Jon, I think that John Cowan hit the nail on the head. Researchers would not be interested in -- or be able to get funding for -- a straightforward "F# for JVM" because it's not likely to yield any particularly interesting new results that have not already been discovered by Don Syme's group at MSR Cambridge.
Researchers need to find novel fields to explore, which is why Martin Odersky and his colleagues at EPFL are able to get funding to work on Scala. It is precisely because Scala is different from F# that they are able to work on it. This leaves two other options. A commercial organisation could do it, but they would need to be persuaded that they it could make money. For example Sun might be prepared to do it in the interests of enriching the Java ecosystem... however they seem to be betting on Ruby rather than OCaml at the moment. Another possibility would be a tools vendor like Borland (or CodeGear), but it would only work for them if they could build an entire suite of tools around the new language and demonstrate far higher developer productivity versus Java. I do not think that the above is likely to happen, so the third option is that the open source community does it. Probably it would require a sufficiently committed individual developer to write the core as a proof of concept, which will draw in other developers. Who better than Jon Harrop, one of the most outspoken proponents of OCaml in the English-speaking world? ;-) Good luck Jon! Regards, Neil On 2 Nov 2007, at 15:51, Jon Harrop wrote: > > On Friday 02 November 2007 13:26, David Pollak wrote: >> One could always use Scala... :-) >> >> I have not used F#, but my understanding is that the F# team and >> the Scala >> team are personal friends and research collaborators and will often >> integrate the best of each language into the other. > > There are various aspects of languages like OCaml that I consider to > be of > critical importance. Pattern matching, type inference and static > checking are > among those. > > F# does quite well in reimplementing them with the exception that its > statically checked pattern matching gives poor error messages > (simply stating > when a pattern is incomplete but not describing how by giving > examples, as > OCaml does). > > However, Scala is far behind in many of these respects and also > incorporates a > lot of needless verbosity from Java. For example, Scala only recently > acquired the ability to perform the most rudimentary forms of > exhaustiveness > checking. The Eclipse plugin for Scala is behind F#'s (rather lame) > Visual > Studio plug-in which is, in turn, far behind the capabilities of > OCaml and > Emacs. > > These differences put OCaml a league ahead as far as I'm concerned. > Thanks to > Microsoft, .NET will catch up and I think it would be great if the > JVM could > catchup as well. For this to happen, someone must take the results of > previous research on ML and CAML and apply it to the creation of a new > language and development environment that targets the JVM. > > -- > Dr Jon D Harrop, Flying Frog Consultancy Ltd. > http://www.ffconsultancy.com/products/?e > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "JVM Languages" group. To post to this group, send email to jvm-languages@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/jvm-languages?hl=en -~----------~----~----~----~------~----~------~--~---
