pulseaudio/todo

*** $Id$ ***

Build System:
- Remove symdef files and use macros (like most other projects)
- Use own name mangling scheme instead of ltdl's, which will eliminate the
  need for .la files or extra trickery.
- build pulsecore only statically by default, it's not a public API yet

Porting:
- rtp module ported to Win32 (sendmsg/recvmsg emulation)

I18N:
- iconv stuff sent from utils to server (UTF-8)
- iconv sample loading in server
- Document utf8.h, timeval.h and util.h
- gettextify polypaudio

Cleanups:
- drop dependency of libpolyp on libX11, instead use an external mini binary
- merge module-oss-mmap into module-oss
- module-tunnel: improve latency calculation
- use software volume when hardware doesn't support all channels (alsa done)
- silence generation should be moved into the core to avoid races and code
  duplication in the backends

Auth/Crypto:
- ssl
- key rings for auth
- challenge response auth
- sasl auth 

Features:
- alsa mmap driver
- alsa driver with hw mixing
- allow passing data with shared memory between local clients and the server
- "window manager for sound"
- hooks for pa_source_output_new, pa_source_disconnect()
- chroot()
- use scatter/gather io for sockets
- CODECs to reduce bandwidth usage (plug-in based)
- multiline configuration statements
- paplay needs to set a channel map. our default is only correct for AIFF.
  (we need help from libsndfile for this)
- Fix a way for the threading API to handle state and subscription callbacks
  in a nice way.
- examine if it is possible to mimic esd's handling of half duplex cards
  (switch to capture when a recording client connects and drop playback during
  that time)

Long term:
- pass meta info for hearing impaired
- X11: support for the X11 synchronization extension

Backends for:
- portaudio  (semi-done)
- sdl
- xine (needs update)