> Shen is a splendid compromise, but I found myself wishing it came with Haskell's uncompromising purity and Standard ML's straightforward type system.
I guess Haskell/ML minded people have a reason to think this way, but I'd be more interested to know whether someone is doing anything interesting with Shen. Have you got any "production" code written in it and running to show off?
EDIT: sorry! ...fixed the gender bug.
This could also be a factor in the BODOL author's dismissal.
It's not like e.g. Python or Haskell have a fragmentation problem despite permissible licenses. Scheme has fragmentation problems but that seems to be a symptom of simple implementation and lack of a common set of libraries.
I think Shen is interesting, but if I used it I'd only be mining it for ideas to apply in other languages. It would make far more sense to me if he allowed derivatives iff they did not use the name.
It's the long winded, says the same things at least 5 (!) different times and ways, and therefore subject to many interpretations license, which is then subject to whims of the author or an external board (compare to Java, where in theory passing a suite of tests allows you to offer a version of Java(TM)).
I'm not going to invest serious effort in something that might get killed simply for political reasons including different interpretations of the license; even if I trusted the good will of everyone involved, sooner or later the players will change.
She's pretty fast to dismiss Shen.
You could create a more functional version as long as it had a switch or default to the normal behavior and it was blessed by the powers that be (see comments on license), but you'd be "fighting the system" and all that.