February 25, 2003

Operating systems will soon be obsolete

- By Robin 'Roblimo' Miller -
Linux, Windows, Mac. All have their place, and before long that place will be in history books. The seeds of a new style of computing device have been sown, and as they grow they will inevitably lead to a world where the computer operating system as we know it today is as dead as the planetary transmission that drove the Model T Ford.

This train of thought was spurred Saturday night when I turned on my TV while reading an article about how Phoenix Technologies is starting to add things like Internet connectivity to some of its latest BIOS products. The press release, while written in appallingly dense marketese, contains this key paragraph:

"Phoenix cME addresses the needs of increasingly connected and digitally dependent users," said Albert E. Sisto, Chairman, President and CEO of Phoenix. "Now they can access diagnostics and self-healing capabilities, Internet access, and recovery utilities, even after a major system malfunction. In conjunction with our industry partners, Phoenix is changing the user experience at its very core while allowing enterprise systems providers to help reduce support costs for their customers and to deliver brand-differentiated products across a wide variety of platforms."

Think of what Microsoft did to Netscape by making MSIE part of the operating system (or claiming it had, anyway). Now think about incorporating operating system functions and higher-level apps -- like Web browsers -- directly into the BIOS. Suddenly the need for an "operating system" disappears. We remove a layer of code interpretation. We have a faster and more efficient device. We have a whole new family of software to develop. We finally have a chance to move beyond today's clunky computers and dream about ones that do more work, in less space, with fewer parts and less electricity usage, than we can possibly make with today's hardware and software platforms.

Turn it on and start working

What made solid-state (transistor) TVs an immediate mass-market success was the fact that you could turn them on and start watching right away instead of waiting a minute or more for tubes to warm up. I remember this transition. I remember tubes, and turning radios and TVs on, then listening or watching only after they had gotten to operating temperature. I remember the intermediate evolutionary stage in the design on "instant on" TV sets, that kept a little juice flowing through tube filaments all the time so they appeared to be "instant on" although what really happened is they were never entirely off, but were more like a computer in "suspend" mode.

Imagine the in-store demo: A customer is given a choice between two computers with similar capabilities, but one turns on *bang* right now while the other goes through a bootup routine. Which one is the average person going to buy?

Even if the "on" isn't quite "instant" but takes a second or two, I suspect the vast majority of home and office users will take choose it over what we have now.

Other future computing ideas

Right now, we don't have much in the way of "peer to peer" since we rely on connections through big companies to reach our "peers." when my (wireless) computer talks directly to yours, 300 meters away, and yours can talk to another one 300 meters in the other direction and pass messages back and forth, we'll have real peer to peer networking.

There have been a few stabs in this direction, but none of them have gone very far yet. One device allowed kids to SMS each other directly. It was promoted with a bang a year or so ago, but I haven't heard even a whimper from that vendor since. The technology is here, just not available in consumer form. It will come. It is inevitable. It will also be fought by "content producers" (RIAA, MPAA et al) like mad, because the devices will inevitably be hacked to remove any digital rights management built into them at the factory, and if they can talk directly to each other there will be no way to control files they pass back and forth the way there is now with ISPs and other network operators.

The other big deals that are going to revolutionize computing are improved input and output devices. Right now, the biggest laptop/handheld size bottlenecks are displays and keyboards. Replace our current displays with "eyeglass" LCDs or holographic projections, and replace keyboards with gesturing systems or voice recognition (or both), and we'll see some devices that bring new excitement to the computer industry.

What operating system will we see?

I doubt that any of today's popular operating systems will survive a transition to a BIOS-like standard. Even QNX and the various "real time" Linux systems probably won't be tight enough. I expect to see something entirely new -- something that isn't widely-known now, that may be under discussion in academic circles or under (secret) development by some company somewhere -- and not necessarily one in the United States or one that is well-known, either.

Since I haven't heard anything, have you? Have you spotted any new research developments that might lead to an instant-on, peer-to-peer, small computing device that isn't an extension of what we already have, but may move us in a totally new direction?

If you have, please include a link to information about it in a comment below. I suspect a lot of people are interested in learning what may be coming along next. (I know I am.)


  • Migration
Click Here!