The Future of Meshtastic and Community Involvement

[daviesgeek]

First of all, I want to make it very clear what this is not. This is not a post intended to criticize the developers or past decisions. I want to thank the developers for their hard work and dedication to the project. This also isn’t a plea for me to join the core team or any sort of similar misunderstanding. I want to be very clear about my intentions: this is simply an effort to understand how we, as the greater community, can both better comprehend and better contribute to Meshtastic.

All that said, a number of decisions have been made with little to no communication about the change, why it’s made, and where the direction of particular features may be headed. I don’t know what the actual solution is, but some of these decisions need community input as well. I understand that collecting this feedback could be difficult, but that’s what an RFC-like process is for (maybe the meshtastic/rfc repo could be of some help in this). In particular, the router_client role change left a lot of people confused and with no recourse for options.

The future of Meshtastic at the moment is relatively unclear, at least to me. Maybe there are other channels of communication that I’m missing, but I only know of future and upcoming work from random Discord messages that are sent that I happen to see or that happen to be passed along. I’m aware of the 3.0 work and the routing changes that are happening, but most people probably are not. Again, I’m not entirely sure the solution here, but as the larger community, we need to understand the future vision of the project.

Lastly, there are problems around documentation. I understand, documentation is always a thorn in the side of developers. As I mentioned in earlier conversations, though, I feel like the UIs are missing information that could be vital to the end user’s configuration and setup. For instance, some of the public MQTT kerfuffle could have been solved by better messaging around the MQTT configuration.

Again, I greatly appreciate the hard work and dedication of the developers, but there are areas where improvement is needed. The lack of clear information about changes and future directions can leave community members feeling disconnected and confused. By involving the community more in decision-making and enhancing documentation, we can ensure that everyone has a better understanding of the project’s trajectory and can contribute more effectively. I look forward to contributing to the solution and helping bridge the gap between the community and the project leaders, ensuring we can all work together towards a better future for Meshtastic.

[kennytaylor661]

Well said, Davies. The router_client deprecation caught me by surprise as well. From what I've gathered, it was intended to prioritize well-placed repeaters and reduce congestion in large urban environments. But not every location is as dense as San Francisco or Los Angeles. In my area, there are zero hilltop repeaters, and it makes sense to designate home rooftop nodes as router_client until we get some well-placed nodes. We lost that flexibility and don't quite understand why.

The MQTT issues were similar to the debate over APRS-IS bridging. Some users really want to be off-grid, and MQTT bridging prevents it.
The solution could be as simple as adding a no-bridge flag to messages and position reports. The app could even prompt a user for permission to bridge to the Internet on initial node setup. Polling the user base would provide some insight into how widespread the concern is and how to handle it.

Communicating a clear vision is critical to open discussion with the community. The platform is awesome, and I have no doubt the dev team knows what they're doing. There's a community of users out there investing heavily in building the mesh, and it's important to keep them involved in the discussion.

[rcarteraz]

In my area, there are zero hilltop repeaters, and it makes sense to designate home rooftop nodes as router_client until we get some well-placed nodes. We lost that flexibility and don't quite understand wh

It actually doesn't. This is the issue, too many people did not understand this and it was seriously degrading a lot of networks. If every node has prioritized routing, then no nodes have it. This is why that change was made. People who complained of issues and then completely ignored suggestions because they felt they know best. That change received A LOT of compliments due to resolving a lot of network stability issues for larger networks.

[dmpayton]

These are just my initial thoughts, but in essence, I applaud this post. I run a community hackerspace/makerspace where interest in Mesh has been growing, and there have been a lot of discussions at the space that are full of questions that would be addressed by the points raised here.

An RFP process would be great. I'm mainly familiar with Python's PEP process for enhancement proposals and that seems to work well enough. I know that other projects have similar processes.

More transparency and visibility from the core team would be nice. Discord is great for community engagement, but core discussion should happen somewhere less ephemeral. Mailing lists are one tradition, but GitHub Discussions could work well – I've seen other F/OSS projects using them effectively.

While I'm excited for some of the features in 2.5, but I also don't think that, "on a whim, someone spent a week hacking on this old PR and now it's being merged in" is a great development methodology. Having a roadmap for the project would be great.

I know that shipping code is fun and planning is less so, but planning is necessary work. And so, for what's it's worth, is documentation.

[garthvh]

You can easily go dive into the docs, what is stopping you?

[rcarteraz]

An RFP process would be great. I'm mainly familiar with Python's PEP process for enhancement proposals and that seems to work well enough. I know that other projects have similar processes.

Well, there's an existing RFC Process that quite a few people acted like it was needed, but yet it's hardly been used since being introduced.

but I also don't think that, "on a whim, someone spent a week hacking on this old PR and now it's being merged in" is a great development methodology. Having a roadmap for the project would be great.

I'm sorry, but I disagree. At least as it relates to Meshtastic. Some of the best features added have come from whims. Recent examples: linux native support, location precision, compass module, and the PKI changes.

And so, for what's it's worth, is documentation.

This is absolutely the easiest way for anyone to contribute to the project, but yet there's only about 3-4 of us who do so regularly. The project is a fast-paced one, it's sometimes hard to keep up with the changes. If anyone wants to submit a PR for review, we welcome it.

[garthvh]

You are unlikely to find a large open source project it is easier to contribute to than Meshtastic. There are tons of ways to contribute, I encourage all of you to dig in.

There is an edit link on the Botton of every documentation page.

There is an existing RFC process https://github.com/meshtastic/rfcs

[KE6VDW]

Dear Meshtastic Development Team,

I hope this message finds you well. I am reaching out as a dedicated advocate for Meshtastic, with a deep commitment to both promoting and improving the platform. My involvement spans content creation on my YouTube channel and contributions such as the Meshtastic Anthem. Through extensive participation in various forums, social media channels, and chat groups, I have observed some concerns within the community that I believe are crucial to address for the continued success and engagement of Meshtastic.

One recurring issue is the disconnect between the community and the development team. The communication about updates and changes can be inconsistent, leading to confusion and frustration among users. This issue is compounded by the sporadic nature of firmware updates and the varying messages conveyed across different platforms.

A primary concern is community attrition. I have noticed an increasing number of community members expressing frustration and considering leaving due to these communication gaps and uncertainties. To mitigate this, I propose the creation of a centralized SIMPLE single "Firmware Alerts and Status" page. This resource would serve as the official source of information regarding firmware updates, including reasons for changes, rollback instructions, and highlights of new features or stability improvements. Such a page would provide clarity and ensure that all users are on the same page, reducing confusion and frustration. Key is simple. One page you scroll down with the newest information up top so users can go there to get a quick catch-up.

Additionally, I believe that transitioning firmware releases out of the beta phase would greatly benefit the community. Acknowledging a particular version as officially stable would offer users reassurance and help in decision-making, even if it is not entirely free of imperfections.

I understand and appreciate the hard work the development team is putting into Meshtastic. We are all committed to the advancement of this innovative platform and seek to contribute positively. By improving transparency and communication, we can alleviate anxiety among users regarding firmware updates and ensure a more cohesive and supportive community experience.

Thank you for your attention to this matter. I look forward to your response and am eager to assist in any way that could help enhance our collective efforts.

Best regards,

Norman Jester
The Alien (AYB Network)
KE6VDW

Shameful plugs:
https://www.youtube.com/@ke6vdw
https://www.instagram.com/normanjesteriii

[KE6VDW]

I know the groups and forums etc are alot to keep up with. It is alot of reading and im sure the dev team cant read it all. Because of that, I think they miss a lot of these things were seeing that could ultimately alter the course of meshtastic. I believe they should appoint community ambassadors who ARE out there in the trenches, who can identify trends, needs, curiosities and concerns. This can help them with the planning stages and identify what the community needs to keep them on the ship. I can self-appoint myself as an ambassador as I have been doing but it would be nice to be made official, and also bring in more of guys like us to be official Meshtastic Ambassadors.
We can summarize whats going on out there in the wild for them.

Norman Jester
The Alien (AYB Network)
KE6VDW

[thespillmonkey]

The Router_Client issue was something that actually affected the infrastructure. That was a decision made due to real-world congestion/communication problems. Too many people were just setting it to this mode and screwing up the infrastructure.

For those that actually need it in this mode, there is a method to set it to router and disable BLE sleep. Leaving this option availible prevents the masses from misconfiguring while giving those that need it (base stations that are also in solid router positions) functional. It was a slightly hot topic, but this isn't related to the current conversation.

The MQTT GPS issue is a very different story. There are no issues with the mesh/infrastructure.

The sole controversy here is that the devs believe it is their duty to save the masses from themselves.
Rather than just making changes to the various UIs to inform people about publicly broadcasting their GPS positions on channel 0/the default key, they are working towards preventing all percise GPS positioning over that channel and breaking our map systems.

It is a very personal political stance.

"Because some users don't take the time to read before doing things, we will break functinality for what we believe is needed saftey"

Vs

"People have self-accountability for their own actions and are smart enought to know that transmitting your GPS position publicly means you have given that information to the public"

I have seen some debate on the EU's PII laws and California's CCPA, but these only relate to handling/storage of PII data between a client and business-like entity. It also has nothing to do with information scraped from public transmissions made by a person. There is no expectation of privacy in public spaces. Be that in person or RF.

It comes down to forced regulation for a non technical issue. And as this entire project is encrypted-offgrid-comms, forced regulation for perceived saftey is something that the majority of platform users are adamantly against...

The devs can obviously do what they want, but if this goes ahead, it is definitely the sort of decision that will lead to forks in the project, which will then lead to decisions by those fork owners which DO end up impacting the stability of the large meshes.

If the average person looking for encrypted comms is looking at two versions of firmware, and one is called "Meshtastic Freedom Edition" with enhanced features or "Meshtastic Unleashed", which do you think they will be loading on all their devices?

All of the traffic trimming on the official fork will eventually be for nothing.

[garthvh]

Awesome to have so many volunteers interested in doing tedious bureaucratic work! Excited to see all the new contributions from you guys, there are hundreds of hours of work to do all over the place, we are happy to point you to the correct repos.

[madeofstown]

I noticed the heavy sarcasm but I don't believe this is an issue of man-power. It's more of a coordination thing. I see that the devs tend to have meetings on discord chat, maybe record minutes so others who can't sit in know what was discussed. Recoding what is being discussed is important because there is no way to search through voice chats..

[rcarteraz]

I see that the devs tend to have meetings on discord chat, maybe record minutes so others who can't sit in know what was discussed. Recoding what is being discussed is important because there is no way to search through voice chats..

I would say 90% of these voice chats have very little if anything to do with changes — honestly a high percentage have nothing to do with Meshtastic. The bulk of them are just hangout sessions. Just FYI. The majority of discussions related to changes happen within the discord text channels.

[Marx1]

This is totally uncalled for.

We are trying to have a legitimate discussion about serious topics and you, the dev, is purposefully derailing the conversation. This is an example of the "devs" vs "the users" that we're talking about.

We understand you're super busy, with only a handful of devs. I've personally tried to get assistance understanding how to get around the MQTT JSON issue on the NRF chips, and was told to read the code and pound sand.

As others have said, how the devs respond/react/interact with the public will directly affect what happens. Either a fork with the features the public wants put back, or people leaving and going to other similar systems that are starting to be developed.

You commented "You are unlikely to find a large open source project it is easier to contribute to than Meshtastic. There are tons of ways to contribute, I encourage all of you to dig in."

This is pretty arrogant. There are many other large open source projects that are easy to continue to. I've contributed directly to AREDN via documentation and bug reporting and deep troubleshooting/analysis to get issues resolved. When I tried to do the same here, I was greeting with gruff responses because I requested a older radio be added that was widely available at the time.

[garthvh]

Lets see the pull request.

[GUVWAF]

This is an example of the "devs" vs "the users" that we're talking about.

Devs are users and every user can be a dev (or contributor). While there may very well be reasonable requests here and for the project in general, if you don’t help to make it a realization, you can’t expect others to do it for you.

I've personally tried to get assistance understanding how to get around the MQTT JSON issue on the NRF chips, and was told to read the code and pound sand.

You are probably referring to my comment here #2804 (comment). I mentioned there that this issue is hard to resolve. I spent a couple days trying to fix this (to me fairly unimportant) issue, but it seems the only way is to use another JSON library, which would be a huge undertaking. Since you were working on a bridge already, I then suggested to do the decoding yourself outside of the firmware (like others are doing) and provided a link to an example. Not sure what else I could have done in the limited time I am able to spend on this project.

I was greeting with gruff responses because I requested a older radio be added that was widely available at the time.

You are probably referring to this comment here #3682 (comment). To me this is a very reasonable answer and surely not “gruff”.

[thebentern]

"People have self-accountability for their own actions and are smart enought to know that transmitting your GPS position publicly means you have given that information to the public"

I believe the big issue here with the MQTT changes is uninformed consent for your mesh peers. For instance, you might choose to share precise locations over the internet, but your mesh neighbors may not (and may not fully realize that this is happening). Since we (Meshtastic) are responsible for the official broker and any reported incidents, we have to start taking a very conservative approach to upholding as many reasonable privacy safeguards as possible.

It's unfortunate that this particular decision needed to be made very quickly, but there were many privacy concerns expressed on the discord server by community members who were very alarmed at the long data retention periods, position history, and other "creepy" features exposed as a part of the feature set of some of the community maps built around data from our broker. We feel that the changes needed to be made quickly to address the issues in light of this.

All that said, a number of decisions have been made with little to no communication about the change, why it’s made, and where the direction of particular features may be headed

I will say that many of these decisions are publicly discussed with community members, but within the context of our Discord. Perhaps we can improve on funneling that information out to the edges of the greater Meshtastic community more effectively. I do believe Discord is the best place to engage folks who are invested in the project though, as it is our largest and most active. Some decisions are simply more tactical in nature (like the recent MQTT one) and aren't necessarily a part of some larger vision for Meshtastic.

[thespillmonkey]

The words in that sentence don’t make grammatical sense. I assume you miss typed?

But yes, there exists zero expectation of privacy when you publicly transmit information. Maybe that is different outside the USA, but here we have definitive federal case law and even fcc regulation that covers all forms of public communication.

[Effective-Gate]

There is no agenda, the privacy is an illusion theory you are married to does not exist in the real world.

Mate, reading your responses versus the other devs here it's obvious you have an agenda. Also, the project should probably not let you interact with the community, you are terrible at communication.

[garthvh]

The only agenda is protecting user's privacy, which most users appreciate, those scraping all the data don't.

[Effective-Gate]

The only agenda is protecting user's privacy, which most users appreciate, those scraping all the data don't.

Look, no one's criticising better privacy options. We're talking about saying one thing and doing another, making dodgy moves without any warning or communication, and the inconsistency in the actions and responses (particularly from you) in this discussion. Most users want their nodes to share their exact location because most users want their nodes to show up on maps. If you don't want it shared by default, that's fine, mate—just don’t take that option away from those of us who do. Educate those who don’t get it a bit better, because the majority of us are aware and okay with it.

[thespillmonkey]

^this 1000%

[jp-bennett]

Hey all, glad to see so much excitement and passion around Meshtastic. I wanted to make some notes about the thoughts raised here. I am speaking as a developer, but not for Meshtastic officially.

First, like every other functioning Open Source project, the Meshtastic feature set and roadmap are set by the developers, with feedback from the community. And that's something of a tautology: nobody is paying the developers to work on things, and we got to be Meshtastic developers by simply deciding to roll up our sleeves and do the work. You, too, can upgrade yourself from the user community to the developer community. And yes, if there is an incurable disagreement over something, you can fork the Meshtastic code. And just like with many other Open Source projects, you get to choose a new name for your fork. No calling it Meshtastic++.

Second, none of the 2.5 changes were done on a whim. PKI, in particular, I've been planning to get working for nearly a year now. Because there were severe deficiencies with both the DM and Remote Admin systems without it. Once the old patch was working, we spent hours kicking around ideas and approaches for whether to use it, how to use it, etc. That discussion happened on Discord, because that's where the developers spend time. GitHub is great for some things, but not live chat or calls. Not a whim, but carefully planned and rolled out.

Third, the MQTT changes are not all sunshine and roses. I get that. But I also get that when one of the public maps added route saving to the location tracking, the people who were on the map called it creepy. It seems to me that Meshtastic has not done enough to be clear to users that the default channel is essentially unencrypted, and that their location data may end up online. But regardless, something needed to be done right away. Things were done right away, and now that the fires have been put out, we can all come and figure out what the public MQTT server needs to be moving forward.

[thespillmonkey]

Thank you Jon!
Solid response.

And we do thank you all for your hard work.

And can’t wait for PKI!

[madeofstown]

I just have a simple request: put a notice on the front page of Meshtastic.org (or in the blog) whenever new changes are coming down the pipeline with links to the appropriate GitHub pulls/issues. Put the information front and center so people with a stake in the game can comment before the changes are made.

[garthvh]

There is a feed of the GitHub changes in discord, it is way too much for the website.