On Nov 15, 2007 5:09 PM, <[EMAIL PROTECTED]> wrote: > "Vesa Karvonen" <[EMAIL PROTECTED]>: > > Any chance that Alice ML might switch to the de facto standard (and > > unspecified, AFAIK) semantics of use? > > Brief answer: no. The slightly longer answer is that the "de facto > standard" semantics heavily relies on certain implementation techniques > that are not used in Alice ML. It also is largely meaningless in the > presence of concurrency. > > For an extensive answer, please see this posting, where I explain the > reasons for the semantics of 'use' and related features like 'eval': > > http://www.ps.uni-sb.de/pipermail/alice-users/2007/000783.html
Yeah, that is pretty much what I suspected. The reason why I'm asking about this is that I recently wrote a simple "Use Library" (http://mlton.org/cgi-bin/viewsvn.cgi/*checkout*/mltonlib/trunk/org/mlton/vesak/use-lib/unstable/README) to help with porting my libraries and programs to multiple SML implementations. The point is that by writing a single set of "use" -files for a library/program (e.g. http://mlton.org/cgi-bin/viewsvn.cgi/*checkout*/mltonlib/trunk/com/ssh/generic/unstable/lib.use), one can more quickly try/get an initial port to most interactive SML implementations. Fortunately, I think that it is possible to work around Alice ML's more restricted use function in this case. I'll have to try it tomorrow. (BTW, currently SML# has an even more restricted form of use. SML#'s use is a new form of declaration and not a first-class function.) -Vesa Karvonen _______________________________________________ alice-users mailing list [email protected] http://www.ps.uni-sb.de/mailman/listinfo/alice-users
