Yes, a lot of people like to talk about how easy it would be to significantly better AX25 as we know it, and several have given it a try... I have seen several such attempts, and been involved in a couple of them.
But guess what? It's kind of like HF Packet, I guess. 73 DE Charles, N5PVL ----- Original Message ----- From: Brian Lloyd To: Charles Brabham Cc: Lee McDaniel ; [email protected] Sent: Saturday, February 11, 2012 5:07 PM Subject: Re: FLDIGI and Packet Operation: openpty() On Sat, Feb 11, 2012 at 2:46 PM, Charles Brabham <[email protected]> wrote: To hear some people's political opinions about Packet, you would think that it would not work at all. Only thing is... It does work, and has worked FB on HF for decades. Not only that, but AX25 Packet does things that most other digital modes simply cannot do. If you leave off the politics and just stick with observed reality, the fact is that no other digital mode has been developed yet to do what Packet does - but better. Q15x25 was a nice try, but in on the air tests, regular old HF Packet was more robust, and provided better throughput despite the lower baud rate. We had great hopes for Q15x25, but it did not prove to be better than 300 baud Packet. The old urban myth about Packet being technically unusable on HF was started in the late 80's by amateur TCPIP buffs who were frustrated because nobody liked amateur TCPIP. The situation hasn't changed... HF Packet is still going strong, and the myth-makers are still frustrated because nobody likes amateur TCPIP. And here I thought that the old AX.25 vs. TCP/IP wars were over. I know I moved on. I guess there are die-hards everywhere. :-) And for what it's worth, if we compare the success of TCP/IP with the success of any flavor of X.25, including the mis-named AX.25, well, the jury is definitely in on that topic. :-) But getting back to AMPR on HF, consider: that something works does not mean that it is optimal and cannot be made better. And my comments about both the FSK modem and the AX.25 protocol specifics are true no matter who speaks them. Phil Karn and I did an excellent analysis of the AX.25 shortcomings back in the 1987 or 1988 ARRL Computer Networking Conference. We ran computer simulations and then backed them up with real-world observations of live networks. And the observations and conclusions haven't changed, even after 25 years. (Wow, has it been that long?) So, just because it works doesn't mean it can't get better. And just because someone who likes TCP/IP says it, doesn't mean it isn't true. Frankly, I couldn't care less about running TCP/IP over an HF link but I do think that it is possible to craft a better protocol that will do what AX.25 is doing, but do it better and more reliably. It would be quite interesting to see progress in this area rather than arguments about how adequate it is. 73 OM -- Brian Lloyd, WB6RQN/J79BPL 3191 Western Dr. Cameron Park, CA 95682 [email protected] +1.767.617.1365 (Dominica) +1.916.877.5067 (USA)
_______________________________________________ fldigi-alpha mailing list [email protected] https://lists.berlios.de/mailman/listinfo/fldigi-alpha
