3 ms·

Oh, really? You're actually making the opposite argument: divided technical leadership is what drove most progress in personal computing. Unless you yearn for t

Oh, really? You're actually making the opposite argument: divided technical leadership is what drove most progress in personal computing. Unless you yearn for the days of the IBM mainframe... The Economics of the Microsoft Case, by perhaps the best technology economist there is, Timothy Bresnahan of Stanford: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.24.7515 http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.24.7...

CamperBob 16y ago
Divided technical leadership was, indeed, important, up until the point where the hardware platforms converged. At that point it became a stumbling block to progress in application software. The problem was that the hardware platforms were converging in terms of functionality, but not at the all-important register and ISA level. Consequently the OSes were not converging at all. Let's assume I don't have time to read a 27-page .PDF -- do the authors address the economic impact of the lack of a common platform for application development? It's possible that somebody would have duct-taped some sort of graphical shell to CP/M and gained some traction, but at best, they'd have ended up doing exactly what Microsoft did by duct-taping a graphical shell to MS-DOS. People who walk into computer stores want to be told what to buy. In the end it won't matter, if you believe that HTML is the next platform. In that case we'll all end up running dumb terminals anyway, and it won't matter what chips and what OSes are handling the UI. But that was never something that was going to happen in the 25 critical years between 1985 and 2010, when the commodification of high-performance PC hardware took place. We will not see another "window" of opportunity like that, and we absolutely would not have seen that one happen without Wintel, or something very much like it.

CamperBob 16y ago

Divided technical leadership was, indeed, important, up until the point where the hardware platforms converged. At that point it became a stumbling block to progress in application software. The problem was that the hardware platforms were converging in terms of functionality, but not at the all-important register and ISA level. Consequently the OSes were not converging at all. Let's assume I don't have time to read a 27-page .PDF -- do the authors address the economic impact of the lack of a common platform for application development? It's possible that somebody would have duct-taped some sort of graphical shell to CP/M and gained some traction, but at best, they'd have ended up doing exactly what Microsoft did by duct-taping a graphical shell to MS-DOS. People who walk into computer stores want to be told what to buy. In the end it won't matter, if you believe that HTML is the next platform. In that case we'll all end up running dumb terminals anyway, and it won't matter what chips and what OSes are handling the UI. But that was never something that was going to happen in the 25 critical years between 1985 and 2010, when the commodification of high-performance PC hardware took place. We will not see another "window" of opportunity like that, and we absolutely would not have seen that one happen without Wintel, or something very much like it.