[ Pobierz całość w formacie PDF ]
.When the theologian-turned-statistician wasn t debating the motivation of divine acts or defendingIsaac Newton s theory of differential calculus, he tried mathematicallypredicting the future, based upon past events and new observations.AnEssay Toward Solving a Problem in the Doctrine of Chances, which was pub-lished after his death in 1761, introduced Bayes theorem, a statistical ver-sion of soothsaying.Imagine, for example, having six coins in a bag, ofwhich one is double headed.You take a coin from the bag and flip it fourtimes, getting heads each time.Bayes theorem enables you to calculatethat there is a 76 percent chance that you ve picked the double-headedcoin.(As British journalist Simon Goodley explains, Bayesian thinkingwould also help you to figure out that the word Amazon on a Web pagerefers to the river rather than the retailer, based on previous references tothe rain forest and South America.)Bayes s ideas were widely used over the centuries.Religious skep-tics used them to infer that the amount of evil in the universe made God sexistence improbable.Business-school professors taught executives how KNOW HELPFUL FROM ANNOYING 187to use them to predict the probable demand for their products.In the1990s, software designers hit upon another application.A software pro-gram could employ Bayesian methods to figure out what a computer usermight be trying to accomplish, based upon the steps that the person hadfollowed up to that point, and intervene to offer him what probably wasthe appropriate assistance.That breakthrough came along at a critical moment for computerand software makers.By the mid-1990s, despite the introduction ofgraphical interfaces that made computers vastly easier to use, only about aquarter of American households owned PCs, according to the NationalScience Foundation.Other studies found lingering unease and suspicionof them.A 1994 Gallup survey found that 59 percent of white-collar work-ers were reluctant to try new technology.Nearly a third of the workersincluding four out of ten women confessed to being outright afraid ofthe powerful, still-mysterious machines on their desks.Seeing a threat to its growth, the industry searched for ways to helppeople get over their cyber-anxiety.Stanford University researchers Computers as Social Actors theory seemed promising.CASA held thatusers instinctively treated computers as if they were people, not tools, andthat they might feel more comfortable if a computer acted more humanparticularly if its virtual personality was similar to the user s.As Micro-soft manager Chris Pratley explained in a 2004 Web log essay: Thetheory was that if you could provide an interface for the computer thatexpressed emotion and that you could interact with, you would be lesslikely to develop animosity toward your PC.and would actually beencouraged to learn and interact. Thus, software designers wedded CASA with Bayesian probabilityto create social interfaces peopled by agents who would guide actualhumans efforts to use the software.Instead of making characters realis-tic, however, they gave them the exaggerated, childlike qualities of Satur-day morning cartoon characters.(Intentionally or not, that fit theories of OOPS 188Nobel Prize winning behavior researcher Konrad Lorenz, who saw cute-ness as a potent cue for behavior, and noted that people were nicer to ani-mals whose physical features reminded them of human children.)Microsoft s initial attempt at a social interface, Bob, was devel-oped under the supervision of Melinda Gates, a Microsoft executive whoalso happened to be the wife of company chairman and cofounder BillGates.It was released in early 1995.When installed on early versions ofthe Windows point-and-click user interface, it converted the Windowsdesktop into a house with ten rooms including a family room, a study,and an attic that a computer user could decorate in a variety of styles,ranging from medieval to postmodern.The furnishings were actuallyicons that opened various applications, such as a letter-writing program, acheckbook, a calendar, and an e-mail program.The interface featured anumber of animated inhabitants whose personalities ranged, in the wordsof software historian Dan Rose, from happily enthusiastic to down-right obnoxious. Bob s cast included cartoon animals such as Hopper thestuffed blue bunny, Digger the worm, Blythe the firefly, Rover the dog,and Chaos the cat.The program also offered Orby, an anthropomorphicworld globe, and a William Shakespeare caricature who spoke in floweryElizabethan language.The most bizarre character was Scuzz, a surly ro-dent who took occasional bites of a package of rat poison, tormented Chaosthe cat, and dismissed the other guides as losers. (Oddly, there was nocharacter named Bob.)Bob had at least a few fans.Wall Street Journal technology criticWalter Mossberg lauded it as a bold departure that attempts to give non-technical people more control over their computers [ Pobierz całość w formacie PDF ]
zanotowane.pl doc.pisz.pl pdf.pisz.pl matkasanepid.xlx.pl
.When the theologian-turned-statistician wasn t debating the motivation of divine acts or defendingIsaac Newton s theory of differential calculus, he tried mathematicallypredicting the future, based upon past events and new observations.AnEssay Toward Solving a Problem in the Doctrine of Chances, which was pub-lished after his death in 1761, introduced Bayes theorem, a statistical ver-sion of soothsaying.Imagine, for example, having six coins in a bag, ofwhich one is double headed.You take a coin from the bag and flip it fourtimes, getting heads each time.Bayes theorem enables you to calculatethat there is a 76 percent chance that you ve picked the double-headedcoin.(As British journalist Simon Goodley explains, Bayesian thinkingwould also help you to figure out that the word Amazon on a Web pagerefers to the river rather than the retailer, based on previous references tothe rain forest and South America.)Bayes s ideas were widely used over the centuries.Religious skep-tics used them to infer that the amount of evil in the universe made God sexistence improbable.Business-school professors taught executives how KNOW HELPFUL FROM ANNOYING 187to use them to predict the probable demand for their products.In the1990s, software designers hit upon another application.A software pro-gram could employ Bayesian methods to figure out what a computer usermight be trying to accomplish, based upon the steps that the person hadfollowed up to that point, and intervene to offer him what probably wasthe appropriate assistance.That breakthrough came along at a critical moment for computerand software makers.By the mid-1990s, despite the introduction ofgraphical interfaces that made computers vastly easier to use, only about aquarter of American households owned PCs, according to the NationalScience Foundation.Other studies found lingering unease and suspicionof them.A 1994 Gallup survey found that 59 percent of white-collar work-ers were reluctant to try new technology.Nearly a third of the workersincluding four out of ten women confessed to being outright afraid ofthe powerful, still-mysterious machines on their desks.Seeing a threat to its growth, the industry searched for ways to helppeople get over their cyber-anxiety.Stanford University researchers Computers as Social Actors theory seemed promising.CASA held thatusers instinctively treated computers as if they were people, not tools, andthat they might feel more comfortable if a computer acted more humanparticularly if its virtual personality was similar to the user s.As Micro-soft manager Chris Pratley explained in a 2004 Web log essay: Thetheory was that if you could provide an interface for the computer thatexpressed emotion and that you could interact with, you would be lesslikely to develop animosity toward your PC.and would actually beencouraged to learn and interact. Thus, software designers wedded CASA with Bayesian probabilityto create social interfaces peopled by agents who would guide actualhumans efforts to use the software.Instead of making characters realis-tic, however, they gave them the exaggerated, childlike qualities of Satur-day morning cartoon characters.(Intentionally or not, that fit theories of OOPS 188Nobel Prize winning behavior researcher Konrad Lorenz, who saw cute-ness as a potent cue for behavior, and noted that people were nicer to ani-mals whose physical features reminded them of human children.)Microsoft s initial attempt at a social interface, Bob, was devel-oped under the supervision of Melinda Gates, a Microsoft executive whoalso happened to be the wife of company chairman and cofounder BillGates.It was released in early 1995.When installed on early versions ofthe Windows point-and-click user interface, it converted the Windowsdesktop into a house with ten rooms including a family room, a study,and an attic that a computer user could decorate in a variety of styles,ranging from medieval to postmodern.The furnishings were actuallyicons that opened various applications, such as a letter-writing program, acheckbook, a calendar, and an e-mail program.The interface featured anumber of animated inhabitants whose personalities ranged, in the wordsof software historian Dan Rose, from happily enthusiastic to down-right obnoxious. Bob s cast included cartoon animals such as Hopper thestuffed blue bunny, Digger the worm, Blythe the firefly, Rover the dog,and Chaos the cat.The program also offered Orby, an anthropomorphicworld globe, and a William Shakespeare caricature who spoke in floweryElizabethan language.The most bizarre character was Scuzz, a surly ro-dent who took occasional bites of a package of rat poison, tormented Chaosthe cat, and dismissed the other guides as losers. (Oddly, there was nocharacter named Bob.)Bob had at least a few fans.Wall Street Journal technology criticWalter Mossberg lauded it as a bold departure that attempts to give non-technical people more control over their computers [ Pobierz całość w formacie PDF ]