Valtteri Halla, the Nokia member of the MeeGo Technical Steering Group recently attempted to allay the concerns of the open source community by underscoring the importance of openness in the development of the MeeGo operating system.

In a blog post last week, the veteran Nokia executive who has been working to get Linux onto Nokia phones since 2000, announced that the MeeGo repository will be opened by the end of this month. The source and binary repo will provide a raw baseline for building MeeGo on the Nokia N900 and Intel Atom-based netbooks.

What’s more interesting is the flurry of debate surrounding the details – or lack thereof – related to the processes that will enable the open source development model to flourish around the MeeGo platform. So far, Halla has said little about the decisions made by the technical steering group to choose RPM rather than deb as the package manager for MeeGo:

While code is certainly the most important question, the most frequently asked, however, has been about technology selections. The big ticket items like Qt, OBS and RPM were already communicated at the launch and as we expected, kicked off a few small avalanches of debate! These selections were, of course, pre-agreed and I can assure you that the amount of effort spent in resolving these was not small. After all, these are the points driving most of the investment cost and transition pains for Nokia, Intel and the Moblin and Maemo communities. Further selections are mostly still under discussion and beyond a few obvious ones (X, connman, ofono, gstreamer, dbus,…) can be considered as working assumptions for MeeGo 1 release. Now that the internal responsibilities within Intel and Nokia are becoming clear I expect that the people behind these selections and assumptions will start appearing in meego.com pages, mailing lists and wikis during the coming days.

This has generated some concerns that developers are being left out of the conversation, thus increasing their skepticism of the purported “openness” of the MeeGo platform. To be fair, it was necessary for the steering group to make some hard decisions at the beginning of a big open source initiative that merges two existing platforms. Otherwise, we could argue till the cows come home and nothing will move.

But what is the process that governs these decisions? Are we talking about a process similar to JCP (Java Community Process), where there are clearly defined procedures for the development and revision of the Java’s technology specifications?

Other tough questions that need to be addressed: driver support from hardware and device manufacturers, DRM support that is compatible with operators’ content business and application support for potentially different variants of MeeGo that could emerge. The technical steering group should address these issues early on if it wants to seed a rich developer ecosystem that is crucial for the success of any mobile platform.

Linux users are often at the mercy of hardware vendors when it comes to device drivers. The open source community often needs to turn to reverse engineering to churn out drivers from proprietary ones. As long as the majority of hardware is made for Windows and OS X machines, Linux users will need to wait until the community figures out the nuts and the bolts of a piece of hardware before a Linux driver can be written.

Take Nouveau for example. The open source project started in 2006 with the aim of building high quality drivers for Nvidia graphics cards. Although Nvidia provided a Linux driver several years ago, it was a basic driver with no 3D support. The Nouveau project gained momentum and a year later, its driver soon outperformed Nvidia’s in 2D performance.

While some Linux drivers can be as good as proprietary ones from hardware vendors, others only allow basic functionality with sometimes abysmal performance. My interest in Linux was rekindled recently when I installed Ubuntu 9.10 on my three-year-old Macbook. The basic hardware like the keyboard and graphics worked right out of the box after the installation, with the exception of the iSight webcam which only worked after I installed a software that reverse engineered the right driver out of Apple’s proprietary iSight driver.

Driven by commercial interest through a stranglehold over the unique features of its hardware, it is not in Apple’s interest or any hardware vendor to release open source drivers. A recent feature in Linux Magazine on the Nouveau project pointed out that Nvidia “still gains far too much advantage by keeping their driver closed. They get support for brand new models, extra performance, better power management, extra features like VDPAU, and certain technology components can remain a company secret”.

Meanwhile, Linux users have to continue tinkering with their boxes to make things work, but with the support of the community through hundreds of thousands of forums and interest groups on almost every Linux distro, there’s bound to be a workaround out there. But hey, it’s also what makes computing fun isn’t it?

By now, few people in the IT industry would dispute the benefits that the open source development model brings to businesses: flexibility and to some extent security, though it isn’t always the cheaper option if you do your TCO (total cost of ownership) sums.

The best open source software, notably the LAMP (Linux, Apache, MySQL and PHP) stack, is commonly used by enterprises large and small, with many businesses running mission-critical applications such as sales transactions on Linux.

Yet, the merits of open source software should not be judged based on its “open source” label. While there are a handful of stars like the LAMP stack, there are thousands out there that will never gain the prominence of Linux and remain difficult to use and maintain.

Last week, the open source community was abuzz with conversations on Microsoft’s contribution of 20,000 lines of code to the Linux kernel. While some are skeptical over the software giant’s move to get its code into the Linux kernel, others, including Linus Torvalds, welcomed Microsoft’s contribution just like the code from anyone else in the community.

The code, which includes three Linux device drivers, will enhance the performance of the Linux operating system when virtualized on Windows Server 2008 Hyper-V or Windows Server 2008 R2 Hyper-V. On July 20, the code was been submitted by Microsoft to the Linux kernel community for inclusion in the Linux tree.