In October, MSI's Director of US Sales delivered an interesting statistic that Linux netbooks were returned four times more often than Windows versions. It didn't seem, perhaps, an unreasonable number, but it was a bit ambiguous what data it was pulled from. I had speculated it was perhaps a market-wide number, pulled from other netbook manufacturers (and incorporating MSI's sales data on Linux netbooks internationally, as a Linux version of the Wind has not yet been released in the US).

On Monday last week, you saw us continue to convert that brand awareness to revenue - with the introduction of a full line of MySQL optimized systems. By our estimates, there are about 11,000,000 MySQL users on earth - our new systems can triple their application performance. So we've made free evaluation units available to MySQL users (via our Try and Buy programs). Click the image to the right to listen to Marten Mickos and John Fowler talk about the opportunities ahead.

I remember taking an operating systems class in college and marveling at the fact that operating system design seemed less about elegant engineering and more about [what I viewed at the time as] performance hacks. I saw a similar sentiment recently captured by Eric Florenzano in his post It's Caches All the Way Down where he starts describing how a computer works to a friend and ends up talking about the various layers of caching from CPU registers to L2 caches to RAM and so on.

Well, I am a programmer so here’s my take. There are two main issues: a simple problem with describing a set of CPUs to run on, and a more complicated problem with lock contention. Let’s look at the simple one first.

It was possible to use generic thunks in 16-bit code to allow it to call into 32-bit code. Why can't we do the same thing to allow 32-bit code to call 64-bit code? It's the address space.

Described as a multicore-friendly, lightweight parallel framework, fork/join uses the strategy of recursively splitting a task into smaller subtasks; forking the subtasks into separate processes or threads, so that they run in parallel on multiple cores; and joining all subtasks to compose a result to return. Expected to be added to Java 7, it will boost the ease of writing parallel programs in Java.

Today Apple released the new Macbook and Macbook Pro lines of laptops. They are nice, but seem to miss something fundamental for anyone sitting for more than 5 minutes looking at their laptops: matte screens. "Not only every smudge or reflection draws immediate attention, but when pair programming you're constantly face-to-face with your fellow software developers."

JCard kit is a solution for Java Card development, supporting developers using Eclipse or NetBeans with supported smart cards or simplely using a card simulator.

In the not-too-distant past, ENIAC was programmed with switches and a plugboard. Stored program computers soon followed that allowed one to write a program, load it into the computer memory, and run it. Initially, those programs had to be written in or manually translated into binary machine code, but soon assembly languages and assemblers were developed to simplify the process.

A little while ago I was asked to make a small circuit based around a microcontroller and a Digital to Analog Converter that would output various sounds corresponding to different stimuli. These stimuli would be commands received by the microcontroller via UART, and the output sounds would simulate a switch, or a closed door, or other ambient sounds from a working environment

Apple announced this morning via a posting on the front page of its iPhone developer site that it has decided to discontinue the non-disclosure agreement preventing developers from discussing iPhone programming.

When you get successful, you may find that you move from someone else looking after your hosting, to running a complete rack (or racks) yourself. Standing with that shiny new rack towering over you, there seems so much space and flexibility, and anything not quite right now can be fixed later, right?

Wondering how threading thrives in the real-world? The enormous and highly threaded code base at SAS is an example of an application that benefits from the spelunking and debugging the Intel Thread Checker and Thread Profiler tools offer.

Reasoning why and why not to use multi threading. Pretty informative.

University of Rochester researchers have developed the first true 3-D processor and it runs today at 1.4 GHz. Previous attempts to build 3-D chips simply stacked identical processors on the top of one another. On the contrary, the new 3-D chip, dubbed the ‘Rochester Cube,’ was specifically designed to optimize all key processing functions vertically. And each layer could have a different function. For example, this kind of 3-D processor could have a layer dedicated to conversion of an MP3 file and another one to provide information about light to your digital camera.

In a session at its Google Developer day in London this morning, Google executives demonstrated a working Android mobile phone for "the first time in Europe."

Using better tools is good. In this post I want to take a look at a better workbench to use these tools, the machine itself. XP or Vista are not bad, far better than the Windows 98 days where failing to set a property would be enough to plunge Delphi and the complete machine into a blue screen. But there is still room enough for improvement.

Hardware for virtual reality and virtual worlds

Products like the Kindle from Amazon.com and Sony Reader are becoming more readily available, cheaper and more popular. I will never buy one and if you gave me one as a gift I would likely prop up a shorter table leg. If you want to know what I find so repellant about this technology, read this.

Hello, my name is Sam Ceccola, I have been involved with use of Java technologies in the industry and the government communities since 1998. For me this started with Bluestone software (later purchased by HP), then with BEA and now I have the privilege of joining Azul Systems (this is my first blog entry for them). At Azul we have been working with an increasing number of military customers, which in-and-of itself is a testament to what we see as growing use of Java within the defense community. One point to add to this discussion is in terms of overall infrastructure that is available – that’s where we see a significant gap between Ada and Java. A couple examples that highlight this are the availability of programmers over time, and the number of COTS solutions available to fine-tune Java applications. Over time we see the pool of well-trained developers shrinking for Ada, versus continued growth for Java. This is matched by the vendor and tool base situation – in 2003 Venture Development Corporation (VDC) estimated that the market would be flat for Ada at $49M/year, as opposed to Java’s growth from $1B/year in 2003 to over $1.5B in 2007.

My home office has just about everything needed for a productive development environment. Its stocked with the essentials (test server, quality monitor, good lighting), a few "nice-to-haves" (Wacom tablet, comfy chair) and a some fluf (sound system, Betamax and VCR). It has at least one item I would rather do without (phone). When Data Robotics announced the new Firewire800 Drobo, the time was right to fill a home office void: reliable versioning and backups.

Follow the illustrated step-by-step assembly of a liquid-cooled, 8-core development machine based on the Intel SkullTrail motherboard, blazing fast Intel QX9775 Core 2 Extreme processor, and Adaptec RAID controller with 3 10K Velociraptor hard drives. The assembly is preceded by the goals for the system, as well as cost-benefit analysis and justification for spending $5,000 - $6,000 on a development machine. Full parts list and pricing, and a brief photo tour of the home office where it all took place.

Software running -- with no process separation -- on a BAE Systems RAD6000 single board computer that's configured with one 33 MHz RISC CPU and 128 MB of RAM.

If you've got a few servers you are responsible for, check out these tips to keep yourself from accidentally causing problems. Good stuff!

Khronos Group has announced the release of the OpenGL 3.0 API specification and the GLSL 1.30 shading language specification.