Recent versions of firmware will be removed

[VK3KYY]

Whatever, it is a pity that this project is dead. This was the only open source community radio firmware in a well usable state. And it is probably the last. Most projects that started died because of the complexity of the work. So, we will be left with OEM firmware at the mercy of the manufacturer. Best bet is probably Anytone, which at least sometimes listens to its user base and does some firmware updates here and then. Most other, like Yaesu, Kenwood etc do never publish any updates. If we want new features, we'll need to buy a new handset again. So sad

Well, according to KC7RBW, he is actually "hopeful" now (and also "relieved" and some other sentiments that he wrote). He said that it does not mean that the project is dead. But he also added that he was not going to become the new maintainer.

Personally I feel sorry that it has to come to this. But oh well.

I do thank all the contributors to this project, especially Roger.

I've posted in another thread, that I think this is dead if the project has to remain GPL, becuase :-

1. GPL does not seem to allow any additional conditions of use to be applied, which are essential before anyone will spend their precious free time working on the project
2. The project is almost certainly in breach of GPL, e.g use of hacked closed source binary from the official firmware, and for including other components and source code with various other potentially incompatible licenses.
I'm not an license expert, so I have no idea what else is wrong with it.



Perhaps (2) can be resolved and stay within GPL, e.g. by making loads of changes to the firmware, but (1) is a deal breaker for everyone as far as I can tell, and can't be resolved without removal of GPL

[VK3KYY]

It's hard to fight GPL. It's got a nickname as "copyleft" for a good reason!

I don't know about flighting GPL.

From what I have read it may be possible for GPL to be removed, if the copyright holders all agree, or if all copyright holders don't agree, then the code that is copyright those people need to be removed.

See.
https://softwareengineering.stackexchan ... sed-source

Albeit this is stack overflow and not some legal adjudication

I'm pretty sure when KC7RBW first raised his copyright on the firmware, I thought that all of his submissions were so old that they had been superseded and rewritten, because he had not contributed since August 2020, and in the process of some big overhauls of the code, mainly undertaken by Daniel, a lot of the code got written.

And from what I can remember, he agreed that if all his code has been rewritten and superseded than he didnt have a claim of copyright to enforce the GPL

But when I analysed the current code, and checked whethr there was still items from KC7RBW I could see that he had a valid copyright claim.
I already posted this somewhere that I was incorrect in my initial ascertion that his code had been superseded



Additionally, if the code already falls foul of GPL, e.g. because of conflicting licenses or the elephant in the room of the 30% of the binary hack from the official firmware, was it every really GPL.

I"ve no idea.

I'm pretty sure attaching a GPL license to someone elses code (or binary) does not make that code GPL, and but other individual files will be GPL if they the GPL header.

Huge grey area.

I'd however be interested to know if it was possible to remove GPL and still remain as something I think gets called "Source available" - though this seems a vague term, and is not often used.
or perhaps creative commons..

Again I have no idea.

Edit.

Removal of GPL from one version, does not mean discontinuation on a GPL version if people thing they are willing to work on the old GPL version

This is what has happened to various Linux distributions, which became closed source commercial

In this case the proposal was not commercial or closed source, its has to be non-commerical, and if possible "source available" or something similar.

But I seriously doubt, considering KC7RBW has not contributed since Aug 2020, and has said he will not maintain it, that you will find any skilled devs to take it over. All the Italian devs are already on their OpenRTX project "inspired" by the OpenGD77 and I would have thought are now unlikely to want to take over a project with a dodgy license.

[KC7RBW]

This was the only open source community radio firmware in a well usable state.

It stopped being open source when the maintainers stopped sharing the source code. I never wanted the project to die, I only wanted it open again.

GPL does not seem to allow any additional conditions of use to be applied, which are essential before anyone will spend their precious free time working on the project

From section 6:

You may not impose any further restrictions on the recipients' exercise of the rights granted herein.

But that doesn't have to mean that nobody will spend their time working on the project. Someone else making money from your hard work doesn't take any money out of your pocket unless you were making money too. Millions of developers contribute to free software projects merely because it's a fun hobby, or because it solves their own problems. Meanwhile hundreds of companies like Red Hat make money directly off those projects, and millions of other companies save billions by using that software instead of some commercial alternative. In many cases no viable commercial alternative exists. And yet we keep coding, for our own reasons, because none of that economic activity takes anything away from our own enjoyment of the hobby. At least that's how it works for me and for millions of others.

You could, like I do, shrug and say "well, someone else is making some money off my work, so what? I don't care."

The idea behind copyleft isn't that people shouldn't make money, but that when they do they need to share their work just like you did. It's a bit weird, but it hasn't killed Linux - in fact it's helped it thrive!

And given that several of the developers do sign on to the non-commercial clause, despite it's conflict with the terms of the license itself, it would be risky business to try to make real money off it.

The thing I still don't get is how keeping the source code private can stop someone from selling firmware (or installation services) if the firmware builds are still available.

From what I have read it may be possible for GPL to be removed, if the copyright holders all agree, or if all copyright holders don't agree, then the code that is copyright those people need to be removed.

If you were to choose a suitable copyleft license (one that will keep the source code open) with a non-commercial clause I could sign on to that. No need to re-implement my meager contributions. The only license I'm aware of that I'd find suitable is CC BY-SA-NC but there could be others out there I haven't read yet. CC licenses weren't meant for software (but for photos, music, and video, etc.) but lots of people do apply them to software anyway (including the company I work for, though they usually choose Apache-2.0 which is far more permissive, or MIT-0 which has no conditions at all).

And there are several other developers you'd need to get permission from. I can't speak for them.

But I don't think that solves any problems you care about that aren't already solved or just can't be solved. The NC clause, incompatible though it is, is going to keep Big Co away, and as long as we're sharing firmware builds nothing is ever going to stop the little eBay leeches, as much as they suck. I have no idea if Radioddity or some other Chines company would have the guts to sell our firmware on their radios, but as long as they also give me the source code so I can make my own changes and share them with this community I don't mind because I still get what I want - a radio that works better for my use than the one I bought.

I won't reiterate my arguments that the AMBE codec problem is solvable.

As long as the source code stays open so that I can make my own changes on my own, and so can we all, I'm satisfied. Literally the only thing I objected to, on the basis of my own copyright, is that you stopped sharing the code but kept sharing the firmware, and since you shared the code I no longer have any objections at all. If you want to stop taking contributions of any kind that's fine too - the license doesn't force you to review pull requests or resolve anyone's issues.

[DU2XXR]

I think Andrew has got the spirit of open source software and collaboration-- and licensing, as far as GPL goes-- in the right interpretation. It's about freedom as in free speech (freedom to distribute, build, enjoy) and not necessarily freedom as in free beer.

[DJ0HF]

As long as the source code stays open so that I can make my own changes on my own, and so can we all, I'm satisfied. Literally the only thing I objected to, on the basis of my own copyright, is that you stopped sharing the code but kept sharing the firmware, and since you shared the code I no longer have any objections at all. If you want to stop taking contributions of any kind that's fine too - the license doesn't force you to review pull requests or resolve anyone's issues.

Hi Andrew, are you saying I can put my package for the blind back up as it is based on the last source that was released by Roger ?

Ian
DJ0HF/G3ULO

[oh1fss]

I think Andrew has got the spirit of open source software and collaboration-- and licensing, as far as GPL goes-- in the right interpretation. It's about freedom as in free speech (freedom to distribute, build, enjoy) and not necessarily freedom as in free beer.

I agree.

Open source gives you freedom to get name and fame - not necessarily free beer. Because of their achievements, Linus Torvalds and Joe Taylor K1JT are highly respected. They could get revenue from consulting and lecturing - and could get a job with high salary if they wanted.

[do1jml]

Sorry to come to that subject again. Am I right in my understanding that the present GD-77 firmware includes a binary blob from the manufacturer and that this blob corresponds to the voice codec (ambe2+)? That means that the code to compress and decompress audio data runs on the MK22FN512 ARM processor, right? So, in theory, it would be possible to replace that blob with different code to compress and decompress audio data and we would have an entirely open firmware?

[ok1pt]

Sorry to come to that subject again. Am I right in my understanding that the present GD-77 firmware includes a binary blob from the manufacturer and that this blob corresponds to the voice codec (ambe2+)? That means that the code to compress and decompress audio data runs on the MK22FN512 ARM processor, right? So, in theory, it would be possible to replace that blob with different code to compress and decompress audio data and we would have an entirely open firmware?

Hi!
I'm afraid that it's not possible, because the algorithm is patented, and you can't reimplement it neither by HW or by SW means unless you have an agreement with the patent owner.
With regards,
Pavel

[do1jml]

I'm afraid that it's not possible, because the algorithm is patented, and you can't reimplement it neither by HW or by SW means unless you have an agreement with the patent owner.

First, the original patent for the ambe codec (US6199037) is expired and, second, the DMR standard does not mandate the use of a particular codec.

I would still like to have an answer to the question whether the codec in the original firmware runs on specialized hardware or runs on the arm processor.

[F6GVE]

I would like to say a big big big thank you for spending many years to make this fabulous project.
This is the OM spirit, as it was in the old years.
I don't know/understand what the problem is, but I guess it's mainly a human problem, from acidic and jealous people who have easy criticism but don't know how to do much concrete and really useful with their 10 fingers. Such people always look for legal ways to put obstacles in the way of those who are creative and inventive.

You did something great, very great, huge, and always with a lot of patience towards all of us

THANK YOU THANK YOU