Could Haiku offer BeOS a future?

When Be closed it’s doors back in 2001 I thought the future for a promising operating system (OS) looked very bleak as without the creator and the main driving force behind the OS it would surely die within a year or two. It didn’t take long for the BeOS community to announce some Open Source efforts to recreate the OS from scratch or using bits of other OSes like Linux and though they seemed to be the logical step forward to save the OS from certain death it seemed a tall order to recreate 10 years of OS development. Many promising OSes have come and gone over the years, what would stop a BeOS clone suffering the same fate of grinding to a halt at some point?

The first problem I saw was fragmentation, at least 3 different projects appeared with the promise of recreating BeOS but OpenBeOS was the most prominent, this was later renamed Haiku for legal reasons. While other projects never seemed to make it off the ground Haiku did start to gain some traction by offering drop in replacements for various parts of the existing free release of BeOS R5. This seemed to be a smart move, recreating an OS from scratch is going to take a long time just as it took a long time for BeOS to build up traction originally. Fortunately Be had released a free version of BeOS before the company collapsed and this allowed the OS to live on legally if only in a frozen state of development. Because the original OS was stable and mature it allowed application and driver development to continue for those who were prepared to stand by their favourite OS though almost all major commercial application development under BeOS was terminated fairly soon after Be closed. The BeOS community was also given a boost by the appearance of a company called YellowTAB who developed a commercial version of BeOS called Zeta from where the original BeOS was left off just before the R6 release, this helped maintain interested although some BeOS fans were sceptical about the legitimacy and viability of Zeta.

The fact that the BeOS community has managed to stay alive and healthy in the 5 years since the original OS ceased development is its biggest asset because it has kept applications alive and driver development reasonably in line with current hardware. Other than Linux I don’t think any other free OS has this kind of support behind it. I’ve been following the developments of Haiku on the Weekly Haiku blog for quite a while and it seems the development of Haiku has reached a stage where the lower level efforts in recreating BeOS are starting to bear fruit. While Haiku is nowhere near complete or bug free I get the impression that some stability is starting to show through and some fairly big applications are starting to run, e.g. Quake 3 and VLC. This is excellent progress and hopefully the developers will stay committed to achieving their goal of recreating BeOS to where the original closed source R5 is now. The developers have taken the wise decision of not releasing anything half-finished to the public and thus keeping the progress and development fairly low key. The reason for this is to limit the publicity until the software is ready otherwise a lot of people will download a half finished OS that may not work well and leave a lasting bad impression of the OS. In many ways Haiku operates much like a commercial organisation would by keeping things relatively contained, keeping the team focused on the important tasks and staying quiet on the media front until they have something really big to announce.

When Haiku does reach its goal of releasing version 1.0 their software will fully replicate BeOS R5 providing a fully Open Source easy to use desktop focused operating system. It will be mostly suited to home users and small offices where complicated IT systems aren’t used since BeOS is largely single user and not designed for mass deployment in a managed environment. If Haiku can gain good driver support and a set of good quality free applications then I can see it displacing Linux as the default OS on ultra-budget PCs which choose to reduce costs by not bundling Microsoft Windows. I can also see power users using it as an alternative to Windows without the headache of managing Linux. It may also be promoted amongst less computer literate people who want computers simplified and do not care for fancy graphical interfaces like those promised by Windows Vista.

When Haiku does reach version 1.0 this will only be the beginning of a BeOS revival, much has changed since 2001 and there was still plenty of room for improvement back then. There are many advances that could be implemented to make Haiku a more competitive OS, here’s a few I’ve thought of:

Automated updating of the OS, drivers and other major components could be a key feature to reduce manual maintenance of the OS. Windows Update is a good example of an update system that works well.

Integrating the OS with community websites and resources such as the BeBits application repository could help make the process of getting help and applications much easier than leaving the user to seek these out themselves.

Ensuring there are plenty of no risk methods to evaluate the OS before jumping right in. Initiatives such as live CDs, virtual machine distributions, easy dual boot installations immediately come to mind. Virtual machines could help encourage users to use the OS from within Windows without the risk of hardware incompatibility while the OS builds up its native driver set to accomodate a larger variety of hardware.

Ensuring that the online experience is as equal to Windows and the Mac as possible, that involves having a state of the art web browser (Firefox is already the browser of choice on BeOS), online media playback, Flash, Java, etc.

Integrate support for cross platform development systems such as Java and .NET to help bring more applications to the platform.

Supply some well featured but simple general applications for common tasks such as word processing, e-mail, video editing, etc with Haiku so users can be productive straight after installation.

Whatever the developers choose to do once they reach their milestone it will be entirely up to the development team to decide where to go next and this is where Haiku will really make its mark. The success will depend on how radical the changes are, BeOS offers an excellent foundation to build on but the OS will require plenty of changes and there will always be the question of how far can you go without breaking too much and how far do you go to maintain compatibility. Will the developers want to keep the changes in line with what Be would have done or talked about doing or will the developers try their own things. The community have been thinking about the future beyond version 1.0 and they call it GlassElevator with much discussion of what should be in it, hopefully it won’t be too long before the community get a chance to implement it and we can find out if Haiku is the Open Source desktop OS everyone has been waiting for.