M17 had promise, but looks like the foundation is in the hands of opportunists who weaseled their way in. Given this history, I’m glad it’s being removed.
https://groups.io/g/OpenDV/message/2311
It is with regret, and a certain amount of relief, that I have removed M17 from the MMDVM Project.
I have two sets of issues with M17, administrative and technical.
Firstly the administrative side of M17 is very worrying, and even more so in recent months. A couple of years ago, M17 received $478,900 in grants from the ARDC to develop M17. I very much feel that ARDC should take a closer look at how this money was spent.
The new M17 Foundation isn’t much better. A number of the M17 stalwarts were excluded from it when it was formed, rather more sadly, the M17 Foundation make no mention of a number of people or organisations that helped them to get where they are now. This is particularly troubling as it is a rewriting of history and not apportioning praise where it is due. A lot of people put a lot of time and effort into M17 and to not get their due is dishonest of the M17 team. An example is the fact that only one commercial entity has put their money where their mouth is, and that is Jerry of Connect Systems Inc. however his contribution to the project has been belittled online by the M17 team (as has mine), despite it being the only way to get a commercial M17 radio, he should be praised not dismissed out of hand.
I have heard rumours that the M17 Foundation is looking at charging commercial entities a royalty to include M17 in their equipment and to use their logo. This is not in the spirit of open source, and has not been the route followed by the MMDVM.
Secondly technical. When started, M17 was proudly created by people who said that they brought fresh thinking to digital voice, I would say that it was characterised more by a combination of arrogance and stupidity. The fact that none of them had ever operated a digital voice radio, let alone studied a DV mode, was seen as a positive. One of the key members designed it like a packet radio system where each block (or packet) of information needed to be received perfectly. This is not a useful attribute to have. As originally designed the synchronisation patterns were literally one bit different from each other, which is useless in an environment where signals are routinely corrupted to a greater or lesser extent. Over the first six months of my involvement I managed to get them to change the synchronisation vectors to be something more reasonable, and got them to add the CAN (Channel Access Number) to allow for some sort of channel sharing between M17 systems.
I also added things that we take for granted in other DV modes like embedded GPS data and short text messages. These both run in parrallel with the audio like in D-Star and other DV modes. There were still five big issues:
– A networking protocol designed before the RF protocol had stabilised and included ideas that didn’t make sense later in development.
– A very weak end of message indicator, a single bit, which I argued against and got them to change but then they retained in the official specification. This is an exceptionally weak part of the protocol and the fact that they couldn’t see that showed that many of the M17 team needed to attend Digital Voice 101.
– Inclusion of optional strong encryption, this is against regulations in most countries and shouldn’t even be thought about. Oddly enough encrypted M17 wouldn’t pass through an MMDVM based repeater.
– An orphaned vocoder, it doesn’t sound very good either. AMBE in all its forms is considerably better. Indeed later versions of Codec2 which take cosiderably less bandwidth than the 3200 mode chosen for M17 sound far better.
– The wrong type of FEC, applied badly.
At various times I have brought all of these items up, but no-one seemed to be interested.
As of last Wednesday M17 has been completely removed from the MMDVM Host, the M17 Gateway and M17 Client have already been removed from my GitHub, however it still hangs on in the modem and hotspot firmware, but this too will be excised soon. Future hardware, software, and firmware will not include M17 at all.
It shouldn’t really be a problem since the M17 team have said that my software isn’t important, so no one should really notice its removal from the MMDVM. It is being removed from WPSD and I believe that Pi-Star is going down the same road.
I would like to express my appreciation and admiration for the OpenRTX project and team. They have worked hard to get M17 running on undocumented hardware and have done wonders. I hope at some future time I’ll be able to collaborate with them on some future open source DV mode. They have been the real workers behind the M17 project and have probably not received the praise, nor the renumeration, that they so richly deserve. To my mind, and to others, they are the real developers behind M17 and they deserve better.
There is a need for an open source DV mode, however M17 is not that mode.
Does anyone want to buy a CS7000-M17?
Jonathan G4KLX