Entry No.128

IT Writers Awards

When it comes to computer crashes, we're like Pavlovís dogs. The machine freezes when we hit the reboot key sequence and hope that nothing important has been lost. Then it happens again, and again. The modern PC (or Mac) is by and large unreliable and inefficient, and yet we keep going back for more.

Is there a better way? Linux zealots would claim so, but you still need more technical knowledge than the average 14-year-old to get that system up and running. However, there is another, less well known, option. BeOS is one of the unheralded treasures of the PC age, yet few people know anything about it or about Jean Louis Gassee, the man behind it.

The seeds of BeOS were sown back in the late '80s when French-born Gassee helped establish Apple in Europe. A couple of years later he was made responsible for overseeing the development of Pink, one of two attempts by Apple to develop an operating system to replace the venerable Macintosh OS. Technically, BeOS has no relationship with the stillborn Pink, however both shared similar conceptual goals ñ modularity, efficiency and reliability.

Gassee left Apple, not because of the failure of Pink, he says, but because he wanted some new challenges. And so instead of leaping into a new venture, he cooled his heels, contemplated the purchase of Commodore, which held the rights to the revolutionary Amiga operating system, rejected that idea and kept thinking.

"After looking at Amiga I realised that there was an opportunity for a 'clean Amiga', and that if we managed the business semi-intelligently we could be successful," says Gassee. He then approached colleague Steve Sakoman who, much to Gassee's surprise, said he thought the idea was a good one. "As soon as I heard him say yes, I stood on the brakes," says Gassee. "Part of knowing how to make a sale is knowing when to be quiet!"

He says that he took away from Apple the experience of attempting to build a new operating system, and could identify the pitfalls. There was still a lot of learning to do.

Initially BeOS, as it was to become, was designed to run on inexpensive microprocessors, similar to those found in PDAs. That didn't work so well because they weren't common enough. Then Be botched its first attempt at writing a file system for the new OS, and subsequently had to redo the file system. Gassee also made the mistake of initially tying the OS to proprietary hardware, the BeBox, a twin processor computer that shared some design similarities with contemporary Macintoshes. With hindsight that decision can be seen as a legacy of his time at Apple, a company that insisted (and still insists) on hermetic relationship between hardware and the operating system.

When it became clear that the BeBox strategy wasn't going to work, Gassee focused the company on developing for the PowerPC platform, the same platform that Apple Macintoshes used. That strategy was aborted when Apple refused to release the specifications for its new G3 processor, leaving Be stranded with an operating system that would only work on the older (and hence slower) PowerPC computers.

"We made mistakes and addressed those mistakes," Gassee says. "We're still refining the process."

Itís also significant that at this time Apple and Be were discussing the possibility of Apple acquiring BeOS as its next-generation OS. Ultimately the deal came to nothing, in part because Apple co-founder Steve Jobs had rejoined the company, and packed in his baggage was NeXT, the company he founded after being hounded out of Apple in the mid-'80s. Apple ultimately acquired NeXT, and its NeXTStep operating system from Jobs for the princely sum of $US450m ($780.6m), and used it as the basis for a next-generation OS. The child of that deal, MacOSX, is slated for public release at the beginning of next year.

The only option left for Be was to move its software again, this time to the open Intel-based architecture. It's in this space that Be has made the most impact, but its share of the OS space is still negligible. During the same period, another upstart operating system, Linux, also came along, and while figures vary it's safe to claim that Linux is now used on about 4% of computer desktops (as opposed to server systems), while Windows (all types) takes 90% and MacOS 5%.

So is Gassee a techno-Don Quixote, tilting at the glowering windmills of Redmond and Cupertino? Is he, to put it more bluntly, a madman?

Bruce McCabe, research director for Gartner, is blunt in his assessment of the world's need for another desktop operating system. "It's not needed," he says. ìNot even Linux, which is getting so much attention, is making much of an impact in the desktop market. Itís a server operating system.î

Given that Be's share of the desktop market is minute, what's to get excited about? The fact of the matter is that it's great software ñ software that doesn't crash every other hour, that's fast, portable and reliable should be acclaimed. Be has done what Microsoft, Apple, a resurrected Amiga, the open source community and others have not been able to do: it has developed an industrial-strength operating system that anyone can use.

Despite this, Be hasn't set the world afire, which is why Gassee has reoriented the company yet again. And surprisingly this reorientation, focusing on devices dubbed internet appliances, is one that meets with approval from the analyst community. "There is plenty of room for OS technology in specialised applications," McCabe says. "By this I mean on devices which are designed for a single purpose or narrow range of purposes (eg imbedded in handheld devices, set-top boxes, thin-servers etc). If any OS vendor does a good job designing and marketing for this space they can still win themselves a niche."

The motivation behind the move towards internet appliances was pure pragmatism, Gassee says. "We're entering a new phase of computing, from the mainframe to the minicomputer, and then from the mini to the PC. The appliance is the next step, and we're in a perfect position to take advantage of this market."

Recent deals between Be and Compaq will see BeOS used in a Compaq-branded appliance called Stinger. Part of the attraction of the BeOS is the fact that the nuts and bolts of the system are compact and modular, meaning that an appliance manufacturer can imprint their own look and feel on the system while still preserving the underlying efficiency and strength of Be. "The appliance market is one where people won't put up with unreliability. They want the machine to come on and work and keep on working," Gassee says. "Be is the answer for this."

The appliance strategy doesn't signal the end of the Be PC operating system. "BeOS is the root of the appliance strategy, and it's also the perfect platform to develop applications for the appliances," he says. "Aside from that, the move towards releasing BeOS personal edition has been wildly successful, and it has prompted many more people to have a look at alternatives to Windows."

It's estimated that at least one million people have downloaded BeOS and given it a spin. It's easy to do, because it will launch from within Windows, rather than requiring an elaborate and danger-fraught installation routine and, most importantly, it's free.

Gassee says that itís difficult to estimate the number of people who are actually using Be after installing BeOS Personal Edition, as opposed to the people who are just kicking the tyres. ìWeíre happy for people to use it, or to just check it out, because it increases our profile and the awareness of Be,î he says. ìNaturally, however, weíd like to think that most of these people will continue to use BeOS in some way.î

BeOS isn't going to replace Windows, but by the same measure, it's unlikely that Linux will do that for average computer users either. What it does is provide an alternative for people who want more reliability and want to use a completely modern piece of software. You won't be able to run Microsoft Word, but getting Be and Windows to live happily side by side is easier than replacing a washer in a tap. Just don't be surprised if, having taken the leap, you find yourself spending more and more time using the little OS that could.

Back to Voting Form
Back to Best Features  
Top of page

Content Copyright © the author/publisher listed above

Design Copyright © Consensus Pty Ltd

This web-site uses frames, click here for the full picture