God Wants Me Dead?

According to this convoluted blog, the Almighty is greatly enraged with the fact that I use Linux.

These Microsoft fanboys are something else. Now they are pretending to be Christian bloggers. You Windows lovers have no shame!

IT Conservative: WTF?

You know, there are moonbats in the IT field too. Besides the obvious, such as Steve Ballmer, Paul Thurrott, and John Dvorak, there is also IT Conservative.

To be fair, he shares my disdain for Richard Stallman’s antics. That’s great. I’ll buy him a beer on account of that. Unfortunately, he goes from being pragmatic to just plain bizarre in claiming:

• Open Source software is the work of Communists hell-bent on destroying capitalism.
• The OpenMoko phone is somehow related to the threat of global terrorism and September 11th.
• DRM-free music is somehow a bad thing

Besides these bizarre claims, this guy’s blog is riddled with glaring factual errors. For example, he claims that Linus Torvalds named Linux after himself. (Anyone who knows anything about the history of Linux knows that it was a friend of Torvalds who provided the name Linux by creating a /linux/ directory on his FTP server to upload the Linux source code- Torvalds had intended to call the project Freax.)

His site is a great read if you are looking for some tongue in cheek humor or the occasional honest and fair criticism of certain aspects of Linux/Free Software. On the other hand, this gentleman is a Redmond fanboy who hates more programming languages than he likes, believes Open Source software is a Communist conspiracy, and believes Microsoft is here to save us.

Fundamentalism is scary and bad and should be fought. We’re in war in Iraq and Afghanistan right now to hunt down fundamentalists. We should do the same in our own country. Because, as I said many times before, the real threatening fundamentalists are the ones that are killing the very foundation our country is built on: capitalism. Giving into the freetard fundamentalists is going back to communism. Is that what we want?

I hope he has been declared mentally incompetent and thus ineligible to own a firearm.

Seven Things I Hate About Linux

Linux is great technology. It clearly brings to the table a great number of technological advantages at a more economical price than proprietary technologies. It’s Open Source foundation is great for developers and the licensing quite generous. On the other hand, there are several things that I hate about Linux.

1. Too much energy spent on “evangelizing.” and not development. Yes, talk about the capabilities of Linux. Of course you want to make a pitch to organizations in order to convince them to adopt Linux. At the end of the day though, those who will end up using Linux and staying with Linux will be those for whom the underlying technology proves to be superior.

2. Piecemeal documentation. Yes there are sites like the Linux Documentation Project. Yes you can use Google to see if Linux supports the Acme12345XT Motherboard. But with the exception of Redhat and SuSE’s HCLs, it is difficult to find comprehensive, accessible, and up-to-date information on hardware compatibility in an easy to navigate format.

3.  Mediochre GUIs. Not that GNOME or KDE are bad by any means, but when compared to Apple’s Aqua interface, its like viewing a paint-by-numbers  next to a Rembrandt. There is little polish to the interfaces of either. GNOME, as Linus Torvalds himself has said, makes the assumption the user is dumb. KDE has the appearance of a interface that has been cobbled together and lacks any sense of unity. When I first install a distribution, the first thing I do is tweak KDE to something I can tolerate looking at for an extended period of time.

4. Tries to fill too many niches. Whether it is desktop Linux, embedded Linux, server Linux, real-time Linux, or whatever Linux we are talking about, the Linux community has tries to shove Linux in anything with a hole. In trying to fill every conceivable niche, Linux has become a jack of all trades, master of none.

5. Asinine release schedules. Whether it is releasing version 2.3.1 of a distribution a month after the 2.3.0 version, or waiting three years between releases, I always hated this aspect of Linux. At least with Microsoft and Apple, there was an appreciable amount of time between major releases. Six months between Ubuntu or Fedora releases is dumb.

6. Richard Stallman. The Linux community largely ignores him, but they need to denounce him and all of the Stallmanites who have taken up residence in the community. He is unwilling to work with vendors except on his terms and he has totally mischaracterized Trusted Computing. The Linux Community needs to tell him to have a cup of shut the hell up.

7. Poor quality control versus BSD. OpenBSD in particular. With OpenBSD only having two remote exploits in a default install in ten years, this should be something Linux developers are aiming for. Don’t rush out a kernel update just to include some shiny new feature that less than ten percent of the community will use. Sure, Real Time Linux would be great, but if the code is buggy, that just means you can be owned in Real Time.

High Ho, High Ho, Off to Bankruptcy Court for SCO!

Beautiful news.

But let me temper my glee a little bit. Darl McBride clearly misled his employees and investors into thinking SCO had a chance in court against Novell and IBM. These people were loyal to him and were more than willing to put their livelihoods into his hands. All that they wanted, I am sure, was to produce quality software and reap the fruits of their labors.

Novell has an opportunity to show its generosity by inviting the best of SCO’s software development team, whatever remains of it, to come aboard. Red Hat would also be well served in showing some magnanimity. The responsibility of SCO’s financial ruin rests on McBride’s shoulders. He pursued his quixotic campaign against Linux so its nasty end, but what many forget is that McBride wasn’t just going after Linux. He threatened Apple. BSD. McBride even had the chutzpah to threaten Microsoft.

At the time, McBride could do nothing more than congratulate himself for his scheme of selling snake oil UNIX licenses to Linux users for $199 a pop for workstations and $699 for servers. Now he is probably busy looking for cardboard boxes to pack up his office with. Surely SCOs boardmembers have not taken the Chapter 11 filing laying down…

Jacksonville (but Raleigh first) or Bust!

I love roadtrips. Especially ones that involve a stop at a major Linux company. A month ago I filled out a feedback request form on Red Hat’s website to see if it would be possible to visit the Raleigh headquarters, as I was planning a trip to Jacksonville, North Carolina and Raleigh is on the way. To my pleasant surprise, I received a very quick and courteous response from a senior project manager (who would have thought that I’d get a response from anyone save an intern?) who indicated that it was very much possible and should I ever wish to submit my resume, he would be more than happy to take it.

It seemed that due to my wife’s work schedule, our trip to Jacksonville (and thus Red Hat headquarters) would not be possible, but it would seem that I will be able to go after all. With that in mind, I am looking forward to the trip.

I need to get away for a little bit. Between school, my internship, and work, I haven’t had any time to really unwind. I am debating leaving my laptop here in Indiana so I can take a break from that. At the same time, I know I am going to want to post the pictures from my roadtrip as soon as I upload them.

I’m hoping to be able to head down to Wilmington and see the USS North Carolina again. During my last visit, I didn’t have the time to see the entire ship, and I’m hoping to see the upper platforms of dreadnought.

Darl McBride, Go to Hell

This is the photo of a man in denial. If the board members at SCO have any sense, which I doubt, because they allowed this buffoon to call himself a CEO while growing the company only through the threat of lawsuits, they will can his ass fast.

Dear Darl,

You lost. SCO should reorganize as a fast food chain or a used car dealership ASAP. Even then someone would be a moron to buy anything called stock from you. Nyah, nyah!

Sincerely,

The Linux and computing community in general.

Why Windows 98 Matters Even Though It Shouldn’t

Although this is only my second week as an intern for the school system, I have already made several observations regarding the infrastructure behind the computer systems as well as the deployment of the same. Some of these observations I will keep to myself, as they concern best practices and I am not at liberty to publicly advertise any information about systems that will result in a security compromise of any kind.

With that said, I will make some remarks regarding my observation regarding the continued deployment of Windows 98 on many of the systems in the district. As support for Windows 98 ended on July 11, 2006, relying on this operating system for production systems has it’s drawbacks.

Why not upgrade to Windows 2000?

Windows 2000, officially known as NT 5.0, was intended for deployment in the enterprise environment. It was originally hoped that Windows Millennium  edition would serve as an able successor to 98 for the average user. Unfortunately, Millennium was 98-improved-worse and the costs of Windows 2000 often prohibitive for mass deployment by small school districts like the one I work for. The result was a reliance upon an operating system (Windows 98) that continued even after the support cycle was terminated.

Windows XP was meant to fill the enterprise, home, and educational markets and did so quite well.  Unfortunately, a lot of systems deployed by school districts could not support XP. Forced to rely on Windows 98, it was a matter of time before the other foot dropped.

Finding software that still runs on Windows 98 is becoming increasingly difficult. Internet Explorer 7, the most recent version of Microsoft’s flagship web browser will only run on Windows XP SP2 and above. Office XP was the last version of Office to support Windows 98. Without applications that support it, Windows 98 is becoming quickly obsolete. Yet we are servicing machines that short of resurrecting them of Linux boxes, have no future beyond Windows 98.

Ordering replacement hardware components for these machines is becoming an issue as well. Last week I replaced an old NIC in a Win 9x box. When I stopped at our local supplier to pick up the new NIC, I happened to read the system requirements for this card to work.

Windows 2000 SP4 and up.

Do not pass GO. Do not collect $200.

So what did we end up doing? We cannibalized a machine for parts and Googled for the driver. We lucked out and found a working NIC and the driver. It saved the district some money on a NIC, but what will happen when we can no longer cannibalize parts from other machines?

There are very few attractive solutions to this problem. Sure, the district can spend money on brand new PC’s that are preloaded with Windows Vista. Then we have no guarantees  that we won’t run into the same problems with software and hardware compatibility. The only difference this time is that it would be a forwards compatibility issue.

We could standardize on Windows XP Professional SP2/SP3 and hope

1) SP3 does not break apps in quantity like SP2
2) SP3 is not the end of the line for XP Pro after our investment.

We could also refurbish our current hardware with Linux and pretend that the learning curve and the lack of some niche applications for the educational market for Linux are non-issues.

Despite the technological advantages of Linux and the ability to deploy an operating system that still supports Pentium II and low-clock speed Pentium III machines, this is not an option. The number of calls we would get when a user tried to do something that required root access to do and the obligatory explanation of why we “can’t just give out the root password” alone makes this unfeasible.

The final option, despite the expense, would in the end be most satisfactory:

Buy a bunch of damn Macs. 

The End of the Road for SCO

With the price of their stock at a miserable 45 cents per share, and Judge Dale A. Kimball’s ruling that SCO never owned the UNIX copyrights it has tried to use against IBM, Red Hat, and Novell, I think it is safe to say that SCO is headed for insolvency.

SCO brought this on themselves by initially filing a lawsuit against IBM in March of 2003 that they should have known they couldn’t win. After failing to intimidate IBM, they were then sued by Red Hat (the decision of this lawsuit is on stay until the final resolution of SCO vs. IBM).

SCO was unable to articulate what it claimed IBM had stolen from them, though claiming they had “millions of lines of code” as evidence.

The most recent ruling on August 12th, recognizes Novell as the true UNIX copyright holder, not SCO. As SCO was merely a licensee of the UNIX copyright, it owes Novell 95 percent of all UNIX revenue that it earned from licensing the UNIX copyright to Sun and Microsoft, minus a five percent cut for “administrative costs”.

With this in mind, investors have been bailing from SCO like a sinking barge and in the unlikely event there is anything left of SCO when Novell collects, the legal expenses of Red Hat vs. SCO will finish off Darl McBride’s quixotic campaign.

This is what happens when you try to threaten companies with lawsuits that have no merit. Now SCO has no product to sell. No Linux. No UNIX. (While they theoretically could still distribute Linux if they did so in accordance to the terms of the GPL, no one, after seeing the mess SCO made when trying to leverage their alleged IP against the community, would buy from them. Without the copyright to UNIX which they once claimed to have, they are done.

Red Hat Global Desktop

 Remember when you could buy Red Hat Linux as a shrink-wrapped box at your local Best Buy? Not since Red Hat Linux 9 (2003) has a Red Hat operating system been available as a boxed media set that could be purchased from a retailer.

Well that hasn’t exactly changed, although the month of September will make the Red Hat Global Desktop available, which fills in a niche that Red Hat had long neglected.

While Red Hat Enterprise Linux 4.5 and 5 are targeted at large enterprise customers in need of high level security, smaller customers who found the price prohibitive were forced to use rebranded versions of Red Hat such as White Box or CentOS. The only other alternative was to use the Fedora brand of Red Hat’s operating system, which although featuring bleeding edge code, suffered from the relative immaturity of the code.

While it will only be available through Intel white-box channels, it will certainly be an imrpovement over Red Hat’s current distribution strategy. Furthermore, Red Hat is in negotiations with Microsoft to make available codec such as WMF fo various media formats.

From my point of view, this could be an excellent opportunity for Red Hat. The customer will benefit from having a machine with Red Hat Linux that doesn’t require the often difficult to axquire cooperation of manufacturers such as Dell or HP. More than that, the customer will have a machine that comes with support, that is quite often murky where Linux on x86 is concerned.

While the shrink boxed Red Hat Linux has not returned, we can take at least some assurance from Global Desktop that Raleigh has not forgotten about the little guy.

Honest Talk About Linux on the PC

APCMAG.com has an excellent interview with Con Kolivas, who recently stepped down as a kernel developer due to disagreements over the state of Linux’s CPU scheduler.

Kolivas also touches on a far more broad-reaching issue, the poor state of the PC in general and how the Golden Age of PCs, reached in the late 80′s, has given way to a dearth in not only variety among PC manufacturers, but innovation. The PC architecture as it is today, he argues, is the result of the dominance of inferior technology that has been propagated due to a philosophy that hardware isn’t marketable and that sales should be operating system driven.

I have to say that I feel much the same way as he does in this regard. While I don’t remember the Amiga, I do remember the Classic Macintosh OS. Most everyone who knows me knows how I rave about these little machines, so suffice to say, the experience on a Classic Mac running at 16, 25, or 33Mhz was better than that afforded by today’s dual-core, 64-bit processors.

While using a system that had little or no memory protection could sometimes cause problems, more often or not, the memory protection was something that the average desktop user could afford to be without. Usually a restart would fix any problems and it taught the user to adopt good computing habits (save often, backup your files, etc.)

With that in mind, a good question that few in the Linux community seem to be asking is, “In light of the fact that the x86 architecture is now almost 30 years old, how are our efforts towards improving the users overall experience bearing fruit?”

It’s a question I hope Linus Torvalds, Andrew Morton, and Co. are willing to answer honestly.