mic_e.c:
  - Add PIC-E support: /P>mon
  - Is this necessary anymore given aprsd?  Look at putting kernel ax25
    support into aprsd.
  - Try to understand javaGate, get the latest version, and see if it's
    still OK with Alan to bundle it with this distribution.

aprsdigi:
  - debug this kprintf: protocol 0000 is buggy, dev ax0
    AF_PACKET vs. AF_AX25???
  - supress loops/dupes per-interface???
  - test kill_loop
  - Implement HOME:
     - use UDP instead of TCP?  Get an assigned number for the port.
     - sendto digi.aprs.net:port
    Example:
      N2YGK-9>APRS via RELAY,WIDE3-3,W2AEE-1
    gets sent via UDP to w2aee-1.aprs.net:12345 as ax.25 frame encapsulated in
    in the UDP payload:
      N2YGK-9>APRS via RELAY,WIDE3-3,W2AEE-1
    which is then received on W2AEE-1's port 12345
    This requires adding an interface type of UDP and that's about it for
    the UDP->AX.25 side and a hack lookup table (or just searching any non-
    alias call from the last digi via DNS) for the AX.25->UDP side.
    The home node may have further local routing information, e.g.:
      N2YGK-9 via WA2JNF-4,WB2ZII since not all IP->RF transitions are
    going to be single hop.
  - Implement compressed and/or decompressed GPS strings and posits.
  - Generally go over the latest DIGIS.TXT and make sure I match.
    - add maxhops for TRACEn-n.
  - Look at my notes from DCC proceedings APRS papers.
  - should SSID-routing always work or only for mic-e packets?

General:
  - test build with "standard" libax25 instead of my patched one.
  - parse_kiss/gen_kiss are misnomers.  Rename them. The data is 
    actually a raw ax.25 payload preceded by a byte of 0.  There are 
    no FESC/TFESC tricks or flags happening....
  - Look hard at libax25 and talk to maintainers about features wanted.
  - Make a new distro rpm and tarball.


