Glassfish is a Java-based application server. As such it competes with other quality open source products.
Schwartz' bet was that Sun will, in the long run, make more money by leveraging an open source community and selling support than by keeping its code proprietary. Financial types might want to measure support revenue, compare that to the previous year's sales revenue, and then look hard over the tops of their eyeglasses.
Schwartz will argue that would be wrong. Look at the revenue number's direction. If support revenues are on an upward trajectory, not just for Glassfish but for all the tools in Sun's Web Developer Pack, that's a turnaround. If losses are even narrowing, month to month, that's a turnaround.
Add in the value of community contributions, the goodwill from enterprise developers, the translation into hardware sales down the road, and please look out with a longer time horizon than 11 months before giving out any dirty looks.
Then recognize that without these moves Sun's software business would still be nosediving, he might say. Oh, and it's a one-way trip. You don't come back from the GPL. The best (financially) thing you can do is cut off support to the community, to walk away from the software entirely.
So, is Jonathan Schwartz' open source turnaround working?[poll id=34]