>>>I have no idea whether Gates knew about the TCP/IP work but Ballmer did. The networking team was part of the OS/2 development group at MSFT, not the Windows team. We had a much closer relationship with the OS/2 team but more on that below.
<<<
This is an odd debate, inasmuch as people think this TCP/IP business has some relevance to whether and when Microsoft had some aggressive, well planned Internet strategy back in 1991-1992.
Facts:
1. Customers, like myself, were yelling at MSFT to get this protocol into Lan Manager since the start of LM.
2. We wanted it primarily for LAN use then, tying our Suns to our PCs, something they (MSFT) didn't want to do particularly, since we would be using Sun servers and UNIX.
3. MSFT and IBM reps up to at least 1993 were trying to sell us Netbios networks and the proprietary token ring architecture. They were not driving the TCP/IP initiative, their big accounts were. One element of the final capitulation to customer desires was when Sun started to make it's own version of Netbios, so we could use Sun servers in any event.
4. OS/2 development, where the first good IBM/MSFT TCP/IP stack came from AFAIK, was driven by IBM even early on (you should hear IBM programmers diss MSFT on that point - IBM was even resorting to using the Borland C++ compiler and lying to MSFT about it!)
5. TCP/IP, used in a LAN, is just another, albeit better, LAN protocol, and that is how many customers saw it in 1992. Some people were thinking Internet, but most were thinking about how to set up barriers to keep it *off of* their internal TCP/IP networks. However, people were going to use it, were going to give lots of money and share to FTP, Wollengong, Banyan and other network protocol companies that did support TCP/IP, and MSFT had to stop that, so they made their own. Notably, Novell missed this wave and got punished.
MSFT did not want TCP/IP, were forced to implement TCP/IP, did not have TCP/IP because of any internet strategy, but because at that time they needed to squeeze the purveyors of alternative networks. Just talk to people who were doing LANs for big companies in those days, or to IBMers, for the truth.
That this fig leaf is being presented as evidence that MSFT was anything but unconscious where the Internet was concerned in 1992 is preposterous. The buzzword for intercompany operation and distributed computing in those days was WAN. The TCP/IP stack enabled them to have an Internet strategy later, but the fact is they were forced into it, feet dragging. At that point they still thought they were going to have something like Compuserve, totally proprietary. One only had to talk to them in those days to know how totally clueless they were about the Internet as a corporate culture. They had contempt for both it and the 'hippies' who ran it.
The best evidence is the unrevised first edition of 'The Road Ahead' by Bill Gates, versus the second run (which was published without explicitly called it a second edition, I believe), which really shows better than any other evidence the exact moment in time when they realized that the Internet was important.
Cheers,
Chaz |
