Re: Java (bytecode) execution speed

Supersedes: <heap-20070429204231@ram.dialup.fu-berlin.de>

  I would like to add two quotations to this thread:

      ?Your essay made me remember an interesting phenomenon I
      saw in one system I worked on. There were two versions of
      it, one in Lisp and one in C++. The display subsystem of
      the Lisp version was faster. There were various reasons,
      but an important one was GC: the C++ code copied a lot of
      buffers because they got passed around in fairly complex
      ways, so it could be quite difficult to know when one
      could be deallocated. To avoid that problem, the C++
      programmers just copied. The Lisp was GCed, so the Lisp
      programmers never had to worry about it; they just passed
      the buffers around, which reduced both memory use and CPU
      cycles spent copying.?

<XNOkd.7720$zx1.5584@newssvr13.news.prodigy.com>

      ?A lot of us thought in the 1990s that the big battle would
      be between procedural and object oriented programming, and
      we thought that object oriented programming would provide
      a big boost in programmer productivity. I thought that,
      too. Some people still think that. It turns out we were
      wrong. Object oriented programming is handy dandy, but
      it's not really the productivity booster that was
      promised. The real significant productivity advance we've
      had in programming has been from languages which manage
      memory for you automatically.?

http://www.joelonsoftware.com/articles/APIWar.html

Supersedes: <heap-20070429204231@ram.dialup.fu-berlin.de>