Cheryl,
Although in your words from a previous post, I'm one of the "tweezerheads" from Redmond, I do appreciate your attempt to appear unbiased.
I'd like to respond to a few statements you made in the interest of promoting balanced discussion.
Assuming that MSFT had marketed MS-DOS on their own, proprietary Microcomputers (ala Apple), I don't see how anyone could complain about anti-trust (given the current complaints). They could have integrated any pieces into the OS & sold the it bundled with pastrami, if they wanted.
It's interesting that you believe if only Microsoft sold the hardware in addition to the OS that runs it (like your employer, Sun) then, and only then, should all be forgiven. I wonder how that level of additional bundling would have enhanced the innovation that's occurred in the PC industry, the same innovation which is threatening Sun's overpriced server business with high-powered, inexpensive hardware and software.
They [Microsoft] had a tendency to keep the OS specs closed. I belonged to an informal coterie of determined hackers who uncovered the "mysteries" of MS-DOS & tried to inform the public (pretty much a waste of time).
...
Now, however, they have more ways of locking out a competitor than just manipulating the api's that they publish.
Interesting. I also reverse engineered a great deal of MS-DOS and wrote a multitasking system that was used by a major shipping company to operate a 3270 emulator along with a handheld device package tracking system. The multitasking system was also used by a large number of end users. In all of my reverse engineering, I never found undocumented APIs that were exploited by Microsoft applications to some great advantage.
Since that time, I worked on Win95. I can assure you that it was our strict policy to document all APIs that could be useful to applications, Microsoft or otherwise. The only ones we did not document were used by system components, may have needed modification in later OS versions, and were sometimes potentially destabilizing to applications that used them (not so with system components). If some application used these APIs, it was not due to any concerted "lockout" attempt by Microsoft. Usually, I consider the "undocumented API" complaint to be an excuse for those who can't compete on application functionality alone. Since you were a member of this informal coterie, I'd be interested to hear of any API you found that was exploited by a Microsoft application to the detriment of its competition.
This would require MSFT to publish the OS specs to everyone, the way Java is published.
Windows API specifications are better documented than any comprehensive API of which I'm aware. In addition, there are an incredible number of 3rd party texts on Windows and its APIs. If Microsoft published specs the way Sun published Java's, then anyone who read them would be at risk of being sued. That is effectively what Baratz told Hewlett Packard when he said that by reading the spec, they were bound to conform to Sun's rules even in their clean room development, no?
Finally, IE on Unix does not have the level of OS integration that IE on Windows does. I think that's one of Microsoft's points.
Thanks,
Mike |
