What was BeOS?
BeOS was an operating system founded by ex Apple executive Jean-Louis Gassée in 1990, in hopes of creating a completely new computer operating system created for media professionals from the ground up. BeOS was originally created to run on a specific set of hardware, called the BeBOX, however Be Inc. was forced to port the OS to run on a variety of machines. It directly competed with both Apples Mac OS, NeXT's NeXT Step, Windows 95, and Linux. While it was UNIX-like, the BeOS contained only new and proprietary code, a selling point of the system as it was not based off of 20 year old code and programming models. It was said to have been operating system which was future "to be the ultimate operating system for processor-intensive multimedia and Internet applications", and had many innovative features not seen in rival operating systems which have only been met in recent generation of operating systems. The general idea for the operating system is captured from a section of the mirrored Be Inc. website from 1999:
"The company wanted to step beyond the evolutionary approach to personal computing architectures; to see what could be accomplished if you built a personal computer using new assumptions, based on cutting-edge software design concepts, and designed for the next decade's applications, rather than the last decade"
BeOS failed for a number of reasons, most of them being tied to business related environmental failures. At its heart, the operating system was unable to secure a large variety of developers to continue to make a wide variety of software which to run on the machine. Another early limitation was the processor specific architecture, as BeOS was not originally portable across different hardware platform, a folly of a business decision when they tried to bundle their hardware with their software (a la Apple 2008). In early 2002, founder Jean-Louis Gassée left after being sold to Palm Inc, and shortly afterward was BeOS commercially discontinued and was open-sourced. The focus of this dossier is not on the company Be Inc. but the technology they sold. In some cases, the fate of the business was the fate of the technology, but I will focus mostly on the failure of the technology.
Relevence of BeOS as a Dead Media
One of the main reasons BeOS is an interesting Dead Medium to study is that it was created with such political and deterministic goals in mind. From these goals, one can extrapolate both the intent within the context of when BeOS was created, and well as to place it within a context of a greater technology history.
What BeOS said about technology
BeOS was created with a specific purpose: to be "the Media OS" (BeOS website). All of the technical advancements, which today seem somewhat old hat, were really pioneering features which shaped the current PC landscape. Selling the very idea of an operating system which was completely "rebuilt from the ground up" and not based of code which in many cases was more than 20 years old. Coming only a little short on "year-zero" rhetoric, the BeOS was created "today" for "tomorrow".
This purpose, coupled with the fact that Be was so focused on creating completely new code, without and of the "cruft" of old operating systems, marks a drastic change from how software is normally produced. In much of the technology landscape, progress is a function of modernity, with product iterations building from previous versions, being systematically improved, repackaged, and sold as brand-new. Technology innovation then, while it may be marketed as "all new," are actually restrained by the pre-existing limitations and conception of previous iterations of the product.
In a computer operating system, these hang-ups have drastic effect on what the computer and its users conceive the computing meta-machine can achieve. Many choices within the system architecture can result in end user limitations, as can more user facing, interface elements which prevent users from exerting full control over the machine. Other operating systems are patched and re-purposed to achieve tasks that they were not necessarily created to do, in the name of deadlines and not re-inventing the wheel.
BeOS was creating with a technological determinist position in mind: to create something which rethought these innovations not as progressive improvement, but as a clean slate, rethinking past innovations as starting points, rather than the result of years of progress.
For example, looking at some of the advertised underlying technology benefits from a technical white paper, we can see these innovations being touted much before BeOS's competitors (Forgive me for the overly technical nature of this quote) :
In their place was put an architecture specifically designed for handling intensive tasks, such as digital audio and video. The building blocks include:
- Symmetric multiprocessing - Take advantage of two, four or more processors in a single machine.
- Pervasive multithreading - Multithreading takes large tasks, such as applications, and breaks them down into a myriad of smaller tasks. Pervasive multithreading means that this approach is used throughout the BeOS, from the kernel, through the graphics and I/O systems, and through BeOS applications.
- Preemptive multitasking - The BeOS works on and rapidly switches between dozens, often hundreds, of smaller tasks. These tasks can be deployed on a single processor, providing a smooth multitasking environment, or across any number of processors in a multiprocessor system.
- 64-bit journaling file system - This enables extremely large volumes and files, of terabyte size and more, enough to handle even raw uncompressed, high-resolution video and audio -- the foundation for high-quality editing systems. The Be file system goes even further, providing database capabilities allowing the storage of multiple attributes and indexes along with files.
- Object-oriented APIs - The BeOS API makes programming easy and efficient. " - (BeOS Technical Whitepaper)
While this may sort of seem like technological gibberish, the BeOS was able to use these low level technological improvements to fundamentally change what was possible with a computer. End user features, such as playing multiple video and audio files simultaneously, virtualizing other operating systems, and fully integrated networking and internet systems, all were features which were not seen in other computing platforms until several years after displayed in the BeOS (Whitepaper). However, the platform itself never caught on with consumers, and while this main have been primarily a failed business proposition, it could also signal interesting truths about progress, modernity, and the public's technological acceptance.
What BeOS said about users
The BeOS was a classic example of a product which had a very structured, deliberate use value which it proposed for consumers, and perhaps was too limited for mainstream adoption. Being a revisionist product, BeOS hoped to capitalize on user actions which were being misrepresented in other computer products. Be Inc. in a sense saw into the future that people would need high performance internet terminals, as they started to move their media communications, and expression to the computer. Rather than simply being a machine for business or productivity, the BeOS was supposed to revolutionize what a computer really could be to an individual. Being the first machine to really bundle video, audio and image editing with full internet support made BeOS out to be one first operating systems where all of the features we come to expect from a computer today existed, and this was almost ten years ago from today (2008).
However, there are two fundamental differences between what we see today and what the BeOS may have fell short on, and these are made apparent when comparing BeOS to Mac OS 10, and it's parent (and primary BeOS competitor in terms of ideas) NeXT Step. Primarily, while Be was offering all of these features to it's consumers, the considered this group to exclusively be media power users, those who were outside the mainstream computer users. So while they made great machines that actually felt like modern computers, they thought individuals would never need such a system, a failure not in vision, but in scope.
Another primary failure in the BeOS was the lack of aesthetics. BeOS was conceived as the ultimate tool for people to do amazing things, but very little focus was placed on user experience outside of being technologically elegant and powerful. This was in stark contrast to Steve Job's approach at NeXT, which shared some of the "rip it up and start again" ideologies (to a lesser degree), but was much more focused on being a powerful, elegant system across all uses. Gassée was even quoted to say:
"For God's sake, don't compare us to NeXT. We want to be a better tool for developers, not to be tasteful. We don't cost $10,000. We have a floppy drive. We do not defecate on developers" (Apple Confidential).
In fact this difference may have been the difference in the success of the company. Be Inc. was extremely close to actually being purchased by Apple to be the basis for the next Apple OS, but Be asked for too much money, and Apple ended up purchasing Job's NeXT after the deal fell trough. The rest is history, as NeXT became the basis for one of the most popular and most advanced operating systems today, and Be fell into techno-obscurity.
BeOS may not be in use today, but may of the technological ideas, and insights about our computing society are visible. Even though Apple Computer did not buy BeOS, they still shifted their prospective with Mac OS X to fill this void of the computer for Media Consumers, not just Computer consumers. The computer as a complete media portal is now the primary paradigm for personal computing, rather than just something for power users. Additionally, many of the lower level innovations have been taken up by both Apple and Microsoft, although in a way which could be argued is much less elegant, and more of the idea of the that these new features are simply lipstick and a jetpack on a pig. Additionally, BeOS may say something about how society views progress: although sometimes it may technically be the best idea to "rip it up and start again," consumers may just not be ready to give up what they already know and are comfortable with.
all images from BeOS website
Be Inc. BeOs Data Sheet. 1999.<http://www.beatjapan.org/mirror/www.be.com/products/beos/beos_datasheet.html>
Be Inc. BeOS Technical White Paper. 1999. <http://www.beatjapan.org/mirror/www.be.com/products/beos/mediaos.html>
Be Inc. BeOs Data Sheet. 1999.<http://www.beatjapan.org/mirror/www.be.com/products/beos/beos_datasheet.html>
BeOS Promo Video. Prod. BeOS. 1996. <http://www.youtube.com/watch?v=9eMGbDJmgv0>
Linzmayer, Owen W. Apple Confidential : The Real Story of Apple Computer, Inc. New York: No Starch P, Incorporated, 1999. <http://macspeedzone.com/archive/art/con/be.shtml>
Markoff, John. "A New Computer Dazzles a Jaded Industry Crowd." The New York Times 4 Oct. 1995. <http://query.nytimes.com/gst/fullpage.html?res=990CE2D9133EF937A35753C1A963958260>
"Mirror of Be Inc. Website Circa 1999." Be Inc. 1999. Be Inc. <http://www.beatjapan.org/mirror/www.be.com/>.
Sydow, Dan P. Programming the Be Operating System. O'Rielly, 1999. <http://oreilly.com/catalog/beosprog/book/>