> Michael Turner wrote:
> <snip>
> > > Java isn't an interpreted language.
Ben Hutchings:
> > A matter of perspective, I guess. To me, virtual machine
> > code is interpreted; JIT compilation is an optimization of
> > an interpreter.
Ben:
> Sure, but Java bytecode is something different from the Java
> language.
Yes, as I pointed out in an earlier posting: you could probably
run Visual Basic apps over JVM. (*Shudder*.)
This is academic, though. Almost all Java is compiled to JVM
bytecodes, and almost all JVM bytecodes come from Java
compilers. Language-subculture membership is 9/10ths of
the law.
> Garbage-collection rather than explicit freeing generally saves
> time, actually.
Any numbers on this? Surely it's application-dependent, and
GC-style-dependent.
I love garbage collection as a concept, and it was a major
undergraduate research preoccupation of mine at one time.
There is, however, still something about it that smacks of
the 12-year-old boy who, upon being asked to clean his
room, starts designing robots to do it instead. ("Jimmy...
are you done yet?" "Almost, mom....")
In professional life, most of the memory leaks I've fixed
have been either in very new code or in very bad code.
If you don't know enough about your data structures to
know when they are collectable, you probably have a
mess on your hands anyway; leaks are just a symptom.
I'm sure this is the stuff of which holy wars are made, so
maybe I should let it rest.
And anyway, how is this relevant to the thread topic? I'm
no longer sure.....
-m
leap@gol.com
[ Did you check the archives? http://www.appelsiini.net/keitai-l/ ]
Received on Wed Apr 25 10:47:40 2001