Tuesday, January 19, 2010

Whatever Happened to Cloud Computing?

Whatever Happened to Cloud Computing? For a while it was getting big play in books and magazines, but the noise seems to have faded. Here's one take on it. (This entry is a replay of a piece I wrote for TechRepublic/CNET a while back.)


When I first heard about cloud computing in 2007, it was with a feeling of déjà vu. I had indeed been there and done that — with thousands of other personal computer users — as far back as 1983. That’s the year high-volume cloud computing was kicked off by the debut of the Radio Shack TRS-80 Model 100.

Portable, but lacking storage

You may not have seen or heard of the Tandy Model 100. TechRepublic has a Photo Gallery on it if you want to see what it looked like - - inside and out. It was a minimalist microcomputer, a not-quite two-inch thick slab that measured about 12 inches by 8-1/2 inches.

The Model 100 came with built-in firmware applications: text editor, calendar/scheduler with an alarm clock function, BASIC, an address book, and terminal program — plus a 300-bps modem. Four AA batteries powered it for about 20 hours. It could display 40 or 80 columns on a monitor or television set and had its own 40-character by 8-line LCD display.

This miracle machine’s only real shortcoming was storage, particularly if you traveled for work. Like most, my Model 100 had only the basic 8K of memory, which didn’t quite hold 12 pages of text. There was no slot in the side for a disk or card. I worked with a lot of files — articles, short stories, books — so I needed external storage. So did most other Model 100 owners. Many who spent hundreds of dollars for 24K of RAM still didn’t have enough memory.

A cassette interface made tape storage possible, but the medium wasn’t always perfect. Besides, that would have meant hauling along a tape player that weighed more than the computer, and the reason for the Model 100 was portability. A Model 100 disk drive presented the same lug-along problems as the cassette and cost lots more. (Cost aside, I would probably have lost, damaged, or forgotten the peripherals, cables, or media at some critical point.)

But a storage solution was built in to the little machine in the form of its modem. It was easy to get an account on an online service such as CompuServe or The Source. (In fact, I had free press accounts.) So before I left home on a trip I would upload my work files to CompuServe, using the service’s own data network. When I was at home and wanted to work somewhere other than my office, I transferred files between my laptop and desktop machines using my virtual disk drive.

In essence, I was doing what they now call cloud computing 25 years ago.

Granted, the applications weren’t on a server, but that could be an advantage. If you had no telephone link, you could still work with the data in the Model 100’s memory. And even in the early days you could use online applications to write or crunch numbers if you really wanted. In fact, according to CompuServe founder Jeff Wilkins, using applications and storing data online were the two major attractions of the service when it as opened as MicroNET in 1978. (Most microcomputer owners couldn’t afford a floppy drive anyway, and transferring data to and from a mainframe was about as fast as and more reliable than using a cassette tape.)

Going back to the future

Of course, people were cloud computing with dumb terminals and mainframe computers long before 1978. It was called “timesharing,” and General Electric opened the first commercial timesharing service in 1965. And today one prediction about cloud computing is that we’ll soon be using minimal terminals to run online apps and retrieve and store data — just like in the 1960s.

I wonder what else from those days hasn’t been reinvented yet.


Michael Banks is the author of On the Way to the Web: The Secret History of the Internet and Its Founders (APress, 2008) in which he writes about the histories of timesharing services, databases like Dialog, and the consumer online services that paved the way for the Web: CompuServe, GEnie, The Source, Viewtron, AOL, Q-Link, Prodigy, Prestel, and many others around the world.

Tuesday, January 05, 2010

Bold, Brave Predictions of 1994

Those of you who have read On the Way to the Web know that Knight-Ridder was pretty busy with experiments in the online world in the 1980s, beginning with its Viewtron Videotext service. As I note in the book, Viewtron was an advanced system, much like the Web today. Its advanced graphics carred news, entertainment, and a host of interactive services, including email.

In 1994, ten years after it shut down Viewtron, Knight-Ridder produced the embedded video program above. It's worth taking a look at--and wondering where the hell all these wonderful ideas went. For many reasons, neither Knight-Ridder nor its competitors "got it," and failed to follow up on the developments of the era. Even though their own spokesman says, "It may be difficult to conceptualize the idea of a digital paper, but in fact we believe that's what's going to happen."

That had to have been a minority view, because Knight-Ridder failed to forge ahead with "digital paper." Again, watch the video--you'll see devices that showed up at last week's CES.

Dumbing Down the Web

If the Web is going to replace print, get ready for some real dumbing down. Apprently some publishers who put reviews on the Web (among them a former print mag or two) figure that Web surfers are functionally illiterate, and in terms of text require nothing more than strings of characters next to images and ads. It's kind of an insult to those of us who grew up on real magazines and books.

I was browsing hardware reviews earlier today, and much of what I encountered made about as much sense as this:

"So, I got this thing, I'm cool, and you plug this gadget into your computer, and set your phone next unto it, plug in your refrigerator, initiate windowed XMODEM protocol and wait for the furnace to switch on your printer's FAX unit. So then the USB chukka boots plan nose hair down to the CCITT RS232C diagram."