In the embedded segment, names are getting tossed around willy-nilly as we try to define how things have changed and are continuing to change. I suppose this has something to do with the nave assumption that by naming something, we have defined it, understand it, and thus control it.
A number of terms have emerged to describe the various connected, small-memory-footprint devices and applications: connected computers, connected computing devices, consumer appliances, embedded Internet devices, information appliances, information-centric embedded devices, Internet appliances, Internet devices, Internet-centric embedded devices, Internet-enabled embedded devices, mobile Internet devices, pervasive computers, smart devices, ubiquitous computers, and, of course, one of my favorites, net-centric computing devices.
My initial feeling about this whole naming thing is, why not keep it simple? So, the name I propose is “personal computers.”
It is not as goofy as it sounds. I was present when the term “personal computer” was probably first used. That was in 1971, long before the desktop system that now has the designation, and it referred to something entirely different, something much closer to what we call “embedded computer” today.
I had just arrived at the California Institute of Technology and was working with Carver Mead, coauthor of a seminal book on semiconductor technology, VLSI Design , and then a professor of electrical engineering. We were developing an article for the in-house magazine Engineering and Science .
The article was about the impact new microprocessors, such as the hot new Intel 4004, would have on society, and it was titled, “Computers That Put The Power Where it Belongs.” The article was about what Mead called “personal computing devices,” which he defined as special-purpose processors that would be incorporated into the workings of ordinary machines: automobiles, ovens, refrigerators, and other appliances, to replace the electromechanical mechanisms used previously. He differentiated these personal computers from general-purpose minicomputers and mainframes, which were totally focused on computational and data processing problems.
After leaving Cal Tech, I spent the rest of the '70s writing about what I thought were special-purpose personal computers. So, imagine my surprise in about 1980 when IBM popped out of the woodwork with what it called a “personal computer,” a general-purpose computational and data processing engine that fit on the desktop. It was personal in the sense that it fit on or under a desktop and one person could use it. And it was within the price range that an individual could afford. But most of the functions it has performed for most of its life have been all of those general-purpose computational and data processing functions that were previously the concern of centralized mainframe and mini-computers.
While it would be delicious irony if the greatly expanded embedded market got back its original (and I think appropriate) name, I must admit that my candidate for a new name has too much legacy associated with it. The desktop computer is still a PC to everyone else. But most of those other names are not suitable, either. They have too much specificity.
No, the new name we want to give to this new segment of the net-centric computing market has to be very “high concept.” That's a Hollywood term for a short description that a writer or director must come up with so that a producer or a marketing or advertising executive will think he understands the theme of a movie.
Well, how about “embedded computing?” While it once had a very specific meaning, the pesky habit that embedded companies have about going after any application for which their services, software, and expertise will find a use has expanded its borders into markets that have little or no relation to its original meaning. Now, the term describes something so broad that the word is devoid of any useful information content.
And while a lot of the devices being built by embedded designers using embedded tools, languages, and software have a small footprint, require high reliability, and, increasingly, deterministic operation, few of them are embedded, really. But I guess reality is not a good enough objection.
After a lot of thought, however, I have finally come up with a name that fits the bill: “Popular Computers.” These new computers will be popular in every sense of the word that I can find in the dictionary: they will be widely used by the whole population, will be “regarded with favor, approval or affection” by the populace, and they are “adapted to the ordinary taste or intelligence.”
What do you think? What are your candidates for a name? What is your opinion about some of the names I like and others that I obviously don't? This could be fun. And we may come out of the process with something that is actually useful.
I think that a fitting name would be 'invisible computers'. Desktop (or personal) computers today are very obvious. When you look at a portable mp3 player or a cell phone you do not see a computer, you see the functionality that the computer inside is enabling. Although when you look at the beige box on your floor, you instantly think 'computer'.
I think that embedded applications of computers are exactly that, invisible computers. The computers can not be seen, but their impact can be felt. They're embedded within another device, which is something very different than what most people think of as a computer.
I welcome you to the age of invisible computing.
Software Engineering Intern
RSA Security Inc.