At 3:34 PM +0900 3/5/02, Matt Armstrong wrote: >Dan Sugalski <dan / sidhe.org> writes: > >> At 6:37 PM +0900 3/4/02, Paul E.C. Melis wrote: >> >>>Of course, it will take some time for Parrot to mature, but I wonder >>>what the future plans are for Ruby with respect to Parrot... Matz? >> >> We've already covered this, and I expect by the time this makes it >> out you'll have at least a dozen references to Matz's decision on >> this. Parrot won't be the reference engine for the next version of >> Ruby. (A decision I agree with, honestly) Neither will it be the >> core engine for the next version of Python. >> >> Parrot *will* run both Ruby and Python code in addition to Perl (and >> Scheme) code. My goal is to run them all faster than their reference >> implementations (well, except for perl 6, as we'll be the reference >> implementation). If we manage that we all win, as there'll be a fast >> interpreter for them all. If we don't, well, everyone still wins, >> since that'll mean the other interpreters will be even faster. I >> like fast. Fast is good. :) > >I notice Parrot has string operations, and I have this impression that >Parrot understands Unicode only. True? False. :) >I.e. most of the world punts and goes to Unicode only for the internal >representation of strings, while Ruby will be able to deal with >strings in their native charset. \ Parrot's going to handle strings in their native character sets. We're going to punt and use Unicode as a pivot when we need to convert between two charsets we don't have a direct conversion for, and it'll be the lowest common denominator for truly mixed-set character usage, but unless you go doing cross-set things you'll stay in a native set. -- Dan --------------------------------------"it's like this"------------------- Dan Sugalski even samurai dan / sidhe.org have teddy bears and even teddy bears get drunk