Re: Java virtual machine as target language for C/C++

Our moderator writes:
>[I may be an old grouch, but this is starting to smell like the UNCOL black
>hole. Yes, you can turn your intermediate code into native code with some
>amount of effort on some number of platforms, but the details always end
>up killing you. -John]


John, methinks that you are _not_ an old grouch. You've just been here
before (as have many of the rest of us).


UNCOL is a beautiful idea which has been shown to never work in practice.
But it has such allure that people keep coming back to it time and time
again. In just the recent past, we have been offered ANDF, which is dying
(has died?) without too much fanfare (i.e. a whimper rather than any spec-
tacular bang). Now Java bytes codes are being held out to the computing
community as the new solution to universal code portability. I gather that
Java byte codes have already acheived more success in the marketplace than
have most of its predecessors, but that fact doesn't imply that the funda-
mental problems with the age-old UNCOL idea have been eliminated. As usual
with these schemes, the inherient problems have just been cleverly side-
stepped (for the moment and for a limited set of targets) through the
judicious application of appreciable quantities of labor.
--


-- Ron Guilmette, Roseville, CA ----
---- E-mail: rfg@monkeys.com -------
[If you're willing to live with the cost of interpretation and a somewhat
constrained environment, there have been lots of quite effective
machine-independent byte code environments, of which UCSD Pascal and Pick
have been probably the most well known. -John]
--

Return to the comp.compilers page.
Search the comp.compilers archives again.