For this round-up we had used a Dell Inspiron 1525 notebook (PM965 + ICH8M Chipset) with an Intel Core 2 Duo T5800 processor clocked at 2.0GHz, 3GB of DDR2 memory, 250GB Hitachi HTS543225L9A300 HDD, integrated Intel 965 graphics, and a screen resolution of 1280 x 800. On the Windows side we were using Windows Vista Home Premium SP1 while with Ubuntu we were using Ubuntu 8.10 and the stock packages (Linux 2.6.27 kernel, X Server 1.5, etc). Each OS was left to its default settings, including the use of the standard desktop effects.

Larabel’s graphs and detailed technical specifications and analysis tell the full story, but the good news to Linux fans is that Ubuntu was a clear winner, and OpenJDK seemed to hold its own compared to Vista.

Well, Java on Ubuntu was pretty much the hands-down winner compared to Microsoft Windows Vista Premium SP1. Running the Java tests on Ubuntu had experienced significant advantages when it came to file encryption, Fast Fourier Transforms, Successive Over Relaxation, Monte Carlo, and the composite Java SciMark performance. In only the Sunflow test were the results between Ubuntu and Windows even close. With the Java 2D Microbenchmark, Windows was faster but that likely falls on the Intel Linux graphics driver having little in the way of performance optimizations and Java on Linux not yet utilizing the X Render extension.

Comparing Sun’s Java and OpenJDK / IcedTea on Ubuntu had roughly the same performance between the two except for a few areas (FFT and Monte Carlo) where the official JVM was noticeably faster.

Last time we reported on these tests and it appeared that Windows had the lead, you responded. Larabel’s article includes a discussion link, and a brief review of some of the comments reveals that some Windows fans don’t like the results.

Linux got owned where it matters the most: Graphics.

But RealNC’s comment about graphics doesn’t quite align with a recent article reporting the results of the 2008 Linux Graphics Survey compiled by Phoronix.

What do you think? Are Linux graphics lacking? Are you surprised by the Java VM performance?

Microsoft used last week’s ApacheCon as a platform to reach out to the open source community in a public way.

In his keynote last Friday, Sam Ramji, Microsoft’s senior director of platform strategy, told the Apache faithful that Microsoft is serious about partnering with the open source community to create open standards and interoperability. Collaboration on the fundamentals will promote healthy growth, competition and a new round of innovation and will enable customers to allocate IT dollars for constructive uses instead of overcoming infrastructure bottlenecks.

The most recent example of Microsoft’s collaboration with its open source counterparts was its recent decision to join the Advanced Message Queuing Protocol (AMQP) Working Group for improving message interoperability at the application level, which is currently very difficult without expensive proprietary solutions. But Microsoft is also boosting interoperability with open source in Web services, security, databases and network monitoring, Ramji said. This past spring, Microsoft reached an interoperability “milestone” between Soap and Apache’s Axis Web services protocols, he added.

“I’m an eternal optimist. I’d have to be after three years of leading open source at Microsoft,” Ramji said. “There’s been a big shift in a short period of time,” involving hundreds of steps in a company with 93,000 employees, he said.

Ramji’s embrace of open source was echoed, if somewhat less strongly, in a speech last Friday in Australia by Microsoft CEO Steve Ballmer and reported in a CNET blog by Matt Asay. In response to a question about its Internet Explorer browser, Ballmer said Microsoft is unlikely to make Explorer open source because of its proprietary extensions, but he didn’t reject the suggestion out of hand. The measured tone of Ballmer’s response, Asay wrote, “could well be the most rational, pragmatic, open-source-related comment from Ballmer that I’ve ever read.”

Ballmer’s comment suggests that Microsoft has finally recognized that open source can be a useful component of its overall software strategy, Asay concluded.

In other words, Microsoft may finally have decided to stop fighting open source and instead begin to find areas where the two communities can help each other. And that’s a good thing.

According to an updated Linux Foundation study, to build from scratch today, the Linux kernel would cost $1.4 billion; a typical Linux distro, $1.2 billion. In addition, Fedora 9, the current community version of Red Hat’s operating system, would cost a whopping $10.8 billion to replicate in current dollars.

The study also quoted a report from Framingham, Mass.-based IDC that appraised the collective value of the entire Linux computing ecosystem at $25 billion. That’s quite a trajectory for Linux Torvalds’ kernel in just 17 years.

The conclusion underlines the obvious: Linux has become a computing powerhouse, running everything from tiny mobile devices to the largest banks and supercomputers. While the software’s open code and modular construction are inherent advantages, the massive Linux community of individual and corporate developers who share the task and cost of improving the software are key to the platform’s success, the report concludes. In contrast, proprietary software companies, which must shoulder their development costs in isolation, will ultimately be hard put to compete with the open source model, the report concludes.

No kidding. As far as this blog is concerned, the report and its conclusions preach to the converted.

According to a ZDNet.co.uk article, Whitehurst told a Singapore business forum that companies using open source software address their major software pain points right away and can then share the results quickly with the rest of the community. Proprietary software companies, on the other hand, are slowed down by the need to solicit user feedback and then fix the problems through a top-down, planned development cycle. That model, proclaimed Whitehurst, is on the decline, he said.

Look at the recording industry, Whitehurst said. They spend far less to market American Idol winners, whose appeal has been proven up front than they do on the multimillion-dollar marketing campaigns for Britney Spears that lack a similar advance-popularity litmus test, Whitehurst said.

Whitehurst’s analogy couldn’t be clearer (which assumes that the typical Asian businessman is familiar with American Idol and is old enough to remember records). Anybody else still got any old 45s kicking around?

A SearchEnterpriseLinux reader from Queensland, Australia, wrote that my recent article asking whether open source is really cheaper than proprietary software missed the point.

“Most users are locked into the concept of Microsoft and this is the problem,” wrote Trevor Hughes. “For Mr. Average, Linux has much to recommend it. If I can do it anyone can. I am 57 and a newcomer to computing. For Internet communications and normal use, Windows is NOT worth the money.”

Hughes really takes issue with the article’s conclusion that users require more technical expertise and a certain openness to risk-taking to maximize the financial savings of open source. However, the article actually addresses Linux on servers in the data center, not Linux on home desktops.

Nevertheless, Hughes expresses the positive, can-do, problem-solving attitude that would lead to successful operation of Linux in the data center — and Windows, too, for that matter.

It’s pretty obvious that Linux can be a big money-saver over Windows. Why else would so many big corporations make the switch? It’s like anything else. There are always trade-offs, and companies and individuals are able to vote with their dollars on what they want to do. And choice is always a good thing.

In the meantime, however, U.K.-based TechRadar.com recently published Linux: The Girlfriend Test, and a tough test it was indeed. The goal: to find out if a first-time Linux user, presumably a female college student, could accomplish nine familiar Windows tasks using Fedora 9, the community version of Red Hat.

Here are the tasks and how she fared on each one:
1. Bookmark a website in Firefox. No problem.
2. Write and print a letter in OpenOffice. The first part was easy but the letter wouldn’t print and no error message appeared with a reason or resolution.
3. Rip a CD. Task accomplished. But Fedora failed to identify all the output options.
4. Send an instant message. After several unsuccessful attempts, she succeeded by going to msn.com and inputting user data from her Windows Live Messenger account.
5. Create a pie chart in OpenOffice. No problem.
6. Transfer the ripped CD to her iPod. Attempt failed because of a protocol problem. Again, no error message appeared to identify or fix the difficulty.
7. Move a photo of her head onto a photo of her boyfriend’s body using Photoshop. Easy.
8. Watch a video on YouTube. Failed because Firefox was unable to install Flash player due to a malformed file. There was no work-around explanation.
9. Make an international phone call using Skype. Application installation was successful but audio playback problems prevented communication.

The writer concluded that Linux needs to do more with wizards and pop-up instructions to help new users without a technical background successfully transition from Windows or Macintosh to the Linux desktop.

I agree. But I have to say it was a pretty tough test, and the writer never mentioned how he fared in the “Boyfriend Shopping Test” that was his part of the bargain. Inquiring readers want to know.

Microsoft’s SUSE certificate sales, in fact, have grown faster than expected, exceeding $157 million in the first 18 months. Novell, in turn, agreed to boost its investment in tools and customer training to help make the two systems more interoperable and user-friendly at the joint Microsoft/Novell research facility in Cambridge, Mass.

Justin Steinman, Novell’s director of product marketing for Linux and Open Source, said SUSE Linux core sales have continued to grow rapidly and the Microsoft certificate sales represent just another channel to get into the market.

Al Gillen, research vice president of system software with Framingham, Mass.-based International Data Corp., said that the extension is “positive for both companies” and reassures users that the pact “isn’t going to run out of gas” before its scheduled termination.“The market for nonsupported Linux is strong, and Microsoft is trying to penetrate that market with these certificates,” Gillen added. “Microsoft doesn’t want to compete against free Linux software, and by selling SUSE support certificates, Microsoft creates a level playing field.”

Although open source is touted as a lower-cost alternative to proprietary software, Gillen said the cost difference is insignificant when comparing the tab for acquisition and support over a five-year period.

Since when has Microsoft been in favor of choice? Like the choice of eat or be eaten. . . .

How can Microsoft be so kludgy as to create an operating system so bad that users race back to the previous OS (Windows XP), and Microsoft’s response is to discontinue support for XP? she asked.

“James [her husband] is learning to use Ubuntu. Actually he’s still learning to install it,” our friend writes. “So far, it recognizes that we have a printer but not that it ought to run it.”

As James struggles to learn Ubuntu, he keeps reminding himself that Vista “is a flaccid, overstuffed OS that has crashed in the middle of every task he’s tried to do. [Microsoft] will be lucky if the whole planet doesn’t go Mac and Linux.”

I’ll bet there are many Penguin fans out there who would be happy to help our North Carolina friend. Anyone?

Corporate software vendors like Microsoft have been all over the branding game for a while, but open source vendors have been slower to embrace dogmatic self-promotion.
There is still a gap between IT consumers’ perceptions of open source software and the ideological fuel driving the work of its developers. Corporations are hesitant to weaken their brand appeal by associating their software with open source.

As pointed out in a recent article on customer relationship management software branding, the danger for open source in this environment is that it’s a buyer’s, rather than a seller’s, market. The saying “If you build it, they will come,” doesn’t hold anymore, says Yegge.

Open source’s next big move will not be the decision on whether to go commercial. Rather, it will be its ability to produce an identity that buyers — ideological and commercial — can brand as a positive thing.