This email from Nikolaj Nyholm from Ascio in Denmark reflects a few I have had. The idea that even trying to compete with Microsoft is unachievable if you say in advance that is your goal.
Netscape is a common example.
I always believe that Netscape failed not because it was too audacious but because it was not audacious enough. When Mark said this [see the email below] it was bravado to be sure but more problematic was that there was no execution against this goal. To be successful it would have been necessary to develop more than an application – even one as important as a browser. Sure, it would have helped if Netscape was free. But even then, it was just an application.
Instead of developing an API for all access to the network Netscape moved from the browser to server software for the enterprise. It was then competing with Microsoft on the desktop and IBM/Sun/Oracle etc in the enterprise.
These were fundamental mistakes. Although the $5bn sale to AOL looks good with hindsight, it could have been so much more if Andresen’s statement had been a vision with appropriate mission and goal driven execution.
My idea of an OpenWin API is not an application. It is a layer of API’s that can be hooked by many applications and services. The only applications that would be forcefully hooked would be IE, Outlook, Office Apps, and others that make calls to the underlying wininet.dll.
The explicit goal would be to seperate Microsoft’s Apps from its OS. And to build a layer that other apps could choose to call and to customize in their own way.
So in short, I don’t think it is impossible to compete with Microsoft, or to declare that your goal is to do so. It is a problem if you don’t go the whole way in executing on that goal.