NEW YORK - For decades, the tablet computer was like a mirage in the technology industry: a great idea, seemingly reachable on the horizon, that disappointed as hopeful companies got closer. Microsoft has experienced this cycle of hope and disappointment many times.

The device unveiled by the Redmond Wash.-based software giant on Monday - the Surface - isn't the first tablet it envisioned. Indeed, the company's engineers have been trying to reshape personal computing for as long as there's been a PC.

The first PCs had keyboards, borrowed from the typewriter. But people quickly started wondering whether pens, which are more comfortable writing tools, wouldn't be a better basis for personal computing.

Several companies worked pen-based computing in the late 1980s, and Microsoft jumped on the trend. By 1991, it released "Windows for Pen Computing," an add-on to Windows 3.1 that let the operating system accept input from an active "pen" (really a stylus). Several devices used Microsoft's software, and are recognizable as the ancestors of today's tablets: They were square, portable slabs with a screen on one side. They weren't designed to respond to finger-touches, however: the reigning paradigm was that of the notepad and pen.

The pen-computing fad subsided in the 90s. While PenWindows tablets got a lot of attention, mainstream computing remained stubbornly keyboard-based.

In 2002, Microsoft founder Bill Gates said these early tablet ventures were "almost painful to recall," but not to worry. He had something much better, a device that would fulfill "a dream that I and others have had for years and years," he said. It was Windows for XP Tablet PC Edition. This time, hardware makers like Hewlett-Packard Co., Samsung Electronics, Toshiba Corp. and Acer Group played along, producing tablet PCs.

Like the earlier generation, some of these looked like today's tablets, but inside, they were really PCs. Compared to an iPad, they were expensive - at around $1,500 - heavy, and didn't last long on battery power. Buyers paid a lot for the ability to enter things on the screen with a pen.

Another problem was that the pen-based adaptations were skin-deep. Windows remained a thoroughly keyboard-and-mouse-based operating system, and many functions were simply hard to get to with a pen. Third-party applications weren't converted for pen use at all. As a backup, many of these tablets had keyboards, just like laptops.

The tablet PCs found homes in a few business settings, where a PC that could be used while standing, at least for short periods, was welcome. But they remained a niche product, and the number of manufacturers who made tablet PCs steadily shrank.

In parallel with the Tablet PC push, Microsoft prompted partners such as Fujitsu and ViewSonic to create Smart Displays. These were big tablets intended for home use, and each one was linked to a PC through Wi-Fi, making it something of an expensive monitor with short-range portability.

This was supposed to be a cheaper alternative to a full-blown tablet, but the devices reached shelves at $1,000 and more in 2003. While a Smart Display was in use, the associated PC could not be used. Very few were sold, and Microsoft cancelled the project the same year.

Microsoft gave tablets another try in 2006, launching "Project Origami" with some of its partners. The idea was to make really small PCs with screens sensitive not just to pens, but to fingers. This time, fewer companies followed along. One of them was Samsung, which had high hopes for its "Q1".

But Microsoft hadn't learned much from its Tablet PC adventure. Windows was still hard to use with anything other than a keyboard. The "Ultra-Mobile PCs" were still expensive and suffered from very short battery life - the Q1 could surf the Web for about 2 hours. One thing they did get right was weight - the Q1 weighed 1.7 pounds, just a bit more than a first-generation iPad.

In 2008, reports emerged of yet another tablet computer, or rather a "booklet computer," being developed by Microsoft. Code-named "Courier," it had two screens joined by a hinge, and facing each other. It was designed for pen and finger input. Microsoft cancelled the project in 2010, saying it was just one of many projects it tests to "foster productivity and creativity."

One touch-based computer that did see the light of day in 2008 was Microsoft Surface. It was more of a table than a tablet: the computer was a big box that sat on a floor, with a big, horizontal screen on top. It was intended not for home use but for store displays and similar applications. Unusually, Microsoft didn't rely on hardware partners for this product, but made and sold it on its own. Intended as a niche product, it has remained one.

Microsoft has had one notable success in the tablet space - if you apply a broad definition to the term. Its "Pocket PC" operating system, which is distinct from Windows, ran on phone-sized hand-held "personal digital assistants" starting around 2000. The devices were powerful compared to Palm's PDAs, the market leaders of their time. The Pocket PCs supported color screens, and could recognize casual handwriting. Compaq made good use of Microsoft's Pocket PC software in its popular iPAQ line. But PDAs were a small market, and when Pocket PC moved over to smartphones and was renamed Windows Mobile, it soon found tough competition in the shape of BlackBerrys and then iPhones.

The company that finally cracked the tablet code in 2010 was Apple, not Microsoft. Apple made the iPad a success by scaling up a phone rather than scaling down a PC, which is what Microsoft had been trying to do with the Tablet PC and Origami. Phone chips are cheap and last much longer on batteries, which meant that the iPad was both light, inexpensive and had good battery life. In addition, the iPhone software it used was designed from the ground up for touch input.

Microsoft's new strategy is similar. For Windows 8, it's borrowing design features from Windows Phone, its new smartphone system. Most importantly, one version of the software is designed to run on phone-style chips, rather than the PC-style chips that have been the mainstay of Windows since it was created in the 1980s. It remains to be seen whether Microsoft can make its tablet vision a reality, or if it will stay a mirage.