In his blog, Lorenzo Puccetti asks the question, why isn’t JavaSpaces as successful as it should be? To answer this question he turns to his readership to ask them that use it a few questions. Lorenzo has been working with JavaSpaces for the last 12 months and could no longer see another way of solving the same business problem

how come there is only one proper open-source implementation (Blitz) and one proper commercial solution (gigaspaces) ? why does it seem that only few patterns have been described to work with spaces ?

Lorenzo states that he can see that one needs to adjust their mindset when working with this technology. There are also other problems that one has to deal with that do not appear in non-distributed systems or with local computations. What ever the problem, Lorenzo is asking for your input for answers to the question, why?

My guess is that JavaSpaces implementations: 1) may not be abstracted enough -- there may be guts or bones showing that people don't want to deal with. 2) still have that Jini smell to them that is hard to get rid of, even with the strongest air freshener. 3) are one solution of many, more popular solutions in their "pattern-space".

JavaSpaces ? 1. Because Sun is not promoting Java Spaces on main page news. Last time I heard about this was 7 years ago. Many people like me don't even know there is any progress going on on Java Spaces or JINI.