Most mission-critical applications still run on mainframes. Almost every bank, insurance company, airline, hospital, and government entity that provides vital services uses a mainframe. Which is not to say that distributed systems aren't growing in importance, or that they won't eventually edge the big iron out. But at this point, legacy still rules.
In 1981. the IBM PC made desktop computers credible with business; the Apple }{ was popular in homes and schools, but had no corporate cred.
IBM and IBM COMPATIBLES made inroads into business. Xerox, TRS, and Texas Instruments were already there, but couldn't carry the clout IBM did. When Blue stepped into the game, the PC came of age. Unfortunately for IBM, its own models were overpriced and underpowered, so they eventually lost their market share to the Dells and Gateways and Compaqs. However, the operating system they used -- compatible with the Intel chipset -- became a de facto standard.
Microsoft leveraged its association with IBM, the big swingin' Richard of the day. And for two decades, that was enough to keep them on top
Yep. Then IBM decided it didn't need Microslop so it came up with OS2, a better OS by the way. But nobody bought it, even though it was part of IBM's Enterprise Service Architecture. DOS, which was really just a rip-off of Unix, continued to run most of the world's PCs in one form or another.
Microsoft prospered not because it had the best answer, but because it had the most popular. Now, when machines and people are closing in on common languages Microsoft cannot control, its influence is fading.
Exactly. Java has had a lot to do with that, as well as Microslop's arrogance about its marginal products. My next machine will run Linux; most of the machines at work run some form of Unix. Microsoft's days on top are numbered.
1) You're describing big business. 75% of the economy is in small businesses.
B) The definition of "mainframes" has shifted radically. your average "mainframe" today is a row of rack-mounted PCs. Even Cray has gone to massively parallel multiprocessing. The vast majority of IP addresses are allocated to machines that would fit in the trunk of a Volkswagen.
IBM and IBM COMPATIBLES made inroads into business.
IBM opened the door. The clones rode the coattails. That was what made Microsoft the juggernaut it is today -- when it sold DOS to IBM, it retained the right to sell it to others.
Xerox, TRS, and Texas Instruments were already there, but couldn't carry the clout IBM did. When Blue stepped into the game, the PC came of age.
Don't forget Apple. Desktop computers were seen as toys, or as ways for the little woman to organize her recipes; they didn't become serious business tools until they had the IBM name. IBM stands for International Business Machines. That neatly summarizes their reputation.
To make a breakthrough, a new technology needs a "killer application" -- one use or one piece of software that, all by itself, justifies the cost of the device. For the IBM PC, the killer app was Lotus 1-2-3. Word processing, spreadsheet, and -- primitive as they were back in the day -- graphics.
Unfortunately for IBM, its own models were overpriced and underpowered, so they eventually lost their market share to the Dells and Gateways and Compaqs.
That description of what happened is a little simplistic. IBM thought it could control the desktop market as it had controlled the mainframe market. Compaq was the first company to reverse-engineer the IBM BIOS -- not to make a cheaper PC, but to make a "portable" one (suitcase-sized and about 20 lbs.)
IBM fought that through the courts, and lost. So the clones walked in the door IBM had opened.
However, the operating system they used -- compatible with the Intel chipset -- became a de facto standard.
True. IBM made three key mistakes that caused it to lose its ownership of, and eventually any position in, desktop computers. (1) they used off-the-shelf components. (2) they used a BIOS that wasn't copyrighed and could be imitated, and (3) They bought an OS and didn't buy it outright. So they built a machine anyone could build, running an OS anyone could buy.
I was a kid spending the summer in Silicon Valley in the summer of 1981. We were spending the summer with a friend of my mom's, a engineer at Intel. At dinner one night, the wild rumor was that IBM was planning to sell a microcomputer. That's what they were called then. A few months later, here came the PC.
Yep. Then IBM decided it didn't need Microslop so it came up with OS2, a better OS by the way. But nobody bought it, even though it was part of IBM's Enterprise Service Architecture. DOS, which was really just a rip-off of Unix, continued to run most of the world's PCs in one form or another.
After the failure of OS/2, IBM joined Apple to form another OS -- called, at various times, Taligent or Pink. After that effort collapsed, Apple went running to Steve Jobs and bought NeXT. Jobs came back to Apple with not only a new OS, but with his old mojo.
From the day OS X was released, there was a hush-hush parallel effort to develop it for Intel processors. The kernel was ready there, in the form of NextStep. When they finally got sick enough of Morotola's failure to deliver speed gains, Apple was ready to switch.
According to Silicon Valley lore, IBM was ready to buy the rights to CP/M, but its developer, Gary Killdall, was out fluing his Cessna and didn't want to interrupt his fun to meet with the suits from Armonk. So they went to Bill Gates.
>>>Microsoft prospered not because it had the best answer, but because it had the most popular. Now, when machines and people are closing in on common languages Microsoft cannot control, its influence is fading.
Exactly. Java has had a lot to do with that, as well as Microslop's arrogance about its marginal products. My next machine will run Linux; most of the machines at work run some form of Unix. Microsoft's days on top are numbered.
Java is certainly a facto. I think bigger ones are open standards in general -- POP, SMTP, HTTP, HTML, XML, CSS. Most of the things that most people do on their computers most of the time, they can do on any platform without breaking a sweat. So M$'s main marketing tool -- "It's compatible" -- is moot.