I very much prefer having it ‘fade out’, vs. having to hover over nodes to see the “Last Updated” date in the metadata. It would allow someone who is zooming around in a region, to see ‘grey’ (old/offline?) nodes vs. ‘green’ (up) vs. ‘red’ (off, but current), without having to open each node.
1 Like
No. It would be a PITA to have to come back and “refresh” nodes that haven’t moved. Especially if one has a number of nodes dispersed throughout. I have 5 nodes out there right now and more that will be deployed soon. Plus some of us travel on business and Internet connectivity isn’t a given in some places (like Africa or parts of the Middle East) and expecting me to be able to refresh a node when I may have dubious Internet access for weeks at a time is not reasonable. Especially if those nodes are in the same place they were when I left and have no need for refreshing. And even auto-updating locations of nodes may not be possible as some nodes may be located where there is no Internet access.
I’m also not keen on contacting owners of nodes. There is something to be said for privacy and I wouldn’t want a bunch of people pinging me asking about my nodes “freshness”. It could easily become a spamming situation and I’d simply send those requests to the spam folder automatically.
I agree. Nodes can come and go, but once placed they generally stay. One of my nodes has been down for several weeks and I’ve tried to be persuasive about the need to reboot it. Ultimately, people get around to fixing things and those of us building networks have to rely on persuasion, not coercion to get things accomplished. This is a volunteer effort and that can be like herding cats at times, but trying to build “discipline” into the system is probably something that is counter-productive to having more nodes in the long run, for privacy and other reasons, even if it seems to offer a shortcut to better documentation.
Good thoughts aspexin.
Even if all of your map pins are still valid I think we both can agree that some percentage of pins on the map are effectively “abandoned” by now. Someone bought a set of Mesh devices six months ago, got excited and placed some map pins, and then later on lost interest and put their Mesh devices into a box somewhere and forgot about them. I know that I have seen at least one post on this forum from someone who did just that after never getting a reply to a Shout.
If pins never go away then the map data will become less accurate over time. Without intervention the map will become a form of “false advertising” and a liability for goTenna if prospective customers visit the map and use the number of pins in their area as a gauge of how many active Mesh devices are out there.
If I was a true cynic I would question if existing pins on the map are all from real customers. Since pin identifiers are random what is keeping goTenna (or a malicious user) from just placing lots of pins randomly across the world to artificially inflate numbers?
Anyway, new proposal – Map pins fade out over time but are automatically refreshed every time the user:
- visits the meshcommunity forums or the map (while logged in).
- opens the goTenna smartphone app (if they entered in their e-mail address into their profile in the app)
1 Like
Did you remove your broken node’s pin from the imeshyou map while you work on getting it fixed? Why or why not?
1 Like
I think you’re not taking into account things that are pretty clear with a bit more care in interpreting the types of map pins that are present. The most significant is assuming all those pins represent the same thing - or at least mushing things together in a way that doesn’t do justice to the data. That’s exactly what was going on with…
There are about 10K pins on the map. A little less than one in 10 is a stationary node, i.e. available 24/7 to relay messages. The other ~90% are devices that are used anywhere from daily to very intermittently or as the definition in the map input form says, “Mobile nodes show general locations where meshers regularly use their devices.”
As was explained in that other thread, assuming that all those pins should represent some wannabe as-if-CB-equipped trucker standing by to give you a radio check just doesn’t fit the actual circumstances of how people use mesh networking or what people are representing when they lay down a pin. In other words, the pins represent places where people may use their goTennas. Nothing else is implied or promised.
Stationary relay nodes do represent at least the hope that one might find something to bounce messages off of, because they are “locations where meshers have set up fixed, powered-on devices.” In some cases, these form a network. But in many cases, there’s an isolated relay that would serve those nearby, but is too far away from other stationary nodes to do anything but provide longer links to those using them.
I’m still trying to understand what the actual concern is with the absolute accuracy of each and every pin? The closest I can find to a concrete expression of that is this nugget about why people should just be able to ping away at node holders
While everyone has there own comfort level with privacy, both those questions and most that would arise in between them would be rather off-putting to me personally. I’ve got some contact info listed and am happy to meet and discuss what we’re doing locally, in general. Take it to the node level here and it’s mostly people who’ve agreed to do this as a way of returning a favor or just because they like the idea. To recast these efforts in the language of failure as some seem inclined to do would tend to discourage more from getting involved. Once again, the intention might be to “help” here, but I tend to see it as unhelpful to encouraging people to participate.
That’s a very good example of why I think this argument is more about throwing shade on goTenna than it is about building mesh. Are you even affected, first of all? And the deficiency is noted in the node notes and in the general notes for the network map, so if someone had a concern about knowing what was going on they could satisfy their curiosity, the info is out there. Besides, it is essentially a 15 minute job to fix this, except that we need to install a grabhandle for safety reasons to make getting up on that roof safer. My schedule and her schedule just haven’t aligned yet, but could at any moment. She might even get up there and fix it herself in the meantime, but I was wanting to be there in order to analyze how things were found to determine whether it was a power problem or a device issue. Things are still somewhat experimental, so we shouldn’t be rushed in order to get better data for the long run. And the radioscape maps and testing show some service in the area still. That’s the way mesh works, there are multiple ways for a user in an area to get service once the mesh thickens.
And that, more nodes, is the ultimate answer to this “problem” such as it is.
How do you even define “valid node”? Either it works or it doesn’t as a relay. Depending on the density of nodes in an area, how do you even determine which node it is you’re pinging away at? You might THINK you\re pinging a specific node, but there’s really no way to know. What’s important is that it works or doesn’t. And if it doesn’t them moving to a better location seems a more useful remedy than setting up people to be spammed.
To me, a close reading of the text here is that goTenna knew going in that crowdsourcing a map wouldn’t be totally accurate, but did have other useful results, so they were careful not to make big promises of accuracy that are largely superfluous to a map meant as general guide to finding mesh. Then people seem to want to ignore that and proceed to claiming that exceptional accuracy was both claimed and should be expected because of their making this essentially strawman argument.
Relax, this appears to neither be a big problem, other than for a few who seem strangely collectively obsessed about it, nor one of much significance. Things happen, as the one node that is at issue here was first put in place because the only stationary relay in the map before I started our network - just a couple of blocks away, seemed to be out of service when we needed to use it.
My solution? Put up another node to fill in for it. So it’s somewhat ironic this comes up, as doing that seems more practical and useful than suggested a wholesale invasion of the privacy of those motivated enough to go to the trouble of putting up nodes. Which, in the long run, is going to bring more service with greater redundancy and reliability to more people? Schemes to impose a greater “discipline” on relay owners? Or just forging forward and putting up more relay nodes that have people caring for them? And that is what my solution may be, because I have one that is planned for the next slot to the west from the down node’s location. This will not only fill in the lack of service, but expand it. The nodes are less important than the service area that they create happens to be.
First of all, this is the most respectful forum I have ever been on. Thanks for that everyone.
I’m not “throwing shade” on goTenna – I love the idea of mesh. I just bought another set. What I am “throwing shade” on is the current implementation of the map and I’m really confused on why you seem to be defending it.
Personally when I visit the map I filter out all the “mobile” nodes as I don’t think they should even be allowed to exist. How useful is a pin that says “i might be here” anyway?
This is not ham radio where everyone is a volunteer. goTenna is a commercial entity trying to make a profit. As a hypothetical example, it is not ok to post maps that show 80% coverage of Chicago when in reality it is 20%.
Do you really expect potential new goTenna customers in your city to click on every map pin and read the notes to see if the node is currently working? That’s not realistic. Pins that are displayed on the imeshyou node map will be assumed to be valid by most potential customers… and the more pins in their city the more likely they are to purchase a Mesh device.
(The above comment assumes that potential customers visit the map first. I claim that’s true for a subset of customers because the map has been mentioned in every goTenna review I have read so far.)
That’s too simplistic. New customers shouldn’t be required to “verify” the accuracy of the map before they buy a Mesh. I don’t think you are really suggesting that I drop a $80 Mesh on top of map pin that is not valid? Isn’t the proper solution to remove the map pin that is not valid?
…I am not “strangely collectively obsessed”. I am slightly upset that the imeshyou map in my area is objectively wrong about the existence of relay nodes and I want to make sure that future customers won’t have the same experience. That is why I state that nodes on the imeshyou map should be removed if they no longer exist in the real world.
1 Like
In this one respect, goTenna IS a lot like ham radio. There are people who make equipment and there are people who use it.
But you’re obviously not getting the point about crowd-sourcing. Sure, accuracy is of some importance, but equally important is the effort to involve people to see beyond their own interests. As we know from many other human endeavors, this is not always perfect, “objective,” or some other standard that tends to fit square pegs into round holes. There is trust enough in people to do the right thing, because trusting people to make the call for themselves is oftentimes just as accurate as a managed, top-down scheme would likely be. Without sending survey teams around to verify every entry, no map by anyone would be substantially better. I’m OK with letting users decided when, why, and how they post up on the map. It may not be perfect, but I bet it’s just as good as an alternative that suggests automatically eliminating nodes according to an arbitrary standard that doesn’t reflect the real world, but instead an obsession with centralized neatness and order.
BTW, I’m not sure how you see 80% coverage of Chicago. Maybe try zooming in as the magnification changes depending on the setting. That’s a mapping artifact, not a claim about coverage that I think you’re seeing.
Objectivity is over-rated at times. This is one of them. You’re seeking a centralized solution to problem that isn’t much of a problem. People vary, so the results may vary. I trust the crowd at least as much as I trust your solution, which would likely take far more functioning nodes off the maps as it would non-functioning ones. How is that objectively more accurate? It’s not.
If you had that level of concern about your privacy, you wouldn’t list your personal node on a publicly accessible map on the live Internet, would you? The two seem at orthogonal purposes to each other.
A few good ideas did come out of this discussion, however:
- Node color on the map should represent its ‘state’ (stationary vs. transitory)
- Gradually dimming nodes to indicate a level of ‘expiry’ is also a good idea. Nodes that were put up 4 years ago and never updated, may likely no longer be in use, so dim them. If they are, notify the owners to update their pin, and bring them back into currency.
- Not everyone uses nodes in the same way, nor considers their use equally, so accommodate that, but not at the expense of the usefulness of the “Mesh Network Map”.
Right now, the Mesh Network Map is… relatively useless, other than “Well that’s neat to look at” and nothing more.
If 75% of the nodes are down/transitory/not-in-use, what would the map look like then? If 10% are stationary and 90% are mobile nodes, does that change the map?
If you want the map to be a useful tool, then make it so. Right now, it isn’t.
1 Like
One can have privacy through obscurity. Doesn’t mean people don’t want to publish their node locations. But that also doesn’t mean they want to be contacted. Remember, people could opt to not post their node locations at all which really makes the map useless (see below).
TL;DR due to people not posting their nodes to the map or an emergency that lasts days/weeks/months as people lose power the map is mostly useless no matter how much effort is made to keep it up to date.
To me owning a mesh is more about emergency preparedness than how accurate the map is. When I looked at the map before buying goTenna mesh there was one pin within 5km and that is just over a km North from us.
As for the map it is just that. A map where nodes may or may not be active. Like a list of 2m repeaters. They may or may not be up and running. You don’t know until you try to transmit through it. But it doesn’t mean we delete the repeater from the list if it happens to be down at that moment.
Imagine a prolonged emergency that stretches days into weeks into months. The map could be mostly useless in that location within 24-72 hours as people run out of backup power. If they don’t have power they most likely do not have Internet connectivity to update the map saying their node is down. So the map is still useless in that affected area.
1 Like
I’ll make my own decisions about privacy and security, thank you. That’s one of the things I like about goTenna, as you can make a wide array of choices to suit my needs. I don’t possibly see how your one-size fits all approach works for anyone but you, but I won’t be suggesting I know better than you what your needs are.
It already does that, although you seem to be arguing for some sort of arbitrary standard narrower than the current definitions.
You’re assuming a lot here, essentially trading what you see as the uncertainty of false positives into the certainty of deciding to eliminate nodes on no better basis than false negatives. Is that more accurate, objective, even realistic? Not as far as I can see. It trades faith in the user base for some entirely arbitrary metric that has no relation to a node being on the air or not.
Now you’re dipping into the pool of fabricated statistics. Where do you come up with either 75% useless or 100% useless? You’re not even trying to have a conversation about this, you’re just being silly. 
Oddly, this is pretty much what is on the map right now as I explained earlier. Now I’m not sure you’ve even actually looked at it closely. It’s not perfect, but it is useful if we avoid things that make it more inaccurate. I’m sticking with the crowd on this one. Humans do vary, but they also trend to average out. I’m fairly certain that meshed humans average out to something better than 75% bum nodes.
Good points all. I would add that whatever faults may lie in the map, it is probably more accurate than any alternatives. Maps improve over time, but they are never completely accurate. Obsession with that sort of thing marks you as either a very talented manipulator of geospatial data or just plain obsessive. For the rest of us, the map is what it purports to be: a fairly accurate guide to where you might find mesh. At least around here it is. Is it perfect? No but it doesn’t need to be. Mesh tends to overlap and go into unexpected places. One bad node gets ignored in the bigger scheme of things, because there is so much inherent connectivity as mesh grows.
Here’s a proposal that won’t satisfy those wanting to hold a dead node firing squad, but might be a more positive nudge in the right direction. Instead of trying to minus out someone’s account rep (doesn’t that sound like what a bank would do?) we should make good nodes more visible or highly rated on the map. There’s blockchain or the old fashioned survey, but the problem then is figuring out who wrote what about what node once stuff is sprung from everyone’s private mailboxes. It might be more happen-chance your nodes even get used enough to be rated, even though they’re great, well-placed and reliable nodes. That’s not fair, but we may have to accept that, but overall even just getting rewarded with some ridiculous-coin thingee may be enough to incentivize people to put up nodes. I just like the thrill of seeing the mesh expand as a reward personally. But positive reinforcement like that goes a lot farther than antagonistic and arbitrary processes.
One other thing. As accurate as the map is it doesn’t factor in structures and foliage that can interfere with the signal propagation. I have 5 nodes within two blocks of each other at a distance of 1km from 1 end to the other. You’d think within that 1km 2-5 blocks you’d be able to send a message over X number of nodes to the other side. Not the case. I was out range testing to day to measure the impact of the new 5th deployment and standing in certain places where buildings or a thicket of trees blocked a path to any node in the mesh and no node picked up the message to relay it. So while the map may be accurate it still doesn’t ensure the signal will reach that node (or any other node) nearby.
1 Like
Another very important point. It’s also food for thought for anyone who is dissatisfied with the performance of the nodes that others report placing in the map. I’ll bet that you probably know of better locations for a relay node that are close by, too. Knowing about and being able to access better locations are, however, different things and often impossible to resolve.
Should the marker be whisked away because someone reports it as non-functioning? I’d hope not. It might be just enough to do what the owner needs to do. Maybe it does great in the winter, but when the leaves erupt in the spring it’s range is sharply lower, yet still suffice’s for the needs of those who installed it? Unworthy of a pin on the map? I would argue it’s as worthy as any other node.
Your 5th node might even turn from bit of a frog into a prince once the network grows around it and signals reach it from the backside as the mesh thickens. However, if that’s the best location possible for it right now, it certainly gives more service from there than it would sitting in a box in the closet. Depending on the location’s ability to “see” other nodes it could turn out to be a much more useful relay than at first it seems. What is certain is that steps to grow the mesh are inherently useful, even when less than perfect, while suggesting that subtracting from the mesh as a means to do that seems rather a dead end, impractical means to achieve that goal.
1 Like
I feel like MikeL and aspexin are ganging up on people making good faith suggestions on improving the imeshyou map and I’m not sure why that is.
Regardless of how MikeL and aspexin feel about their own pins I think most people would prefer a map that is more up-to-date/accurate.
There is nothing that requires MikeL or aspexin to keep their pins on the map if they want to guard their privacy and/or decide that “refreshing” pins is too much of a burden. The map will survive losing the ~20 pins placed by both of them of they decide to remove them.
We all can both acknowledge their concerns and also move forward with suggestions on how to improve the accuracy of the map for the rest of of goTenna Mesh community.
1 Like
I’m sorry you feel that way. That isn’t the intent. You have to understand how inaccurate the map is always going to be. I can literally move 10’ and lose connectivity to my 5 node mesh because of a building. That doesn’t mean the pins need to be refreshed.
2 Likes
They may be in good faith, but they miscast the facts and arguably would be less accurate than the current map if they systematically eliminate nodes from the map based on a formula that is itself based on seemingly nothing more than a hunch or perhaps you being comfortable with a whole new set of inaccuracies that more clearly exist than the weak case you’ve mustered about existing flaws with the map.
If I saw some of these concerns being addressed, I might view what you’re proposing more favorably in terms of the intent if the valid concerns I raised about your proposals weren’t dismissed as “ganging up.” Facts don’t gang up on anyone, although when they are ignored they do tend to accumulate and become inconvenient. People can review it above…
Hmmm, that’s an example of exactly what I just pointed out in general, so might as well address it and a couple of other points you reiterate but do nothing with to advance your argument. My nodes are detailed as to their status. They’re still up on roofs doing their thing and the one that has a problem will be back, as well as having it’s service area partially covered by adjacent nodes. I’m not going to take it off the map because of your apparent feeling that eradication is the only “accurate” solution to the “problem.” Frankly, It’s better that it’s a topic of discussion than being dead and buried. When it comes back, it also won’t be an additional node serving new territory, as that would be also inaccurate.
This has nothing to do with my privacy, except for your notion that anyone with a node should lose as much of their privacy as it takes as a means to further your goal of “accuracy.” The map was built by goTenna Mesh users under a defined set of privacy assumptions. You want to change that set of rules in mid-stream. They say that Facebook is arrogant about the privacy of its users’ data (I don’t know because I avoid that sort of thing) but you seem intent on one-upping that – and ignoring the fact that goTenna just doesn’t operate like that. They give their users a range of options, then respect their decisions on privacy
Why would I remove any? They’re all working, except the one, which will soon be back up, as noted in its details (in place since before you ran this red herring up the pole to see if anyone was willing to salute it.)
Strangely, you don’t blink an eye as you promote this kind of inaccuracy, thinking that people won’t notice you seem comfortable with the notion that making 5 functioning nodes disappear is worth it to put one wounded node out of its misery in your eyes. That’s a rather strange agenda if what you say you want is accuracy. I have no intent to remove any nodes (unless they are permanently taken down or relocated for some reason), in fact I aim to add to my total. The map accurately reflects what’s happening here. You’re welcome to use whatever standard you desire when and if you build your network.
3 Likes
The “save” button isn’t working when I add a my powered relay node. Grrr.
Multiple phones can share a GTM, just not simultaneously? If so, I think this is great. So, would the GTM save messages for both phones to which it has been paired and sync them accordingly when each phone pairs, like a phone answering machine with multiple Inboxes? Or does GTM delete stored messages when it is unpaired?
Can it work the other way around as well (feature request), where I have all of my GTMs “known” to each of my family’s phones for many to many usage? The idea is that anyone can take any GTM when leaving - oops, I forgot to charge my GTM, let me use another GTM seamlessly if previously configured.
Each of our GTMs would receive and save messages for all phones it knowss and sync respectively when connected via Bluetooth? Of course, the phone app would just discard duplicate messages already received by another of our GTMs.
This may take up quite a bit of storage depending on number of phones known to any GTM, so maybe allow the mobile app to set an “owner” / admin for any particular GTM, who can set message quotas for anyone paired / known to that GTM.
It. does neither. It holds the messages sent to the GUID, which would be the same for any phone paired to it. There is not a separate address (or memory or mailbox) associated with each phone, just the singular GUID of the GTM.