pointers

lucid at secret.org lucid at secret.org
Wed Mar 12 14:27:25 PST 1997


>as far as i can tell, every stated reason for not including pointers in
>Java, particularly `run-time safety' is bogus.  not all pointers are C
>pointers, which have given references a bad name, and there are lots
>of languages with perfectly safe pointers (often under a different
>name), and even C pointers are actually allowed by Ansi C to be rather
>more constrained
>by the Ansi C standard than many people think.

Out of general frustration, some off topic comments:

  I still continue to think that not including pointers in java
  was a seriously stupid move... I have needed pointers so many
  times it is unreal. *sigh*

  I still have this wired feeling that java programmers were intended
  to be idiots, thus they thought to remove the burden of anything
  so troublesome as pointers. Argh!

  And why the hell is all of the audio code in java.applet.*
  it should really be in java.awt.audio.*, the bastards...
  I dont even want to start about how poor their HTTP code is
  I seriously doubt the person who wrote it every read RFC 2068
  or even bothered to think there was ever going to be a HTTP/1.1

Now to a related topic:

  I wonder how hard it would be to start blowing away some of sun's
  code on the net in general... I would like to see a full HTTP/1.1
  networking package that worked in a way the meshed well with
  object creation. I would love to know how the heck to create
  an AudioClip object from a DataInputStream. If anyone out there
  has any insight into object creation, which seems totally and
  intentionally undocumented, please let me know. I am willing to
  put in the time to build the library and slap on a BSD license
  with manditory attribution, except for the obvious Object creation
  problem.

-bill




More information about the kaffe mailing list