OStatic

I like hardware. When I say this, I don't mean gadgets (although I don't turn those away either), I mean hardware. I like to build desktops, from chassis to power supply to motherboard to peripheral cards. I like desktops because I feel I have control over what goes in them, as well as what I run on them. They're easy to upgrade -- and they seem far more durable than their lap-sized peers. I understand this is in no small part due to desktops not being portable.

I made a vow a few years back I would never use a laptop as my primary machine. This was before the advent of solid state disks and netbooks. But because portability has its appeal, Linux is my operating system of choice, and because it fit both the hardware and gadget criteria, I picked up an EeePC a year ago. It works well as a quick way to get online and do basic tasks. Who needs a full-sized notebook?

I do. I've found I need to break my vow, even, and use it as my primary work machine. It needs to run Linux, and what's more -- it needs to be pre-installed. Fortunately, I have options.

One of the most appealing aspects of open source software is its ability to be almost everything to everyone. Fine, it's hyperbole, but flexibility and access to the source code means a wide array of needs can be met for diverse groups of people -- and the shared efforts benefit developers as much as those who've never written (or seen) a line of code. I've been enamored of the Moblin Project for some time, simply because it so clearly demonstrates that open source can be as hands-on (or off) as you'd like.

On Monday, the Moblin team announced its Moblin V2 Core alpha release was available for testing. Tariq Shureih, Moblin Linux Engineering Manager, says that this alpha release is the first chance the wider community has to really kick the tires on a number of new features, including package and core interactions, the Fastboot feature, new user interface development and connectivity tools, and a new iteration of the Moblin Image Creator.

Hindsight, it is said, is always 20/20. The OLPC has traveled a turbulent path for quite some time, with its latest stumble coming in the form of drastic staff and development cuts.

Four years on, many are analyzing the choices the OLPC made, and some contend it was a bad idea all around.

In late December, the Linux Foundation announced its I'm Linux promotional video contest. Though early submissions were welcomed, today is the day the contest officially gets under way, sporting a new name and a panel of judges hailing from all over the open source and media industries.

Maybe it's personal bias, but a few of us here at OStatic have wondered why using open source software isn't a foregone conclusion in education.

The BBC has posted an insightful piece on open source software in schools that not only touches on the strengths of the software, but also the complexities of deploying it in these environments.

By way of PCWorld comes word that Mozilla Labs will release its Test Pilot project in the next few weeks.

Test Pilot's ultimate goal is to collect data on how people use Firefox (or any Mozilla based code, extension, or add-on). While word-of-mouth and formal usability testing are effective in developing and improving features, they rely heavily on getting the word out, getting feedback in return, and tend to attract the same demographic (power-users, or early adopters, for instance). Test Pilot aims to make usability testing private, open, and easy -- with participation and the type of information shared remaining completely optional and up to the user's discretion.

Bugs, system conflicts, and errant bits of code add unique challenges to the technical area of open source development. They also affect a project's community -- and as any community manager can tell you, developing a healthy community is often more difficult (and has higher stakes) than rogue code.

Management is tough all round, but managing open source projects is different still. Most developers are giving their time because the project interests them, and non-developers join because they find the project useful, and they want to share their enthusiasm. But a community not being any one remotely homogenous group means that passions sometimes run high, and it's not always easy to keep a project's community -- it's life -- moving forward.

It may have just gotten easier. Ubuntu's Community Manager, Jono Bacon, announced his upcoming book, The Art of Community will be available later this year.

The Marketcetera team is as aware as the rest of us that economic changes are coming fast and furious, and that open source software can have an impact on a company's -- or individual's -- financial future. Honestly, one could say Marketcetera is twice as aware of open source software's financial potential.

Today, Marketcetera released the first full production release of its open source automated trading platform. Aimed at hedge fund managers, traders, brokers and dealers, the system is standardized, open, scalable and modular. This, says Marketcetera CEO Graham Miller, offers users faster deployments, better integration, and the ability to customize everything from the public APIs to data models.

In its Quixotic effort to never admit defeat, the SCO Group has (again) filed reorganization plans in Bankruptcy Court. It has been a drawn out procedure that first raised the ire of the open source world, and later the courts, as appeals were filed and reorganization plans were rejected and resubmitted.

The SCO Group is giving it another shot, this time proposing it can rise from its ashes by selling some of its assets in a public auction. The money made from these sales will contribute to the operating budget and fund the OpenServer product line and mobile business development. A press release issued earlier says that there are several investment groups interested in a sell off, but even so, they'll still have to wait to hear what the Court says.

By way of Heise Online comes word that Tor, the internet anonymization system, has hit a milestone of nearly mythical proportions -- there are currently no known bugs in the Tor code.

Tor began scanning its development releases in September using Coverity, a bug detection application developed by Stanford University in collaboration with the US Department of Homeland Security. In September, Coverity revealed 171 issues in Tor's code base, ranging from annoying yet not critical sloppiness to bugs capable of causing crashes that would prove challenging to debug. By December, Tor had lowered this number to 15, and last week, Coverity testing revealed that the project had successfully eliminated the last known issues.