the same packages.

What's really needed is to give the system libraries priority over GCC, while letting GCC keep its own versions for when it needs those internally. I chose to do that through the system paths. Other options are installing GCC in a separate package pool, or making an exception for it when symlinking in Packager.

Note that other packages in /resources/ still have priority over /system/resources/, so apart from GCC, it's still possible to overrule system components.

The symlink pools were moved from /system/indexes/ and /resources/indexes/ to /system/index/ and /resources/index/ to shorten paths throughout the system. The old locations are still available as compatibility symlinks to the new locations, to support packages that were built before Syllable 0.6.7.

/usr/ is now a symbolic link to /resources/ instead of the other way around. In the next step of this migration, they will be separated into two directories.

Implemented, in Packager, extra symlinking in the symlink pools to point to the main version of installed resource packages. This allows to point to packages in versioned pools without mentioning the version. The package symlink points to the main/currently selected version.

The links are made in the symlink pools, not in the installed packages. This allows to have multiple symlink pools making different selections of the same packages, for example for different user accounts or software build environments.

Symlinks that are made from the system into resource packages (usually for compatibility reasons) can be completed by Builder. So far, Builder added the complete path of the specific resource package, including the version for packages in the system area. This is not robust against upgrades of a versioned package or a change of the package name, because these symlinks are not included in single binary packages.

Now, Builder completes incomplete paths in a recipe based on the symlink pool of the package. Thus, the symlink does not contain the package name and version anymore and remains valid when those change.

Jumat, 12 Oktober 2012




Gnu/Linux distribution

You'll need at least 80GB of free disk space, and preferably a very fast Internet connection (you'll be pulling 40GB of data). This can be run on any Debian derivative capable of bootstrapping Ubuntu Hardy (8.04). All commands must be run as root.

Step 1: GPG key

More recent versions of apt require GPG signed release files in a repository to ensure the integrity of the distribution, so our first step is to create a GPG key.
 gpg --gen-key
will let you do this. Make sure the key has a blank password. Take a note of the fingerprint of the key as you'll need it later for the config file.

Step 2: Checkout builder

To run the build scripts, you will need to get a bazaar checkout.
Make sure you have bazaar installed:
 apt-get install bzr
Then download Builder (it's about 35 MiB):
 bzr branch
cd builder
To also get the source packages specific to gNewSense, check them out with mr (this can be a big download, depending on the packages that have been added):
 apt-get install mr
cd packages
mkdir deltah{,-security,-updates,-backports}
mr -c gnewsense.mrconfig checkout

Step 3: Configuration

Copy the config file to config.local, then open config.local in a text editor. The settings you need to worry about most are listed below, with a short description. Changing the others will lead to better personalisation (see the comments in config/config.local for what setings do).
MIRRORLOCAL is the path to your Ubuntu mirror (used in step 4)
MIRROR is either the mirror you setup in step 5, or an Ubuntu mirror being served via http. It should contain the security packages.
DISTRONAME is what you want to call your distribution which should consist of only letters and numbers.
SIGNINGKEY is your GPG key id from step 1, it should not contain any spaces.
ALL_REPO_ARCHES Lists all architectures you want in your distribution. only i386 is supported currently.
REPOAPT is where the build scripts will pull packages from, so you should setup Apache to serve this.
DOMAIN is used to create the default sources.list when installing. We assume an Ubuntu-like setup where you have archive.DOMAIN and security.DOMAIN and subdomains for all the country codes.
REPODST is where your distributions mirror will be built in the filesystem ($BASEDIR/$DISTRONAME_L by default).
Ensure you remove the following lines from config.local, or you will cause a loop in the scripts:
 # Allow for local customisations
if [ -r config.local ]; then . config.local; fi

if [ -r ]; then .; else exit 1; fi
These lines must remain in the file config.

Step 4: Debmirror

Here we create a keyring for the mirror, so you can stay in 'secure apt' mode.
 sudo mkdir /the/target/directory-keyring
Import the Ubuntu keyring, this is what the upstream packages are checked against.
 gpg --no-default-keyring --keyring /the/target/directory-keyring/trustedkeys.gpg --import /usr/share/keyrings/ubuntu-archive-keyring.gpg
This mirror will require approximately 50GB for i386 only, and 70GB for both i386 and amd64. This mirror should be created on a hard drive with at least 10GB of space free after the mirror has been created.
export GNUPGHOME=/home/mirrorkeyring

debmirror --verbose --progress --method=http --arch=i386 --source \
--dist=hardy,hardy-security,hardy-updates,hardy-backports --section=main,main/debian-installer,universe \
--root=ubuntu /the/target/directory

/the/target/directory should be the path you set in $MIRRORLOCAL

Step 5: Packages

You'll need a number of packages for Builder to run properly.
 apt-get install reprepro debmirror build-essential apache2 subversion cdebootstrap debootstrap imagemagick apt-utils
apt-get install squashfs-tools netpbm syslinux bittornado fakeroot devscripts equivs sharutils mkisofs
If there is packages missing from this list, please let us know.

Step 6: Setting up $REPODST

Hard linking of $REPODST (distribution target directory) is handled by gen-repo, so you are not required to do it by hand.
To do intial repository setup, run gen-repo
For gen-livecd to work, you will need to serve $REPODST with Apache (or another http server). The simplest way to do this is symlink $REPODST into /var/www/$DISTRONAME_L, and leave the $REPOAPT variable alone.
 mkdir -p /var/www/
ln -s $REPODST /var/www/$DISTRONAME_L
At this point its also sensible to make your local Ubuntu mirror (made in step 4) available via http
 ln -s /the/target/directory /var/www/ubuntu/

Step 7: Generate the repository

This will take a while. The repository will possibly be in an inconsistent state while this is running, which is why you only push your changes to a mirror in Step 7. Every time there's new versions upstream (e.g. security updates) re-run do-update (and debmirror if needed).
If you are running a 32bit userland on a 64bit kernel, install the linux32 package and use
 ./gen-repo && linux32 ./do-update

Step 8: Generate the LiveCD

The created image will be placed in $LIVECDDIR/$DISTRONAME_L-livecd-$RELEASE-$ARCHTOBUILD-$LIVECD_VERSION.iso
Will prepare a source tarball and push the iso to $LIVECDDIR

Step 9: Push your repository to a mirror and publicise

You can now publish your repository (the dists and pool) directories as well as your LiveCD to your mirror and publicise your new distribution to the world.

Debian 5 Lenny: New Features

Debian: New Features
Debian 5 Lenny is an excellent desktop or server option. Check out the latest Debian Tutorials for some more up-to-date reading.
I have been running the Debian 5 Lenny candidate for awhile and have been very pleased with the stability and features.  It actually functions and acts more like the distribution I need and work on than Ubuntu.  Not so say  that Ubuntu is bad, just that I typically do not need or use the latest applications.
Debian has a philosophy that is more like Red Hat/CentOS in that the focus is more on testing to insure stability of the whole system than to provide the latest and greatest applications.  Debian comes with 18,733 packages.  Debian 5 Lenny will have 5 DVDs to download with those packages, that is simply beyond any other distribution. These packages are maintained by volunteers who cover the globe providing insight and thinking from various cultures and backgrounds.
Debian Desktop

Debian Desktop Web Browser

Iceweasel is really Firefox that has been backported with security fixes to the Debian 5 stable version.  All of the trademark artwork has been removed from it.  Iceweasel is version 3.05.  Iceweasel functions the same as Firefox so it is a great web browser.

debian iceweasel

vi on Debian

As I am really used to vi on CentOS servers I am forever using the command vi and then unable to edit the way I do on CentOS.  So I am constantly creating aliases so all the machines I am on work the same.  This is a real pain...

So I installed vim-full. Now vi works like I am used to with version 7.1.314.  For me this is an important part of any transition.

I quickly found the /root terminal in the menu and found that was a great option for me as I am usually working on a server and it is a quick way to get access.  Quite franlky I get tired of doing sudo all of the time with Ubuntu.
debian 5

Debian Desktop Icons

The icons are simple but pleasing.  I want more simplicity and less "bling" because all I do is work on these machines with very little play.  So faster, simplier is better for me.  OK ...sometimes the Debian icons are not as pleasing as Ubuntu, to me it does not matter, function is more important.

Debian 5 icons

OpenOffice on Debian

Though Debian 5 uses version 2.4.1 of OepnOffice instead of 3.0, the most important feature of 3.0 is available and that is the ability to read the new version of Microsoft Word 2003, XML.  As you can see in the example, here the test file created in 2.4.1 is able to save in the new format.  Version 2.4.1 is also more stable which is certainly an important aspect.
OpenOffice on Debian
Lenny uses Gnome 2.22.3 which is compared to Gnome 2.24.1 on Ubuntu 8.10.

Lenny is using the 2.6.26-1 kernel compared to the 2.6.27 kernel in Ubuntu 8.10.

New Debian Features

Install Debian from Windows. Debian now supports an application that allows you to install Debian right from a Windows desktop.
SATA RAID support. 

Support for volatile which is the archive that will host data that changes frequently like anti-virus signatures.  This is now supported by the installer.

Support for hardware speek sysntheses devices which will aid those which are visually-impaired.

You can now synchonize your clock with NTP, Network Time Protocol at installation time.

SELinux support is not enabled by default but is available with:

apt-get install selinux-basics

When you install SELinux it is enabled as permissive, see /etc/selinux/config.  The permissive option does not enfoce SELinux voilations but it does report them in the /var/log/messages so you can learn how it works and what needs to be fixed but you are not locked into the restrictiosn immediately.

New versions of programs include these update.  If you follow any of these packages you can see it is a major improvement.

Apache 2.2.9
Bind  9.5.0
Dia   0.96.1
Ekiga  2.0.12
Exim  4.69
GNU Compiler  4.3.2
GIMP  2.4.7
MySQL   5.0.51a
OpenLDAP  2.4.11
OpenSSH  5.1p1
PHP  5.2.6
Postfix  2.5.5
PostgresSQL  8.3.4
Python  2.5.2
Tomcat 5.5.26
Debian Users: Don't stop now. We've got plenty of useful Debian Linux Tutorials for you to read.

Selasa, 02 Oktober 2012

Absolute 14.0 released


 Just Released, I burnt A copy Right away to Test it out...... So far it works Nice This is an operating system that I must have.  And keep it Installed In one of my computer  and customize it, to my needs. 

Coincides with the Slackware 14.0 release, keeping in step with libraries, toolchain and basic apps. Absolute has moved away from udisks to use the lighter spacefm file manager, which takes advantage of native kernel polling. So as always, Absolute will run fast on modest hardware. Network manager is taking care of internet connections by default. Java and multimedia add-ons need to be installed (via system tools, as root) post-installation.

Chrome browser is now the default (taking the place of chromium and/or firefox.) I distribute the browser with several plugins and the Chrome browser comes with a more up-to-date flash release than what is available as a generic plugin for Linux machines. If you install the firefox browser (from Extra/internet) you might want the flash plugin package located there as well. Last note about the browser -- I had briefly switched to Firefox as the Absolute default, but a couple "hang-ups" made me switch back to the Google app. Sorry about any confusion. I just like to use whatever works best.

About Absolute Linux

  • Is a modification of Slackware

  • Fast, stay-out-of-your-way desktop. No Gnome or KDE. Uses well integrated collection of the best lightweight software to give you most of the things the "big desktops" give you, only at at much, much faster pace.

  • The software folks use on a desktop: Firefox, OpenOffice, Pidgin chat, GIMP image editor, WPClipart, Thunderbird mail, K3B CD/DVD burning, Frostwire P2P, Deluge BitTorrent, 2 dictionaries, a bunch of games -- basically any of the latest, most popular Open Source software.

  • Many script utilities (with friendly little GUIs) to make configuration and maintenance of system easier and still allows manual (text-file based) configuration, if desired.

  • Package compatible with Slackware minor versions. (ex. Absolute 12.2 -> Slackware-12.2), with a few exceptions: kernel packages, KDE-libs, madwifi, ndiswrapper, svga-helper, a/etc.

  • Root user can install software or set system configurations. That's it. Perfect for parents and IT guys. If you want to do something root user is allowed to do, log in as root. That's what the account is for.
I hope you Download a Copy and Enjoy it.

Thanks for stopping by.