This number tells so much about the improvements in developer productivity over the years.
Adjusting for TVM, the same revenue is now supported by far fewer engineers (obviously not all 3,200 employees where engineers). The Netscape engineers were highly skilled, so the change in the number of engineers cannot be explained away by differences in engineering skill/capability.
It doesn't really say much about developer productivity. What are you comparing this to? You might also be forgetting the insane number of (many 'enterprise') products Netscape was trying to churn out at the time. See:
The browser alone with all of its extra doodads ran on three radically different platform families (Windows 95/NT, Mac OS, Several proprietary Unixes). And that's just the browser. Throw in running one of the top traffic sites at the time.
The internet advertising market which drives a lot of the revenue at many current similar startups was far smaller and less mature, as well. It's difficult to see how we can draw any sensible conclusion about developer productivity from all this.
There should be a "NSFW" equivalent for articles with potentially harmful levels of smug. (I've read a lot of articles by jwz over many years, and he is how he is, but still...)
This number tells so much about the improvements in developer productivity over the years.
Adjusting for TVM, the same revenue is now supported by far fewer engineers (obviously not all 3,200 employees where engineers). The Netscape engineers were highly skilled, so the change in the number of engineers cannot be explained away by differences in engineering skill/capability.