Losing control of the device

I’ve been fascinated by the space where communications meets computing for quite some time now, ever since the early 1980s. When I started working for Burroughs Corporation in 1981, I hadn’t really seen a “proper” computer before; my experience of computers was limited to playing hours and hours of the text-based Star Trek on a wealthy friend’s Commodore Pet between 1977 and 1980.

The first real computing experience I had was in working on Burroughs mainframes (specifically the B1000 series and B6000 series) via dumb TD830 80×24 terminals; for me, computers and networks therefore went together right from the get-go. Having been brought up on a traditional journalist diet rich with Smith-Corona and Remington standards and portables, it seemed magical to me that I could type things using a simple “language” (a text management system based on something similar to SGML) into a keyboard on the 1st floor of a building, and documents would come out on gigantic printers on the 9th floor of the same building.

I became used to the concept of a network, of being able to move data around, of being able to attach and detach devices and to address them, to direct stuff to them, to receive stuff on them. I became used to being able to run software on boxes on one floor and to view what was happening on another floor. But the network was limited to the building I was in, the protocols were proprietary, and the devices were dumb and locked down.

A few years later, my astonishment grew. I found out that I could run software in one building and demonstrate it in another using a display unit, a modem and a phone line: I was introduced to dial-up (and, incidentally,  to a world where 4800 baud was normal and 9600 baud was super-fast). The “network” was now more than just one building, the buildings could be miles apart, but everything else remained the same. Proprietary architectures, proprietary devices, proprietary protocols.

Incidentally, it was around then that someone recommended the Steven King film to me, but the video was not easy to find. It is now. And I would recommend that any of you that wants to know a little bit about computers and communications watches the film, Computer Networks: The Heralds of Resource Sharing, which was made in 1972.

Then a few things started happening all at the same time. First, the PC came along. [At Burroughs, we had networked multitasking colour PCs in 1983-84, from Convergent Corp, running something called CTOS; we marketed them as B20s and B25s, running our own variant BTOS). The advent of the PC meant that there was now considerable intelligence within the device, even if memory was limited and fixed storage almost nonexistent. We were beginning to lose control of the device, which was no longer a dumb slave to the mainframe. But it was only a beginning: the only PC was the IBM PC, and the only operating system was MS-DOS.

Two other things happened, both possibly due to regulatory pressure. First,  IBM seemed to be allowing people to build “IBM Compatible” PCs; the AT bus became the ISA bus, and clones started appearing everywhere. Around the same time, “open systems” began to make their mark, apparently driven by the availability of SVID from AT&T; organisations like X/Open, with their XPG and later POSIX specifications, really started to drive the open systems world forward.

It was around the same time that I started getting interested in “the internet”, I was a fairly late starter. In fact I didn’t even have an e-mail account until that year, nor did I belong to any internet groups or bulletin boards until then. In fact, it wasn’t until much later, 1994 in fact, that I had regular access to the internet, as the Web was emerging. And it took till 1997 before I had regular internet access at home.

So in a relatively short space of time, maybe a decade, I watched dumb proprietary “tied” devices connected to proprietary architectures transform painfully into intelligent open distributed devices that connected to open architectures and operating systems. The computer industry were beginning to lose control of the device.

To my way of thinking, the same thing had already happened to the telecommunications side of the industry, albeit more slowly. First it was “any colour you like, as long as it’s black” Bakelite phones you could only rent from the telco. Then the telco gave you coloured phones, but you still had to rent them. You were “subscribers”, not customers. After a while they started letting you buy the phone, but only from them. It took a long time before you could buy your edge device from anywhere, anytime.

Computers and telecommunications. Two industries with a history of monopoly participants, both losing control of their “device”. The first time around, it wasn’t too painful for either industry. Control of the hardware was soon replaced by control of the software — in the case of IT, this meant locked-down desktops, centralised “builds” for the desktop, complete emasculation of the technology while flying the flag of security and confidentiality. Thou shalt not. Thou shalt not. Thou shalt not.

That was then.

Today, in a converged world where digital natives have entered the workplace, device lockdowns are no longer sustainable. The objective — that of keeping enterprises secure from unauthorised access and keeping confidential information confidential — need not change, it is perfectly rational.

What has to change is the way the objective is being met. The historical way of doing this is pretty asinine when you think about it. I’m going to loan you a Ferrari and then take the wheels off and immobilise the steering wheel and block access to the tank. Because that way you won’t steal the car or hit something or cause any damage.

So now new ways are needed. New ways that begin with services that are delivered on a device-independent basis. Services that work out the hardware and software specifications of the device at the point of delivery, and adjust everything accordingly.

This time around, the converged industry has really lost control of the device. And the customer has gained that control, in the name of choice and preference and style and affordability. And even freedom.

That’s why I’m still excited about the promise of networked IT. Because we’ve lost control of the device. Any fool can make money in a bull market. I’m excited that I’m with a bunch of guys who are going to make money despite the market.