Thank you 🙏

The comparison to video games never occurred to me, but perhaps as someone who grew up with video games, there is some intrinsic motivation in me that makes it look like that. Now that you have mentioned it, I see it too.

For my personal data, I use GPSLogger; it is fine and does what it is supposed to do. I never look at it, so I do not need any fancy features, and the main selling point for me is that it also saves the GPS data on my phone so that, in case of an error, I can still pull it and import it. OwnTracks also works fine; if you have multiple users, one benefit could be that Reitti can send the location of your other users back to OwnTracks so you can see their locations directly in the app. Finally, if you want more features, there is also Colota, which is the most modern of them all.

Thank you for your kind words. Let me know what you think after the upgrade.

I love seeing these updates, Daniel!

That counts as a vote for the upgrade notification system. :) Right now, I’m trying to reduce the number of posts about reitti across all social platforms for different reasons, and the notification system would help get the word out about new versions.

Yeah, good catch. I didn’t think about that anymore. There was a change while introducing custom map styles to Reitti so they can also be proxied through the cache. That was from 4.0 to 5.0 https://www.dedicatedcode.com/projects/reitti/latest/upgrade/

Reitti should work with a couple of different tracking setups. The latest release improves the handling of apps that track only when movement is detected.

If you tell me your setup and the settings of the app, I could try to help you make it work.

Thank you 🙏

Ok, we need to find out what the culprit is. First we check the status of the tile-cache if there is any:

sudo docker inspect $(sudo docker compose ps -q tile-cache) --format='{{.State.ExitCode}} - Status: {{.State.Status}} | Error: {{.State.Error}}'

This shoudl tell us if the config is wrong. It should not be, but you never know. If that returns 0: Status: unhealthy then the healthcheck is propably the problem. It calls this command inside the container:

curl -f -s -o /dev/null -H "X-Reitti-Upstream-Url: https://tiles.dedicatedcode.com/" http://localhost/custom/planet/latest/0/0/0.pbf || exit 1

So, basicly it tries to fetch a tile from tiles.dedicatedcode.com under this url: https://tiles.dedicatedcode.com/planet/latest/0/0/0.pbf

Hi Estebiu, i am sorry for the trouble. Lets try to find out what the problem is. The tile-cache is a simple nginx container. I never experienced any problem with it.

• From which version did you upgrade?
• Any mounted directories for the tile-cache?

Thank you 🙏 I am trying my best.

Thank you 🙏

I would try to use GPSLogger, it is capable of sending data to dawarich and is really reliable. AFAIK, the Owntracks app has a queue and normally sends the data back when it is online, but I also found that this is sometimes not the case, and there is no way of restoring it then. GPSLogger has the advantage that it stores the GPX file also on your phone; if something goes wrong, you can still grab that and upload it.

Colota also seems like an awesome alternative. I’ve only heard good things about it.

Oh, in theory Reitti could function in such an environment. You would need your own tile server to display the map and maybe your own reverse geocoder. Then you can create your own map style (get an existing one and replace the source tile URLs) and point Reitti that way to your hosted services. It will be a lot of work and, depending on your needs, will need a lot of space.

For map tiles, I host an instance of https://openfreemap.org/ And for reverse geocoding, there is Photon or my own, Paikka.

If you want to follow up, let me know in a GitHub discussion. I could sketch out a way to do it, and maybe someone else could also use that information then.

I personally mainly use GPSLogger on my phone. I had one problem with it but this is remedied in V136 of it and now it can use the HTTP File Upload which brings the battery usage down a lot. So in my experience this works really well. There is also an autoconfig button. You can open the Integrations Settings page on your phone and press the button. This will configure GPSLogger so you do not have to enter the address and the auth token by hand.

I would check the settings afterwards. I have set mine to not report anything with more than 25m accuracy. And upload the data once an hour.

Hello and thank you. I have no idea how an app gets onto Yunohost. But I have now seen that you can nominate an app to be packaged, and someone put Reitti in it. Wishlist Reitti, but it seems they have some restrictions on who can vote for it.

Maybe there is another way to get it on, like a generic Docker thing. If that is a way, I am more than happy to assist.

Hello! If you ask me personally. I am happy with how reitti is working today. It suits my own movement patterns well enough, but everyone is different, so your mileage may certainly vary. I am improving it step by step though, so it should keep getting better over time.

On a general comparison with dawarich: I honestly do not know enough about it anymore to give you a fair assessment. I know they have a comparison on their website, but as with all of these, it is not completely correct. That is simply the nature of such comparisons, and it is also why I will not make one myself. After all, I would be hopelessly biased.

So here is my suggestion: if you have Docker and some gpx files handy (you can export them straight from dawarich), spin up an instance with docker-compose and throw the gpx files at it. See how it goes and whether it fits your needs. If it does not, no worries at all, just shut it down and never think about it again. Or maybe come back and give some feedback. 🙂

Yeah, understanding the data beneath the point would help alot. The problem with that is that we first can not query some system for every point. And second, since the raw data is already jumping around alot, even if we could it is still way off.

Commute tracking could be a way for the statistics. They need to improve a lot.

For the logic to detect valid points, we basically do that already for the visit detection. I am quite happy with it right now. Maybe the second year of reitti will be focused on the trips. 😀

Google did something like the confidence thing in their old timeline (Records.json) format. There for every path they had stored the confidence of the transportation mode.

And finally the mobile app. I doubt it will come the next year but we never know. I would have some ideas like tracking bumps (smoother rides are propably a car or train, elevation above x meters is propably a plane and so on) and such things to have more confidence in the trip detection. Right now, there is an app for everyone available which can report into reitti. So there is no pressure. But I don’t know, I also did not thought about writing my own reverse geocoder but that also came out of a necessity to have some smaller hosting on a small vps.

otherwise a way to reclassify all trips between the same places via the same route at once would be great

After thinking about this part a bit: I could add something like a rule system to reitti. I will do it either way since I am searching for a way to add notifications to reitti. One rule then could be, if there is a trip between place A and B, during the workweek between 07:00 and 09:00, re-classify this as train ride. But this would override then the walk to the trainstation. So the rule would need to be:

• Trip between Place A and Place B
• during workweek
• between 07:00 and 09:00
• trip is classified as driving

-> switch it to transit

Another idea: adding hints for the classification in the transportation mode setting. Like: during the workweek, between 07:00 and 09:00 it is more likely I take the train instead of the car. The classification could then pull that information in and decide based on the hints which one to choose

Hello! Thank you so much for your kind words, they really mean a lot! 😊

I already have some ideas for improving transportation detection down the line. One option is understanding the track beneath the path, sort of a “reverse route planner”: we have the route and then figure out which mode of transport was used along it. Another idea is analyzing the movement pattern in more sophisticated ways. Train rides might be recognizable by their consistent speed with full stops in between (I’m no train operator, but in my mind they accelerate to max speed, cruise, then stop), whereas car rides tend to fluctuate more.

I completely understand your point. Especially when switching between trains and cars, speed alone makes it really hard to pick the right mode. I’ll definitely keep working on improving this!

As for your second point: there’s colota-forwarder, which can fan out GPS points from one system to multiple others. Maybe that would work for your use case? I don’t currently see reitti itself implementing something like that, as it would mean keeping up to date with other services’ APIs and reacting whenever they change.

Thanks again for taking the time to share your thoughts.

That’s exactly the use case someone reported recently.

Right now, you can press a button on an individual image in Reitti if it’s aligned by time rather than location. And if you’ve got multiple images and you’re feeling brave (in other words, you trust Reitti 😄), there’s also a button to do it all at once from the gallery in the map bubbles.

Give it a try, and let me know how it goes! If anything needs tweaking, I’m happy to adjust it. Happy to hear feedback either way!

Hi Jolly, that is an awesome question, and I did not give it much thought. At the moment, there is no search on that data. Even displaying it is hard since you have to open the dialog again.

Right now, it is more like a “gathering phase.” You can collect the data, and we will figure out how this can be used. My main motivation, and I repeat myself for the last three major releases, is to enhance the memories feature with more context. Maybe, if the technique ever becomes feasible for the average homelab, we could generate more personal memories by incorporating some local LLM with the additional information of the tags and descriptions. This currently breaks because there is no single multilingual LLM that works well.

But I like the idea of searching through that data. This could be a nice addition; feel free to create a discussion on GitHub. I really like having multiple views on these topics. The device feature would not have existed if there had been no discussion about it. I always refused it before because it would not fit.

Since the data there is really open and quite free in its structure, I already have some ideas, like putting the songs I heard through Plex, scrobbled through MultiScrobble, into Reitti. That way, I could see what I listened to at that moment. For whatever reason, that sounds like an awesome idea, maybe for a side show when creating a memory of our trips.

Thank you for your kind words. That really warms my heart. 🙏