You can read up on a few past discussions on bytecode vs. source, and even find Volta references on the contributor's list http://groups.google.com/group/Google-Web-Toolkit-Contributors
Fred Sauer [EMAIL PROTECTED] On Sun, Aug 24, 2008 at 8:55 AM, Tim <[EMAIL PROTECTED]> wrote: > > last I heard MS' Volta was buggy and slow and unacceptable for any > serious development. Have things changed? > > from MS' own description, Volta hasn't undergone yet any performance > optimizations yet http://livelabs.com/volta/docs/issues/. I wonder how > many underwater road blocks they'll face with that, since MSIL was > designed before the MSIL to javascript compilation process was > conceived. > > > > On Aug 24, 7:59 am, "Ian Bambury" <[EMAIL PROTECTED]> wrote: > > As I see it: > > > > Bytecode is intended for a JVM. As such you would either have to emulate > a > > JVM in JavaScript or at least partially reverse engineer the bytecode to > a > > point where you could optimise it. The compiler out put would be pretty > much > > unreadable (no -pretty available) so developing the compiler would be n > > times more difficult. Never mind less optimised and more stages to the > > compile. > > > > There would also be no way to stop people trying to compile any old byte > > code they produced whereas now if it ain't emulated, you can't use it. > > > > I'm sure there are people who would prefer to write in Fortran or Visual > > Turing, but Google will have a budget for GWT and can get more features > in > > GWT faster this way. > > > > You're hungry, someone offers you a free lunch, and you start moaning > that > > you can't pick the restaurant. :-) > > > > Ian > > > > 2008/8/24 Juan <[EMAIL PROTECTED]> > > > > > > > > > > > > > > I don't see how using byte code would be any better than using source > > > > code, since the source code has just as much information. > > > > > It's not about having more information, it's about the possibility of > > > using other JVM supported languages. Language lock-in seems like a > > > severe limitation for a platform like the JVM -that it's clearly > > > evolving into a multi-language platform- > > > > > > GWT probably couldn't generate fast JavaScript if it used Ruby > instead > > > > of Java, because it wouldn't have the type information it needs for > > > > optimization. > > > > > I assume this is only your guess. Anyway MS seems to be following this > > > way so soon we will see how fast it's (anyway I don't see why it > > > should be slower than Ruby -as they are both dynamically typed > > > languages-) > > > > > Anyway I was just wondering if anyone did know some solid reason as > > > why they didn't use bytecode interpretation. > > > > > Juan > > > > -- > > Ianhttp://examples.roughian.com > > _______________________________________ > > > > Stuff the environment - Print this email > > _______________________________________ > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Google Web Toolkit" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/Google-Web-Toolkit?hl=en -~----------~----~----~----~------~----~------~--~---
