Linux Certification – Preparation (Part 3)

In Part 1 of my Linux+ preparation, I selected study materials. During Part 2, I selected a Linux distribution and listed the reasons behind my choice. In Part 3, I will reveal my target system.

I have three preparation tasks before I start my quest for certification:

1) Select Study Materials
2) Select a Linux Distribution for Study
3) Select a Target System for the Linux Distribution


Preparation – Part 3: The Target System

Disclaimer: What follows are my very personal and specific reasons for selecting a particular type of hardware to run Linux. Your needs and wants may be very different from mine. Use whatever hardware you want to run Linux, and don’t hold my choice against me. Choice is what Linux is all about.

When I set out to obtain a Linux certification, one of my primary goals was to make the switch to Linux full time. To make such a transition would be tough. My primary machine has many proprietary programs that are required by some of my other hobbies (Digital Photography, GPS and trekking, …), as well as software I use for work. Windows is hard to get away from. Since scrapping my primary machine was not an option, I had to think creatively to come up with a solution for a Linux system that I would use daily. The answer for me was a netbook.

The thought of having a small computer that could sit on my nightstand (or an end table) and be available whenever an inspiration strikes was quite appealing. Due to their small footprints, a netbook would be perfect for this. A netbook would be an excellent travel companion as well. Since they are so lightweight, I could bring it almost anywhere.


Choosing a Netbook / Mini-Laptop / Ultra-portable / Smartbook

Tiny notebooks go by many names, and with the rise of the “netbook” there are many choices out there. For my purposes a few factors would come into play:

1. Cost. I wanted to budget around $250 for my purchase. My better half said $150. That meant I would more than likely need to look at used laptops. The myth of the (new) $100 laptop has not become a reality yet. I would have to look on the used market.
2. Size. I wanted small, but with a keyboard I could type on. I can work down to about 90% full size without much issue. However, the larger the keyboard, the better. A screen with a resolution greater than 800x480 is another size factor.
3. Memory. I wanted at least 512MB, enough to work with a descent light-weight Linux and have more than one application open at once. A lot of older ultra portables have 256MB or less, but most new netbooks ship with a minimum of 512MB.
4. Processor. Not Important. I would prefer x86-compatible, but with ARM processors and some of the "soon-to-be-released" ARM-based smartbooks on the way I am open to any platform as long as conditions 1-3 are met. HD and streaming flash content are not needs for me. Productivity is the primary goal.
   
5. Hard Drive. I want storage -- enough to store a fair amount of music and photos. The first generation netbooks with their 2GB, 4GB, and 8GB flash drives would not work for me. I wanted 30GB at the very minimum. Again, this requirement pushed me to the used market.

My search began with me taking stock of what I already had in my house. Over the years, I had accumulated several mini-notebooks: 3 Toshiba Librettos and 1 Soyo PW-9801. Unfortunately, none of these had enough horse power to run a modern Linux distribution. The Soyo, with its 128MB of RAM came close, but its Cyrix MediaGX processor was troublesome. I did try to find a distribution that would work these pieces of “atticware”. Other than DSL (which is too limiting), I could not find a suitable distribution. Absolute Linux 11.0.92 came very close. It runs everything well on an old Libretto 100CT (64MB) with one exception: web browsers. Even old versions of Firefox would take 30-60 seconds (or longer) to load. I tried lightweight browsers, but the experience was not all that great. None of these old minis would work as a side-table browsing and writing PC. I would need to search for something a little more modern.

There are so many netbooks on the market today, most with the Microsoft designated 1GB of memory and 160GB hard drives, that it can be a bit overwhelming trying to select a device. I chose to eliminate Windows-based netbooks (unless used) from my search and concentrate devices that came pre-loaded with Linux.

Laptops that come preloaded with Linux should:

• Run any Linux (the pre-load serves as proof-of-concept)
• Be slightly cheaper (at least in theory, due to no $20-40 Microsoft XP Home tax)

Here are a handful of the devices I considered:

• Dell Inspiron Mini – preloaded with Ubuntu Netbook Remix
• Everex Cloudbook or the Sylvania Netbook – preloaded with gOS
• Sylvania G Netbook Meso – preloaded with Ubuntu Netbook Remix
• HP 2133 Mini-note – preloaded with SUSE Linux Enterprise Desktop (SLED)

Try as I might, the only netbooks in this list that routinely would appear used for under $150 were the two laptops based on the VIA C7-m platform: the Everex Cloudbook and the original Sylvania Netbook. I had seen the Everex variant in the flesh at a Circuit City store (before they went out of business) and found them to be the perfect size. In fact, they reminded me a lot of my old Libretto 100CT. Only the screen resolution (800 x 480) kept me from taking the plunge. HP 2133 mini-notes, which have one of the best (if not thee best) netbook keyboards around, routinely appear for just under $200, but I wanted to break the $150 barrier.

I began to look at devices with odd hardware:

• The toy-like XBurst CPU netbooks (32-bit MIPS-based)
• Always Innovating’s Touchbook (ARM-based)
• Emtec Gdium Liberty 1000 (64-bit MIPS-based)
• Nohrtec’s Gecko Edubook (x86; modular; can use AA batteries)

The XBurst CPU is MIPS-based. While netbooks based on this platform are typically very cheap, and are a natural curiosity, the challenges of working on the unusual MIPS platform would not be conducive for studying for my exam. These netbooks -- with names such as the "Belco Alpha 400" and the "3k Razorbook -- were more a curiosity than anything else. They are cheap. I will give them that ("cheap" is why people are drawn to these toys). I’ve seen many of these netbooks sell used for well under $100. But the limits from the low hardware specifications and the lack available distributions for these were obvious: low memory (128MB), low storage (2GB or less), low resolution (800x480), no modern browser… and the list goes on.

Always Innovating’s Touch Book is a device I want, plain and simple. This device inspired “gadget lust” the moment I saw it. Early reviews of the beta units have not scared me away (“tippy”, software issues, etc.). This is a device based on the open source BeagleBoard platform, which features an ARM processor. It is a dockable tablet that can get amazing battery life (10 hours or more). You may have heard that ARM-based netbooks are coming (I believe “smartbook” is the term being applied to these). One technology prognosticator predicted that ARM-based netbook would make 2009 “the Year of Linux” (laughable… as Linux and ARM-based netbooks are still not on the market beyond beta). And if you follow “little laptop” trends, then you will most likely have already head about Ubuntu’s announcment to develop for ARM-based platforms. Slackware (my distro of choice) has an official ARM port as well (ARMedslack). Great, I had found the device that will allow me to transition to Linux. ARM is the future of armchair computing and development on the go. Not so fast! Being a first generation device, the A.I. Touchpad current runs $400 with the optional keyboard dock. Ouch! That puts this well out of my price range for the time being.

Finally available in the U.S., the EMTEC's Gdium Liberty 1000 netbook is another device that does things just a little bit differently. For one it runs the Loongson processor, which is a 64-bit MIPS CPU. Next, it runs everything from a form-fitting flash drive, dubbed the “G-key”. The theory here is that multiple users can run using the same base, but use different G-keys. So, a classroom could provide a netbook chassis to different students during different sessions. Or, for family use, each member could keep a personal key for the single family base. This is a very interesting concept. Laptop Magazine reviewed an early unit (not the final production unit) and basically thought the same thing: interesting concept; let’s see where they go with this. Right now, the Gdium Liberty is available from Amazon for $265 (or less) – not a bad price for a Linux netbook and one very unique piece of hardware.

If the name Loongson seems familiar, it may be because you have heard about its use in the open hardware netbook, the Yeelong. The Yeelong is the netbook that is (or was) used by FSF’s Richard Stallman. Unlike the closed hardware on the Gdium Liberty, the Yeelong is completely open, with an open source BIOS and full hardware specifications available. Add an FSF-supported operating system, like gNewSense and you would have the first completely “free” (as in non-proprietary) hardware AND software. If Free and Open Source Software (FOSS) are your religions, then right now the Yeelong is one of the only ways to go. Unfortunately, you’d have to import it, and it is not cheap to do so. With shipping from overseas a Yeelong will most likely run you $350-$400. In this case, freedom isn’t free, and it is also more than my budget will currently allow.

The Nohrtec Gecko Edubook is another interesting device. It is modular. It can use AA rechargeable batteries (of which I have a stockpile) and it is low cost -- currently $200, and less for the barebones version. Unfortunately, the Gecko is a unit designed to be sold in bulk, and shipping on a single units to the United State appears to be expensive. For example, on Nohrtech’s MicroClient and MicroServer products single unit shipping is over $50. If the same shipping rates apply, this would place the Gecko more in the $250 dollar range and there are better netbooks to be found at that price point. Still the idea of an upgradable and modular platform is quite intriguing.


The $123 Netbook – Cameron Maxmedia NB-1060




After searching for months for a suitable target netbook, I came across a very odd netbook on eBay that was originally designated for the Russian market. The "Cameron Maxmedia NB-1060" is a Geode-based (LX800) netbook with 512MB of memory and a 60GB. The netbook was being sold by a state-side seller (in Pennsylvania) and had a starting bid of $123 with free shipping. It was a very odd piece of hardware, but I threw a low-ball bid in on it and ended up winning the thing for the starting price. I now had my sub $150 netbook, but what did I just buy? Here are the listed hardware specifications (click specifications to enlarge view):



The Cameron Maxmedia NB-1060 marked all my important checkboxes:

• Price <= $150 – CHECK
  
• A keyboard I could type on – CHECK (approximately 90-92% full size)
• Resolution > 800x480 – CHECK (NB-1060 has the netbook “standard” 1024x600 screen)
• Memory >= 512MB – CHECK (NB-1060 has 512MB DDR; upgradeable to 1GB)
• Hard drive >= 30GB – CHECK (NB-1060 has a 60GB hard drive)

Further research revealed that the NB-1060 was most likely a rebranded OEM of the Malata PC-81002, a Chinese-made netbook. I managed to track down a Russian forum with links to drivers, disassembly instructions, Linux X-windows configurations settings, and other useful information too (I had to use Google to translate the pages). I would be ready when the little netbook arrived.

The laptop arrived with a 2007 version of NeoShine Linux Desktop (version 3.0) converted to English (a bit strange for laptop meant for the Russian market). NeoShine is based on some form of Red Hat and was very slow on this hardware. Items such as the wireless adapter were not functional and the fit and finish of the custom user interface left something to be desired (no task manager or tool bars, for example). Finding out exactly what hardware was in the NB-1060 was a task. The included manual just listed the basics and without the proper drivers setup in NeoShine, complete hardware was not visible. I ended up wiping the drive. I briefly loaded Windows XP (oh, the humanity!), loaded up all the Windows drivers, and then exported all the information in the Device Manager and System Information to text files. Now I had a clear view of the odd hardware in this system.

• Processor: AMD Geode LX800 clocked at 500 Mhz, coupled with a AMD Geode CS5536 “companion device” (x86 Family 5 Model 10 Stepping 2)
• Memory: 512 MB (with 24MB set aside for shared video)
• Video: AMD Geode custom driver (could be problematic)
• Audio: AMD Geode WDM Audio Driver (could be problematic)
• LAN: Realtek RTL8139/810x Family Fast Ethernet NIC
• WLAN: ZyXEL G-202 (ZD1211B) IEEE 802.11 b+g USB Adapter (could be problematic)
• PCMCIA: Texas Instruments PCI-1510 CardBus Controller

The system also features a “Geode GX3 AES Crypto Driver” which I guessed at the time would not be supported in Linux (later I would find that AMD did have a Linux version available).

Searching around various sites (AMD, ZyXEL, Maxmedia), I was able to obtain all the drivers for Linux, including the AMD Geode platform drivers, the ZD1211B drivers, and X.org configuration settings.

I now had all the pieces necessary to attempt a new Linux install on this odd netbook. The reality of creating a Linux study platform and end-table computer had come together. With my study materials selected, distribution picked, and target system acquired, I was ready to start my ‘Year of Linux”.


Links to More Linux and Netbook Information:

• Amazon - one of the best Netbook selections on the Internet.
• ARMedslack - Slackware for ARM processors
• eBay - allows for very customized searches (< $150, <>
• FSF Hardware - Free Software Foundation's list of supported hardware
• FSF Operating Systems - FSF's list of recognized "Free" operating systems.
  
• Gdium Community Page - for all things related to the Gdium Liberty.
  
• Liliputing - one of the better news sites for netbooks
• Little Linux Laptop - want to try an XBurst netbook? Start here.
• Russian Maxmedia Forums - not a lot of content; but what's there is good.
  

Linux Certification – Preparation (Part 2)

In Part 1 of my Linux+ preparation, I established that I would employ self-study and I selected my study materials. Continuing on in this series of posts is Part 2, in which I will select a Linux distribution.

I have three preparation tasks before I start my quest for certification:

1) Select Study Materials
2) Select a Linux Distribution for Study
3) Select a Target System for the Linux Distribution


Preparation – Part 2: Linux Distribution

Disclaimer: What follows are my very personal reasons for selecting a particular Linux distribution for use. Your needs and wants may be very different from mine. Use whatever distribution you want for study or daily use, and don’t hold my choice against me. Choice is what Linux is all about.

I am not going to bother with suspense here. I will be using some form of Slackware or a derivative for study. This should come as no surprise, as this blog does have a Slackware focus. I am also the most comfortable with Slackware and Zenwalk, so the old adage, “stick with what you know”, applies here.

There were many options to choose from for a study distribution:

• Preconfigured Linux (Linux live CDs, et. al)
• Targeted study distributions
• CompTIA referenced
• Traditional install

Let me cover “the how and the why” of each briefly:

Preconfigured Linux: Yes, Live CDs, USB distributions, and running a virtual distribution from within another OS (using VMWare, VirtualBox, QEMU,..) would provide a very fast method for getting a working Linux distribution going. These three methods provide the benefit of keeping the host OS intact, allowing the user to run Linux and another OS on the same PC. Unfortunately, these methods would not work for me. My goal here is to make this the “Year of Linux”. That means running Linux on a dedicated system. There are no shortages of repositories of virtual images and excellent Linux live CDs out there – SLAX, KNOPPIX, and DSL to name a few – but I wanted to go the dedicated route. Is that not what “making the switch” is all about?

Targeted study distributions: There are a couple of older (deprecated and possibly discontinued) Linux distributions specifically targeting education. The Privare Learn Linux Course (also found under the names “Freeloader Linux”, “eLearnix”, and “The Wanderer”) provides a bootable live CD with tutorial movies for Linux basics. The Privare live CD was released in 2006 and runs best on older hardware. The tutorials are all available online at the project’s SourceForge site, so there is really no reason to run any of these Live CDs. Another more recent project was elpicx. Elpicx (notice the LPIC in the name) was a live DVD designed to help students prepare for the LPIC-1 exam. The last release (v2.0) was in August 2008 and the website is currently offline (as of September 2009). The trainer, Karl Schock, apparently closed down shop in October 2008. Before its close, the site offered a “Premium” version of the DVD and LPIC 10-week courses as well. You can still grab the ISO here: elpicx_20_20080925_EN.iso if you want to check it out. Once again, both of the targeted study distributions seem interesting, but neither is in the true spirit of “making the switch” full time to Linux.

CompTIA referenced: In the exam objectives for Linux+ version XK0-002 (remember these were written in 2004 for the 2005 version of the exam), CompTIA states:

“The candidate is not expected to know how to install a specific distribution, but should be familiar with setting used by installers on the major distributions. The scope of the exam is limited to software and settings common to Linux software from Red Hat, SuSE, Mandrake, and TurboLinux. Students will need to know one, not all, of these distributions.”

I can see why Linux+ was in need of update. Three of these distributions mentioned do not exist anymore* and TurboLinux has not been popular in the United States for years.

* Let me clarify the “do not exist anymore” statement. Red Hat 9 was the last “Red Hat” release before the split into RHEL (Enterprise Linux) and the community based Fedora project. SuSE was acquired by Novell in 2003, and at the time I believe only had a commercially available version of SUSE (SLE) or a network installable version (note: I could be very wrong on that). The community fork, openSUSE came around in late 2005. Mandrake ceased to be Mandrake when it merged with Conectiva Linux in 2005 and thus became Mandriva.

So, in the forms mentioned on the 2005 objectives, Red Hat, SuSE, and Mandrake do not exist anymore. Anyone wishing to study based on these recommendations, might want to either grab an old ISO and old piece of hardware to go with it (not recommended), or use the current distributions based on the CompTIA recommendations.

Current Commercial Distributions:

• Red Hat Enterprise Linux (RHEL) Server or Desktop
• Novell SUSE Linux Enterprise Server (SLES) or Desktop (SLED)
• Mandriva Enterprise Server, Mandriva Directory Server, Mandriva Powerpack (Desktop)
• TurboLinux Server, TurboLinux FUJI (Desktop)

Current Free or Community Distributions:

• Fedora (community fork of Red Hat; sponsored by Red Hat)
• CentOS (community-based binary compatible version of RHEL; not associated with Red Hat)
• openSUSE (community fork of SUSE; sponsored by Novell)
• Mandriva One (“pure Free Software edition, no proprietary software/drivers”)

All of these distributions do have at least one thing in common. They are all RPM based. Even though CompTIA has aimed for a vendor neutral exam, the objectives suggest a Red Hat (or Red Hat Package Manager) bias. The rise of Debian-based distributions, specifically Ubuntu, had not really begun when the 2005 objectives were penned. I suppose this was something to keep in mind as I selected a distribution for study.

Traditional install: Finally we come to the method I would ultimately select, the traditional Hard Drive-based installation. Installing directly on a device gives more permanence to my goal of switching to Linux. It also provides easier access to kernel source and developer tools. With hundreds of Linux distributions to choose from, I selected a Slackware-based distribution over an RPM-based Linux or DEB-based Linux. The reasons for this selection were easy.

1. I have used Slackware and derivatives more than other Linux family tree. Familiar means I could dive right into studying without worrying about learning a new system.
2. Most of the objectives in Linux+ appear to be based around the command line which is readily available in just about any Linux distribution (vendor neutrality).
3. RPM support can be added to Slackware quite easily, making the bias of CompTIA’s recommended distributions a non-issue.
4. I wanted a very current distribution, and at least four Slackware-family distributions have been released within the past month.

So how come I did not consider an RPM-based or DEB-based distribution? For RPM-based distributions that is an easy answer. For Fedora, and openSUSE in particular, I would never run the current versions of these releases on anything less than a 2Ghz processor and 1GB of memory (2GB of memory is even better). Fedora and openSUSE are slow on the hardware I have at my disposable. I do like openSUSE and have even spent a year with SLED while on past job, but even on a Pentium 4 HT 3Ghz processor and 2GB, SLED (which is a very stripped down SUSE), felt slow.

As for DEB-based distros… To be honest, I’ve tried various Ubuntu family distributions off and on since Ubuntu 6.06 LTS. I have never really been impressed with the way Ubuntu does things. I miss my separate root account. I have always had hardware issues (mainly audio and wireless) that took hours or days to resolve. I have never been impressed with the performance of Ubuntu (although that said, Ubuntu does seem peppier than Fedora and openSUSE). Finally, the distribution just looks ugly to me with almost no thought going into polishing the look and feel of the distribution. I realized that most of these issues can be changed or customized (or I could just use stock Debian), but I find the speed and customization of Slackware-based operating systems to be better suited to me. Slackware and friends are fast and run better on low-end hardware. Also, I like to tinker under the hood. Ubuntu assumes you won’t need to, or won’t want to play around at the command line.

So, there you have them, my reasons (not yours) for selecting a distribution based on Slackware. The next question was which Slackware-based distribution to select. I considered four of the most recently released distributions:

• Slackware 13.0 (release 27-AUG-2009)
• Absolute 13.0.1 (released 12-SEP-2009)
• Zenwalk 6.2 (released 6-SEP-2009)
• Salix 13.0 (released 16-SEP-2009)

I also considered other Slackware offspring of interest, including:

• Slax 6.1.2 (released 5-AUG-2009)
• Vector Linux 6.0 (Standard 22-FEB-2009, Light 2-APR-2009, Light “Live” 30-AUG-2009)
• Kongoni 1.12.2 (released 12-JUL-2009) -- recognized by the FSF
• NetSecL 2.4 (released 20-MAY-2009) -- a security focused distribution

Let us quickly take a look at these slightly older distributions, before taking a look at my “final four” contenders.

SLAX: SLAX is a one of the better (and longest running) Linux Live CDs out there. The fact that it is a Live CD, eliminates it from my list. I did consider it briefly as a method to get up and studying quickly. In the end, I wanted a more permanent hard drive-based install.

Vector: I have used Vector in the past. As far as performance goes, it flies. I decided to wait until their next version arrives, based on Slackware 13.0, before taking it for another spin. For desktop use, I have flip-flopped quite a bit between Vector and Zenwalk over the years. Only netpkg keeps me going back to Zenwalk.

Kongoni: Kongoni is an interesting distribution. Kongoni means “gnu” in the Shona language, and only includes software that complies with Free Software Foundation's definition of software freedom. Because the distribution is based out of South Africa, they can still include media codecs that other countries cannot. Here’s what their website has to say about this:

“Since Kongoni is developed in South Africa, we are subject to neither the DMCA nor Software Patent laws and are therefore able to offer our users these codecs which other distributions are often forced to exclude for fear of legal action.”

Kongoni is the process of switching over to the Slackware 13 tree. Until that transition is complete, I will watch but will not test drive.

NetSecL: NetSecL is a lesser known Slackware off-shoot, but has very interesting focus: Security. To harden the distribution, the NetSecL team removed all the servers, closed incoming ports, and turned off unnecessary services. NetSecL also includes a handful of security tools, a source-based portage system (CruxPorts4Slack), and grsecurity patches. Its only weakness that I see at this point is the lack of an update system (due to Slackware changing package formats). I believe an update system is in development for future releases (which will most likely be based on Slackware 13.0). I’ll keep an eye on this distribution. The idea of having a hardened system out of the box was something I though only OpenBSD users enjoyed.


The Final Four – Slackware and Friends from the Past Month

Slackware: How could I not put Slackware on the list? I am a Slackware subscriber (and have been since about version 8 or 9). The subscription is my way of giving back, and I give to the parent so that the children will continue to be supported. Slackware makes an excellent base distribution because of the “Keep It Simple” philosophy. Slackware comes with the bare essentials and leaves most of the configuration up the user. I have certainly learned a lot over the years due to this approach. The idea of configuring a system largely from scratch scares away a lot of people, when in reality the fears are largely unfounded. Slack does do some things for you (hardware detection,…), but leaves other choices (machine purpose, security, desktop environment, …) completely up to the user. I love the approach, however, for purposes of getting up and ready to study quickly, I would choose to go with a distribution that came a little more pre-configured and ready to run for desktop use.

Absolute: Absolute Linux IS Slackware. Don’t like that description? Well, it’s the best way for me to describe Absolute. It is a Linux distribution that does a lot of the post-installation tasks that Slackware leaves off. Absolute could almost be distributed at as add-on package for Slackware that quickly turns the system into a desktop (if I’m not mistaken, it formerly was released a series of add-on packages). For those of us that are too lazy to install such packages or to configure Slackware for desktop use, creator Paul Sherman has rolled his Slackware modifications into a distribution for easy installation and done an amazing job. Absolute is growing by leaps and bounds (it even has its own repository now). Absolute keeps things lightweight (IceWM), adds custom configuration tools, and some great documentation. I also really like the old-school decision to keep root and user accounts separate (rather than messing with all this “su” or “sudo” nonsense that so many modern distros employ). I have used an older version of Absolute -- based on Slackware 11 and the 2.4 kernel -- on really old hardware (Pentium 166Mhz) with great results. I have not tried any of the releases based on the 2.6 kernel yet and I do not plan to for my studying. I’m looking more at an Xfce-based distribution. Xfce, at the moment, is my preferred desktop environment (at least until LXDE matures a little more).

With that in mind, my install decision came down to two contenders: Zenwalk and the result of a Zenwalk spat, Salix OS.

Zenwalk: Zenwalk has been around quite some time (2004), started out as a stripped down Slackware called Minislack, but developed into its own distribution. As such, Zenwalk has not always remained 100% compatible with the parent OS, but is this really an issue? Zenwalk is simple, fast, has great hardware detection, and an excellent package manager with dependency resolution, netpkg. To me, Zenwalk is Slackware-like, not just a modification of Slackware, and netpkg is the glue that holds Zenwalk together. Slackware packages are hit and miss with Zenwalk, however, Zenwalk has deep repositories for making changes with netpkg. Zenwalk follows the “one application per task” philosophy, which keeps the distribution within the size of a single CD (Slackware’s new compressed file format has shrunk the ISO size even lower). Zenwalk really is a great distribution and I have probably logged more hours on this distribution than any other except for Slackware, but I’ve always had several issues with Zenwalk:

• “One application per task”. What if they are not your applications of choice? Just install what you want then you say. Okay, fine, but now I no longer have one application per task.
• Standard vs. Spin Offs. To correct the application choice issue, a spin-off called Zenwalk Core was launched. Core was (is) very bare bones, with only command line and no X-windows. Would it have been that hard to install X-windows on Core or at least provide a Wiki article on which packages were required to setup a basic system with Core + X? Doing so manually is trial an error with Zenwalk Core. Core disappeared for a little while. I am glad to that it has been resurrected with version 6.2. Even better than offering Standard and Core, why not let the user choose at install: Core, Core+X, or Full (Standard) Zenwalk. Bundle it all on one CD. Now that would be really cool.
• Stable vs. Current vs. Snapshot. In the Slackware world, there are two trees or development trunks: stable and current. Coming from a Slackware background, Zenwalk’s use of “current” (for stable release) and “snapshot” (for the current developmental packages) is a bit confusing (Note: I've long adjusted, but I see other people complaining about this as well)
  
• Partially Rolling Releases. Because Zenwalk is a partially rolling release, there are no patches or updates for older versions. A user is expected to run the latest “current” release. There is no such thing as Long Term Support (LTS) on Zenwalk.
• To Be Free ("libre") or Not to Be Free. Zenwalk takes a weird stance on proprietary software that I do not fully understand. For example, Firefox is left off for having proprietary blobs within it. Unfortunately, Firefox just works on the 5% of sites that IceWeasel cannot handle. Firefox is not even in the repository as an option (note: there is a contributor that offers a packaged version for Zenwalk; check the forums). Why such a lack of love for the Fox, when Adobe products, Intel drivers, various wireless drivers, and such all make it in to Zenwalk. Furthermore, the repository offers quite a few packages that may have proprietary or legal issues (such as, media codecs). If these are in the repositories, why not Firefox? IceWeasel works OK for me, but after surfing for a little while, it always causes funky rendering issues (double buttons, screen smearing, etc.). The issue has plagued just about every Zenwalk installation I’ve run and I suspect it is still in the 6.x series.
• The Benevolent Dictator factor. Slackware has single head (Patrick Volkerding), but maintains a defined vision: Keep it simple, never over configure things, and leave major decisions up to the users. This approach has worked for Slackware. For some reason the Dictator approach does not work for the majority of other distributions. Culture clashes within Linux distributions seemingly make the Linux press once a month and this year Zenwalk got hit with a pretty big clash. This could be because Zenwalk’s vision is a litter murkier (a best guess follows): work on netpkg, keep a never ending rotation of the “one application” for each task, introduce spin-offs, but never merge features. It is difficult to determine the real goals of Zenwalk. In the spring of 2009 a rift between Zenwalk’s Benevolent Dictator (Jean-Philippe Guillemin) and some major developers caused an exodus of some of Zenwalk’s heaviest contributors (you can read a little more about this HERE). In my outsider opinion, the primary cause for the split was the Zenwalk project's lack of clearly communicated goals. Without direct communication, everyone had their own idea of what Zenwalk should be. In reality, the vision of most distributions rests with the original creator (or "Benevolent Dictator"). Tempers and egos flexed, and some people got mad and left the project. Zenwalk moved on thankfully (some distributions don’t make it after such a split) and a new distribution was born from the developers that left: Salix.

Salix: Salix is an interesting new animal and I think the remaining Zenwalk crew should look carefully at their first release. Some of things Silax did right with their first release:

• Full, Basic, and Core from one ISO. How hard would this be for Zenwalk to do?
• Full DVD Source, which a rarity this days. Source is not requirement per se, as many distributions simply provide the source for their modifications to the parent, and then link to the parent for the original packages. This is a practice that saves bandwidth for the offspring, but makes tracking down the source trickier the user. Full Source was nice to see.
• Firefox included in the default install (both Full and Basic)
• I really like the default single toolbar setup. I typically setup my Zenwalk boxes with a single bar across the top to 1) remind me that I am not using Windows and 2) to save screen real estate.
  

Stuff I am not sure about with Silax:

• Quality Control of Slackware Packages (note: I am not refering to Salix's repository). Salix aims for full compatibility with the upstream parent, Slackware 13.0. I think backward compatibility is interesting in that it opens up a huge library of packages for Silax users. However, the quality of some Slackware packages is questionable (for example, some of the packages from linuxpackages.net and Slacky.eu need "caveat emptor" warnings). Sometimes it’s easier to just compile from source.
• Available Packaging Tools. The package tools for Slackware, slapt-get (and gslapt), have broken more than they have fixed for me in the past. I find them largely unreliable (possibly due to the package quality issue). It will be interesting to see if Silax can build up its own repositories large enough to prevent this issue. In contrast, netpkg (and xnetpkg) have “just worked” for me (In fairness, I know of people that have had issues with netpkg. I am not one of those people).
• Age of the Distribution. Silax is new -- really new. As of this writing, it is wait-listed on Distrowatch. The website doesn’t have a lot of meat on it yet. The community and forums are there, but not heavily populated yet. I think there will be interest in Silax, so time will tell.

So when faced with a decision between Zenwalk and the new distribution made by ex-Zenwalk contributors, which would you pick? Would you go for the tried and true that is still going strong despite the developer split, has a great and reliable packaging system with dependency resolution, still has great community, and gets pretty much everything right? Or, would try the upstart, the unproven new kid on the Slackware block – a distribution that has an excellent installer, includes the most popular browser for Linux in the default build, and makes the source DVD readily available?

It was a really tough choice for me. In the end I decided to stick with what I know, which is Zenwalk. Zenwalk is what I am most familiar with at this point in time, so for study purposes it will allow me to get up to speed faster than Silax. Also, Silax reminds me very much of Absolute Linux in that it IS Slackware – Slackware pre-configured for a user to get up and running quickly. This is not in any way a knock against Absolute or Silax. I am merely pointing out the differences in approaches. One approach aims to provide an easy to use and unique desktop experience with its own package management tool. The other approach aims for full compatibility with the parent and a simple pre-configured setup. Neither approach is right or wrong, just different. With all this in mind, I took the easy road and will stick my familiar friend, Zenwalk.

I will say this: Silax will be the first distribution I test on my VirtualBox setup once I am finished with my exam (I already have the ISO ready to go).

For some more perspective, here are a few early opinions of Zenwalk 6.2 and a release announcement pertaining to Salix.

• Zenwalk 6.2 Review (from LinuxBSDos.com)
• Zenwalk Linux 6.2 Released (from ZDnet.co.uk)
• Zenwalk And Open Sores: Life Goes On (from Red Devil): Discusses the developer split briefly and then covers Zenwalk… which did not miss a beat.
• Enharmonic Zen: Former Zenwalkers Do Salix OS Fork (from Linux Magazine)

Update (9/24): I found an early review of Salix and interesting thread on LinuxQuestions.org:

• Branching Out with Salix (from Red Devil): the First Review of Salix?
• Salix 13.0 Linux OS Announcement on LinuxQuestions.org

The LinuxQuestions thread contains a discussion of how Salix differs from Slackware. The Salix project lead states:

“There are no real differences”.

Salix appears to be simply...

“an easy way to install an Xfce (or console based) Slackware system fast and easily using a single CD, without having to go through the package selection in the Slackware installer.”

So my guess was correct. Salix IS Slackware (similar to way Absolute Linux IS Slackware). The most compelling reasons to try Salix would be 1) if you wanted to save time versus installing and configuring Slackware, and 2) you like the way Salix has set things up.
.

Linux Certification – Preparation (Part 1)

I have already established that Linux certification is in my future, and this would be the “Year of Linux” for me. The next step in my Linux education progression is to select my study aids.

I am going for Linux+ (rather than LPIC) so my decisions will be based on studying for the XK0-002 exam. Typically, when I prepare for an exam I like to take a course. In the past, my employer would send me for a week-long class. In the current economy, everything will be on my own (self-study). With this in mind, I have three preparation tasks before I start my quest for certification:

• Part 1: Select Study Materials
• Part 2: Select a Linux Distribution for Study
• Part 3: Select a Target System for the Linux Distribution

Preparation – Part 1: Study Materials

Choices:

1. Course: TestOut Linux+
2. Book: Linux Administrator Street Smarts, by Roderick W. Smith

THE COURSE

My previous employer gave me a very small education credit (in addition to test vouchers) when I was downsized. The amount was not nearly enough for any sort of cool boot camp (CISSP for example) or even a week-long course on a single topic (Linux+). The credit was enough to purchase some self-study materials that mimic a classroom, so that is the route I took.

To get the closest experience I could find to a classroom at home, I used my education credit (called a grant), to purchase TestOut’s Linux+ course. I was familiar with Test Out from way back in my Novell CNE days some 10 years ago. Back then, they had some of the best self-study courses available. With this in mind, I figured I would give them a try again. All I can say is: my how things have changed.

TestOut now uses a DVD training simulation (called LabSim) and a web-based activation system. The DVDs have multimedia lectures that are supposed to mimic a classroom setting. They do, in theory, mimic classroom lectures. However, after watching the first module, I can safely say that I feel the implementation is a bit off. Here is a screenshot of the TestOut LibSim software and the lecture for section 1:



Notice anything that just doesn’t look right? If you think the video window looks tiny, that is because it is tiny (note: for some reason the actual video could not be captured with a screenshot; an example of Windows Media quirkiness or DRM, perhaps?). The lecture videos sit in small centered window that cannot be resized. I am guessing the TestOut did this in order to remain compatible with any screen resolution right down to 640 x 480. I would prefer to be able to watch the lectures fully filled to the LabSim interface, but that is not an option here.

Update: you can right click on the video and make it full screen. You cannot, however, increase or maximize the video within the TestOut interface. I feel it would be a nice feature to view the video as large as possible and still have the module outline and table of contents visible. It would make navigation easier than the current method of flipping back and forth (between LabSim and full screen).

The next issue I have is with the lecture content. In my opinion, TestOut did not put much thought into the editing. The sections are read, sometimes poorly (lots of “um” and “uh”), in a very dry manner. The speaker is not engaging and does not seem at all comfortable in front of the camera and it shows. So, in addition to the small display, the lecturer is hard to watch. I hope that I as I get deeper into the LabSim, that the lectures will improve.

The final issue I have with TestOut is that you have to be connected to the Internet to use it. The application has to “phone home”. Apparently you can only view the content for 3 years, after which time you can no longer view the content on the DVDs. This activation and online requirement limits portability and is a sharp contrast to the TestOut versions of 10 years ago, but I can work around it. Thank goodness I have a wireless connection. I like to study whenever and wherever the mood strikes me.

The three current issues I have with TestOut (size of video playback, quality of lectures, Internet requirement) will be largely moot as long as the content included in the course is good. I will report back as I work through each section of the course.

THE BOOK

In my opinion, one source of study is never enough. I typically take a 3-tiered approach to certification exams: 1) Classroom Course, 2) Textbook or Books, and 3) Practice Exams, which typically are bundled with the Course or Book.

For my text book, I chose Linux Administrator Street Smarts, by Roderick W. Smith (October 2006).



The reasons behind this choice were easy:

• It is one of the shorter Linux certification texts available containing just over 300 pages of “meat”. I did not want a 1000+ tomb that tries to contain everything Linux, including command references. I already have a command reference book (“man” works quite well too), and I already have Linux experience. I just need a refresher.
• The book labels itself as a “hands-on” book with scenarios and exercises, rather than a purely academic or traditional text. I tend to stay better focused when I have actual tasks, and not just text to read. Hands-on seemed like a very good way to go.
• The book also labels itself as an adjunct text to a more traditional course or your typical 700-1000 page textbook. I consider it an exercise book to go along with TestOut’s Linux+ course.

Are there weakness to this book and approach?

Sure. For one thing, Linux Administrator Street Smarts is very broad and does not go into a great deal of depth. This is not a real issue for me, as any topic requiring more study can easily be researched online. However, some people do like to have all their information in one place. If that describes you, look for a different text (I list some below). Secondly, this text attempts to be generic enough to provide an introduction to both the Linux+ and LPIC-1 exams. Its focus appears to be on common exam objectives, so some topics will not be covered. Again, this was not an issue for me. I have the TestOut course to fill in any gaps and the Internet if I still need to research.

You can never have too many sources when you study for an exam. I will report back if the two sources I selected prove to be the correct media for my certification attempt.


Other Books on Linux+ Certification:

There are other books I considered that specifically target Linux+. There might be general Linux texts out there that cover the topics in the test objectives better as well, but I list these here for reference since they are specifically labeled as XK0-002 study guides. As I mentioned, for my purpose, the smaller and hands-on “Street Smarts” text fits my style of studying better than some of these thick tombs.

Jason Eckert’s Linux+ Trio

Looking at these three books there did not appear to be too much difference except for about 100 pages between them, price, and publication date (March 2006, April 2005, March 2005).

Getting Started with Linux: Novell's Guide to CompTIA's Linux+ (March 2006)

First up, we have 848-page Novell course book (Course 3060). This book is thick and expensive. Most likely due to two reasons: 1) it is probably used in Novell training courses, and 2) it published by Course Technology (who rarely publish anything under $50).

Linux+ Guide to Linux Certification, 2nd Edition (April 2005)

Clocking in at 784 pages, this is slightly shorter than the Novell text, but still plenty expensive – especially expensive if you add in the corresponding Lab Manual.

As mentioned, this guide has a corresponding Lab Manual that increases the overall price. I cannot dispute the quality of these types of texts. In the past, I have used Course Technology books in a classroom setting and found them to be pretty good. In this case, it is the price with which I take issue. These are priced more like college textbooks.

Lab Manual: Linux+ Guide to Linux Certification, 2nd Edition, by Ed Sawicki (March 2005)


Linux+ 2005 In Depth (March 2005)

Finally, a cheap text from Mr. Eckert. It runs 704 pages, but is labeled as “Large Print”. Given that is was the first of his Linux+ trio of books, it appears the lightest on content.


Traditional Linux+ Tombs

Linux+ Certification Study Guide, by Robb Tracy (August 2007)

Do you like your study books thick and full of extra fluff? As one of newer texts available for the XK0-002 and the longest at over 1000 pages, you would think this might be the best text for the exam. Well, reviews are largely mixed. Many reviewers feel it is good for an introduction to Linux, but only average for test preparation. In other words, do not make this your only source for the exam. I don’t know about you, but if have to wade through more than 1000 pages I would certainly expect it to be my one and only source.

Linux+ Study Guide, 3rd Edition, by Roderick W. Smith (February 2005)

Roderick W. Smith does an excellent job at being the first out the door with a new book when a new version of an exam comes out. This was (I believe) the first text out for the XK0-002. The Study Guide is referred to in Linux Administrator Street Smarts as the more traditional text, where as “Street Smarts” is the adjunct guide or hands-on lab manual.

Mike Meyers' Linux+ Certification Passport, by Michael Jang (February 2008)

Michael Jang has written quite a few Linux books and Linux certifications texts. This is the latest (and last) text written on for the XK0-002 version of the Linux+ exam. The book follows more of an “exam cram” style and is just short of 400 pages. It comes with a CD with practice questions. I did consider this text over “Street Smarts”, but looking at both texts, I preferred the more general hands-on approach of “Street Smarts” to the direct information cramming of the Passport series. Both are good texts. Choosing between them is a matter of taste and study style.


The 2009 Exam Objectives

Afraid of certifying with XK0-002 since they are discontinuing the exam very soon. Well, you are in luck. First out of the gate (again) with a book pertaining to 2009 Objectives is Roderick W. Smith.

CompTIA Linux+ Study Guide: 2009 Exam (October 2009)

This book follows the typical form and function of the earlier Sybex guides for the XK0-002 and XK0-001 exams. If you do not think you will have enough time to finish XK0-002 before it is discontinued (end of 2009), then as of right now (late 2009) this is your only option for the next version of the exam.
.

2009: The Year of Linux

You hear that declaration a lot: "this is the year of Linux". It seems silly to say. Every year experts predict Linux will take off in some way. I don't understand why people hype it up so much. It has become a punch line, a cliché, and it is just not necessary.

Why is it not necessary? Every year since its creation there has been steady Linux growth. On the server side, Linux has an excellent foot hold (thanks no doubt in part to Apache), and in the corporate world, Red Hat and Suse (Novell) have been doing brisk business here in the States. Ubuntu LTS is gaining in popularity in corporate markets as well. So what to what does the mythical "year of Linux" even refer? Year of Desktop Linux? Ubuntu has already exploded on that front in my opinion. Year of the Linux embedded platform? I think the Nokia tablets and Zaurus devices have already proven that excellent devices can be made with an embedded versions of Linux (see Linux Devices for more). So whether it be servers, desktops, or embedded devices, Linux is already here and here to stay. If the true meaning for the declaration is "the year Linux over takes Microsoft Windows on the desktop", then that is a bold prediction. It will probably never happen in my lifetime, nor would I expect it too. Microsoft has already proven that a very lousy Operating System (Vista) did not create a mass exodus to Linux (as some experts predicted). Unless something changes, Linux will never overtake Windows on the desktop. You see, companies like Microsoft and Apple both have:

1. Marketing Budgets
2. Standards in terms of APIs, Libraries, Drivers
3. Major Software Vendor Backing

...and I could go on. All these points have been said many times before. So I will not elaborate anymore. Instead, I will turn myself into a cliché and report that:

2009 is the Year of Linux

...at least, it is for me. I had planned to make the switch in at the beginning of 2009. My former employer gave me some training vouchers on the way out the door (awfully nice of them to do so), and I decided to use the training for Linux+ certification. I figured studying for a Linux certification would be a good learning opportunity and a way to grow and formalize my Linux skills (and associated IT skills). I planned to blog about the experience (via this blog here) and I would truly be making the switch for at least a year. However, a couple of things happened that delayed everything:

1. I was waiting for the new version of Zenwalk (v6.0), which was delayed due to a delay in the launch of a new version of Xfce 4.6. Zenwalk 6.0 finally launched in March 2009.
2. I got a new job in February 2009, which has pretty much zapped all my time and resources since.

Fast forward to September and several items prompted me to get finally get started:

1. My training voucher expires on October 2nd, so I only have a month left to study for Linux+
2. My voucher is for XK0-002 (2005 Objectives), and I believe CompTIA is discontinuing this version at the end of this year in favor of XK0-003 (2009 Objectives).
3. Zenwalk released version 6.2 after a very lengthy Release Candidate period, most likely due to problematic items such as the new Grub2 release and legacy Intel driver issues (to name a couple).
4. I now have a spare target system for Linux. I managed to pick a $100 netbook on eBay (impulse buy). It's a very odd piece of hardware, but it should be fine for study (and travel) purposes.

What this all means, is that my posts over the next few weeks will cover

• Installing Zenwalk on very rare netbook: Cameron Maxmedia NB-1060
• A lot of Linux+ study information

After the one month rush, the true "Year of Linux" will begin.

About Walking with Zen - From Concept to Print

Ah yes, the first blog post. I have never kept a personal blog. I am at that age, where every one of my younger co-workers spends half of their day (at work) in the social network space or in the cloud. They blog, they update their status, they trackback, they twit, they have 1000 “friends” in their networks.

I just do not get it. I fail to understand the need to tell people what you are doing every minute of the day (like, the woman that twittered through giving birth). I have enough trouble keeping up with my offline friends. Having thousands of online friends seems like a lot of work, superficial, and pointless to me. I just don’t understand the draw and the aggressive need for acceptance of the social space.

At least I know about all the current trends. The older staff here just thinks a “twit” is a fool or an annoying person. I know about Web 2.0. I know about social networking, but something deep down inside me kept me away from it… as if it was in some way evil.

Fast forward to the fall of 2008… I found myself overpaid, with deprecated computer skills, and supporting legacy systems – not a good place to be. When the economy crashed, I got my walking papers. This was good thing. Stressful? Yes, but absolutely essential in reigniting my passion for technology and updating my skill sets. I also started walking and hiking more (walking helps me organize my thoughts).

My new technical journey all started with LinkedIn and some social job and career management websites. I was joining The Cloud. When I landed my new job, it would require me to keep a blog of everything I do, an announcement board for staff if you will. Other social sites would follow. Content management would be a target goal.

In additional to online skills, I also had to find a way to quickly update my skills from the command line, Novell servers (eDirectory), and Windows-only clients to Windows servers (Active Directory and Group Policy), Linux servers (and Samba), and multiple client operating systems. I purposely took a new job that has me doing everything IT related and more to force me into new technologies and out of my comfort zone. I manage the servers, desktops, network, website, phone system, security system, and even the electrical breakers and electronic timecard system. If it has electricity running through it, I manage it. I also have to write technology plans, educational course work for the staff and clients, and even grant and bid proposals. In all, if a computer-support discipline exists (whether it be purely technical or more managerial and administrative) , I will most likely have my hand in it. My friends called the career move "suicide". I call it a learning opportunity. Being a Jack-Of-All-Trades does have its advantages and disadvantages.

• Primary Advantages: flexibility; variety; chance to figure out what disciplines you enjoy
• Primary Disadvantages: being pulled in too many directions; staying organized; staying sane.

Because I have so many topics to learn for my career, I decided to narrow down my studies outside of work and really concentrate on my interests. These will be the primary subjects of this blog and include:

1) Linux
I’ve been an off-and-on user since 1995-6
More Specific: Slackware and its siblings: Zenwalk, Vector, Absolute, …
(Slack was my first distribution. You never forget your first)

2) IT Security
CompTIA’s Security+, which I have, planted the seeds of interest in this topic.
I am working on my CEH next.

3) Technical Training and Writing
Will eventually be a big part of my current job.
This blog will develop some of these skills as well.

4) Walking / Hiking
Hey, it's called "Walking with Zen" for a reason. Can you combine a tech blog with the outdoors? Can you combine digital and analog? I guess I will find out.

I may stray from time to time away these topics, but I will do my best to keep focus on the primary subjects. Hopefully, as I learn a thing or two, anyone reading will too. Thanks for stopping by.

Welcome to Walking with Zen: Technology. Philosophy. Slackware.
Born Analog. Raised Digital. Kicking and Screaming into the Cloud.
.